SSSR 2011 Conference Abstracts
Click on the name of the first author and see the abstract, click on Top to go back.
Abusamra - Adrian - Ahmed - Ahn - Aicher - Al Otaiba - Allaith - Allan - Allen - Allen-Tamai - Anthony - Anthony - Apel - Aram - Arciuli - Au-Yeung - August - Balass - Barbosa - Barker - Barquero - Barron - Barth - Beattie - Bell - Ben-Yehudah - Berthiaume - Betjemann - Biancarosa - Bingham - Binks-Cantrell - Birgisdottir - Blanchard - Boggs - Bowles - Branum-Martin - Braze - Breznitz - Bridges - Brimo - Burgess - Burke - Burnham - Bus - Cain - Calfee - Calhoon - Cao - Capotosto - Caravolas - Cardoso-Martins - Carlile - Carlisle - Carroll - Carson - Cartoceti - Chang - Chen - chen - Chen - Chen - Chesson - Chiu - Cho - Cho - Christodoulou - Christopher - Chung - Cirino - Compton - Conner - Conrad - Conway - Cordewener - Core - Cormier - Cowles - Coyne - Craig - Cronin - Crowe - da Mota - Daigle - Dalton - Davis - Day - Demir - Denton - Desrochers - Dickinson - Dombek - Douglas - Dunning - Dynia - Eidson - Elbro - Elleman - Escribano - Evans - Farnia - Farquharson - Farran - Farrington - Fejzo - Feng - Folsom - Fonseca - Foorman - Fragman - Fraser - Fricke - Frijters - Fuchs - Fuchs - Fumagalli de Salles - Furnes - Gathercole - Gathercole - Gellert - Georgiou - Gerde - Gilbert - Gillam - Godoy de Oliveira - Gomes - Goodrich - Goodwin - Gottardo - Grant - Green - Greulich - Grover - Guardia - Guo - Guthrie - Hagan-Burke - Haigh - Hamilton - Hannon - Harris - Hart - Hassan-Yari - Hebert - Heggie - Hiebert - Hill - Ho - Holliman - Holmes - Hornickel - Horton-Ikard - Houston - Hsuan - Hua - Hume - Hurry - Ijuin - Invernizzi - Jackson - Jarmulowicz - Javier - Jiménez - Johns - Jubenville - Justice - Juul - Kaiser - Kantor - Karemaker - Kearns - Keenan - Kegel - Kemp - Kershaw - Kieffer - Kilpatrick - Kim - Kim - Kim - Kirby - Kleemans - Ko - Koponen - Korat - Koutsoftas - Koyama - Kukona - Kuperman - Laks - Lambrecht Smith - Landerl - Lassonde - Lauterbach - Lederberg - Lee - Lembke - Lemons - Lervag - Leseman - Levesque - Li - Li - Li - Lien - Lin - Lin - Lin - Lin - Lindberg - Litt - Liu - Logan - Lombardino - Long - Lonigan - Loveall - Lovett - Lu - Luo - Luque - Macedo - Magdales - Magliano - Mak - Mancilla-Martinez - Marinus - Martin-Chang - Marty - Masterson - McBride-Chang - McCormick - McDonald Connor - McGeown - McGinty - McKeown - McMaster - McNorgan - Medford - Meenan - Melby-Lervag - Mengoni - Miciak - Middleton - Miller - Moore - Morrison - Morrison - Mostow - Munger - Murray - Nagler - Nagy - Navas - Nelson - Nicholson - Nielsen - Noe - Noordenbos - Oakhill - Olinghouse - Olofsson - Olson - Onochie-Quintanilla - Oslund - Ouellette - Park - Parrila - Patton Terry - Pearson - Pegado - Pendergast - Peng - Perfetti - Perin - Petersen - Petrill - Petscher - Phillips - Phillips - Pinheiro - Pisupati - Pitchford - Plisson - Poelmans - Pollard-Durodola - Pollo - Porta - Poulsen - Powell - Powell-Smith - Priebe - Protopapas - Puranik - Purpura - Quemart - Quinn - Radach - Ramirez - Rastle - Reta Sánchez - Riddell - Rivera - Roberts - Robins - Rodriguez - Rodríguez - Roskos - Ruby - Russak - Sadeghi - Saiegh-Haddad - Salmi - Sarsor - Savage - Sawyer - Schabmann - Schatschneider - Schirmer - Schneider - Schraeyen - Segal-Seiden - Segers - Seidenberg - Seipel - Serrano - Sevcik - Shafman - Shany - Shapiro - Silverman - Simpson - Sims - Skibbe - Smith - Soden - Soderberg - Soltész - Sparks - Speece - Spencer - Spencer - stacey - Staels - Stainthorp - Steacy - Strait - Street - Struthers - Swanson - Szwed - Taibah - Tani - Tarczynski-Bowles - Tellings - Templeton - Thompson - Thomson - Tibi - Tighe - Toledo Piza - Tong - Tong - Tops - Torppa - Traficante - Treiman - Tsai - Ugen - Uhry - Ullman - Uno - Valdois - Van Daal - van den Bos - Van den Broeck - Van der Kooy- Hofland - Vandermosten - Vaughn - Vaughn - Veispak - Verhallen - Verhoeven - Vincent - Wade-Woolley - Wagner - Wagner - Wang - Wang - Wanzek - Washington - Wasik - Webb - Weng - Wheat - Willcutt - Williams - Wiseheart - Wolf - Wolter - Wood - Wright - Wu - Wydell - Xiao - Yang - Yang - Yeh - Yulia - Zaretsky - Zhao - Zucker
Valeria Abusamra (Universidad de Buenos Aires); Romina Cartoceti; Andrea Casajús; Aldo Ferreres, Rossana De Beni; Cesare Cornoldi - Text comprehension as a multicomponential ability: The test "Leer para Comprender".
INTRODUCTION Text comprehension is not a single ability but rather includes independent skills. In this way, a multicomponential perspective is essential when considering text processes from theoretical and empirical points of view. PURPOSE The aims of the present work were: 1) to present the normative data of a test to evaluate text comprehension that identifies 11 components of the textual dimension; and 2) to consider the influence of variables such as sex, school grade and educational opportunities on the students performance. METHOD Some 2092 students of schools in Buenos Aires between 9 and 12 years of age participated. All of them were evaluated with the test Leer para Comprender (Abusamra et al, 2010). On the basis of the results we calculated: (a) the mean, standard deviation, percentiles and cuartiles of each area of the protocol and (b) the influence of three independent variables - school grade, educational opportunities and sex - on performances in the 11 areas of the test. RESULTS It was verified that the students' performances improved significantly the higher their school grade (p<,000). A Kruskal-Wallis test showed that the differences by level of educational opportunities were highly significant in the 11 areas that make up the test (p<,000). Finally, the girls showed a significantly better performance than the boys in 9 of the 11 areas (U of Mann-Withney). CONCLUSION We have obtained a very useful instrument for the fields of both clinical (neuro)psycholinguistics and education. The information provided by the test may be useful for designing a customized intervention that improves the child's potential.
Pasquarella Adrian (Ontario Institute for Studies in Education (OISE) at the University of Toronto)Xi Chen; Esther Geva ; Alexandra Gottardo - Universal and language-specific processes in word reading: Comparing cross-language transfer between Spanish-English and Chinese-English bilinguals
Purpose: This one-year longitudinal study examined cross-language transfer of word reading accuracy (WRA) and word reading fluency (WRF) in Spanish-English and Chinese-English bilinguals. Method: 105 Spanish-English and 97 Chinese-English bilinguals completed parallel measures, in grades (G) 1 and 2, of WRA (i.e., Woodcock Word Identification), WRF (i.e., TOWRE), and phonological awareness in English (L2), and their first language (L1). Results: SEM analysis was used to examine the relationships between L1 and L2 WRF and WRA from G1 to G2. WRF in G1 had significant cross-language relationships with WRF in G2, for both groups. These cross-language relationships were found in both directions (L1 to L2 and L2 to L1). For WRA, the only significant cross-language relationship was from G1 English WRA to G2 Spanish WRA. Conclusions: Researchers have proposed two hypotheses, "central processing hypothesis" and "script dependent hypothesis" to understand universal and language-specific processes in reading different orthographies (Geva & Siegel, 2000). Our results suggest that reading fluency may operate under a central processing component because significant relationships were found across languages with similar (Spanish-English) and different (Chinese-English) scripts, and in both directions. Alternatively, the significant relationship across English-Spanish word reading accuracy, and the lack of relationships in Chinese-English, demonstrates that transfer of word reading accuracy is script specific. To summarize, cross-language transfer of word reading accuracy may be dependant upon similarities in the scripts while reading fluency transfers universally. Implications for assessment and intervention of bilinguals will be discussed.
Yusra Ahmed (); Richard K. Wagner - Modeling the developmental relationship between reading and writing: a latent change score analysis
Purpose - The connection between reading and writing have been studied extensively. This article applied new advances in dynamic modeling to investigate longitudinal developmental relationships between reading and writing skills. Latent change score models were used to compare unidirectional pathways (reading-to-writing and writing-to-reading) and bidirectional pathways in a test of nested models. Method - Participants included 316 boys and girls who were assessed annually in grades 1 through 3. Data analyzed in this study included word level reading and spelling. Results - The results showed that a bidirectional model fit the data better than the unidirectional models, and a reading-to-writing model fit better than a writing-to-reading model. Significant longitudinal relationships were observed for the reading-to-reading and writing-to-writing stability parameters and between reading and writing in the bidirectional model. The magnitudes of the reading-to-writing pathways (2.36 from grade 1 to 2 and 1.51 from grade 2 to 3) were stronger than writing-to-reading (1.38 from grade 1 to 2 and 1.32 from grade 2 to 3). Conclusions - The results suggest that reading performance increased gain in spelling more than spelling performance increased gain in reading and this association was stronger between grades 1 and 2 than between grades 2 and 3.
Hei Won Heidi Ahn (University of Oxford); Kate Nation; Elizabeth Wonnacott - Differences in morphological decomposition in visual word recognition: A direct comparison between English adults and children
Purpose: A well-documented finding in the visual word recognition literature is morpho-orthographic decomposition, a process posited to operate in the early processing stages of visual word recognition whereby all words giving the appearance of morphological complexity are parsed into their plausible morphemic constituents regardless of their genuine morphological complexity. Though this is well-replicated in skilled adult readers using masked priming, very few have investigated morpho-orthographic decomposition in child readers. This study sought to provide a direct comparison of morpho-orthographic decomposition in visual word recognition between English adults and children. Method: In Experiment 1, forty English 9-10 year old typical readers were run on a visual masked prime lexical decision task. Critical items were pre-piloted and also included in a post-test to ensure familiarity. In Experiment 2, forty English typical adult readers were run on a modified version of the task. Results: Whereas adult readers evidenced morpho-orthographic decomposition, this finding sharply contrasted with that of the child readers who demonstrated a different pattern of decomposition. Child readers appeared to only decompose genuinely morphologically complex words, such as 'builder' into its constituents and not words simply bearing the appearance of morphological complexity, such as 'corner'. Conclusion: These findings implicate the decomposition process to be truly morphological in the child reader and not simply morpho-orthographic. In contrast to skilled adult readers, child readers appear to be sensitive to genuine morphological relationships between the whole word and its components even in the very earliest stages of visual word recognition.
Karen Aicher (University of Connecticut, Haskins Laboratories)Jay Rueckl - The influence of word form on the acquisition of meaning: an adult word learning study
Purpose: Similarity to existing words in the lexicon (wordiness) has been found to influence novel word learning, recognition, and recall (cf. Storkel, et al., 2006; Joordens, et al., 2008; Collisson et al., in prep). The current study examines the influence of the orthotactic similarity of novel words on the acquisition of meaning and subsequent naming behavior for those items. Method: Wordiness was quantified using orthographic neighbors, the summed frequency of orthographic neighbors, and bigram frequency; high and low wordiness pseudoword stimuli were created with no overlap between the categories. Phonotactic probabilities did not differ between the two wordiness categories. Participants learned pseudoword-picture pairs across eight learning epochs using a paired associate paradigm (Sandak et al., 2004). Participants named high and low wordiness pseudowords pre and post-training, and completed the orthographic choice lexical decision task (Olson et al., 1989). Results: Preliminary results of the learning task revealed two distinct groups: those that were faster and more accurate on high wordiness items than low, with the other group showing the opposite pattern. These groups also differed in their patterns of performance overall. Better performance learning high wordiness items was associated with greater changes in learning on both wordiness types, faster and more accurate naming, and better pseudohomophone detection. Conclusions: Results will be discussed in terms of these individual differences in learning patterns and the implications for learning, memory, and current theories of skilled visual word processing.
Stephanie Al Otaiba (Florida Center for Reading Research at Florida State University)Jessica Folsom; Luana Greulich; Yaacov Petscher; Jeannie Wanzek; and Carol M. Connor - A closer look at RTI: The role of language, behavior, home literacy and classroom reading instructional environment on student response to first grade multi-tiered intervention
Purpose: This study examined the effects of child and instructional mediators and moderators in examining response to three-tiered intervention in first grade. Methods and Results: This RCT included 7 schools, 34 teachers and 556 students. After initial screening, students were randomly assigned within classrooms to one of two types of supplemental intervention; subsequently screening occurred every 8 weeks and students in both conditions were eligible to change tiers based screenings. In Dynamic RTI, students were assigned to tier based on their initial screening; 23 students entered directly Tier 3 and 58 students entered Tier 2. In Typical RTI, all students began in Tier 1. Graduate RAs provided tiered intervention in both conditions. Students' reading was tested in fall, winter, and spring; parents completed a survey about home literacy, and classroom instruction and supplemental intervention was observed, videotaped, and coded for quality of reading instruction, individualization of instruction, and for fidelity of intervention implementation. Results: MANOVA revealed a significant difference for Tier and Condition, but no Tier by Condition overall interaction, indicating that students assigned to either RTI condition showed similar growth on reading. Regression conducted with 164 students who received Tier 2 or 3 revealed significant predictors of RTI (in order of strength) were scores on initial blending and elision, spelling, reading scores, and language and memory factor scores. By contrast, fall rapid naming and behavioral measures did not add significantly to the model. Conclusions: We anticipate that additional analyses using hierarchical linear modeling that include home and classroom information and for the nested structure of the data will confirm the relative importance of mediators and moderators to RTI.
Zainab Allaith (); R. Malatesha Joshi; Astri Yulia - How does a diglossic first-language phonology influence spelling in a second-language? The case of the Arabic language
Purpose: The purpose of the present study was to examine the influence of Arabic, a diglossic language, on spelling English words. Method: The spelling performance of fourth and sixth grade Arabic children was compared with English monolinguals on (1) the phonemes /p/ and /v/ which exist in Standard (written) Arabic and English but not Spoken Arabic, and their phoneme pairs /b/ and /f/; (2) the phonemes /tʃ/and /g/ which exist in Spoken Arabic and English but not in Standard Arabic, and their phoneme pairs /dʒ/ and /k/; and (3) control phonemes pairs, /d/ and /t/ which exist in English as well as in spoken and written Arabic. Results: The Arabic participants did not differ from the English participants in spelling the control phonemes /d/ and /t/. However, unlike their English counterparts, they spelled (1) /p/ and /v/ as b and f due to their absence in Arabic, and /b/ and /f/ as p and v because of overgeneralization and failure in perception of voicing articulatory features; and (2) /tʃ/and /g/ as j and k due to their absence from the written Arabic, but not vice versa because of strong perception of voicing articulatory features between the phoneme pairs because of their presence in Spoken Arabic. Conclusion: Diglossia has a positive influence on helping Arabic children differentiate between phonemes and their phoneme pairs in English. However, knowledge of both the written and the spoken forms of phonemes in first-language is necessary to support spelling in a second-language spelling.
Nicholas Allan (); Nicholas P Allan; Christopher J Lonigan - Predicting emergent literacy growth from externalizing behaviors: The utility of multiple informants
Purpose: Externalizing behavior problems (e.g., inattention, hyperactivity/impulsivity, and oppositional behavior) measured in preschool children predict slower growth in emergent literacy skills (e.g., phonological awareness, print knowledge, definitional vocabulary). Because different raters have different strengths and weaknesses in rating behavior, it is important to examine the utility of multiple informants like research assistants only briefly exposed to children's behavior and classroom teachers. It was hypothesized that both research assistant and teacher ratings would predict growth in emergent literacy skills in preschool. Method: As part of an emergent literacy intervention study, 382 preschool children (Mean age = 54.0 months, SD = 5.79 months) were given emergent literacy measures three times during the school year. Research assistants who administered the emergent literacy measures and were in contact with each child for roughly 30 minutes and classroom teachers rated children on their externalizing behavior problems. Results: Growth curve analyses, controlling for intervention condition, were used to examine the the predictive utility of behavior ratings on emergent literacy skills. Separate analyses were conducted with each pair of research assistant and teacher behavior ratings entered in the model as covariates. Both research assistant and teacher inattention and hyperactivity/impulsivity uniquely and predicted slower phonological awareness growth. Conclusions: Results of this study support the utility of multiple informants because of the unique information each can provide. Behavior ratings provided by research informants with only a limited opportunity to observe a child's behavior provide unique information regarding the relation between behavior problems and emergent literacy growth.
Elizabeth Allen (PRO-ED); Donald D. Hammill - Can the Test of Silent Contextual Reading Fluency Accurately Predict Reading Comprehension?
Purpose: This study examines the ability of the Test of Silent Contextual Reading Fluency (TOSCRF; Hammill, Wiederholt, & Allen, 2006), a 3-minute, group-administered measure, to predict reading comprehension as measured by a combination of longer, individually-administered reading measures (i.e., GRADE and WJ III). Method: This study will use data collected by the Texas Center for Learning Disabilities in 2007 (N = 1,612) because, in addition to the TOSCRF, all students were administered four well-known measures of reading: GRADE, TOWRE, TOSREC, and WJ III. This presentation provides three types of evidence for the validity of the TOSCRF: (a) a review of correlation coefficients showing the relationship between TOSCRF and the criterion measures, (b) a comparison of the means of the TOSCRF and those of the criterion measures, and (c) the results of a series of binary classification and ROC Area Under the Curve analyses. Results: The results indicate: (1) that the TOSCRF possesses a Large to Very Large relationship with the criterion measures; (2) that for all practical purposes, regardless of the samples' characteristics or the criterion test administered, the standard scores that result from giving the TOSCRF will be similar to those obtained from giving the criterion tests; (3) that the TOSCRF is highly predictive of reading comprehension. Conclusions: The TOSCRF is a quick and accurate index of overall reading ability and represents a potential screening instrument that could be used in a RTI framework.
Mitsue Allen-Tamai (Aoyama Gakuin University) - How do children develop their reading abilities in an impoverished environemnt?
There are about 23,800 elementary schools in Japan, where English has not been formally taught, and 98.3% of Japanese children attend public schools. The Ministry of Education, Science, Sports, and Technology publicized its decision to introduce a new course called "Foreign Language Activities" to 5th and 6th graders once a week from 2011. The primary purpose of this course is to foster a positive attitude to understand people from different cultures, but not to teach a language. Since the experiment started in 1992, literacy education has been strongly forbidden. The guideline says that at the elementary school stage, it is important to separate spoken and written English and focus instruction only on the spoken. This paper is a part of a longitudinal study (N = 77 in an experimental group and 60 in a control group) to develop reading ability among young Japanese EFL learners, based on the interactive hypothesis, balancing bottom-up and top-down reading skills. Besides measuring their phonological awareness, alphabetical knowledge and word recognition, the affective variables such as willingness to communicate in L1, learning motivation, self-esteem are measured. Statistically significant differences are found and the experimental group made a great progress in their reading. The presenter will discuss the training effects of her original phonics teaching and join-storytelling developed mainly for EFL learners, who have such a limited exposure to English. The changes on affective variables through foreign language teaching will be discussed too.
Teresa Anthony (University of Texas Health Science Center); Jason Anthony; Jeffrey Williams - Evaluation of the Raising a Reader program and supplemental parent training in shared reading strategies.
Purpose: This four year project evaluated a commercial book exchange program with and without supplemental parent education in optimal shared reading strategies. Method: Raising a Reader involved weekly rotation of four books through children's homes and partnering preschool classrooms and families with neighborhood libraries. At monthly Family Nights, parents learned about and practiced shared reading techniques with their children. Content included strategies often associated with Dialogic Reading. Participants included 620 preschool children and their families from economically disadvantaged minority backgrounds. The 94 classrooms were randomly assigned to one of three groups: Control, Raising a Reader, or Raising a Reader plus Family Nights. School readiness skills were assessed at the beginning and end of the school year. Results: Multilevel ANCOVAs were used to predict end of year school readiness, controlling for pretest scores and classroom nesting. Relative to control classrooms, Raising a Reader had no significant effects. However relative to Raising a Reader alone, Raising a Reader plus Family Nights led to significant improvements in expressive vocabulary (t[1, 449]=1.98, p < .05), receptive vocabulary (t[1, 399]=2.19, p < .05), expressive grammar (t[1, 427]=1.93, p = .05), receptive grammar (t[1, 424]=3.42, p < .001), memory span for sentences (t[1, 428]=2.04, p < .05), print awareness and letter knowledge (t[1, 431]=3.08, ps < .01), and phonological awareness (ts[1, 424-432]=2.08-228, ps < .05). Conclusions: Large scale, school-based shared-reading interventions can efficiently promote school readiness only when they include a parent training component.
Jason Lon Anthony (University of Texas Health Science Center at Houston); Teresa Anthony; Jeffrey Williams - Phonological representation abilities and early literacy skills among children with speech sound disorder: new evidence for a phonological core deficit.
Purpose: This study aimed to identify weaknesses in early literacy skills and phonological processing skills among children with Speech Sound Disorder (SSD) that place them at increased risk for reading difficulties. Method: Language, emergent literacy, and phonological skills of three groups of preschool-age children were compared. A group of 68 children with SSD was compared to a group of 68 peers with normal speech who were matched on receptive vocabulary and a group of 68 peers with normal speech and normal language. Results: Children with SSD demonstrated significant impairments in expressive phonological awareness (ts=2.79 to 6.53, ps < .01), receptive phonological awareness (chi-squares=4.22 to 27.53, ps < .05), accessing phonological representations (ts=3.45 to 5.85, ps < .001), quality of phonological representations (chi-squares=5.45 to 35.08, ps < .02), and word reading (ts=1.96 to 4.01, ps < .05). Children with SSD did not demonstrate weaknesses in knowledge of letter names or letter sounds relative to children matched on vocabulary. ANCOVAs found that lower performances of the SSD group on tests of phonological awareness and word reading could be explained by their weaknesses in quality and accessibility of phonological representations. Conclusions: Results are discussed in terms of a phonological core deficit explaining the speech and literacy profiles of children with SSD and the clinical implications thereof.
Kenn Apel (Dept. of Comm. Sci. & Disorders, University of South Carolina)RaMonda Horton-Ikard; Danielle Brimo; Elizabeth Wilson-Fowler - Productive Oral Language Ability is Not Associated with Early Literacy Skills
Purpose - When assessing the relation between oral language and literacy abilities, researchers typically use language measures, such as the Test of Language Development - Primary, to determine associations between these sets of skills (e.g., Catts, 1993; Peterson, Pennington, Shriberg, & Boada, 2009). However, these types of measures require children's conscious thought about language rather than provide a productive measure of language ability. The aim of this study was to determine whether productive measures of oral language predict early word-level reading and spelling abilities compared to measures that require conscious consideration of the linguistic units of language. Method - Participants were thirty-six typically developing kindergarteners (M=72 mos.). A narrative sample was collected from each child using a wordless picture storybook. Narrative samples were transcribed using the Systematic Analysis of Language Transcripts (SALT, 2010) Productive language measures obtained from the samples were overall Narrative Scoring Scheme scores, number of utterances per t-unit, mean length of t-unit, and number of different words. Additionally, measures of phonemic and orthographic awareness, spelling, and word-level reading (nonsense and real word) were obtained. Results - None of the productive oral language measures were significantly related to word-level reading and spelling; however, orthographic and phonemic awareness tasks were related to these literacy skills (rs = .44 -.76). Conclusions - The results suggest that previous claims of the association between oral language and literacy skills have been misguided; researchers' measures of language ability actually assessed linguistic awareness. Discussion will focus on the theoretical implications of the findings and suggestions for future research.
Dorit Aram (Department of Human Development and Education, Tel Aviv University, Israel);Shira Besser - Parental mediation of writing to precocious readers: A comparison with mediation to age-matched and reading-matched children
Purpose. The study investigated the role that home environment plays in the development of precocious reading via an examination of the characteristics of parent-child writing interactions. Method. Participants were 60 parent-child dyads: 20 Precocious Reader (PR) preschoolers, 20 non-PR preschoolers matched by age and gender to the PRs, and 20 first graders matched by reading level to the PRs. All dyads were videotaped at home writing three tasks (a birthday invitation, a story, and dictated words). We analyzed the interactions referring to four aspects of parental support: literacy (grapho-phonemic mapping and printing of the letters), performance (demand for precision, planning, self regulation, etc.); atmosphere (reinforcements, feeling of competence, mutuality, etc,) and cognitive support (distancing, elaborations, etc.). Results. Parents of PRs differed from parents of both age and reading level comparable children. Their literacy support was similar to that of parents of reading level comparable children and higher than that of parents of age comparable children. Their high level emotional support was similar to that of parents of age comparable children and higher than that of parents of reading level comparable children. Lastly, their level of both cognitive and performance support was higher than in the other two groups. Conclusion. Precocious reading and effective parental support appear together. Parents of PRs were more successful in taking advantage of opportunities for turning the writing encounters into meaningful situations. Parents and educators can learn from parents of PRs how to mediate in a more efficient way that could promote young children's development.
Joanne Arciuli (Faculty of Health Sciences, University of Sydney, Australia) - Assignment of lexical stress during reading
Much of what we know about reading is based on the processing of English monosyllables. As over 90% of word types in English are polysyllabic, data from monosyllables does not enable us to understand fully the process of reading. Of growing interest is the issue of stress assignment during the reading of polysyllables. Take, for example, the monosyllable 'can'. Cognitive and computational models ought to address the reading of that same letter string in 'candle' versus 'canal', when the string is distinct in terms of assignment of lexical stress. However, until recently, leading models of read ing (including both single route and dual route approaches) have not specified how lexical stress is assigned for polysyllabic words and nonwords. I will discuss the results of a program of research that has been designed to shed light on how (1) healthy adults, (2) typically developing children, and (3) children with developmental disabilities assign lexical stress during their reading of disyllables. I will provide an overview of corpus, behavioural and computational investigations in this area. The results of these studies suggest that sensitivity to probabilistic orthographic cues is important for assigning lexical stress when reading polysyllables. In particular, word-endings appear to provide the reader with helpful information regarding patterns of lexical stress in English disyllables.
Karen Au-Yeung (OISE UT)Xi Chen, Active Member - Development of Early English Literacy Skills in ESL Children from Diverse First Language Backgrounds
The purpose of the study was to investigate early reading development in English of ESL children in a French immersion in Canada. Although Canadian French immersion programs were traditionally designed for children from English-speaking families, in recent years, there has been a sharp increase in the number of ESL children in these programs. However, little is known about the development of early English literacy skills of these children, as compared to their native English-speaking peers. The present longitudinal study examined this issue. The performance of 40 ESL children from diverse first language backgrounds (e.g., Russian, Hebrew, Chinese, Indian, etc.) and 41 native English-speaking children was examined on tasks assessing phonological processing, word reading, and vocabulary in the fall and winter semesters of Grade 1. ESL children experienced greater growth in expressive vocabulary, and their growth in receptive vocabulary was not significantly different from that of native English-speaking children. The two language groups showed similar growth and achievement on measures of phonological processing and word reading. Furthermore, phonological awareness was a significant predictor of word reading in each language group. The results of the study suggested that, in spite of their L2 status, ESL children did not lagged behind in developing early English literacy skills, even when their English exposure is limited in a third language educational setting. Keywords: ESL, reading development, phonological awareness, vocabulary
Diane August (Center for Applied Linguistics);Lauren Artzi; Chris Barr; Coleen Carlson; David Francis; - Developing high frequency academic vocabulary in young English-language Learners: Results of an intervention study
Purpose: The Purpose: of this research was to assess the effects of an intervention on the vocabulary acquisition of 187, second grade English-language learners. Method: The intervention was implemented in 4 high poverty schools by 10 teachers. The intervention curriculum consisted of five daily 40-minute lessons per week delivered over nine weeks. The study utilized a within subjects design with all students receiving the intervention. The vocabulary words were directly instructed, paraphrased in context or appeared in the text only (exposed condition). The three instructional conditions allowed for a word level analysis with exposed words serving as controls. Within each instructional condition, there were four types of vocabulary words-- abstract cognates, abstract non-cognates, concrete cognates, and concrete non-cognates. This breakdown by type of word allowed for a comparison of gains for types of words as well as a comparison of the magnitude of the gains within instructional condition and between instructional conditions. Results: Within group T-tests found significant pre- to post test gains on two subtests of the Gates-MacGinitie and for all types of words on a curriculum-based measure under the directly instructed and paraphrased conditions, but not the exposed condition. Gains were statistically greater for the directly instructed words in all conditions except for concrete cognates. Effect sizes ranged from moderate to very large. Conclusions: Findings shed light on how the conditions in which students are engaged with words (direct instruction, paraphrasing, exposure) interact with word type in the development to students' vocabulary knowledge.
Michal Balass (Hobart and William Smith Colleges); Charles A. Perfetti - An ERP investigation of word experience effects on vocabulary acquisition and semantic processing
Purpose: Two studies were conducted to test the effects of word experience, (i.e, type of context, variability, and semantic constraint) on new vocabulary learning. According to the Word Experience Model (Reichle & Perfetti, 2003), and similar hypotheses (Ans et al., 1998) different word experiences elicit episodic memory traces of varying strength, thus affecting subsequent experiences with the word, and later meaning processing and retrieval. Methods: In both studies, adult readers of varying comprehension and lexical ability were exposed to 120 rare English words. In study 1, readers learned words in definitions or single sentence contexts of repeated or varied content. In study 2, readers were exposed words in low or high constraining single sentence contexts of repeated or varied content. In both studies, readers completed a meaning-judgment task and a sentence judgment task. Of interest was the N400 effect which is an indicator of semantic processing and retrieval. Results: Both studies indicated that variability of content (i.e., for definitions or sentence contexts) yielded greater meaning learning in the sentence judgment task; evidenced by a larger N400 effect. However, repeated content yielded greater meaning learning in the meaning judgment task, especially when the sentence constraint was high. Overall, readers with better lexical ability showed greater N400 effects. Conclusions: Variability of content in vocabulary learning allows for the flexible use of the word when encountered in new sentence contexts. However, high constraint repeated content results in less flexible, but well-specified meaning representations.
Thais Barbosa (Universidade Federal de São Paulo)Camila Cruz-Rodrigues; Carolina M. Toledo-Piza; Elayne C. M. Pinheiro; ana Luiza G. P. Navas; Orlando F. A. Bueno - Differences in profiles of children with dyslexia and learning difficulties.
This study aims to identify differences in oral language, reading and writing profiles of children with dyslexia, when compared to children with learning difficulties. 80 children of both sexes, ranging from 9 to 14 years of age, were divided into four groups all matched by sex and type of school (public or private). Dyslexia Group (DG): 20 children, diagnosed according to IQ/achievement criteria; Learning Difficulty Group (LDG): 20 children, matched to the DG by age and reading level; Age Control Group (ACG): 20 children matched by age to both mentioned groups; Reading Control Group (RCG): 20 children, matched to the first two groups by reading level. All subjects were tested in oral language (semantics, syntactic and phonological aspects), reading skills (words, pseudowords and texts - speed and comprehension), writing (words, pseudowords and texts), as well as working memory, and declarative memory. Data analysis indicated differences in oral language, reading, writing and memory profiles of dyslexics, when compared to children with learning difficulties. Phonological processing aspects were the most relevant parameters in differentiating the groups. Furthermore, findings demonstrate that criteria based on IQ/Achievement discrepancies are not sensitive enough to discriminate dyslexics from other groups, since reading skills varied from highly skilled readers, to fluent readers and slow readers, with low accuracy. The present study identified reading skills as part of a continuum, in which dyslexics are positioned in the lower end of the curve.
R. Michael Barker (University of South Florida); Rose A. Sevcik; Mary Ann Romski; Robin D. Morris - Modeling phonological processing for children with mild intellectual disabilities: The relationship between underlying phonological abilities and associated language variables
Purpose: Debate concerning the nature of phonological awareness (PA) has focused on whether it consists of a single underlying cognitive-linguistic ability or multiple abilities that are distinct but correlated. Early research concluded that PA indeed consisted of multiple abilities, but more recent research has concluded that a single underlying ability best explains performance on PA assessments. This paper tested whether phonological processing consisted of a single factor, two factors, or more than two factors for children with mild intellectual disability (MID). Method: PA and language were assessed in 222 students from local public schools using subscales of the Comprehensive Test of Phonological Processing and measures of expressive and receptive language. Confirmatory factor analysis was used to test hierarchical models of the structure of phonological processing. The relationships between these components and language were also estimated. Results: The results indicated that a two-factor model with PA and naming speed latent abilities fit the data best. PA and naming speed were correlated at -.52. PA was strongly and positively associated with expressive and receptive language. Naming speed demonstrated a medium, negative association with expressive and receptive language. Conclusion: This paper represents the first demonstration of the structure of phonological processing in children with MID. These results are consistent with previous research with typically developing children, indicating that it is likely that children with MID develop phonological processing similarly. Consequently, it is very important to support phonological processing and language for children with MID to ensure they achieve appropriate levels of literacy.
Laura Barquero (Vanderbilt University); Donald Compton ; Jennifer Gilbert ; Eunsoo Cho ; Bobette Bouton ; Douglas Fuchs ; Lynn Fuchs - Upside-down RTI: reversing tier 2 and tier 3 for the lowest achieving students
Purpose: This study explored an alternative Response to Intervention (RTI) framework for students most at-risk for reading problems. In "upside-down" RTI, struggling readers bypassed small group instruction and progressed directly to intensive 1-on-1 instruction with the possibility of later moving to small group instruction. Upside-down RTI was compared to a standard 3-tiered RTI progression of general classroom (Tier 1) to small group (Tier 2) to 1-on-1 (Tier 3) instruction. Method: Participants were first-grade students with very low achievement on rapid letter naming, word reading, and word identification fluency growth. In a quasi-experimental design, the upside-down RTI group (n=22) were compared to a matched control group that received standard RTI (n=22). In both frameworks, students received 14 weeks of intervention that included letter sounds, sight words, decoding, and passage reading. After 7 weeks, students changed or remained in a tier based on word identification fluency growth. Analyses of variance (ANOVAs) compared groups on word reading, pseudoword reading, reading comprehension, and word identification fluency growth. Results: When controlling for school, students who received upside-down RTI performed better at post-test on pseudoword reading, pseudoword fluency, and passage comprehension, and exhibited higher rates of word identification fluency growth over the course of intervention than did matched students who received standard RTI. The two groups did not differ in word reading level. Conclusions: In decoding and comprehension skills, highly at-risk students may benefit more from an upside-down RTI approach that provides intensive 1-on-1 intervention sooner than from the standard RTI approach that delays individual instruction.
Roderick Barron (University of Guelph);Maggie Addison - Role of phonological awareness in contingent relationships between letter-name and letter-sound knowledge in non-readers
Purpose: The role of phonological awareness in the relationship between letter-names and letter-sounds continues to be controversial (see Evans et al. 2006; Treiman et al., 2008 versus Kim et al. 2010). Controlling for latent cognitive ability, we used different contingent relationships between individual letter-names and their corresponding sounds to investigate this issue among non-readers. Method: 130 English speaking non-readers (CA= 67.9 months; mean raw Word ID = 1.14, Word Attack = 0.14) named letters and recognized sounds associated with letters. Phonological awareness was assessed with oddity and phoneme deletion tasks. The latent cognitive variable consisted of WPPSI-IQ, serial memory, and WRAT-math. Results: When children knew letter-names and recognized their corresponding sounds, phonological awareness was positively related to letter-sound recognition for consonant letters containing sounds in their letter-names (p <. 001). Phonological awareness was not significant for letter-sound recognition of vowels and consonants whose letter-names did not contain their sounds. When children knew letter-names but did not recognize their corresponding sounds, phonological awareness was negatively (p <.01) related to letter-sound recognition. Structural equation models had RMSEAs < .05 and TLIs and CFIs between .971-.996. Conclusions: With cognitive ability controlled, phonological awareness influences letter-sound recognition for non-readers when letter-sounds are present in the initial or final positions of corresponding letter-name syllables but not when they are absent (i.e. in short vowels and H, W, Y). Children with higher levels of "pure" letter-name knowledge may lack the phonological skills for using their letter-name knowledge to learn letter-sounds.
Amy Barth (University of Missouri-Columbia)Tammy Tolar, David Francis, Jack Fletcher, Sharon Vaughn - Factors Affecting the Assessment of Fluency in Middle School Students with and Without Disabilities
Purpose - The purpose of this paper was to investigate how text characteristics affect oral reading fluency within students, how student characteristics affect oral reading fluency, and the extent to which these two sources of variability interact in their effects on oral reading fluency, Method - We used multi-level models that differed from one another in their fixed and random effects, as well as in their specifications regarding the unexplained variance in fluency scores. Results - Between student variance accounted for 76% of the variance in reading fluency. In addition, we found that passage accounted for approximately 55% of the within-student variance in reading fluency. Passage characteristics such as Lexile, text type (expository or narrative), and page length all had significant effects on reading fluency and explained virtually all variance due to passage. Student characteristics (grade, gender, reader type, verbal knowledge, decoding skill, and word reading fluency) also had significant effects on reading fluency when controlling for all other student characteristics. There were interactions between text and student characteristics. Grade 6 students and students with higher word reading fluency decreased their passage fluency more than older and lower performing students as text became more difficult (i.e., higher Lexile, expository text). Conclusions - Both student and passage level characteristics impact oral reading fluency.
Rachel L. Beattie (University of Southern California); Franklin R. Manis - The role of linguistic stress sensitivity in the transition from pre-reader to reader
Theorists have proposed that prosodic sensitivity plays a role in early reading acquisition by facilitating the detailed representation of the sound structure of language and vocabulary growth. However, based on recent studies, it is unclear whether prosodic sensitivity and segmental phonological awareness are independent or linked abilities (Goodman, Libenson, and Wade-Woolley, 2010). Moreover, there is debate about how prosodic sensitivity and segmental phonological awareness are related to the development of receptive and expressive vocabulary (Wood, 2006; Cardillo, 2008). This study examined the independence of prosodic sensitivity and segmental phonological awareness and the relationship between these two constructs and vocabulary. As a part of a longitudinal study, forty-nine preschoolers were assessed on stress sensitivity using two measures: the Tennessee Test of Rhythm and Intonation Production and the Preschool Repetition Test. Phonological awareness was measured using four subtests: the Test of Preschool Emergent Literacy phonological awareness elision and blending subtests, a rhyme detection task and an alliteration detection task. Expressive vocabulary was assessed using the definitional vocabulary subtest of the Test of Preschool Emergent Literacy and receptive vocabulary was measured using the Receptive One Word Picture Vocabulary Test. The results of the confirmatory factor analysis indicated that prosodic sensitivity and segmental phonological awareness were independent, yet related factors. Prosodic sensitivity accounted for variation in segmental phonological awareness and vocabulary, but not beyond that accounted for by the relationship between vocabulary and segmental phonological awareness. Implications for emergent literacy models and those at-risk for developing reading problems will be discussed.
Lindsay Bell (University of Michigan Department of Psychology); Su Li; Frederick Morrison - Executive functioning and academic skills in Chinese and American kindergartners
Purpose This study compared the executive functioning (EF) skills (working memory (WM), cognitive flexibility (CF), and attentional inhibition (AI)) of Chinese and American kindergartners and examined the relationships between these skills and children's literacy and mathematics achievement. Method 408 participants (212 from 21 American classrooms and 196 from 14 Chinese classrooms; average age was 65 months in both groups) were assessed during the fall of kindergarten on CF (using the Dimensional Change Card Sort) and WM, AI, literacy, and mathematics (using Woodcock-Johnson tasks). Results Hierarchical linear modeling analyses showed that Chinese children outperformed American children on all EF and mathematics assessments (effect sizes (ES) ranged from 0.54 to 1.14). In American children, AI (ES = 0.17) and WM (0.29) predicted literacy skills; in Chinese children, only WM predicted literacy (ES = 0.45). AI (ES = 0.12) and WM predicted mathematics in both cultures, but WM had a stronger effect on mathematics for Chinese children than American children (ES = 0.38 and 0.16, respectively). Conclusions These findings show that Chinese kindergartners are advantaged in AI and CF, extending prior research (i.e., Sabbagh et al., 2006) with preschoolers showing similar results. Our results also show that Chinese kindergartners have superior WM skills. Furthermore, our findings connect EF to academic skills, showing that AI and WM are important for mathematics and literacy achievement in both cultures, but WM may be particularly important for Chinese children.
Gal Ben-Yehudah (The Open University of Israel)Mariya Lozin - Turning words upside-down: investigating perception-language interactions in visual word recognition
Purpose: Word inversion is a form of visual degradation in which an item is rotated by 180° (i.e., turned upside-down). In previous work, we found that bilinguals' ability to read inverted words in their second language was influenced by the characteristics of their native writing system (alphabetic, nonalphabetic). Moreover, performance was affected by the interaction between the orientation of an item and its lexical properties, suggesting that a piecemeal correction mechanism rectifies inverted linguistic items and that this mechanism is influenced by lexical information. The purpose of the present study was to test this piecemeal-interactive account of a perceptual correction mechanism, using the emergence of a word length effect (WLE) as a behavioral marker for piecemeal processing. Method and Results: 29 adults performed a lexical decision task in their native Hebrew language. Repeated measures ANOVAs revealed significant main effects of display orientation (upright, inverted), written frequency (high, low) and word length (3, 4, 5 letters) on RT and accuracy. The WLE was absent in the upright condition, but emerged in the inverted condition. A significant 3-way interaction was found due to RT differences between low- and high-frequency inverted words; high-frequency 3- and 4-letter words had similar RTs, whereas in low-frequency words RT increased with string length. Conclusions: The interaction between display orientation, written frequency and word length supports a piecemeal-interactive account of the perceptual mechanism that rectifies inverted words. Word inversion may, therefore, be a useful tool to investigate the interface between perceptual and linguistic processes.
Rachel Berthiaume (Département de didactique)Daniel Daigle - Phonological processing in deaf children: the role of the instructions attached to the task
Many deaf students encounter great difficulty in learning to read (Gallaudet Research Institute, 2004). Phonological processing is mainly pointed out as the source of their reading deficit (Musselman, 2000). The objectives of this study are to investigate deaf readers' graphophonemic (GP) and graphosyllabic (GS) processing using tasks varying in terms of cognitive constraints (GPCC-, GPCC+, GSCC-, GSCC+) (Bialystok, 2001). Deaf subjects (n=21) aged 9 to 12 were matched to hearing subjects of the same chronological age (CA) and of the same reading age (RA). In the GPCC- and the GSCC- tasks, subjects had to determine which of two pseudo-words (e.g., tauponir or teuponir / marpi or mapri) resembles the most a target item (e.g., toponir / mapli). In the GPCC+ and the GSCC+ tasks, subjects had to choose, out of four pseudo-words, the item that starts with a different sound/syllable (e.g., calime cirone cépale sanire / racro raplé ratru ralto). For all groups, results are generally above the threshold of chance. Results from ANOVAs indicate that all groups had better scores on the CC- tasks than on the CC+ tasks, and that deaf subjects and CL have comparable results, but perform more poorly than CA. These results constitute the basis for further research on the role of the cognitive constraints attached to a task when investigating phonological processing in deaf children's reading development. They also contribute to provide better tools to the teachers and SLPs who deplore the lack of assessment and intervention tools for deaf children with reading difficulties.
Rebecca Betjemann (Department of Psychology, Regis University)Erik Willcutt; Richard Olson; Jan Keenan; Sally Wadsworth; John DeFries; Bruce Pennington - Etiology and neuropsychology of comorbidity between reading disorder, math disorder, and ADHD
Purpose: To use neuropsychological and behavioral genetic methods to test the etiology of comorbidity between Reading disability (RD), math disability (MD), and ADHD. Method: As part of the Colorado Learning Disabilities Research Center twin study, twin pairs (8-18 years old) were recruited in which one of the twins met criteria for RD or ADHD (N = 514) or neither twin had RD or ADHD (N = 350). In addition to measures of reading decoding and comprehension, mathematics, and ADHD, each twin completed an extensive battery of cognitive tests that included measures of executive functions, reward processing, and response speed and consistency. Results: Composite reading and math scores were highly correlated with one another (r = .63) and moderately correlated with measures of inattention (r = .37 - .39). Similarly, significant categorical comorbidity was observed between RD and MD (48%), RD and ADHD (31%), and ADHD and MD (34%). Multivariate twin analyses suggest that common genetic influences that lead to slow and inconsistent cognitive processing speed account for most of the variance shared by all three disorders, whereas separate genetic influences contribute to working memory difficulties, RD, and MD. ADHD is associated with unique genetic influences that appear to be mediated by weak response inhibition and dysfunction in circuits involved in reward-related processing. Conclusions: These results add to a growing literature suggesting that RD, MD, and ADHD are due to both shared and independent genetic influences, and suggest that slow processing speed may account for much of the overlap among the disorders.
Gina Biancarosa (University of Oregon); Scott Baker; Hank Fien; Yong Han Park; Lana Santoro; Priti Haria; Susanna Williams; Janet Otterstedt - Closing the vocabulary and comprehension gap in first grade through read alouds
Purpose: One suggested approach to build young children's vocabulary and comprehension is through "read alouds" (Beck & McKeown, 2001). We studied the effects of read alouds as a means of improving first grade students' vocabulary and comprehension in whole class (Study 1) and small group (Study 2) contexts. Study 1 sought to determine the efficacy of whole class read aloud instruction, while Study 2 sought to determine the efficacy of a read aloud intervention offered in addition to whole class read aloud instruction. Methods: Study 1 involved 638 children in 39 first grade classrooms in 13 schools, and classrooms were randomly assigned to a whole class read aloud treatment or a no-treatment control. Study 2 involved 182 students in 24 first grade classrooms in 9 schools, where all classrooms implemented the whole class read aloud curriculum and students identified with low language and vocabulary skills were randomly assigned to a supplementary small group read aloud intervention or a no-intervention control. Both studies used hierarchical linear modeling to examine treatment effects on vocabulary, listening comprehension, and retellings after controlling for pretest listening comprehension (Study 1) or vocabulary (Study 2). Results: Study 1 demonstrated the treatment positively affected vocabulary. Quality of instruction interacted with treatment such that higher quality instruction in treatment classrooms resulted in larger benefits. Study 2 also demonstrated positive treatment effects, but only on researcher-designed measures of vocabulary and teacher perceptions of student understanding and interest. Conclusions: Read aloud curricula have the potential to improve young students' vocabulary and comprehension.
Gary Bingham (Georgia State University)Nicole Patton-Terry, Bridget Dever, Katherine Green, Kizzy Albritton, & Meghan Pendergast - Examining an early literacy professional development intervention: effects on prekindergarten teachers and children.
Purpose: The purpose of this study was to examine the impact of an early literacy professional development program on teachers' early literacy practices and children's early language and literacy development in prekindergarten. Method: Forty-five (23 lead; 22 teaching assistants) teachers and 326 students across ten early childhood centers participated. Teachers came from diverse ethnic backgrounds (55% Black). The majority of students came from low-income backgrounds and were racially diverse (50% Black, 30% Latino, 15% White, 5% Other). Procedures. Teachers received 8 language and literacy focused professional development (PD) training days during the school year and one on one mentoring in their classrooms once a week. Measures. Children's early language and literacy development was assessed at the beginning and end of prekindergarten on the following measures: Peabody Picture Vocabulary Test (PPVT-4; Dunn & Dunn, 2007), The Test of Preschool Early Literacy (TOPEL: Lonigan, Wagner, Torgesen, & Rashotte, 2007). The quality of children's early literacy environments and instruction they received was assessed using the Early Language and Literacy Classroom Observation Tool (ELLCO Pre-K; Smith, Brady, & Anastasopoulos, 2008) and the Classroom Assessment Scoring System (CLASS; Pianta, La Paro, & Hamre, 2008). Results: Results from student assessments in this study are analyzed against expected results of students in the same peer (age) group from the norm reference group (i.e., standardized scores). Repeated Measures ANOVA show significant standard score gains on receptive and expressive vocabulary measures, print and letter knowledge, and phonological awareness. Hierarchical Linear Modeling (HLM) analyses demonstrate that change in teachers' language and literacy practices explain a significant amount of variance in child gains. Conclusions: Findings have implications for the importance of improving the quality of early childhood educators' literacy and language practices and their affect on student learning.
Emily Binks-Cantrell (Texas A&M University)Bestern Kaani; Emily Binks-Cantrell; R. Malatesha Joshi; Dennie Smith - Phonological processing: A cross-linguistic comparison of reading in Chinyanja and English languages
Purpose: This study compares Phonological Awareness (PA) in English (opaque) and Chinyanja, (transparent) languages among school children in Zambia. PA has is an important predictors among reading novices, (Plaza & Cohen, 2003). The Psycholinguistic Grain Size Theory (Ziegler & Goswami, 2005) posits that that reading proficiency in transparent orthographies develops faster and almost effortlessly than in opaque languages. Method: Independent samples of 118 and 121 grades 4-6 respondents were received the Zambia Achievement Test-Reading Recognition (ZAT-RR) subtests in the two languages respectively in selected Basic Schools in Lusaka, Zambia. PA skills assessed included; initial sound matching, initial sound-picture discrimination and initial sound-word matching. Results: The findings revealed that respondents who received the PA assessments in English (initial sound matching, initial sound-picture discrimination and initial sound-word matching subtests) outperformed those tested in Chinyanja on all subtests. A one-way ANOVA revealed statistically significant main effects on all three PA subtests; initial sound matching, [F (1, 238) = 12.85, p < 0.01], initial sound-picture discrimination [F (1, 238) = 43.23, p < 0.01] and initial sound-word matching [F (1, 238) = 40.19, p< 0.01]. Conclusion: The findings revealed that PA was significantly better in English than Chinyanja language. This contradicts Ziegler and Goswami's (2005) Psycholinguistic Grain Size theory, which posits that acquisition of reading skills is easier in opaque than transparent orthographies. From the findings, it may be assumed that PA is not the best predictor reading in transparent orthographies.
Freyja Birgisdottir (University of Iceland)Hrafnhildur Ragnarsdottir; Steinunn Gestsdottir - The role of phonological skills and oral language in the development of reading in transparent orthographies
Purpose: The role of phonological awareness and oral language in learning to read transparent orthographies is not fully understood. While some studies have shown phonological awareness to be an important predictor of reading in consistent orthographies, others have produced opposite results. Oral languge might also affect certain aspects of reading in different ways across orthographies. For example, Share (2005) suggests that readers of consistent orthographies rely less on lexical and contextual information in word-recognition than readers of more inconsistent orthographies. This longitudinal study examined the effect of phonological and oral language in the acquisition of reading in an orthographically consistent orthography (Icelandic). Method: Participants were 111 Year-1 children in first grade who were assessed on letter knowledge, phonological awareness, vocabulary and oral comprehension and their performance related to reading achievement a year later. Results: Letter-knowledge and phonological awareness correlated highly with all of the reading assessments. However, when the effects of age and initial reading skills were removed, these two measures no longer predicted reading comprehension, but still explained between 3 and 4% in word-recognition, non-word reading and spelling. After removing the effects of age and initial reading, vocabulary and oral comprehension explained 3% in reading comprehension, and oral comprehension explained 3% in word-recognition. Conclusions: Consistent with studies of English-speaking children, these results indicate that phonological awareness is an important predictor of word-recognition and decoding in transparent orthographies. Oral comprehension also appears to affect word-recognition, although the predictive contribution of oral language seems more consistent for reading comprehension than for other aspects of reading.
Jay Blanchard (Arizona State University)Jerry Cabrera; Glen Powell; Kim Atwill - What are the requisite levels of Spanish needed to support transfer to English? A longitudinal successive-year, two-cohort study from kindergarten through 3rd grade
Purpose This proposed poster will present a longitudinal, two-cohort study on the requisite levels of Spanish vocabulary skills at kindergarten needed to support the development of English skills through 2nd and 3rd grades. By the start of formal schooling in the U.S. many Spanish-speaking children have accumulated speaking and listening vocabularies in their native language: Some accumulations are larger than others. These accumulations are important because success in English is based on the theory that Spanish skills support English acquisition. However the development and deployment of Spanish to support English acquisition is not clearly understood. Method Eighty children were randomly selected from each of two successive-year kindergarten cohorts and followed through 3rd grade. All children were Hispanic, low-SES, Spanish-speaking with little or no English and attending immersion classrooms. Based on kindergarten-entry Spanish skills, children were divided into two groups: age-appropriate skills or below. Measures of receptive vocabulary and phonemic awareness were administered in English and Spanish, while reading measures, oral fluency, comprehension and language use were administered in English. Analyses include descriptive, correlation and linear regression. Results The most salient result was that Spanish vocabulary skill in kindergarten was the strongest predictor of performance in English reading and language achievement on the TerraNova tests at both 2nd and 3rd grades. Conclusions The current study indicates that the requisite levels of Spanish needed for transfer to English is at least age-appropriate Spanish skills upon enter into kindergarten (Cummins: Developmental Interdependence Hypothesis). Without these skills, there could be difficulty with English achievement.
Jason Boggs (University of Nevada, Las Vegas); Dr. Ralph E. Reynolds (Charter Member); Cristina Salinas - Third graders' understanding of metaphoric language in text: A mixed methods approach to assessing children's understandings
Purpose The purpose of the current study was to see if interviewing subjects concerning their reasons for answering comprehension questions would lead to a better understanding of their metaphor comprehension ability. Children have shown the ability to comprehend metaphors in text at a better than chance rate beginning in third grade; however, third graders still do not comprehend metaphors as well as they comprehend literal expressions (Reynolds & Ortony, 1982; Reynolds and Heddy, 2010). One possible explanation for this difference in comprehension level between metaphors and literal expressions might be the way in which children's comprehension is measured. Method Data was collected from 47 third grade students from a small public school in a large western school district. A quasi-experimental, pretest-posttest design was used to measure the effects of three instructional interventions on students' metaphor and literal comprehension performance. The comprehension dependent measure was performance (number correct) on a multiple-choice test taken after subjects read each of eight short stories. Participants were interviewed at the end of the experimental session. Results Study results showed that interviewing subjects concerning the reasons for their multiple choice responses significantly increased their metaphor comprehension scores. Conclusions The results of this study show that great care must be taken when assessing children's understanding of figurative language because their background knowledge and metaphoric interpretations are different from adult researchers. Using a mixed methods approach was shown to allow a greater understanding of children's metaphoric interpretations of stories.
Ryan Bowles (Department of Psychology, Michigan State University); Lori Skibbe; Gary Troia; Laura Froyen - The Inventory of Phonological Awareness with Alternative Responses (IPAAR)
Purpose: Phonological awareness is a key predictor of literacy development, yet existing measures of phonological awareness may not be appropriate for children with speech production difficulties. Existing measures of phonological awareness are generally designed for typically developing children, rarely include children with special needs in norming samples, and require speech production for most item responses. Method: In this presentation, we describe and demonstrate the Inventory of Phonological Awareness with Alternative Responses (IPAAR), a new measure of phonological awareness to measure the levels of phonological awareness displayed by children ages 3 to 7 who demonstrate speech production difficulties, but which is also appropriate for children who exhibit typical development; this measure is also useful for older children who have lower levels of phonological awareness. The IPAAR requires no spoken responses by utilizing a multiple-choice format and is completely computerized so that it is accessible through a variety of interfaces including augmented communication devices using touchscreen or eye gaze. Results: We report results from a feasibility study which suggest that children find the task engaging and respond appropriately to the items. We also report results from an ongoing validation study with children displaying typical development that considers relations with an existing gold standard measure of phonological awareness. Finally, we describe further development efforts aimed at making the IPAAR more effective for this population. Conclusions: Assessing children with speech production difficulties entails special challenges with accessibility, fatigue, and score interpretation. The IPAAR addresses these challenges through an accessible interface involving no speech production.
Lee Branum-Martin (Department of Psychology); Paras D. Mehta; David J. Francis - Spanish and English literacy in immersion and maintenance classrooms
* Purpose While literacy is recognized as complex, empirical models of literacy as a system are rare and usually ignore the potentially complex role of classroom differences. This research examines the degree to which reading skills indicate consistent latent traits in Spanish and in English for children and classrooms. * Method Academically acceptable schools (32) were selected to represent English immersion and Spanish maintenance classroom programs involving 1,704 Spanish speaking children in 334 teachers' classrooms in grades 1-2. The outcome measures were the letter-word identification, word attack, and passage comprehension subtests in Spanish and English of the Woodcock LPB-R. Multilevel confirmatory factor models were fit to the two programs as multiple groups to examine the validity of the models and program differences. * Results The models with tests of invariance suggest these three tasks indicate a single factor of literacy in each language across programs. On average, immersion programs performed higher in English and lower in Spanish than did maintenance programs. Cross-language latent correlations were positive at the student level, but negative at the classroom level. * Conclusions Among six different outcomes, consistent measurement of language-specific literacy enables us to better frame questions of educational programs while accounting for measurement error and classroom differences. The findings may suggest that classrooms specialize while children generalize their literacy expertise, even within program. Explicitly modeling literacy as a system of skills appears to shed new light on literacy as a goal of both education and research into bilingual learning.
David Braze (Haskins Laboratories)David Braze; Anuenue Kukona; Whitney Tabor; James Magnuson; Einar Mencl; Sergey Kornilov; Julie Van Dyke; Clinton Johns ; Donald Shankweiler - Reader skill differences in apprehending the meaning of simple spoken phrases
PURPOSE: To identify reading-skill contingent differences in comprehension of a ubiquitous grammatical structure (nouns with adjectival modifiers), we used the visual-world eye-tracking paradigm (VWP) to study readers' online interpretation of simple spoken sentences. We test the hypothesis that reading skill and word knowledge are associated with the moment-by-moment comprehension of speech. METHODS: In this VWP, participants viewed 40 quartets of images (e.g. red-apple, yellow-bowl, blue-chair, brown-book) while responding to verbal instructions (e.g., "Point to the red apple."). Eye-tracking recorded proportions of looks to target and non-target images, providing an index of online speech comprehension on a millisecond time-scale. Young adults (current N=49; planned N=60) were recruited from adult education centers and community colleges, and assessed for vocabulary, verbal-memory, visual-memory, listening and reading comprehension, PA, decoding, print experience. Growth-curve analysis modeled proportion of looks to targets as a function of time. Subjectwise random coefficients were extracted and multiple regression identified associations between variation in VWP performance and reading-related skills. RESULTS: Reading comprehension (Gates-MacGinitie) and vocabulary (PPVT) are reliably associated with speed of convergence to target images in VWP. For both, higher scores go to faster convergence. Differences in speed of convergence emerge within 500-600ms of onset of the critical phrase. CONCLUSIONS: We demonstrate reading skill contingent variation in the ability of adult listeners to integrate meaning across words as speech unfolds in real-time, even in the simple case of adjective--noun composition. Comprehension difficulties in speech are likely compounded in print, especially for poorer readers who still struggle with basic decoding.
Zvia Breznitz (University of Haifa) - Brain system synchronization of dyslexic readers as a result of fluency training: Evidence from imaging studies
Purpose: Effective reading requires synchronization between the brain systems that are activated in reading. Brain systems asynchrony in speed and content characterizes dyslexic readers and impairs their decoding and fluency skills. Based on the plasticity of the brain to change, the aim of the present study was to investigate whether training reading fluency can convey brain system synchronization and enhance reading effectiveness. Method: 55 dyslexic readers (at least -1 s.d and below on reading score) and 55 skilled readers, all adult university students (X age 24.15), participated in the study. Subjects were divided into 4 groups according to training program. The experimental group underwent reading training with acceleration (RAP), and the control groups underwent training of reading without acceleration (NAP), non-alphabetic cognitive skills (CPC), and no training. All subjects received 24 training sessions per program (3 times per week for 20 minutes each), and the subjects that initially undergo RAP and CPC switched training programs for an additional 24 sessions(total of 48 sessions) . Subjects were tested with behavioral and brain (ERP and fMRI) measures prior to, post-, and long-term (6 months) post-training. Results: Results on behavioral and brain imaging measures indicated the beneficial effect of the RAP compared to NAP, only CPC, and no training,. The most effective gains were found among subjects that undergo first CPC and then RAP, most prominently among dyslexics. Conclusions: An innovative theory of training the brain of dyslexic readers will be presented.
Mindy Bridges (University of Kansas); Hugh Catts; Diane Nielsen - The use of a dynamic screening of phonological awareness to predict reading outcomes
Purpose: This research will investigate the ability of a dynamic screening of phonological awareness to predict end of kindergarten and first grade reading ability. Method: Two cohorts of students (N=354) were administered the Dynamic Screening of Phonological Awareness (DSPA) at the beginning of kindergarten along with additional measures of early literacy and language. The outcome measures were the Word Attack and Word Identification subtests from the Woodcock Reading Mastery Test-Revised/NU administered in May of kindergarten and first grade. Results: Logistic regression was used to examine the relationship between the DSPA and other predictor variables and the reading outcomes at the end of kindergarten and first grade. Initial results from end of kindergarten indicated that the DSPA contributed to predictability over and above other measures of phonological awareness. Additional analyses will examine the ability of other language and literacy measures in isolation and in conjunction with the DSPA to predict end of first grade reading ability (second cohort will complete first grade assessment in May, 2011). Conclusions: Results will add to growing research indicating that a dynamic component can add to the predictive validity of static screening measures of reading-related skills.
Danielle Brimo (Florida State University);Kenn Apel; Yaacov Petscher - Syntactic awareness: Constructing meaningful tasks.
Purpose: The purpose of this study was to develop two reliable syntactic awareness tasks that measure adolescent students' ability to judge grammaticality in complex sentences and rearrange words to create complex sentences. Specifically, there were two main goals: a) to determine the dimensionality (i.e., factor structure) of two researcher-made syntactic awareness tasks and b) to investigate the difficulty and discrimination of individual items within the two measures by using an Item Response Theory (IRT) analysis. Method: The syntactic awareness abilities of 207, 9th and 10th grade students' were assessed using the researcher-made syntactic judgment and word-order correction tasks. Results: Exploratory and confirmatory factor analyses revealed that data from the syntactic judgment task fit a bi-factor model with a general factor (i.e., syntactic awareness) and four group factors (i.e., sentence type). Additionally, data from the word-order correction task fit a uni-dimensional model representing one factor (i.e., syntactic awareness). IRT analysis was used to select items on each task that displayed good discrimination and difficulty estimates for the selected population. Based on Item Characteristic Curves (ICC) and Test Information Function (TIF) representations of individual items, the syntactic judgment task was reduced to 24 items and the word-order correct task was reduced to 12 items. Conclusions: The present study addressed concerns related to measurement of syntactic awareness by developing reliable syntactic awareness measures developed for high school students. In the future, these tasks may be used to investigate the relation between syntactic awareness and reading comprehension among adolescent students.
Stephen Burgess (Southwestern Oklahoma State University) - Stability of home literacy environment activities for preschoolers
Purpose The preschool home literacy environment (HLE) is related to a variety of cognitive and educational outcomes (e.g., Phillips & Lonigan, 2005). However, very little research has examined the factors related to HLE provision and stability. We conducted a short term longitudinal examination of several elements of the preschool HLE and factors associated with this provision. Method 55 mothers of preschool children completed a survey designed to assess the HLE activities their child was exposed to on the previous day (e.g., read to, alphabet game play, whether parent read a book themselves) as well as why these activities may have not occurred (e.g., not enough time, child showed no interest). Each mother completed the survey either 2 or three time a week for between 10 and 15 weeks. Results HLE behaviors and opportunities varied widely across families. Almost all of the children were read to at least once during the interval examined, but approximately 30% were not read to at least half of the weeks examined. Not having enough time or having more important things to do were the most common reasons reported for not reading to the child. Mother's interest in reading was significantly related to the stability of reading opportunities for the child. Additional analyses examining a variety of reading and letter teaching opportunities will be presented. Conclusions This project is one of the first to examine the stability of preschool HLE provision and parental reasons for this provision.
Victoria Burke (Georgia State University); Amy Lederberg; Christopher Stanzione - The development of word identification strategies of preschool children who are deaf and hard of hearing
Purpose: Evidence for phonological recoding as an early-developing strategy is well-established for hearing children, yet much less is known regarding the extent to which young children who are deaf and hard of hearing (DHH) use phonologically-based strategies to read words. This study was to investigate whether theories of early word identification developed for hearing children are applicable to DHH preschool children by examining changes in their reading strategies across the school year in the context of an intervention that provided explicit instruction in alphabetic knowledge, phonemic awareness, and phonological recoding. Method: Participants were 15 prekindergarteners (mean age = 54 mos.) who were part of a 4-year research project to develop an early literacy program for DHH children. All children used either cochlear implants or hearing aids and were able to identify spoken words. Children were instructed in pull-out groups of 1-3 children, 1 hour/day, 4/days per week. Archival video of word reading activities (approximately 90 hours per child) was coded for reading context (isolated or connected-text), strategy employed and outcome. Results: Preliminary analyses indicate the children used multiple strategies across the year including sounding out some phonemes, sounding out and blending all phonemes, and retrieving from memory. Strategy choice was adaptive to task demands and gradual change was observed in increased use of sounding and blending all phonemes and retrieval. Conclusion: These results suggest young children who are DHH develop and use strategies similar to hearing children and can benefit from explicit reading instruction. Educational implications will be discussed.
Denis Burnham (MARCS Auditory Laboratories, University of Western Sydney)Sudaporn Luksaneeyanawin - Spelling in Thai children and adults: The effect of phonic and whole word instruction on errors on consonants, vowels and tones
Purpose: The purpose of this study was to investigate the hypothesis that phonics reading instruction results in better spelling than whole word instruction , and that this superiority occurs across segment types. Method: 128 children, 32 (16 phonics, 16 whole word) at each of 6, 8, 10, and 12 years, along with 25 adults were asked to write down as many words as possible in a 3-minute period in each of 5 categories. Errors due to Reversals, Additions, Substitutions, and Omissions were recorded. From the last two a measure of the 'Dispensibility' was derived to index the extent to which, if a grapheme was missing, it was more likely to be ommited than subtituted. Results: Analyses showed fewer spelling errors by phonics taught than whole word taught children, and this was most apparent at 6 years. Most errors were made for consonants then vowels then tones with the vowels vs. tones difference ameliorated in adulthood. This pattern was reversed for dispensibility: tones were more dispensable than vowels and they in turn more dispensible than consonants. The difference between vowels and tones was ameliorated in adults, although both were still more dispensible than consonants. Conclusion: The results suggest consonants act as spelling anchors - while they account for more errors than vowels and tones, the number of consonant errors decreases rapidly over age, and consonants are considered necessary, they are much more often sustituted than omitted. On the other hand both vowels and tones are more dispensible, and this is so even for adults.
Adriana G. Bus (Leiden University) ;Marian J. A. J. Verhallen; Maria de Jong - Young second language learners' visual attention to illustrations in storybooks
Purpose. The major purpose of this study was to test how preliterate second language learners use illustrations in storybooks to understand a story. Method. Subjects were 23 5-year old low-SES children, learning Dutch as a second language. Each child was exposed four times to a digital picture storybook. Five books were used and counterbalanced over children and repetitions. During book exposure, eye movements were registered using a remote eye-tracking system. To test whether eye fixations match the text we scored how often and how long children fixated visual elements in the illustrations that the story text highlights. We also scored how often human and non-human objects were fixated. Results. In line with the hypothesis that eye fixations are time locked to referential expressions in the text, visual elements that the text highlights were fixated more often and longer than elements in illustrations not highlighted in the story text. We also found evidence that inspection of illustrations at times follows an autonomous pattern. Conclusions. We discuss how both types of visual behavior can add to vocabulary development and story comprehension.
Kate Cain (Lancaster University)Hannah Nash & Nicola Pooley - Influence of interclausal connectives on young readers' text representations
Purpose: Connectives are cohesive devices, such as 'after', 'because', and 'but', that signal the conceptual links between clauses. Adults' construction of coherence relations between sentences is influenced by the presence of conjunctions: recall for sentence pairs is better when joined by a connective. We investigated how the presence of temporal, causal, and adversative connectives affects memory and the representation of sentence pairs in relation to age and reading comprehension ability. Method: Experiment One: children aged 8, 10, and 12 years, and college students studied two-clause sentences joined by an appropriate connective or the neutral connective 'and', e.g., "Megan packed her suitcase because / and she was going on holiday." Recall was measured after a short delay. Experiment Two: good and poor comprehenders aged 10 years completed the same task. Results: The presence of a temporal or adversative connective resulted in a stronger association between the two clauses indicated by superior recall. Of note, 10-year-olds benefitted more from the presence of these connectives than 8-year-olds. Poor comprehenders did not benefit from specific connectives to the same extent as the good comprehenders. Surprisingly, the weakest effects were found for causal connectives. Conclusions: Connectives are an important influence on the construction of coherence relations, but their presence does not benefit all young readers to the same degree. The weaker effects for causal connectives may arise because readers expect to find causality in text and readily infer such relations.
Robert Calfee (Stanford University);Elfrieda H. Hiebert; - Using Cohort Analyses to Examine Long-Term Effects of Reading Initiatives in California
Purpose: The impact of federal and state policies of the past two decades on reading achievement has been described in various reports, but require more careful attention. This presentation reports on an analysis of cohort trends in California's Standardized Testing and Reporting (STAR) program from 1998 to 2009, using grade 2-8 data in reading and (for comparison) mathematics. Method: The 1998-2009 tests included a variety of assessments, but trend lines allow evaluation of the impact of state (e.g., AB 466 requiring professional training on specific reading content) and national (e.g., Reading First) initiatives. Cohort trend analysis allows the separation of cohort differences from reform impact. Results: As an example of patterns from the analysis, reading performances increased substantially in Grades 2 and 3 from 1998 to 2002 and reached national norms by 2002. No changes were observed in later grades where achievement remained substantially below the national average. Patterns on the STAR were also compared to those of California's fourth and eighth graders on the National Assessment of Educational Progress (NAEP). From 2002 to 2009, percentages at proficient or above on the NAEP increased slightly (Grade 4: 21-23%; Grade 8: 20-22%) with the vast majority performing at basic or below. Conclusions: The size and centralized mandates of California mean that its policies influence textbooks and teachers' guides in other states as well. Analyses of cohort patterns show that mandates on particular reading programs and training improved primary-level performances on phonics tests, but with no transfer to the later grades.
Mary Beth Calhoon (Lehigh University)Mary Beth Calhoon; Julie J. Masterson - Lexical analysis of words on commonly used standardized spelling assessments
Several linguistic skills support spelling and word-level reading including phonemic awareness, orthographic knowledge and the associated ability to establish sufficient mental graphemic representation, semantic awareness, and morphological awareness. Studies of older students indicate a stronger reliance on morphological awareness for spelling. Morphological awareness requires the ability to consider the morphemic structure of words and how spellings result from adding a morpheme(s) or judgments about the relationships between base words and their derived forms. Little research has been conducted on the impact of morphology on spelling ability and even less has been done on the specific morphological requirements of standardized spelling assessments. Words characterized by greater morphological complexity and lower frequency are more challenging to spell than simple base words that occur frequently in academic literature. Knowledge of the morphological and frequency demands inherent in standardized spelling assessments is important for clinicians and researchers. The purpose of this study was to examine the morphological characteristics (i.e., number of morphemes in each word, degree of transparency between a derived morpheme and its root word) and frequency data (Standard Frequency Index (SFI)) of six standardized spelling assessments and alternate forms. Results indicate high variability among the assessments and alternate forms on the use of multimorphemic transparent and opaque words. Variability based on the SFI index between tests and alternate forms was relatively low. Findings are discussed in regard to providing a deeper understanding diagnostically of the morphological and frequency requirements and complexities of these standardized spelling assessments.
Fan Cao (University of Pittsburgh); Marianne Vu; Derek Chan; Jason Lawrence; Lindsay Harris; Qun Guan; Yi Xu; Charles Perfetti - Neural correlates of accommodation in second language learning
Purpose Previous studies found that English speakers showed accommodation in the brain after learning Chinese as a second language. The current study was to determine which learning method produces the most accommodation. Methods 17 native English speaking college students who were taking Chinese level 1 participated in the current study. 30 characters were learned in a character writing condition for 6 days in which participants wrote the character 3 times and other 30 characters were learned in a pinyin writing condition. After training, brain activations were examined in a passive viewing task, a lexical decision task and an implicit writing task. Results We found that in the passive viewing task, there was greater activation in the Chinese network (bilateral MOG, left MFG) for Chinese than for English. In the lexical decision task, we found that there was greater activation in visuo-spatial and sensory-motor regions (SPL, postcentral gyrus) for the character writing condition than the pinyin writing condition. In the implicit writin task, we found that there was greater activation in the Chinese network (right FG, left MFG, MTG) for the character writing than the pinyin writing. Conclusions English speakers can engage additional brain networks to adapt to the new features of Chinese after a short term of training. Character writing training produces a more similar brain network to native speakers than pinyin training, presumably because it improves the quality of orthographic representation, which is crucial for Chinese due to the complex visual configure and lack of grapheme-phoneme-correspondence.
Lauren Capotosto (Harvard Graduate School of Education); Soojin Oh; James S. Kim - The role of student motivation and the home literacy environment in distinguishing between proficient and struggling third grade readers: A mixed-methods study
Purpose: Several studies have explained differences between good and poor readers by factors related to socioeconomic status, such as distinctions in the home literacy environment (e.g., Hart & Risley, 1995). However, exploring differences between readers of varying abilities within the same socioeconomic class can shed further light on factors that distinguish good from poor readers. Examining variation in reading ability within a low-income population, the current study explored differences in child and family characteristics between students who passed and failed a statewide reading comprehension test in Grade 3. Method: 328 students attending five high-poverty urban schools responded to closed-ended survey questions regarding their motivation to read. Structured interviews with 16 of these students, their parents, and their teacher were conducted at two time points to gain further insight into students' reading habits and home literacy environments. Results: Survey results showed that students who failed the state test were significantly more likely than students who passed to read in order to please their parents and teachers. In contrast, interview data revealed that students who passed were more likely than students who failed to be intrinsically motivated, perceive reading as recreational, and engage in self-directed reading outside of school. Furthermore, students who passed were more likely to report that their parents talked with them about books and asked them comprehension questions after reading a book. Conclusion: Findings suggest that a combination of child intrinsic motivation and parenting support operate as critical mechanisms for facilitating low-income children's performance on reading achievement tests.
Marketa Caravolas (Bangor University)Betty Mousikou, Miroslava Schoffelová, Corina Effrim - The relationship between young readers' abilities to read and define words: Is it modulated by orthographic complexity?
Purpose - We tested the hypothesis that learners' reliance on lexical-semantic knowledge in word reading increases as orthographic inconsistency increases, while a reliance on phonological decoding strategies increases with greater orthographic consistency. In the only (to our knowledge) direct investigation of this issue, comparing Welsh and English 2nd graders' abilities to read aloud and define words, Ellis & Hooper (2001) found that: (1) readers of the consistent Welsh orthography read more accurately than English readers; (2) yet, English readers had better semantic knowledge of the words they attempted to read, while Welsh children read many words without knowing their meanings; and (3) Welsh errors indicated phonological recoding strategies (e.g., nonwords) while English errors suggested lexical strategies (e.g., misreading words as other real words). Following Ellis and Hooper's approach, we compared groups of learners of three orthographically differing languages (English - highly inconsistent, French - moderately inconsistent, Slovak - highly consistent), on matched lists of meaningful words. Method - Participants. Monolingual groups of end-1st-grade English (n=169), French (n=143) and Slovak (n=176) children were asked to read aloud and define 25 cognate word lists, also matched for grade level, frequency, length, syllable structure, and imageability. Results - Mixed Anovas by participants and by items revealed: (1) The expected improvements in accuracy as a function of consistency (SK > FR>ENG); (2) However, we found no evidence of greater semantic involvement in word reading in English. In fact, English children virtually never accessed a word's meaning through an incorrect reading (0.6%), whereas the readers of French and Slovak did so approximately 10% of the time; (3) Error analyses showed that English children's most frequent misreadings were other words, or refusals, while French and Slovak children mainly produced dysfluent or nonword misreadings. Conclusions - We suggest that while orthographic consistency affects the efficiency of phonological recoding, and gives rise to different decoding strategies, in the first year of reading tuition, at least, knowledge of the word forms and their meanings generally goes hand in hand; although readers of more (Slovak and French), not less consistent (English) orthographies, can sometimes penetrate inaccurate phonology to access word meanings.
Claudia Cardoso-Martins (Universidade Federal de Minas Gerais)Natalia Rakhlin; Elena Grigorenko - Phonemic awareness, rapid serial naming, and orthographic coding: evidence from an asymmetric orthography
We investigated the contribution of phonological awareness (PA) and rapid serial naming (RSN) to phonological and orthographic coding skills in Russian. Russian orthography is fairly consistent in its grapheme-to-phoneme correspondences, but quite inconsistent in its phoneme-to-grapheme correspondences. This asymmetry renders Russian well-suited for studying cognitive correlates of orthographic coding because here difficulties with phonological processing are less likely to confound measures of orthographic coding compared to orthographies symmetrically inconsistent in both directions. Indeed, difficulties exhibited by children learning to read Russian - typically, slow reading and incorrect (but phonologically plausible) spellings - appear to result from a difficulty in building an orthographic lexicon due to inconsistent phoneme-grapheme correspondences (Kerek & Nieme, 2009). In addition to tests of PA and RSN, 150 Russian children and adolescents (M age = 13.67; SD = 1.59) completed tests of word and pseudo-word reading fluency, pseudo-word reading accuracy, and spelling accuracy. PA was a better predictor of word spelling accuracy than RSN, as for pseudo-word reading accuracy. RSN contributed more variation to word and pseudo-word reading fluency than PA, and even controlling for spelling ability did not wash out its contribution to word-reading fluency. In contrast, it no longer contributed to variation in word reading fluency after we controlled for pseudo-word reading fluency. These findings question the hypothesis that RSN indexes orthographic coding skills, in line with Moll et al.'s (2010) findings for German, and suggest that RSN taps into timing, automaticity/efficiency of processing, and ability affecting fluent reading independently from phonological processing, indexed by PA.
Andrea Carlile (Florida State University)Carol Connor - Teacher dispositions, teacher self-efficacy, and second graders' literacy outcomes
Purpose: Teacher dispositions, a key construct promoted by the National Council for Accreditation of Teacher Education (NCATE), represents characteristics and aptitudes that are supposed to be associated with better teaching (e.g. patience). Teacher self-efficacy is one's belief that he or she can orchestrate the actions necessary to perform a teaching task. The purpose of this study was (1) to examine whether teacher dispositions and self-efficacy represented two distinct constructs and (2) to investigate whether or not each affected student achievement. Method: In this study, researchers working with second-grade teachers (n=33 teachers, 5 schools, 591 students) in a training study assessed teachers' dispositions using the Professional Commitments and Dispositions Rating scale (PCDR) developed at FSU. Each teacher completed a Teacher Self-Efficacy Scale (TSES) regarding their own self-efficacy. Student reading was measured using the Gates-MacGinitie Reading Test. Results: Factor analysis revealed two factor variables for each scale. Factors and total scores were not correlated. Using HLM, we found that the factor Management/Promoting-Student-Engagement (from the TSES), negatively predicted student reading gains (residualized change). The factor, Accommodating-Individual-Student-Differences, (from the TSES) positively predicted reading gains but effects were greater when students' fall reading scores were weaker. Professional-Attributes, (PCDR) positively predicted gains and were greater when students' fall reading scores were weaker. Conclusions: Results of this study indicate teacher dispositions and teacher self-efficacy are separate constructs, appear to affect student achievement, and should be considered when future research examines the role of the teacher in the classroom.
Joanne Carlisle (University of Michigan) - Analysis of measures of morphological awareness
Rationale and Purpose: With the increased interest in students' developing morphological awareness has come the need for assessments that can function well for both research and instructional decision-making purposes. Analysis of existing measures offers researchers and educators guidelines for development of new measures or selection of existing measures. Method: I first consider models of the development of morphological awareness, and analyses of the construct of "awareness" (e.g., Cazden; Van Kleeck) provide a basis for analysis of measures of this area of language and literacy. Then I analyze the task, content, and psychometric properties of a number of such measures previously used in research studies. Results: The analysis leads into consideration of the question: what do we learn about students' morphological awareness from different measures? Results suggest the need for closer attention to both "morphology" and to assessment of "awareness." Conclusions: From this preliminary study, I derive guidelines researchers and educators might use for further development of assessments of morphological awareness.
Julia Carroll (University of Warwick); Laura Shapiro; Jonathan Solity - What skills are impaired in dyslexia at the onset of reading? A prospective study of children with reading difficulties.
Purpose: Cognitive deficits have been shown in dyslexic children in many areas, including phonological, short-term memory, motor skills, balance, auditory awareness and speed of processing. This study examines whether children who later become dyslexic show these difficulties prior to the onset of literacy tuition. Method: 444 children representing the full range of abilities in their first school term completed a large battery of tasks measuring phonological awareness, verbal short-term memory, rapid naming, speech rate, auditory perception, motor skills, print knowledge, balance, IQ and speed of processing. Reading skills were retested at the end of year 1, year 2 and year 3. Poor readers were designated as children who showed reading scores one standard deviation below the mean on at least two of the three test points. Results: Poor readers showed significantly lower scores on each of the Time 1 measures, except balance, and were significantly more likely to have deficits in each area. Only a small minority showed no cognitive deficits. However, no single deficit was a necessary precursor to reading difficulties. Conclusions: Results are in line with theories suggesting that dyslexia has multiple causes.
Karyn Carson (Flinders University); Gail Gillon; Therese Boustead - Teacher-implemented phonological awareness instruction and literacy outcomes in the first year of school
Purpose: Phonological awareness (PA) ability plays a critical role in supporting children's acquisition of skilled reading and spelling. PA intervention studies show significant benefits for children struggling with literacy development; however these studies are often implemented by researchers in one-to-one or small group settings outside the classroom environment. Less is known about the benefits of explicit PA instruction when implemented by teachers within the classroom. The aim of this study was to investigate the impact of teacher-implemented PA instruction on the literacy outcomes of children in their first year of school. Method: A quasi-experimental design was employed to measure the PA, reading and spelling development of 120 5-year-old children from twelve classrooms. One classroom (A) received 10-weeks of explicit PA instruction from their teacher in their second term of school. A second classroom (B) received the same 10-week block of PA instruction in their third term of school. The remaining ten classrooms continued with their usual reading program which did not explicitly target PA. Results: Analysis using ANOVA and ANCOVA methods indicated that classrooms A and B showed a) significant improvements in their reading and writing performance immediately following PA instruction, b) significantly higher scores following one year of schooling in comparison to children who received the usual curriculum and c) significant gains in literacy performance for children at-risk for reading difficulties. Conclusion: Integration of explicit PA instruction into classroom curriculums can help raise the literacy profiles of both at-risk and typically developing readers throughout the first year of education.
Romina Cartoceti (Universidad de Buenos Aires - CONICET- Argentina)Valeria Abusamra; Bárbara Sampedro; Aldo Ferreres. - The contribution of executive functions and suppression to reading comprehension.
Purpose In this study we examined the contribution of executive functions (inhibitory control, suppression of irrelevant information) to reading comprehension in 86 children aged between 10 and 13 years old. The aim of this research was to compare the subjects' performances in tests that require inhibitory control and executive functions to those that evaluate reading comprehension abilities. Method The children participating in this study (N: 716) came from public and private schools in the City of Buenos Aires and Greater Buenos Aires, Argentina. They were initially screened using a test that measures reading efficacy (Tecle, Marín and Carrillo, 1999) and two texts (one informative, one narrative) that evaluate reading comprehension through multiple choice questions (Abusamra et. al., 2010). A sub-group (N: 86) was selected and divided into good and bad comprehenders according to their performance on previous tests. These two sub-groups were then evaluated with a child version of the Hayling Test (Cartoceti, Sampedro, Abusamra and Ferreres, 2009), which assesses verbal inhibition, and with the Stroop Test (Golden, 1991) to measure cognitive flexibility, automatic response and attentional control (Drake and Torralba, 2007). The students were evaluated in their schools. Results Statistical analysis (Mann-Whitney U) showed that bad comprehenders performed significantly worse than good comprehenders in texts comprehension tasks (p<.001). These statistical differences were also significant in the tests that evaluate suppression of irrelevant information (inhibitory control) and executive function tasks (p<.001). Conclusion Group differences were found in the ability to suppress information. The bad comprehenders group showed significant impairment to this ability. The inhibition mechanism influences reading comprehension, mainly because it allows the intrusion of irrelevant stimuli and information to be suppressed, thus giving flexibility to working memory resources. The results will contribute to the field of text comprehension field and to the design of evaluation strategies and interventions in the educational and clinical fields.
Ya-Lan Chang (Ya-Lan Chang)Jie-Li Tsai - Adult's Interactive Strategies Influence Children's Comprehension Process during Shared Book Reading: Evidence from Eye Movement
Purpose - Educators have attached importance to shared book reading (SBR) for young children's literacy development. The benefit of SBR has been emphasized on children's decoding ability (e.g. phonological awareness), but seldom on the comprehension ability. The present study investigated the effects of adult's interactive strategies and comprehension level of interactive content on children's comprehension process during SBR, which was reflected by the eye movement data. Method - Experiment 1 compared reading behaviors of children at different ages (8 preschoolers; 11 first-graders) under two different strategies (question vs. comment). Experiment 2 explored the effects of comprehension level of adult's interactive content (literal vs. inferential) and interactive strategy (question vs. comment) on first-graders. The proportions of fixation duration (FD-rate) in text and target picture from eye movement data were analyzed. Results - Results of experiment 1 revealed that the question strategy draw preschoolers' attention to target picture area (critical for story comprehension), while the comment strategy guided it to word area. Experiment 2 replicated the benefits of strategies in experiment 1. Moreover, children paid more attention to target picture area in the inferential than literal condition and the data on target word showed the opposite pattern. Conclusions - Our results confirm that reading to children with explicit references have more advantages to children's literacy development than simply reading to them. Furthermore, different interactive strategies and contents attract children's attention to distinct areas of shared book and thus enhance diverse aspects of the comprehension process.
Ju-Ling Chen (National Taiwan Normal University); Yi-Fen Su; Ying-Yao Chiang; Li-Chen Lin; Hsin-Ying Tung; Chi-Ching Chuang - Effects of Elaboration Instructions on Strategic Reading and Comprehension in Primary Classroom: Evidences from Second Graders' Think Aloud Data
This study investigated elaboration instruction on students' reading comprehension and strategy use. Participants were 2th-grade students with low reading ability, average 7.8 year-old. A randomized treatment-control group design (18 in experimental group, 19 in control group) was adopted to investigate the effect of intervention. Elaboration training (defined as prior knowledge activation, orthogonal elaboration, and inferences) was provided for 10 weeks with direct teaching and reciprocal teaching. Students were asked to complete word recognition test and standard comprehension tests at pre- and post-test. A think aloud method was adopted to investigate strategies performance. MANOVAs were adopted for data analysis. The result of think aloud data shows a main effect for group, F(3,35)=12.219, Λ＝.474, p < .001, η2＝.526. The experimental group included significantly more elaboration reports (M=8.17, SD=5.35) than control group (M=-.05, SD=4.47).The mean frequency of prior knowledge activation in experimental group (M=2.83, SD=.63) is higher than control group (M= -.11, SD=.611). The mean frequency of orthogonal elaboration in experimental group (M=2.61, SD=2.70) is higher than the control group (M=-.32, SD=2.14). Also, the mean frequency of inferences in experimental group (M=2.94, SD=4.15) is higher than control group (M=.215, SD=3.05). The result of reading comprehension shows a main effect for groups, F(1, 34)=5.569, p< .05, η2＝.141. The reading comprehension score of experimental group (M=16.310a, SD=.386) was higher than control group (M=15.022a, SD=15.022). The results shows that children get improve after strategy instructions. It provides evidence of elaboration instruction is effective as early as the 2nd grade.
shuli chen (); Shih-Jay Tzeng+active member; Yi-Chieh Wu+student member; Hsin-Ying Chen - The effectivness of turnaround reading intervention program in Taiwanese disadvantaged district--An examination of the effects of implement duration
Purpose The low achievement of disadvantaged students in Taiwan is becoming a serious problem. The purpose of this study is to establish a tiered literacy instruction program, to examine the effectiveness of supplemental intervention, and to compare the effects of implementation duration for schools in a disadvantaged district. Method Participants are 856 students in first to sixth grades from 8 schools. Two schools (n=232) implemented a Turnaround Reading Intervention Program in the first year, and another 3 schools (n=624) joined in the second year. The remaining 3 schools served as a comparison group. The Turnaround Intervention Program provided school-wide training to improve the quality of literacy instruction at Tier 1 and two additional full-time teachers to deliver explicit and intensive remedial intervention at Tier 2. All of the students received pretests and posttests on word recognition and reading comprehension each semester. Results of the pretest and suggestions from classroom teachers were used in determining placement into Tier 2. Analysis of covariance (ANCOVA) was used to analyze the effectiveness of intervention and the effects of duration. Results 1. Both treatment groups showed significantly better reading improvement than the comparison group. 2. Students benefited most on word reading and reading comprehension before third grade. 3. Results were significantly better with two years of treatment compared to one year. Conclusions The program of tiered literacy instruction had a significant impact on the literacy improvement of disadvantaged students. The earlier and longer the intervention was, the more efficient it was in reducing the gap between typical and low-performing students.
Shiu-Hsung Huang Chen () ;Chun-Lien Chen;Yea-Mei Leou；Ya-Ying Tseng - The Effect of Electronic Textbook Integrating Reading Strategies on Students' Language Outcome and Motivation
Purpose-This study aimed to develop a Chinese e-textbook unit integrating reading strategies on line and explore the extent to which e-textbook increase students' motivation, reading comprehension and discussing questions with peers. Method-The pilot electronic textbook program lasted for six 40-minutes sessions with 64 fifth graders . 31 participants use e-textbook integrating reading comprehension and the others use traditional, print copy of the same lesson. Two standardized test , Language Achievement Test and Motivations for Reading Questionnaire(MRQ), were applied before and after the instruction. At posttest, the teacher-made test including vocabulary test, fact recall, and reading comprehension were administrated to all students to examine the learning outcome of the 2 groups. Results-In this quasi-experiment study, measure of the unit mastery indicated that students' responses to vocabulary test and alternative comprehension questions showed no significant differences. However, students who use the e-textbook recalled significantly more aspects of the lesson and plot than those who read the print textbook and have better reading motivation and are eager to discuss the lesson related questions with peers on line. Conclusions- The e-textbook with comprehension strategies helps students understand the lesson fully might explain why students using the e-textbook had superior response to fact recall. Suggestion are provided for the design of future e-textbook and teaching with e-textbook.
Xi (Becky) Chen (OISE/University of Toronto);Adrian Pasquarella; Helene Deacon; Karen Au-Yeung; - Exploring cognate awareness in first grade French Immersion children
Purpose: This study examined cognate awareness in Grade 1 children in a French-Immersion school. Cognates are words in different languages that are of a common historical origin and are similar in spelling and meaning. Previous research has shown that cognate awareness facilitates vocabulary development in bilingual children (e.g. Nagy, Garcia, Durgunoglu & Hancin-Bhatt, 1990). However, all previous studies have involved children who were Spanish-English bilinguals and who were in Grade 4 and above. Our study extended this research to first graders in French immersion. Method: Seventy-four Grade 1 children completed a cognate awareness measure requesting them to judge whether a French word presented in both oral and written form had an English cognate (not presented). There were three types of items, French words that had the same spelling as English cognates (e.g. dragon), French words that had different spellings from English cognates (e.g. parc), and "false friends", which refer to words that have the same spelling across English and French but are not cognates (e.g. cent). Measures of vocabulary, reading comprehension, word reading, phonological awareness, and morphological awareness were also administered in French. Results: We found that children were aware of cognates as early as Grade 1, although they scored lower on cognates with different spellings than the ones with the same spelling. Results also showed that cognate awareness was significantly related to French vocabulary and French reading comprehension above the contributions of other reading related variables. Conclusions: The challenges children face in identifying cognates, and the role of cognate awareness in French literacy development are discussed.
Lynette Chesson (Roehampton University); Daisy Powell; Lance Slade; Joseph Levy - The home literacy environment: a direct link from storybook reading to later literacy?
Purpose: Research suggests preschool parental teaching of reading and writing relates to early word reading skills; whereas parent/child storybook reading relates to receptive vocabulary and later reading comprehension (Hood, Conion & Andrews, 2008; Sénéchal & LeFevre, 2002). The current longitudinal research extends these findings to examine the effects of home literacy environment for younger children (mean age = 3:10) before they begin formal literacy instruction, with the aim of establishing predictive pathways to later reading acquisition. Method: Time 1: During their preschool nursery year, children completed a battery of tests assessing cognitive skills and exposure to print. Home literacy environment was measured through a parent-report questionnaire, yielding measures of frequency of storybook reading, level of child engagement in stories, and parent teaching of reading-related skills. Time 2: Children's single word reading and letter knowledge was assessed during the following academic year. Results: Several of the parent-report home literacy measures were significantly related to later literacy. Interestingly, not just parental teaching of letters and words but frequency of storybook reading and children's engagement during storybook reading were significantly related to both concurrent and later reading measures. Conclusions: Several aspects of print exposure and cognitive abilities in pre-readers were related to their developing reading skills. Building on previous research parent teaching related to later reading skills, but interestingly parent/child storybook reading was also linked to later literacy, particularly if the child was actively engaged. This suggests that the quality of parent-child interaction during storybook reading plays a part in later literacy development.
Chia-Hui Chiu (Tunghai University, Taiwan) - Using summarization technology in an EFL reading class
Most Taiwanese university students are required to read textbooks written in English to gain content-area knowledge. However, difficulties in identifying main ideas are not uncommon among the students. Recognizing the effectiveness of summary writing in promoting reading comprehension and seeing the promise of technology in facilitating language learning, the researcher used an on-line summarization technology in her reading class. The web-based program provided an electronic environment, which allowed students to write multiple drafts of summaries of reading assignments and received an immediate automated feedback of how well they had captured the main ideas of the reading passages. The study aimed at examining the effectiveness of the program in terms of how well it facilitated learners' development of reading comprehension, their uses of reading strategies and self-perception as readers. Thirty-six Taiwanese college students participated in this study for 12 weeks. They read one expository text each week and were required to submit first drafts of their summaries in the beginning of the week and refined drafts in the end. Pretests and posttests were administered before and after the training. Based on the learners' performance on a reading pretest, the researcher divided them to two groups (high- or low-level) with equal number of students. Results from the t-tests implied that the two groups of learners improved significantly, after the training, in their reading comprehension ability, uses of reading strategies and self-perception as readers. The results from ANCOVA factorial analyses implied that high-level learners benefited more from the training than the lower-level learners.
Eunsoo Cho ()Donald L. Compton; Laura M. Steacy - Comparing responses of nonresponders in Tier 1 and Tier 2
Purpose: In the three-tiered model of response to intervention (RTI) system, students receive Tier 2 if they are shown to be unresponsiveness in Tier 1 before they receive more intensive and individualized Tier 3 instruction. Yet, there is no empirical evidence whether unresponsiveness to Tier 2 is necessary for students to be eligible for Tier 3. If the patterns of response during Tier 1 and Tier 2 are not different from each other, there is a possibility for identifying nonresponders without putting students into Tier 2. The purpose of this study is to explore this possibility by comparing the growth parameters of poor readers in Tier 1 and Tier 2. Methods: A total of 221 first grade nonresponders to Tier 1 were randomly assigned to Tier 1 and Tier 2. The students' progress on word identification fluency (WIF) was monitored each week for 17 weeks. Students were then identified as having reading difficulties (RD) based on standardized reading measures at the end of second grade. Latent growth modeling with random intercept and linear slope was used to model WIF growth using Tiers, RD status, and interaction between Tiers and RD status as covariates. Results: Results indicated that only RD status significantly explained intercept and linear slope. Neither the Tier 1 or Tier 2 instruction, nor did the interaction explain students' WIF growth. Conclusion: We found no effect of putting students into Tier 2 resulting in different response patterns compare to Tier 1. This suggests a possibility of identifying Tier 3 students based on the Tier 1 responsiveness without monitoring students' responses in Tier 2.
Jeung-Ryeul Cho (Kyungnam University)Sang-Mee Kim - The contribution of morphological awareness to reading and spelling in Korean, Chinese, and English among Korean children
Purpose: This study examined the contribution of morphological awareness on reading and spelling in Korean (Hangul) L1, Chinese (Hanja), and English L2 among 101 Korean 5th graders. Korean language and orthography have relatively transparent phonological and morphological structures. In Korean, compounding morphology is rich as in Chinese, and derivational and inflectional morphology is productive as in English. Method: In this study, children were tested with the comparable Korean and English tasks of receptive vocabulary, syllable and phoneme deletion, derivational and compounding morphology, reading, and spelling. They were also tested reading and spelling of Chinese characters (Hanja). Results: Multiple hierarchical regression analyses showed that English morphological awareness explained significant variance in Korean word recognition and spelling, English reading and spelling, and Chinese reading, whereas Korean morphological awareness accounted for Korean reading and spelling only. In particular, English derivational morphology predicted unique variance in Korean reading and spelling and Chinese reading, suggesting the transfer of morphological awareness from L2 to L1, across different orthographies. Conclusion: Our findings provide evidence of transfer of L2 morphological awareness to L1 word reading. This has important implications on current theory of biliteracy acquisition.
Joanna A. Christodoulou (MIT); Stephanie N. Del Tufo; John Lymberis; Patricia Saxler; John D.E. Gabrieli - Neural correlates of reading fluency in dyslexia and typical reading
Purpose - Reading fluency, the ability to read accurately and rapidly, is often impaired for adults with dyslexia. This study identifies the neural correlates of fluent reading using functional magnetic resonance imaging (fMRI) in typical adult readers and readers with dyslexia. Method - Behavioral performance and brain activation patterns were compared between typical adult readers (n=12, 18-34 years) and adults with dyslexia (n=12, 18-34 years). The fMRI task required participants to view sentences (word-by-word) and judge the semantic plausibility. Sentence presentation rate was manipulated as the independent variable in three conditions of slow, medium, and fast, which corresponded to minimal, moderate, and challenging speeds. Results - For skilled readers, increased presentation rate corresponded to increased activation in bilateral inferior frontal gyri, left fusiform gyrus, left superior temporal region, left parietal region, and bilateral occipital regions. Higher standardized reading fluency scores were associated with greater activation in the left occipito-temporal region. No areas were significantly more active in dyslexic versus typical readers. A series of left-lateralized frontal, temporal, and parietal regions as well as bilateral cerebellar regions were more active in skilled versus struggling readers. Conclusions - Results suggest that in proficient adult readers the ability to quickly and effectively read and understand printed language relies on a network of brain regions typically associated with language, memory, and attention. Typical readers show greater activation than struggling readers across these systems, underscoring a marked signature for dysfluent reading in the brain for adults with dyslexia.
Micaela Christopher (University of Colorado, Boulder);Akira Miyake; Janice M. Keenan; Richard Olson - Predicting reading and listening abilities with executive function and speed measures: A latent variable analysis
Purpose Successful reading requires the integration of several different but correlated cognitive processes. Previous research on the predictors of reading generally focuses on a small subset of these processes. We included several measures of different cognitive processes in multivariate models to account for unique variance in word reading, reading comprehension and listening comprehension. Method Latent variable modeling tested whether working memory, response inhibition, processing speed, and naming speed uniquely predicted word reading, reading comprehension, and listening comprehension, and whether results varied across two age groups (ages eight to 10: n = 265, M = 9.15 years; ages 11 to 16: n = 217, M = 13.39 years). Confirmatory factor analyses for the reading and cognitive variables established that our measures formed correlated but distinct factors. Structural equation modeling then tested which cognitive factors independently predicted comprehension and word reading factors. Results We found that working memory and processing speed independently predicted comprehension and word reading, while response inhibition did not predict either. Naming speed only predicted word reading when alphanumeric stimuli were used. The results were the same for both age groups. In addition we confirmed previous findings that reading comprehension tasks vary in the extent to which they tap word reading abilities. Conclusion By including multiple cognitive processes in one analysis, we could parse out which were crucial for word reading and comprehension. Figuring out which cognitive processes are uniquely predictive of reading and listening abilities can help us better understand individual differences in reading and etiologies of reading disabilities.
Kevin Kien Hoa Chung (Department of Special Education and Counselling, The Hong Kong Institute of Education)Kevin K H Chung; Catherine McBride-Chang; Anita M Y Wong; Ka-Wai Hui; Ada, B Y Law - The impact of SES, home environment, executive functioning and metalinguistic skills on academic achievement in Chinese children.
The present study examined the link between socioeconomic status (SES), home environment, executive functions (EF), metalinguistic skills and academic achievement in preschool children. One hundred ninety-nine Hong Kong Chinese-speaking children at age of 4 from low- and mid-SES backgrounds were tested on the measures of inhibitory control, working memory (WM), visual-spatial relationship, parent-child verbal interaction, phonological awareness (PA), word recognition, and mathematics. Findings showed that the low-SES children had lower scores on most of the measures than their middle-SES counterparts except word recognition and inhibitory control. Regressions also revealed that PA uniquely predicted word recognition and mathematics for both SES groups after controlling for age, SES, and other variables. Furthermore, EF as a combination of inhibitory control and WM significantly accounted for unique variable in mathematics only. These findings highlight the importance of SES, executive functioning and metalinguistic skills for learning to read Chinese language.
Paul Cirino (Department of Psychology, University of Houston, Houston, TX)Lynn Fuchs; Sarah Powell; Linda Ewing-Cobbs, Marcia Barnes, Jack Fletcher - Mathematical and Cognitive Profiles in MD With and Without Comorbid RD
Purpose: Comorbidity of RD and MD is common (Barbaresi et al., 2005). Previous studies have shown differences among these groups, but these results have limitations in terms of age range and task choice. The present study evaluated a range of math and cognitive competencies in two large, independent samples in grades 2 and 3. Profiles were expected to differ according to group (favoring students with MD over MDRD) for math and cognitive measures associated with language, with clearer differences expected for older children. Method: Participants were in 2nd (N=171; MD=69; MDRD=102) or 3rd grade (N=266; MD=134; MDRD=132). Academic difficulty was defined as WRAT-R/WRAT-3 Arithmetic and Reading scores at the 25th %ile. Measures involved pre-computation, computation, and problem solving, and several cognitive variables. Profile analysis was performed in each sample; shape effects were followed by examination of the group canonical structure matrix to identify significant determinants. Results: In grade 2, math profiles were similar (all p > .05). In grade 3, there were effects for elevation, p < .0001, as well as for shape, F(7,252) = 5.82, p < .0001, Wilks' λ = .861. Two measures of pre-computation and single digit story problems separated the groups. Also, cognitive profiles differentiated the groups (favoring MD), primarily for variables emphasizing language and attention. Conclusion: Students with MD outperformed those with MDRD, with the largest differences on measures of pre-computation and problem solving, and for language and attention. Group differences were apparent on cognitive variables at both ages, but were more apparent on math variables for older children.
Donald Compton (Vanderbilt University)Jennifer K. Gilbert; Richard K. Olson - Tugging on the Mustache of the Bearded Lady: Is the Orthographic Choice Task a Measure of Orthographic Processing or Just a Cleverly Disguised Reading Measure?
Purpose It has been argued that orthographic processing explains unique variance in reading skill above and beyond phonological processing. However, one of the most commonly used orthographic processing tasks, pseudohomophone choice in which individuals choose the correct spelling between two phonologically equivalent letter strings (e.g., goat vs. gote), has been criticized as measuring word reading ability as opposed to orthographic processing. The purpose of this study was to test the hypothesis that orthographic choice performance could be completely determined by reading skill and to explore person and item variables that explain variance in orthographic choice performance. Method A crossed random-effects model, allowing person and items to vary randomly, was fit to orthographic coding data of 295 individuals in the Colorado Twin Study. Three conditional models were fit: 1) with individual's ability to accurately read each word as a predictor of item level performance, which conditionalizes the probability of choosing the correct letter sequence on being able to read the word, 2) with person level characteristics and word characteristics added, and 3) with an interaction between person level spelling ability and item level word frequency added. Results If an individual was able to read the word then the probability of choosing the correct letter string was .90, and was .80 if they were unable to read the word. Further, controlling for item-specific word recognition, person and word characteristics, including an interaction between them, were significant predictors of performance. Results suggest that the orthographic choice task relies on reading skill but performance is not completely determined by the ability to read the items on the task.
Peggy S. Conner (CUNY Graduate Center, NY); Loraine K. Obler - Novel spoken word learning in adults with developmental dyslexia
* Purpose Two hypotheses developed to explain literacy difficulties were evaluated as possible explanations for word-learning difficulties in adults with dyslexia. The phonological degraded representations hypothesis proposes that poorly specified phonological representations can largely account for the literacy difficulties experienced by people with dyslexia, and the perceptual anchoring hypothesis suggests these challenges stem from a reduced ability to implicitly benefit from repetitions of phonological information. * Method Eight individuals, four with and four without dyslexia, were taught novel spoken words (that varied orthogonally by phonotactic probability and neighborhood density) for pictures of novel objects in a story. Phonological knowledge (measured by the effects of neighborhood density and phonotactic probability) on word learning and perceptual anchoring (i.e. the effects of hearing repetitions of the spoken novel words across story episodes) served as the independent variables. Word-learning group differences were measured in a picture-naming task. Additional measures of phonological processing and perceptual anchoring were also taken. * Results The learners with dyslexia (LwD) produced fewer correct novel words than the typical learners (TL) on the word-learning task (TL x=71%; LwD x=23%) with noticeable differences across testing times. The TL recalled significantly more words of low phonotactic probability (Median test, p< .05) than the LwD, but the groups did not differ on high probability words. Perceptual anchoring as measured by a two-tone just-noticeable-difference task also showed a significant group difference (p< .05). * Conclusions Atypical phonological knowledge and perceptual anchoring ability thus both contribute to word-learning difficulty for young adults with dyslexia.
Nicole Conrad (Department of Psychology, Saint Mary's University) - Individual differences in oral and silent reading comprehension
Purpose: English language reading research is dominated by measures of oral rather than silent reading (Share, 2008). Reliance on oral measures may inaccurately portray children's comprehension abilities and contribute to an incomplete understanding of reading comprehension. Phonological representations needed for reading aloud may not correspond to the phonological representations actually used to access meaning during silent reading. Oral reading is typically slower than silent reading, suggesting that silent reading may involve less exhaustive phonological processing than oral reading, freeing up cognitive resources needed for comprehension. Empirical studies directly comparing children's oral and silent reading comprehension are inconclusive. Using a simple within-participant design in the current study, it was predicted that beginning readers would show greater comprehension following silent compared to oral reading. Method: Twenty-five Grade 3 children completed two equivalent forms of the Grey Silent Reading Test, one orally and one silently. Forms were counterbalanced across conditions, and order of testing was counterbalanced across participants. Various other reading related skills were also measured. Results: A dependent-sample t-test revealed no significance difference between silent and oral reading comprehension. However, this finding was not a result of equivalent performance on both measures for all children. The majority of children showed a clear advantage for either silent or oral reading comprehension. Further analyses indicated these differences were real. Conclusions: This disparity in individuals' oral and silent reading comprehension is intriguing. We are currently investigating factors that may account for differences in performance. A better understanding of individual differences in reading comprehension will inform best practice in assessment and contribute to developmental models of reading acquisition.
Barbara Conway ();Suzanne Carreker; Mary Lou Slania; Regina Boulware Gooden; R. Malatesha Joshi - Analysis of errors on the TWS-4: What does spelling achievement of fifth graders tell us about understanding of language?
Purpose: A standardized test of spelling ability, Test of Written Spelling - 4 (TWS-4; Larsen, Hammil, & Moats, 1999) was used to explore error patterns of fifth-grade students. The error categories (phonological, phonetic, orthographic, etymological, and morphological) were defined so as to represent progressive levels of knowledge of language necessary for correct spelling of phonemes in English words. Questions addressed in the study were: 1) Do students within one grade level but at different levels of spelling proficiency make different types of errors? 2) Do the differences in types of errors in the spelling of words reflect an increasing level of knowledge of the English language? Method - Participants were 105 fifth-grade students from public and private schools in the southwestern region of the US. The test was administered in a classroom setting by the principal investigator. The spelling errors were scored by the PI and two master reading specialists based on a scoring guide. Inter-rater reliability was established using the responses of 20 participants and was .94. Results - Using standard spelling scores and frequency of error types as variables, results indicated that phonological and phonetic errors contribute more to very low spelling achievement than do errors such as etymological and morphological errors which require higher levels of metacognitive thinking. Conclusions - The study confirmed that spelling is developmental. When students struggle with spelling, it is prudent to look to lower level skills first to ensure that students have basic alphabetic knowledge, even at higher grade levels (5th and above).
Kim Cordewener (Radboud University Nijmegen - Behavioural Science Institute)Anna Bosman; Fred Hasselman; Ludo Verhoeven - Predicting early literacy skills
Purpose: This study focused on the predictors of letter reading, word reading, letter spelling, and word spelling in first grade for children with specific language impairments (SLI). The main question was whether a nonlinear model fit the data better than linear models. Method: A sample of 78 second-year kindergartners was followed until the end of first grade. With respect to predictors, linguistic, phonological, and orthographic skills were considered, but also reading and spelling level at an earlier point in time. To account for all possible kind of processes, three linear and one nonlinear regression models were used to accomplish the analyses. Results: We used a cusp catastrophe model for nonlinear fit to compare this to three linear models, including a pretest-posttest linear model. Preliminary analyses indicate that the best predictor of letter reading, word reading, letter spelling, and word spelling in first grade is the same skill at an earlier point in time. For letter reading and word spelling, only one linear model, the pretest-posttest model, fits the data as well as the nonlinear model. However, for word reading and letter spelling, the nonlinear model fits the data better than the three linear models. Conclusions: From the findings of the present study, it can be concluded that nonlinear models are more suitable to predict early literacy skills than normally used linear models. The results of the present study, and the implementation of the nonlinear analyses, have interesting implications for further research and clinical practice.
Cynthia Core (The George Washington University)Cynthia Puranik, Kenn Apel, Erin Estabrooks - Vowel representation in the spelling of preschool children
Purpose - Some reports of early spelling development suggest that children begin spelling by using letters randomly before demonstrating understanding of the alphabetic principle. However, more recent studies suggest that children's early spelling may not be completely random but instead follow patterns of the writing systems to which children are exposed (Pollo, Kessler, & Treiman, 2009) and statistical properties of the orthography of that writing system (Caravola, Kessler, Hulme, & Snowling, 2005). The goal of this study was to examine the patterns of vowel representation in the spelling of preschool children. Two questions were addressed: 1) Are there patterns in the marking of vowels in spelling of preschool children, e.g., do they use a letter name strategy or use vowels from their own name)? 2) Do the phonological properties of sounds in adjacent letters influence the marking of vowels (e.g. are vowels represented more or less often/accurately when followed by sonorants vs. stops)? Method - Written spelling was assessed in 187 preschool children aged 4-5 years. Two protocols were used to score children's spelling, the more traditional method that captures phonological information children represent in their spelling (i.e., Tangel & Blachman, 1992) and a modified version of the Spelling Sensitivity Score (Masterson & Apel, 2010) that captures phonological and orthographic knowledge in spelling. In addition, spellings were also coded to examine children's markings of onset-rime and body-coda. Results - Preliminary findings suggest that children do not mark vowels randomly; instead they appear to use the letter name strategy. Conclusions - The results will add to our understanding and of spelling development and the scoring of developmental spelling. Results of the study will have implications for early spelling instruction, for example, the type of words and letter patterns that need to be targeted with beginning writers.
Pierre Cormier (Universite de Moncton); Julie Ronikier; Jean Saint-Aubin - The content/function distinction in the missing letter effect in children
o Purpose- There is no consensus as to when children start to show error patterns characteristic of the Missing Letter Effect (MLE). MLE appears when readers miss more target letters in frequent versus rare words and in function versus content words, two critical dimensions of word processing. We designed a text to specifically tap the function versus content word distinction. This text contrasted searching for the letter "n" in the function word "nos" [our] and in the content word "nez" [nose], both of the words having similar rate of occurrence in French children's texts. o Method - 346 French speaking children from grade 3 to 7 had to search for target letters in two texts: One that confounds the distinction between content and function words of varying frequency as used previously in the literature, another that includes content and function words of equal frequency. o Results - As early as grade 3, children showed a word function effect. There were more omissions in function words than in content words of the same frequency F (1, 341) = 105.54, ηp² = .24, p < .05. There was no effect of word frequency F (1, 341) = 2.15. o Conclusions This is the first developmental study that has properly isolated the influence of word frequency and word function by eliminating other confounding factors found in all previous studies in the field. Most importantly, this study shows that the missing-letter effect is due to word function in young children and not to word frequency.
Wind Cowles (Wind Cowles); Sunjung Kim - Comprehension and reading of canonical and non-canonical sentences: an eye-tracking study
Purpose: Listeners have poorer comprehension accuracy for sentences with non-canonical role order (e.g. passives or object clefts like it was the boy who the girl kissed) compared to sentences with canonical role order (e.g. actives or subject clefts like it was the girl who kissed the boy) (Ferreira, 2003). However, reading allows comprehenders to control how they take in information, thus readers could process sentences more effectively, resulting in better accuracy regardless of role order. We tested this hypothesis by replicating Ferreira (2003) with readers. Method: Twenty-four reversible subject and object cleft sentences were taken from Ferreira (2003), as seen above. Eighteen college students read these while their eye movements were tracked. After each sentence, a visual prompt asked participants to verbally identify who played a particular role in the sentence, for example, who the "do-er" was in the sentence (in this case, the correct answer would be the girl). Results: Response accuracy data replicated Ferreira (2003): Subject clefts prompted higher accuracy (94%) than object clefts (86%). Whole sentence reading times revealed that readers took longer to read object clefts than subject clefts (3214ms vs. 2782ms). Further, reading measures of individual words showed that readers took longer to read object clefts starting at the word who, and that object clefts prompted more rereading. Conclusions: Readers show poorer comprehension for non-canonical role orders in (relatively) simple sentences even when they take longer to read them, reread parts of them more, and know that they will be asked about thematic relations.
Michael Coyne (University of Connecticut) - Supporting early vocabulary development within a multi-tier approach to instruction and intervention: A regression discontinuity study
Purpose: This presentation shares results of an intervention study evaluating a 24-week implementation of multi-tiered vocabulary supports with kindergarten students using a regression discontinuity design. The primary research question was "What are the effects of a supplemental Tier 2 vocabulary intervention on vocabulary outcomes of kindergarten students at-risk for language and literacy difficulties?" Methods: Participants included 240 kindergarten students from 19 classrooms across 9 elementary schools serving diverse populations. All students received whole class "Tier 1" vocabulary instruction delivered by classroom teachers for 30 minutes per day, 5 days per week, over 20-24 weeks organized around Elements of Reading Vocabulary, a published evidence-based vocabulary curriculum. In fall, students were screened on a measure of overall receptive vocabulary knowledge (PPVT) and those students with a standard score of 95 or below were assigned to receive supplemental Tier 2 intervention for an additional 20 minutes per day, 4 days per week. The supplemental intervention was delivered by school designated interventionists and reinforced vocabulary introduced during whole-class instruction, supported foundational language skills, and provided students with extended opportunities to engage in interactive discussions. Outcome measures included assessments of target word knowledge, language use, listening comprehension, and overall vocabulary knowledge. Results: Results were analyzed using a regression discontinuity design and indicated that students who received the supplemental intervention experienced greater target vocabulary learning that predicted. Conclusions: Findings support the benefits of a tiered approach to vocabulary instruction and intervention although these supports require a significant investment in instructional time and intensity.
Holly K. Craig (University of Michigan)Stephanie Hensel; Rachel Schachter; Giselle Kolenic - African American English-speaking students: a preliminary examination of the reading skills impacted by style shifting
Purpose Production of African American English (AAE) features varies inversely with reading achievement scores, and students who fail to "style shift" away from AAE features to Standard American English alternatives in literacy tasks appear the most disadvantaged. The purpose of this paper is to provide preliminary evidence that it is word identification skills which are affected. Method Participants were 27 typically-developing AAE-speaking males and females residing in a large, segregated, primarily low-income urban center. Students were followed longitudinally from kindergarten entry through second grade, with three language-literacy assessments administered each year. This yielded nine time points per student for feature production measures and annual performances on language, cognition, and reading tasks. Students were assigned to one of two groups: +shifters, and -shifters based on feature rate differences between language samples generated during spontaneous discourse and an emergent reading task. The groups did not differ on measures of general oral language, cognition, and maternal education. Results Groups differed in reading achievement scores with +shifters outperforming -shifters at statistically significant levels at second grade on the Woodcock Reading Mastery Test (WRMT-R, 1998). Multiple linear regression showed that there was a statistically significant relationship between Word Identification and the groups (controlling for vocabulary scores) but not other subtests. On average, Word ID standard scores were 6.46 higher for +shifters. Conclusions Findings contribute to the accumulating research relating style shifting to reading outcomes for AAE-speaking students, and indicate that the impact is on student identification of letter, sound, and word patterns presented as text.
Virginia Cronin (George Washington University)Lila Yuen - Single- and double-deficit readers and compensation
Purpose: To understand ways of compensating for poor reading. Poor readers in elementary school often become average college readers although they remain poor in some reading areas (Bruck, 1990; 1992; Parrila, Georgiou, & Corkett, 2007; Wilson & Lesaux, 2001). Method: In this longitudinal study 130 children trained by whole language methods were followed from preschool through fifth grade when 84 remained in the study. While most of the children were prereaders they were classified into Wolf and Bowers' (1999) groupings by tests of RAN and PA. Results: It was found that a number of poor readers caught up with average readers in comprehension while remaining poor with pseudowords reading. Lexical learning was an important factor as were vocabulary and socioeconomic status. Conclusions: Lexical learning was a more important factor than SES.
Elizabeth Crowe () Carol McDonald Connor, Fredrick Morrison - Paths to proficient reading: profiles and predictions from preschool to fifth grade
Purpose Numerous studies have focused on typical and struggling students while fewer focus on highly proficient reading skills and the development thereof. While initial status has often been considered an early marker of high proficiency later in life, it is possible that these skills may develop over time. This study investigates students' language and literacy skills at fifth grade to determine latent profiles of literacy. In addition, preschool status and growth from PK-5th grade will be used to predict membership in latent profiles. Method 984 participants from the NICHD Study of Early Child Care, were used for analyses. Woodcock Johnson Letter/Word Identification and Picture Vocabulary were administered at PK, 1st, 3rd, and 5th grade. Passage Comprehension was administered at 3rd and 5th grade. Results LCA revealed 6 profiles of readers with unique constellations of vocabulary, letter/word identification, and comprehension. Logistic regressions suggest that preschool skills are a significantly better predictor than growth. However, differences between coefficients were small. Descriptively, within the highly proficient profile, preschool letter/word scores varied across profiles. In contrast, below grade level vocabulary scores at preschool were not characteristic of highly proficient students at fifth grade. Growth varied by profile, indicating that initial status alone may not characterize a later achievement. Conclusions While initial preschool status appears to best predict, growth is also relevant. Students who are seemingly average at preschool but experience growth over time can also exhibit highly proficient skills while some students who start out ahead demonstrate highly proficient skills at the end of elementary school.
Márcia da Mota (Universidade do Estado do Rio de Janeiro);Silvia Guimarães; Carolina Conti - Morphological awareness and reading in Brazilian Portuguese
Purpose : The alphabetic principle states that letters' word map onto phonemes. Phonological awareness has been shown to be an important skill for learning the alphabetic principle. Another skill that has been shown to be related to reading is morphological awareness. Orthographies that follow alphabetic rules strictly may not require morphological analysis. Portuguese is an alphabetic language with fairly transparent orthography. However, it has many morphological complex words. It is possible that children might not rely on morphology in such a system. However, morphological processing may still help children to read. The focus of our study was to look at the relationship between phonological and morphological awareness in reading Portuguese. Method One hundred and thirty-four second to fourth grade Brazilian children participated in the study. They were asked to read words in two reading tests - a reading in context test and single words test. They were also given a vocabulary test and phonological and morphological awareness tasks. Result: There were independent contributions of morphological awareness and phonological awareness for reading when all other measures were controlled for. Conclusions: These results suggest that both phonological and morphological awareness contribute to reading in a fairly transparent orthography as Portuguese. Overall these results suggest that the relationship between reading and morphological awareness may hold even when the alphabetic principle applies.
Daniel Daigle (Universite de Montreal); Rachel berthiaume - Phonological processing in dyslexic children: the implication of the task
Considering the role of phonological processing in reading development and the phonological deficit characterizing dyslexic readers (Demont & Botzung, 2003; Ramus, 2003; Snowling, 2006), the purpose of this study was to investigate graphophonemic (GP) and graphosyllabic (GS) processing using tasks varying in terms of cognitive constraints (GPCC-, GPCC+, GSCC-, GSCC+) (Bialystok, 2001). Dyslexic readers (n=26) were compared to 26 normal reading development students of the same reading level (CL) and to 30 students of the same age (CA). In the GPCC- and the GSCC- tasks, subjects had to determine which of two pseudo-words (e.g., assidème or annidème / dabri or dabri) resembles the most a target item (e.g., academe / dafli). In the GPCC+ and the GSCC+ tasks, subjects had to choose, out of four pseudo-words, the item that starts with a different sound/syllable (e.g., cékane cavène codile cutare / baflo barto balpé barfi). First, for all tasks and for all groups, results are above the threshold of chance. Secondly, results from ANOVAs indicate, that all groups had better scores on the CC- tasks than on the CC+ tasks, no matter the phonological unit targeted and that, whatever the task, dyslexic subjects and CL have comparable results, but perform more poorly than CA. This study shows the importance of considering the cognitive constraints attached to the phonological task when investigating phonological processing in reading, especially with special populations. These results can nuance the variability observed in results obtained in the dyslexic population (INSERM, 2007).
Bridget Dalton (Vanderbilt University); Elaine Mo;Kristin Robsinson; Paola Uccelli ; C. Patrick Proctor - Multimodal retellings of scaffolded digital narrative: A comparison of students with diverse language and literacy needs
Purpose. The purpose of this study was to investigate fifth grade monolingual and bilingual students' composition of a multimodal retelling of a digital folktale scaffolded with supports for vocabulary and reading comprehension. Drawing on new literacies (Coiro, Knobel, Lankshear, & Leu, 2008) and multimodal composition (Kress, 2003; Hull & Nelson, 2005), we anticipated that the scaffolded reading experience and digital media composing affordances would mitigate writing differences often found between struggling, average and above average readers, as measured by pretest Gates MacGinitie Reading Achievement scores. Method. 83 students read a digital folktale and composed a retelling using a PowerPoint template structured with illustrations. Retellings were scored for media use and quality of prose, prose-media integration, and overall narrative, obtaining inter-rater reliability of .87 on 20% of the retellings. A series of multiple regression analyses were conducted. Results. Overall, students' retellings integrated prose and media at a basic level. Contrary to prediction, there were significant differences on the 3 composition quality scores as a function of reading achievement, with stronger readers generally producing higher quality retellings, and integrating graphics more effectively. However, bilingual and monolingual students achieved comparable quality scores. Conclusion. We know little about students' multimodal composing, and even less about their retellings when they've had a scaffolded reading experience. Reading group differences persisted, with stronger readers better able to communicate multimodally, suggesting the need to develop multimodal composing theory and instructional practice to support all learners in developing multimodal communication skills critical to success in the 21st century.
Nicole Davis (Vanderbilt University);Laura Barquero; Aanandhi Venkatadri; Lindsay Wilson; Sheryl Rimrodt; James Pekar; Laurie Cutting - Neurobiological correlates of sentence comprehension in adolescent struggling readers
Purpose: We examined the neurobiological correlates of sentence comprehension (SC) and single word reading (WR) in participants with reading disabilities and control participants. Though previous work has explored differences in sentence and word-reading activations, this study further expanded upon previous work by examining sentences (without comparison to words), as well as isolated words themselves. Method: Adolescents with reading disabilities (n=9) and control participants (n=8) performed two in-magnet reading tasks in a block design. One task involved reading single high frequency words visually presented on a screen. The second task differed in that sentences were presented. Participants determined whether the words were real or not and whether sentences were meaningful or not and responded by button presses. Whole-brain t-contrasts were performed for condition and group contrasts. Results: Contrast maps for SC-WR showed greater activation for participants with RD in the LH insula. Individual contrasts with sentences and words versus fixation also revealed differences between groups, particularly in traditionally seen areas (e.g., RD showed lower activation in occipito-temporal areas as compared to controls for both SC and WR). Conclusions: Results from the SC-WR contrasts replicate findings from an earlier study (Rimrodt et al., 2009) in showing increased LH insula activation in RD. In addition to these findings, the exploration of sentences as well as words as compared to baseline fixation may reveal further aspects of differences in the neural correlates of higher level comprehension in adolescents with RD.
Stephanie Day (Arizona State University and the Learning Sciences Institute)Carol Connor - Examining the relationship between self-regulation and academic achievement in third grade students
Purpose: The present study sought to examine the relationship between children's self-regulation skills and literacy outcomes using a newly designed measure of self-regulation. Self-regulation includes components of working memory, attention, and inhibitory control. Research has found that poor self-regulation skills can negatively impact academic achievement. Previous research on self-regulation has commonly utilized parent and teacher surveys as a measure of self-regulation skills. While these measures are useful, they are open to threats of observer bias, thus, developing a more direct measure of self-regulation may be a more accurate measure of self-regulatory skills as well as a stronger predictor of academic achievement. Method: In this study, we examined the self-regulation skills of 347 third grade students using a newly designed measure of self-regulation, the Remembering Rules and Regulation Picture Task (RRRP). Children's literacy skills were measured using the passage comprehension subtest of the Woodcock-Johnson Tests of Achievement-III. Children were assessed two times a year on the RRRP and the passage comprehension subtest in the fall and the spring. Results: Preliminary results using hierarchical linear modeling revealed that children's performance on the RRRP in the fall and spring was positively associated with gains (residualized change) on students' passage comprehension skills. Conclusions: Performance on the RRRP seems to be a strong predictor of children's literacy skills.
Ozlem Ece Demir (The University of Chicago)Ozlem Ece Demir; Lauren Applebaum; Susan Levine; Katherine Petty; Susan Goldin-Meadow - Early parent-child book-reading, and its relations to later language and reading outcomes
Purpose:The study aims to examine links between early parent-child book-reading and later child language and reading performance. Method:Book-reading interactions produced during naturalistic parent-child interactions when children were 26, 30 (averaged) and 42 months-old in a sample of 46 parent-child dyads were identified. Book-utterances were divided into 3 categories: text- reading off the text; picture- picture descriptions; and nonimmediate- story discussions (e.g. "why do you think he is sad?"). Children were given the Peabody Picture Vocabulary Test at Kindergarten (PPVT), and three reading subtests from Woodcock-Johnson Tests of Achievement (Decoding: Word Attack, Letter-Word Identification (WJ-D) and Reading Comprehension (WJ-C) at the end of 1st grade. Results:25 parent-child dyads engaged in a book-reading interaction at child age 26 and 30 months. Number of book-utterances was significantly related to PPVT, WJ-D and WJ-C. Specifically, picture utterances were related to PPVT, and WJ-C, text to both reading outcomes, and nonimmediate to WJ-C. At child age 42-months, only 7 parents read books with their children. WJ-D and WJ-C significantly differed between those who interacted with books versus those who did not. Although not significant due to sample size, the pattern of results was the same as the earlier time point. Relations remained significant controlling for overall non-book parental talk. Conclusions:Although rare, parent-child book-reading interactions at young ages, relate to later language and reading success, even controlling for much more frequent non-book talk. In addition, specific links between different aspects of parent-child book-reading interaction and later child outcomes might exist.
Carolyn Denton (University of Texas Health Science Center Houston)Tammy Tolar, Amy Barth, Jack Fletcher, Melissa Romain, and Jennifer Hocker - The effects of teacher-reported ADHD symptomology on reading outcomes for primary-grade children in Tier 3 reading intervention
Purpose: The purpose of this study is to examine the potential effects of inattentiveness and hyperactivity on decoding, fluency, and comprehension outcomes in first-third graders who received Tier 3 intensive intervention within a multi-tiered RTI model. Method and Results: In a larger study, 192 first graders received Tier 1 and 2 intervention; then in Grade 2, 72 students who had demonstrated inadequate RTI in Grade 1 participated in a study of Tier 3 intervention. Students were randomized to receive the research intervention (n=47) or to typical practice. Classroom teachers completed the SWAN ADHD rating scale for each participant. Repeated measures ANOVAs revealed significant main effects of inattentiveness (but not hyperactivity) on decoding, fluency, and comprehension. In follow-up analyses we will identify three subgroups: (a) those with teacher-identified inattentiveness, (b) those with hyperactivity characteristics, and (c) those with neither, and we will compare their reading outcomes and RTI in all 3 tiers of intervention. Group N's become small as children progress through the tiers, so these analyses will be descriptive. Findings will be displayed graphically and will be discussed in context of the literature on children with comorbid ADHD and reading difficulties. Conclusions: These findings are generally congruent with those in the literature on students with ADHD that associate inattentiveness, rather than hyperactivity, with reading difficulties. We expect that our secondary analyses will indicate that students with inattentive characteristics will demonstrate lower RTI than other students. The fact that early reading difficulties tend to predict reading problems across a child's school career suggests that early reading intervention may be more effective if accompanied by interventions for inattentiveness in children with ADHD.
Alain Desrochers (School of Psychology, University of Ottawa)Monique Brodeur; Line Laplante; Eric Dion; Glenn L. Thompson - Implementation of a Tier 1 and Tier 2 intervention program in Kindergarten: Impact on phonological awareness and letter knowledge
Purpose : Evaluate the effectiveness of a universal (Tier 1) Kindergarten intervention (French adaptation of Simmons & Kame'enui, 2002: Brodeur et al., 2008) and a related remediation (Tier 2) intervention (Laplante & Brodeur, 2009) for the development of phonological awareness and letter knowledge. The Tier 1 intervention involved 26 lessons, one for each of the 26 letters of the alphabet; each lesson was divided into 4 15-30 min segments delivered over 26 weeks. The Tier 2 intervention included 4 additional 30-min lessons a week over 24 weeks. Measures of phonological awareness and letter knowledge were taken before, during and after the interventions. Method: A total of 12 schools in disadvantaged urban neighborhoods were randomly assigned to the control condition (the usual K program, n = 120 children) or the experimental condition (Tier 1 intervention, n = 159 children). The children who made the least amount of progress in the Tier 1 group (n = 39) received the Tier 2 intervention. Results: Relative to matched pairs, both the Tier 1 and Tier 2 interventions significantly increased phonological awareness and grapheme sounding, with observed effects ranging from .79 to 2.05 standard deviations. Pretest scores were found to moderate the effectiveness of the program, which was more effective for strong or weak performers depending on the outcome measure. Conclusion: These results extend those previously observed with English-speaking children to French-speaking children: Explicit and systematic instructions are found particularly effective in the development of phonological awareness and letter knowledge but initial performance level matters.
David Dickinson (Vanderbilt University); Jill B. Freiberg; Kerry G. Hofer - Interactions between the type of preschool curriculum Intervention, children's language level and fidelity of implementation
Purpose Effects of intervention type, implementation fidelity and children's language skills on fall-spring growth in language and print knowledge were studied. Method Fifty two Head Start classrooms were in a randomized control study: Curriculum only (C) (n = 17), Curriculum + (C +) Enhanced Milieu Teaching (EMT) (n = 19), control (n = 16). Children were assessed fall and spring of preschool (n = 441) using standardized measures of receptive and productive language and letter knowledge. Videotapes were coded for fidelity of implementation. At preschool entry PLS III scores were used to place children in a low-language group (LL) (n = 179) or typical language (TL) for low-income children (n = 292). LL children's teachers delivered 1-1 EMT for 60 sessions. Results Fidelity of implementation for C and C+ was generally low. HLM analyses controlled for child demographic variables, Head Start centers, and classroom variables anticipated to present greater instructional challenges when examining fall-spring gains. A receptive vocabulary effect of .31 was found for the C condition, but not C+ (-.13); for expressive vocabulary C+ results were stronger (d =.21 vs. -.05 for C), possibly reflecting the expressive language focus of EMT. In the C condition letter knowledge growth was seen in the C group (d = .29), but not the C+ group (-.16). Strong curriculum implementation was related to stronger growth of LL children. Conclusions Teacher capacity to implement new methods affects delivery and children's language level may affect responsiveness to a curriculum.
Jennifer Dombek ()Carol Connor - Grade Retention and First Grade Classroom Instruction
Purpose: Retention is a frequently used strategy to support children who are struggling academically. However, the strategy is costly and research findings with regard to positive outcomes are mixed. This study sought to examine (1) whether participating in an efficacious reading intervention might reduce rates of grade retention and (2) the amounts and types of reading instruction children received was associated with whether they were retained or not. Method: The reading instruction first grade students who were retained received was compared to students with similar reading and vocabulary skills who were not retained (n = 114, 17 teachers, 14 schools). Additionally, the reading progress of these students who and the instruction they received during their retained year was examined. Results: Chi-square analysis reveal that it was significantly less likely for a student to be retained at the end of first grade if their teacher was implementing more efficacious reading instruction. HLM revealed that there were substantial differences in the literacy instruction provided for children, with promoted students receiving instruction more closely aligned to their needs. Descriptive analyses indicated, on average, that children showed grade level reading skills after their retained year (grade equivalents ranged from 1.0 to 4.2). Conclusions: Differentiating literacy instruction in first grade to meet the needs of students whose reading and vocabulary skills differ may result in fewer students retained at the end of first grade. For students who are retained, differentiated reading instruction may help them achieve grade level reading skills.
Karen Douglas () - Funding Opportunities at the National Center for Education Research and National Center for Special Education Research, Institute of Education Sciences
This poster will summarize the research objectives, program topics, goal structure, and due dates for applying to the National Center for Education Research and the National Center for Special Education Research at the Institute of Education Sciences for funding in FY 2012. The presenter will provide clarification on the Request for Applications for FY 2012, information on how to work with the program officer in developing an application, and answer questions about changes in the RFA as compared to previous years.
Darren Dunning (University of York)Susan Gathercole; Joni Holmes - The impact of working memory training on learning: A randomized controlled trial
Purpose Children with poor working memory (WM) typically make slow educational progress in reading and maths due to an inability to meet the heavy WM demands of the classroom. Studies have shown that extensive training on artificial WM tasks benefits performance on other rarefied memory tasks administered under controlled conditions in children and adults. The aim of the current study is to establish whether these benefits will transfer to improvements in children's reading, language and mathematical skills, as well as the practical situations in which WM appears to play a vital role using a randomised controlled design. Method 120 children aged 7-9 years were randomly allocated to one of three conditions: i) active intervention in which children trained on the standard adaptive version of the Cogmed WM training program (CWMT) ii) active control in which children trained on a non-adaptive placebo version of CWMT and iii) passive control in which children received no intervention. All children completed a range of assessments either pre and post training, or before and after 6 weeks of normal schooling. Results Children in the active intervention group made significantly greater gains in non-trained and practical WM tasks than children in either control group. There were significant post-test differences in reading rate, written expression, processing speed and attention. In all cases gains were significantly greater for the active intervention group. Conclusion Training improves WM performance in children with poor WM and may transfer to selected learning activities, including reading skills.
Jaclyn Dynia (Ohio State University); Ying Guo; Laura M. Justice - Print knowledge of children with specific and non-specific language impairment
Purpose: Children with language impairment (LI) typically demonstrate lags in their development of print knowledge (PK). These children can be differentiated into two groups: those with average nonverbal cognition, or Specific Language Impairment (SLI), and those with concurrent cognitive delays, or Non-Specific Impairment (NLI). As the distinction between these two groups has important implications for the early identification and treatment of reading disabilities, we aimed to examine the differences in PK for children with SLI versus NLI, theorizing that children NLI would have lower PK than children with SLI. Method: Children with LI (n = 41) received a standard score of 1.25 SD or more below the mean on the language composite of a standardized language measure. The two groups were differentiated into SLI and NLI based on their performance on a nonverbal cognitive screen. Children with SLI (n = 26) scored < -1SD of the mean and children with NLI (n = 15) scored < -1 SD of the mean. PK measures included print concepts, letter knowledge, and name-writing skills. Results: MANOVA indicated that children with NLI performed significantly lower on PK than children with SLI (Wilks's λ = .74, p = .02); post-hoc comparisons showed differences in print concepts, F (1, 40) = 6.21, p= .02 and name-writing skills, F (1, 40) = 7.67, p = .01. Conclusion: Findings showed a distinction between children with SLI and NLI in PK. This can serve to alert professionals to the degree of risk for literacy problems and assist them in planning intervention.
Sue Ann Eidson (Sue Ann Eidson, PhD) - Predictive indicators of emergent literacy skill in three- and four-year-olds
Purpose 1. To determine which tasks given to 3-year-olds predict their performance one year later on a standardized battery for emergent literacy, and, 2. To conduct first stage in developing a battery for 3-year-olds at risk for reading difficulties by identifying tasks on which they show a normal distribution of scores. Two research questions were adopted for this study: 1) Which experimental tasks met normality criterion for 3-year-olds and again one year later? and, 2) Which of experimental domains and tasks that met normality criterion for 3-year-olds and 4-year-olds best predicted a standardized battery of emergent literacy (Assessment of Language and Literacy) index scores one year later? Method A longitudinal test design was used in which 38 children between 3 and 4 years of age were evaluated with experimental and formal measures that assessed cognitive skills (memory) and emergent literacy skills (letter knowledge and writing). The same children were evaluated one year later with the same tasks and with the standardized tests of emergent literacy found in the Assessment of Language and Literacy (ALL). Multiple Linear Regression analysis was performed using 7 experimental composite predictor variables and their tasks (Phonological Processing, Orthographic, Emergent Writing, Fine Motor Function, RAN, Memory, Language composite scores) and 3 outcome variables (Letter Knowledge, Rhyme Knowledge and the composite Emergent Literacy index score from ALL). Only experimental tasks, which were normally distributed, were used in the final statistical analysis. Results Eight experimental tasks met normality criterion for the 3-year-olds (Rhyme Knowledge, Segmentation, Copying, Visual Short-term Memory, Sequential Memory, Letter ID, Fingertapping, and RAN-12 items). Nine tasks met normality criterion for the 4-year-olds (Rhyme Knowledge, Segmentation, Rhyme Judgment, Copying, Digit-word Span Backward (working memory), Visual short-term Memory, Sequential Memory, Fingertapping, and RAN-12 items). At age 3, the experimental domain that best predicted the Emergent Literacy index score on the ALL one year later was the Memory composite score (t= 2.4; p= .02). The normally distributed tasks at 3 that best predicted performance on ALL at 4 were Letter Identification (t= 2.54; p= .017) and Rhyme Knowledge (t= 1.99; p= .056) and the tasks that were the best predictors at 4 were Rhyme Knowledge (t= 4.21; p=.000), Digit-word Span Backward (t= 3.29; p= 003) and Segmentation (t= 2.13; p= .042). Conclusions Current screening instruments can identify children who are at risk for reading difficulties but no screening tools have been shown to reliably identify the specific domains of weakness in preschool children at risk for future difficulties (Wilson & Lonigan, 2010). We need to examine more rigorously the relationship between emergent literacy skills and cognitive processes, such as memory, as part of a larger goal of "differentiated assessment". Examining cognitive processes that subserve emergent literacy may help us reliably identify specific domains of weakness in preschool children at risk for reading difficulties and differentiate children at risk for reading difficulties due to environmental causes from those at risk due to neurobiological differences (e.g., dyslexia).
Carsten Elbro (University of Copenhagen); Peter F. de Jong; Anne-Mette V. Nielsen - From "w..aa..ss" to "woz". A second step in word decoding?
Purpose: Beginning readers often produce recoded word forms such as "w..aa..sss" for "woz" (was). Then they pause and search for the real word. This search may take some time and is not always successful. In other words, there is a gap between the output from a phonological recoding of a word and its lexical pronunciation. In this presentation, it is suggested that the ease of recognition of words from spelling pronunciations contribute independent variance to word decoding ability with both regularly and irregularly spelled words. Method: This suggestion was investigated in a longitudinal study of 187 children from preschool into grade 1 learning to read in a deep orthography. Word recognition from spelling pronunciations was assessed in a task where participants listened to other beginners who failed to recognise words despite correct spelling pronunciations. Control measures were taken of budding reading abilities, phoneme awareness, and RAN. Regression analyses and path modelling was applied to the data. Results: Word recognition from spelling pronunciations predicted development of word reading accuracy with both regularly and irregularly spelled words - even when standard predictors were controlled. Correlations were stronger for accuracy than for fluency in word decoding. In conclusion, word recognition from spelling pronunciations may form a second step in word decoding for both regularly and irregularly spelled words.
Amy Elleman (Middle Tennessee State University); Endia Lindo; Donald Compton - Can Children Learn to 'Read Between the Lines'? A Meta-Analysis of Comprehension Studies Intended to Improve Inference Skills of School-Age Children
A meta-analysis of inference generation interventions in grades K - 12 was conducted with 24 studies. Inference instruction was found to be effective at increasing students' general comprehension, d = 0.49. In addition, the effects for skilled readers on inference measures were moderate to large, d = 0.61, as was the effect for less skilled readers, d = 0.80. The effect for skilled readers on literal measures was not significantly different than 0, d = -0.21, but the effect for less skilled readers was large, d = 0.84. Findings suggest students can increase their inference generation skills, and that less skilled readers gain the extra benefit of increases in literal comprehension. Findings also suggest that instruction using text clues and small groups was beneficial.
Carmen L. Escribano (Universidad Complutense de Madrid) - The Relationship of Rapid Automatized Naming to reading development in Spanish
Purpose To summarize research findings that analyzes the contribution of Rapid Automatized Naming (RAN) to the prediction of reading. Three questions were addressed: (1) Is RAN a good predictor of reading? (2) How does RAN contribute to reading? (3)Do phonological awareness (PA) and RAN differentially explain variance in reading? Method We conducted a computer search of ERIC, PSICODOC, PsycINFO, REDALyC, DIALNET and ProQuest, using the following descriptors "reading, RAN, and Spanish" We found a total of 14 articles. We grouped the articles into four categories: prediction of reading development (six); prediction of reading disabilities (four); the double-deficit hypothesis (DDH) (three); neuroscience (one). Results The reviewed studies reported the following findings: (1) RAN administered during kindergarten discriminated between groups at risk for later reading difficulties; in a longitudinal study RAN best predicted the execution of reading speed difficulties; (2) The DD subtype showed the most difficulties with reading, and the RAN subtype affected measures of fluency; longer RAN times differentiated between reading disabled vs. non-reading disabled children. (3) Correlational comparison and hierarchical regression analysis determined that PA and RAN differentially explain variance in reading. (4) Differences between the ERPs in the slow naming versus the average naming children, to the vicinity of 320 ms. Conclusions These studies consistently show that RAN predicts reading speed, a salient characteristic of reading development in Spanish. This has implications for diagnostic and treatment programs. Specifically, different methods may be needed for children who are not adequately remediate with programs that uniquely emphasize PA and decoding.
Mary Ann Evans (Department of Psychology, University of Guelph, Canada.);Kailey Reynolds - Effects of shyness and reading ability on parent and child behaviours during of shared book reading
Purpose. The study examined the effect of child temperament and reading ability on behaviour when parents listen to beginning readers. Given research showing overprotective parenting and lower risk-taking in anxious children, it was hypothesized that shyness would predict parents using strategies typically used with less skilled readers, and children less often attempting unknown words. Method. Pseudoword Reading of the WRMT indexed reading skill and the Colorado Childhood Temperament Inventory indexed shyness in 94 first-graders. Parent-child dyads were observed and audiotaped reading at home. Behaviours were coded for type of parent feedback following a child miscue and how children approached unknown words. Results. Boys more frequently guessed at unknown words, while girls more often paused or requested assistance. Regressions entering shyness, decoding, sex, and interaction terms revealed no interaction effects. Main effects were as follows. Child reading skill predicted frequency of child pauses, requests for assistance, and incomplete attempts at words, but unexpectedly, child shyness did not. Decoding skill positively predicted parent encouragement to try the word again and ignoring errors, and negatively predicted provision of graphophonemic and context clues. Child shyness negatively predicted encouragements to try again but positively predicted parents supplying the words and giving context clues. Conclusions. That children's approaches to difficult words was invariant with shyness suggests that self-consciousness/anxiety associated with shyness was irrelevant to their reading attempts at home. Observations with teachers might be different. Second, regardless of child competence at decoding, parents were more directive and protective with shyer readers, potentially discouraging reading self-efficacy.
Fataneh Farnia (OISE/University of Toronto);Esther Geva; - Trajectories of vocabulary growth in subgroups of English-language Learners and their monolingual counterparts
Purpose: ELLs have lower vocabulary knowledge than EL1s, but regardless of ELL/EL1 status not all learners develop their vocabulary at the same rate. In this study we focus on understanding the heterogeneity of the trajectories associated with vocabulary development. Method: We used growth mixture modeling (GMM) that handles longitudinal data (nesting of time within individuals) to identify subgroups characterized by qualitatively different developmental pathways. The sample consisted of 390 ELLs from different language backgrounds, and 149 EL1s classmates. Language, cognitive and reading skills were tracked from Grade 1 to Grade 6. Data were analyzed from the following perspectives: (a) are there subgroups of students with different vocabulary trajectory profiles? (b) Is the distribution of the subgroups similar in ELL and EL1? (c) Are the cognitive and language profiles of these subgroups distinct? Results: We found three distinct subgroups: (1) language impaired (LI), (2) "late bloomers", and (3) normally achieving (NA). LI and late bloomers began with low vocabulary in Grade 1 but in the LI group rate of growth continued to be consistently low, while in the late bloomers group there was acceleration. In the NA group vocabulary was higher from the outset and there was constant acceleration. The proportion of ELL and EL1 in the LI group was similar; there was a higher proportion of ELLs among "late bloomers", and the proportion ELL in the NA group was lower. MANOVA confirmed that the subgroups are associated with distinct cognitive, reading and linguistic profiles. Conclusions: Results are discussed in terms of developmental notions of subtypes, L2 developmental models, teaching, and clinical implications.
Kelly Farquharson (The Ohio State University)Tiffany P. Hogan; Jordan R. Green - Babbling at 9 months predicts letter identification at 66 months
Purpose: The purpose of this study was to examine the relation between babbling and letter identification. Characterizing early literacy as an extension of early language skills, we predicted that babbling would correlate positively with letter identification. Method: This study examined data obtained during a longitudinal study of speech, language, and motor abilities in babies through early childhood (Green, PI; NIH R01-DC6463). Phoneme inventory (i.e., babbling), expressive and receptive vocabulary, and early gross, fine, and speech motor data were gathered from babies as young as 9 months old. Data were collected on 13 children every three months through the present time, at which the participants are 72 months old. At 60, 66, and 72 months, a letter identification measure (WRMT-R Letter ID) was administered. At 72 months, a speeded word and nonword reading task was administered (TOWRE). Results: All correlations were positive with the main prediction confirmed: there was a significant positive correlation between the number of phonemes present in early babbling and the number of alphabet letters identified at 66 months old (R = .785). Further analyses are using mixed models including age and skill growth to determine how each skill is related to each other and early reading. Conclusions: Because we know that letter identification is a strong predictor of later reading ability, a tentative conclusion is that the complexity of babies' babble may be an indication of early reading risk.
Lama K. Farran (Georgia State University)Gary Bingham; Mona Matthews - Predictors of Reading Fluency in Bilingual English-Arabic Children
Purpose: This study examined the predictors of word reading fluency and pseudoword reading fluency in bilingual English-Arabic children. Method: Eighty three bilingual English-Arabic children in third, fourth, and fifth grades participated in this study. The following language and reading measures were administered in English and Arabic: The Comprehensive Test of Phonological Processing (CTOPP; Wagner, Torgesen, & Rashotte, 1999); English Morphological Awareness (Duncan, Casalis, and Cole, 2009); the Wide Range Achievement Test -Revised (WRAT-R; Wilkinson & Robertson, 1984; Word Recognition); and the Woodcock Reading Mastery Test-Revised/Normative Update (WRMT-R/NU; Woodcock, 1987/1998; Word Attack). Parallel measures in Arabic were used or adapted from Taha & Saiegh-Haddad (2008). Results: Arabic phonological awareness and Arabic morphological awareness predicted vowelized and unvowelized word reading fluency; and Arabic phonological awareness predicted Arabic pseudoword reading fluency. English phonological awareness, English phonological memory, and English morphological awareness predicted English word reading; and English phonological awareness and English phonological memory predicted English pseudoword reading. Conclusion: Results suggest a division of labor among the language components which changes based on reading tasks and demands. This study provides support for the extended Triangle model of reading (Bishop & Snowling, 2004) which underscores the multiple contributions of language to reading outcomes.
Amber Farrington (Florida State University); Christopher J. Lonigan; J. Marc Goodrich; JoAnn M. Farver; Kimberly D. McDowell - Item response theory of the Revised Get Ready to Read and differential item functioning for preschoolers who are English language learners
Purpose - Children who are Spanish-speaking English language learners are the fastest growing school-age population in the U.S. However, little research has examined the utility of different assessment measures with this population. Screening tools are an effective means of identifying children who may be at risk for reading difficulties and in need of intervention. The Revised Get Ready to Read (GRTR-R) is a 25-item measure that assesses preschoolers' emergent literacy skills. Using item-response theory (IRT) analyses, this study examined the performance characteristics of the GRTR-R and compared the measure's performance across children who were ELL and those who were monolingual English speakers. Method - As part of a larger study, 269 children between 3 and 5 years old (49.3% male), 165 of whom were ELL completed the GRTR-R in English. Results - IRT analyses revealed that the GRTR-R had good characteristics across the mid- to upper-range of performance. Differential Item Functoning (DIF) analyses suggested potential DIF on one print knowledge item in favor of ELL children, and one print knowledge and one phonological awareness item, in favor of non-ELL children. However, once corrected for multiple comparisons, no item had statistically significant DIF. Conclusions - These results indicate that the GRTR-R is an adequate screening measure with good test characteristics to identify children at average or above-average levels of ability. These analyses also revealed that items on the measure functioned similarly for children who were ELL and children who were not ELL. Therefore, the GRTR-R is appropriate for use to identify children who are ELL who are at risk for reading difficulties.
Anila Fejzo (UQAM)Godard Lucie; Laplante Line - The effects of morphological awareness instruction on second language student's literacy development.
This study examines the effects of morphological awareness instruction on word reading and spelling on third and forth grade readers and spellers in French as a second language. 53 Arabic-speaking students in 3rd and 4th grade of primary school participated in this quasi-experimental research study. During ten weeks of the intervention, 31 students in the treatment group which had more difficulties attended a morphological awareness program that focused on meaning and junction rules of most productive affixes in French with specific attention given to word reading and spelling. Before and after this intervention, students in both groups were tested on morphological awareness, word reading and spelling. Pre and post tests were administered on control variables such as phonological awareness, rapid naming and learning ability in order to isolate the treatment effects. The findings showed that students in the treatment group performed significantly better than the control group on morphological awareness, morphologically complex words reading and spelling measures. These results suggest that development of morphological awareness contributes to improve word reading and spelling of readers and spellers at intermediate level in FSL.
Gary Feng (Educational Testing Service)Jia Guo - Joint attention in shared storybook reading: A dual eye-tracking intervention study
PURPOSE: Joint attention is critical in social learning. However, past research on parent-child shared storybook reading suggests that young pre-(conventional)-reading children look mostly at pictures while parents read from texts. This lack of joint attention likely impedes the learning of print-related skills. In this study we sought to objectively measure the joint attention in shared reading. We investigate two potential causes of the phenomenon: (a) children are unaware where parents are looking during reading and (b) parents are ineffective in regulating the joint attention because they do not know where the child is looking at any moment. METHOD: Three experiments involved 37, 27, and 28 four- to five-year-old English-speaking children and their parents. Eye movements were simultaneously tracked during shared book reading. Behavior analyses were based on video recording and questionnaires and interviews. Expt 1 was the baseline, whereas Expt 2 and 3 involved real-time gaze-position feedback. In Expt 2 we showed children where their parents were looking in real-time. In Expt 3 the parent saw the gaze position of the child. RESULTS: Expt 1 confirmed that there is little joint attention on print texts - defined as parents and children looking jointly at a word (even with lenient criteria). Compared to control groups, the real-time gaze information (Expts 2 & 3) significantly increased joint attention, changed the dynamics of the interaction, and improved children's print-related learning. CONCLUSION: Effective learning requires fine-level joint attention between learning partners. Interventions reducing the uncertainty about others' attention facilitate learning.
Jessica S. Folsom (Florida State University);Stephanie Al Otaiba; Luana Greulich - Kindergarten susceptibility factors: What kindergarten skills interact with kindergarten instruction to predict longitudinal reading achievement?
Purpose: This study examined the effect of child by kindergarten instruction interactions on kindergarten, first, and second grade reading. We investigated the effect of code- and meaning-skills and the amount of kindergarten Tier 1 code- and meaning-focused instruction. Additionally, we investigated child by kindergarten instruction interactions to identify early literacy skills that influence students' susceptibility to instruction. Method: This study followed a cohort of 224 kindergarten students from kindergarten through second grade. Students' literacy skills were assessed twice in kindergarten, and reading achievement was assessed at the end of kindergarten through second grade. Kindergarten literacy instruction was videotaped and coded to calculate the amounts and types of instruction received. HLM was used to investigate the predictive role of kindergarten literacy skills and the amount of code- and meaning-focused instruction, as well as a variety of child X instruction interactions. Results: Code-focused skills predicted kindergarten, first, and second grade reading achievement; meaning-focused skills predicted kindergarten and second grade reading achievement. Kindergarten code-focused instruction predicted kindergarten reading achievement, and kindergarten meaning-focused instruction predicted first grade reading achievement; no kindergarten instruction predicted second grade reading achievement. In kindergarten the effect of meaning-focused skills was stronger for those students with stronger code-focused skills; in first grade the effect of meaning-focused instruction was stronger for those with weaker code-focused skills. Conclusions: Our study extends research on the importance of meaning-focused skills, and kindergarten reading instruction. Further, our study suggests code-focused skills are a factor in determining current and future susceptibility to instruction.
Liliana Fonseca (Universidad Nacional de San Martín)Bárbara Gotheil; Adriana Aldrey; María Pujals; Inés Lagomarsino; Eleonora Lasala; Dolores Pueyrredón; Sandra Molina; Luciana Buonsanti; Alejandra Mendivelzúa; Leticia Freire (UNSAM); Juan Pablo Barreyro UBA CONICET y col - LEE. Proposals for developing strategies to improve reading comprehension in a new Instructional Program in Spanish
Purpose: Based on the results of our previous research about the relationship between decoding and reading comprehension, our aim is to observe the effects of an explicit reading comprehension instructional intervention on 4th grade children. This program, designed through a careful selection of texts and activities, is based on word recognition, vocabulary, inferences making and methacognitive strategies (Oakhill, 2004, 2006, 2010). Method Participants: three whole class parallel groups (8-9 years), 21 children in control group (CG) and 28 children, intervention group (IG). Design: pretest, instruction period, two 80-minute weekly sessions each during 8 weeks, to finish with a post test. The instructional program is based on reciprocal teaching (Paliscar and Brown, 1984) structured around four strategies: questioning, summarizing, clarifying, predicting, visualizing verbalizing (Bell, 1986) and active reading (Mac Namara D, 2006,2007,2009). Results: In the pretest session we found no statistical differences in age, verbal reasoning, verbal comprehension (WISC III), non verbal reasoning (Raven), Word and Pseudoword Reading (LEE), Reading comprehension (LEE), and Listening Span Test. After the instruction period the three groups were tested again with Vocabulary (Wisc III) LEE Test and CLP Comparing pretest and post test measures, we found significant differences in the intervention group but not in the control group in: vocabulary, reading comprehension (LEE and CLP), inferential processes, connective and elaborative inferences, monitoring and methacognitive strategies. This intervention clearly results in a significant improvement in reading comprehension.
Barbara Foorman (Florida Center for Reading Research at Florida State University); Yaacov Petscher - Does morphological knowledge uniquely predict reading comprehension above and beyond spelling and text reading efficiency in grades 3-10?
Purpose. This investigation's purpose was to see if a morphological knowledge test added unique variance to the prediction of reading comprehension above and beyond tests of spelling and text reading efficiency. The latter three tests are computer adaptive and are part of the Florida Assessments for Instruction in Reading (FAIR) administered in grades 3-10 three times a year to over one million Florida students. Method. The morphological knowledge test consisted of 30 sentences per grade (in two forms). Students read each sentence and selected which of three morphologically related forms of a word best fit in the sentence. For example: Mike's (convocation, vocation*, evocation) is motorcycle repair, even though he majored in Business. Words were selected based on printed word frequency and oral vocabulary level, and sentences were assigned to grade using a readability formula. The computer-based test was administered in fall 2010 to 450 students per grade in grades 3-10 in a Florida school district. Results. Using IRT, items were reduced to those without bias and with good discrimination and difficulty. This final set of items was entered into correlational and regression analyses at each grade to determine relations to FAIR's spelling, text reading efficiency, and reading comprehension data for these same students. Morphological knowledge added unique variance to predictions of reading comprehension above and beyond spelling and text reading efficiency, especially in middle and high school. Conclusion. Including morphological knowledge as part of FAIR should promote instruction in word parts and meanings as they relate to reading comprehension.
Alon Fragman (Ben Gurion University, Beit Berl College)Susie Russak, Beit Berl College - The development of spelling skills in Arabic as a foreign language
This study examines the development of spelling skills in Arabic as a foreign language among native Hebrew speaking students after their first, second and third year of study of the written form of the language 8th, 9th and 10th grades (N= 352). The study focuses on the acquisition of four novel sounds (ح, ع, ق, ص) using five experimental tasks: real and pseudo word recognition, auditory discrimination, dictation of real and pseudo words. We hypothesized that lexical representations for novel sounds will be weak in the beginning stages of language acquisition, yet with time and increased exposure, they will gain stability. MANOVA analyses showed (1) after 3 years of exposure and practice with the written form, there is no significant change in the stability of lexical representations for each of the novel sounds. (2) There was a significant difference on performance per novel sound. The collective scores for the sound ص were the lowest (57%) followed by the scores for the sounds ق (62%) , ح (68%), and ع (74%)respectively. (3) Overall, scores for pseudo word recognition and auditory discrimination were the highest and pseudo and real word dictation the lowest. In general, these findings are in line with the linguistic affiliation constraint (Saiegh-Haddad, 2003) and support findings regarding persistent difficulties with processing novel phonemes (Russak & Saiegh-Haddad, 2010).Further, they highlight the importance of direct and explicit instruction of novel phonemes as a necessary building block in the construction of foreign language knowledge and proficiency.
Christie Fraser (OISE/University of Toronto)Christie Fraser; Adrian Pasquarella; Esther Geva - Conjunction junction, what's your function?: Exploring the role of conjunctions in reading comprehension
Purpose. Conjunctions play an important role in reading comprehension in that they signal connections between text segments required for the reader to effectively comprehend a text. Given that reading comprehension is an area of particular difficulty for ELLs, research exploring specific components of reading comprehension, including conjunctions, is necessary. The purpose of this study was to explore the role of conjunctions in reading comprehension. Specifically, do conjunctions constitute a special type of vocabulary for ELLs? Method. ELLs of varying home-language backgrounds were assessed in Grades 3 and 4 (N = 111). An experimental logical relations task was used to measure the participants' ability to process conjunctions; the NEAL analysis of reading ability was used to assess reading comprehension. A number of additional reading-related tasks (e.g., word reading, receptive vocabulary) were also administered. The primary statistical techniques used in analyses were correlation, and hierarchical regression. Results. Results indicated that the ability to process logical relations contributed significantly to variance in Grade 4 but not Grade 3 reading comprehension, after accounting for receptive vocabulary and word reading. Moreover, participants scored better on the logical relations task than on other higher-order tasks related to reading comprehension (i.e., comprehension monitoring, and inferencing). Conclusions. Taken together, these findings suggest that the ability to process logical relations may be a form of vocabulary knowledge rather than a comprehension strategy; conjunctions may represent a special type of vocabulary. ELLs rely on conjunctions to comprehend the connection between text segments and this ability is related to their reading comprehension.
Silke Fricke (University of Sheffield); Maggie Snowling; Claudine Bowyer-Crane; Ally Haley; Charles Hulme - Building a secure foundation for literacy: An evaluation of a preschool language intervention
Purpose: This project aimed to identify children with oral language weaknesses at preschool, evaluate the effectiveness of a language intervention focusing on vocabulary and narrative skills in nursery and Reception, and investigate if this intervention showed carryover effects for reading and writing. Method: The project ran a randomised controlled trial. 12 children from 15 UK nursery schools (N=180, mean age 4;2) were selected based on poor performance on standardised language measures and were randomly allocated to either an intervention or waiting control group. A peer comparison group was also recruited consisting of 6 children per school (N=90, mean age 4;2) matched on age and gender to the intervention and waiting control groups. The Language4Reading intervention programme is a 30-week-programme delivered by trained Teaching Assistants, designed to begin in UK nursery and to continue during Reception. Children take part in group as well as individual sessions and the programme aims to develop three key areas: vocabulary knowledge, narrative and listening skills. Results: We will present findings from post-intervention and maintenance assessments for the intervention and waiting control group. Results demonstrate that the intervention group showed significant gains in measures of skills which were directly trained (e.g., bespoke vocabulary) but also transfers to measures of grammar, oral narrative and spelling. Conclusions: Oral language intervention can be successfully delivered by trained TAs in nursery and Reception. Such early intervention has a positive impact on language and emergent literacy skills which highlights the role of oral language skills as a foundation for literacy.
Jan Frijters (Brock University); Maureen W. Lovett; Maryanne Wolf; Rose A. Sevcik; Robin Morris - Comorbid attention deficit/hyperactivity disorder and reading disability: Evidence of equivalent remedial outcomes
Attention deficit/hyperactivity disorder (ADHD) shares 15 to 40% comorbidity with reading disability (RD), with evidence that inattentive symptoms are most important to the overlap. The present study reports outcomes from a highly structured language-based intensive reading intervention for 201 2nd and 3rd graders with RD. The small-group intervention included phonological skills training and metacognitive instruction focusing on vocabulary and reading strategies (Morris et al., 2010). ADHD was confirmed in 25.9% via the Diagnostic Interview Schedule for Children IV, along with parent and teacher reports on the Child Behavior Checklist. Reading outcomes were evaluated using growth curve models to generate per-participant slope estimates of remedial response. Comorbid ADHD + RD participants demonstrated equivalent mean response across multiple reading outcomes, including word identification, decoding, and reading comprehension. Lack of a statistically significant growth parameter for each participant was used as the primary metric of non-response (evaluated at α = .05). Growth rates 1.5x this criterion (< 2.94 SE) defined good response. ADHD + RD participants demonstrated equivalent proportions of average and good response (12-44% average and 51-86% good responders), with very low rates of non-response across the two samples (1-5% across measures). While predictors of both categorical remedial outcome and ADHD status were observed (e.g., visual discrimination, processing speed, orthography), no interactions between these two factors were observed. Overall, the results suggest that this type of intervention is as effective for RD alone as for comorbid ADHD + RD.
Douglas Fuchs (Vanderbilt University) - Peer-assisted learning strategies in the elementary grades: Bottom-up v. top-down approaches to scaling up a validated practice
I will describe a study of how to bring to scale an evidence-based literacy program. The study involved 116 intermediate grade teachers in 39 schools from 9 districts in 3 states. Three study groups included one group of teachers who implemented the literacy program as they were trained to implement it (top-down teachers); a second group that was encouraged to modify the program in accordance with students' and schools' needs (bottom-up teachers), and a third group of controls. Findings indicated that students of the bottom-up teachers made the most reading progress across the study.
Lynn Fuchs (Department of Special Education, Vanderbilt University, Nashville, TN)Sarah Powell; Paul Cirino; Douglas Fuchs; Jack Fletcher - Differentiating Intervention for MD With and Without Comorbid RD
Purpose: Few randomized control trials have been conducted to assess the efficacy of intervention for math disability (MD) with and without comorbid reading difficulty (RD), despite that this is the most frequent method for subtyping MD. In this presentation, we summarize what we know based on 4 RCTs that prospectively stratified random assignment according to MD vs. MDRD, draw some general conclusions and raise issues concerning new directions for reading and mathematics intervention. Method: Across 4 RCTs, participants were 298 MD and 343 MDRD third graders. Criteria for academic difficulty were WRAT-3 Arithmetic and Reading subtest scores < 26th %ile with at least 1 (of 2) WASI subtest SS above 79. Intervention occurred 3 times per week, 30 minutes per session for 13 weeks. Syntheses of effect sizes for different instructional methods were derived for word-problem and for computation intervention. Results: Word-problem intervention targeting language (a deficit associated with MDRD) appears more effective for MDRD than MD. For computation intervention, strategic practice designed to address cognitive weaknesses associated with MDRD (phonological deficits, inattentive behavior, working memory deficits) appears more effective for MDRD than MD. Conclusion: Intervention by subtype studies may provide productive directions for treatment and provide insights into the nature of MD vs. MDRD. Findings suggest that the more severe and pervasive disturbances of oral language in MDRD help explain overall severity across math skills and that an adequate explanation of math competencies may require language, regardless of whether language is considered to simply facilitate development or thought to be causally implicated. Given that language is also implicated in reading comprehension and perhaps fluency, research is needed to assess whether language provides a focal point for MDRD intervention.
Jerusa Fumagalli de Salles (Universidade Federal do Rio Grande do Sul)Candice Steffen Holderbaum - Semantic priming effects in a lexical decision task in third graders: comparing three different stimulus onset asynchronies
The understanding about how meaning access in word reading evaluation occurs and which variables are involved on this process can be obtained through different paradigms. One alternative is to use the semantic priming paradigm, in which accuracy and reaction time to a target-stimulus that is preceded by a prime (semantic related or unrelated to the target) are measured. A main issue concerns the absence of semantic priming effects on children at short stimulus onset asynchrony (SOA). Purpose: The aim of this study was to compare the semantic priming effect during a lexical decision task in 81 third graders at different SOAs. Method: In this experimental study, participants performed lexical decisions to targets which were preceded by semantic related or unrelated primes in three different SOA (250, 500 and 1000 ms), in a inter-group design. Data analyses compared mean reaction times between the related and unrelated condition and the magnitude of the semantic priming effects (difference between the unrelated and related conditions) among the different SOAs. Results: Semantic priming effects were found at all SOAs in the third graders. The magnitude of the semantic priming effects were 87ms when the SOA was 250ms, 204ms when the SOA was 500ms and 159ms when the SOA was 1000ms. Conclusions: Children showed semantic priming effects at all SOAs, with faster reaction times when a related prime preceded the target. However, the magnitude of this facilitation varied among the different SOAs. It was larger in 500ms SOA. Hypotheses to explain these data are discussed.
Bjarte Furnes ()Stefan Samuelsson - Orthography and its impact on early reading and spelling acquisition
Purpose Most research on early literacy development originates from English-speaking countries. Recently, however, a growing number of studies from other European orthographies suggest that the rate at which reading and spelling develops as well as the relationship between cognitive and language skills and early reading and spelling development are affected by orthographic regularity. Such findings raise doubts about whether models of early literacy development can be generalized across alphabetic writing systems. Method The study reported is part of an ongoing large-scale International Longitudinally Twin Study of early language and literacy growth conducted in Australia, the United States, Norway, and Sweden. The samples consisted of 750 U.S./Australian children and 250 Norwegian/Swedish children followed longitudinally between preschool and Grade 2. By using structural equation modelling and logistic regression techniques we examined the extent to which various cognitive and language skills (e.g., phonological awareness and rapid automatized naming) predicted individual variation in reading and spelling as well as reading and spelling difficulties across more and less transparent orthographies (Norwegian/Swedish vs. English). Results Results suggested that similarities between languages were bigger then their differences. In addition, the importance of phonological awareness and RAN were strong in the initial phase of learning how to read and spell, but gradually disappeared when prior reading and spelling skills were controlled. Conclusions It seem safe to conclude that, at least in the initial phase of learning how to read and spell, the relationship between cognitive and language skills and literacy development appear highly transferable across alphabetic writing systems.
Susan Gathercole (MRC Cognition and Brain SCiences Unit, Cambridge); Joni Holmes; Julian Elliott - Towards a cognitive phenotype of working memory deficit disorder
Purpose - Problems in working memory are known to accompany many developmental disorders of learning including specific reading difficulties. The purpose of this project was to chart the cognitive and behavioural characteristics of children identified purely on the basis of poor working memory. Method - Over 300 children aged between 5 and 10 years were identified on the basis of low scores (>1SD below population mean) on each of 2 standardised tests of verbal working memory. Measures were taken of IQ, reading and mathematics abilities, and executive functions. Teachers rated problem behaviours associated with attention and executive functions. Some of the children were observed in the classroom and their areas of difficulties noted. Results - Over 80% of the children had either reading or mathematics scores >1SD below the population mean, and the majority of these had scores below this cutoff value in both areas. The children had significantly elevated ratings of inattentivity comparable to that of children with ADHD but not hyperactivity, and showed problems in sustained attention. Classroom observations revealed frequent problems in following instructions, in activities that combined mental processing with storage, and in keeping track in complex tasks. Conclusions - Children with working memory have a high risk of poor academic progress, and have a profile of inattentive behaviour twinned with deficits in standardized tests of attention. Low working memory capacity may lead to frequently to memory overload in structured learning activities, and that these results in failure to maintain critical information and hence to activity failure.
Susan Gathercole (MRC Cognition and Brain SCiences Unit, Cambridge) - Learning difficulties in children with poor working memory: Identification, characteristics, and intervention
Purpose - Problems in working memory accompany many developmental disorders of learning including specific reading difficulties. In recent years there have been significant advances in understanding how to identify these children, what their learning problems and classroom difficulties are, and how to help support them in the classroom. Method - We have conducted several studies focussing on children with working memory scores that fall below the 15th centile. Results - These children have a very high risk - over 80% - of poor attainment in either reading or mathematics or, in most cases, both. They are highly inattentive in the classroom, and have problems in following instructions, and in keeping track in complex tasks. Recent studies show that their working memory skills improve dramatically following an adaptive working memory programme, with some evidence that academic performance improves too after a time lag. Conclusions - Poor working memory skills put children at high risk of poor academic progress, particularly in reading, possibly because the children often experience working memory overload during learning. Intensive working memory training may help overcome these problems, although the extent to which training benefits transfer to real-world learning is yet to be fully understood.
Anna Gellert (University of Copenhagen); Carsten Elbro - The predictive validity of dynamic measures of acquisition of reading fluency and vocabulary - a longitudinal study of children from grade 3 to 4
Purpose: The aim of the study was to investigate the predictive validity of dynamic measures of reading fluency and vocabulary. Method: Dynamic measures were constructed to measure children's potential for developing reading fluency and vocabulary. Dynamic measures of reading fluency were based on repeated reading. Separate measures were taken with words, non-words and text passages. Dynamic measures of vocabulary were based on paired-associate word learning. The dynamic tests were administered to 99 Danish children in grade 3 along with traditional static tests of reading fluency, vocabulary and reading comprehension. The static tests were re-administered to the same children nine months later in grade 4. Results: Analyses of grade 3 data indicated independence between static and dynamic measures (Gellert & Elbro, 2010). This independence allows for a test of the hypothesis that the dynamic measures add independent variance to the prediction of the development from grade 3 to 4. Grade 4 data have been collected but not yet analyzed. Conclusions: So far, studies of the predictive validity of dynamic assessment of vocabulary and reading have focused on children in preschool and the first school years. This study will extend this research to students in grades 3 and 4. In addition, it will extend dynamic measurement into reading fluency.
George Georgiou (University of Alberta) - Is rapid automatized naming (RAN) automatic?
Purpose: Segalowitz and Segalowitz (1993) proposed that a process becomes automatic when three conditions are fulfilled: (a) the mean response time decreases significantly across repetitions or time, (b) the coefficient of variation (SDRT/MRT) decreases significantly, and (c) the correlation between the mean response time and the coefficient of variation is positive. The purpose of this study was to examine if the rapid naming components - articulation time and pause time - become automatic from Grade 1 to Grade 5. Method: Two cohorts of children participated in the study. Cohort 1 consisted of 62 children who were followed from Grade 1 to 3 and Cohort 2 consisted of 60 children who were followed from Grade 3 to 5. The responses of the children on Digit and Letter Naming were recorded and analyzed in a sound editing program. Results: In regards to articulation time, the three conditions were fulfilled in Cohort 1, but not in Cohort 2. This probably suggests that articulation time is automatic already by Grade 3. In regards to pause time, the mean decreased across time, but the coefficient of variation increased across time. The correlations between the mean pause time and the coefficient of variation were negative. In addition, for Letter Naming in Cohort 1, none of the correlations between the mean pause time and the coefficient of variation was significant. Conclusion: The findings of this study suggest that with the exception of articulation, the rest of the processes involved in rapid naming become faster, but not automatic.
Hope Gerde (Michigan State University); Trish Finger; Ashley Zientek; Lori Skibbe; Jessica Barnes - Relations between name writing, dictation, and letter knowledge for American Indian children in Head Start
Purpose: Name writing and letter knowledge are closely related (e.g., Bloodgood, 1999); however, there is debate regarding whether name writing represents emergent writing or is simply a memorized drawing (Drouin & Harmon, 2009). Thus, understanding how name writing relates to other aspects of children's writing, particularly for children at-risk for academic difficulties is essential. The present study examines how name writing relates to writing dictated letters as well as letter knowledge for American Indian children attending Head Start. Method: Children (N=75; age M=58 months in spring) were tested individually in the spring of pre-kindergarten. Children wrote their name and three dictated letters (i.e., A, B, and C); they also identified up to 10 randomly chosen letters (Mean=4.39, SD=3.5). Both types of writing were coded from 1 (scribbling) to 9 (name spelled correctly or correct letter/good form). Although there was ample variation in children's writing, on average, children wrote their name using letters and letter-like shapes and wrote dictated letters using letter-like shapes (i.e., a lower level of sophistication). Results: Analyses identified that name writing was significantly correlated to writing each dictated letter (r=.61-.67). Further, both name writing (r=.49) and writing dictated letters (r=.35-.47) significantly related to children's letter knowledge. Further analysis will include comparison to 117 non-Head Start children. Conclusions: Children wrote their name and dictated letters using related forms, although the level of sophistication varied by task. Also, both types of writing related to children's letter knowledge, suggesting that both may tap into similar aspects of emergent literacy knowledge.
Jennifer K. Gilbert (); Donald L. Compton; Devin M. Kearns - Word and person effects on decoding accuracy
Purpose: The purpose of this study was to extend the literature on decoding by bringing together 2 lines of research, person and word factors that affect decoding, using a crossed random-effects model. Method: The sample was comprised of 196 grade 1 students who were at risk for developing reading difficulties. A researcher-developed pseudoword list was used as the primary outcome measure. Results: Because grapheme-phoneme correspondence (GPC) knowledge was treated as person and word specific, we are able to conclude that it is neither necessary nor sufficient for a student to know all GPCs in a word before accurately decoding the word. And controlling for word-specific GPC knowledge, students with lower phonemic awareness and slower rapid naming skill have lower predicted probabilities of correct decoding than counterparts with superior skills. This means, for example, that a student with lower phonemic awareness would be less likely than a student with higher phonemic awareness to decode a word correctly even when both students know the GPCs in that word. Results also reveal that, again controlling for word-specific GPC knowledge, words are more difficult if they contain an infrequent as compared to a frequent rime and a complex vowel grapheme as compared to a complex consonant grapheme. By assessing a person-by-word interaction, we found that students with lower phonemic awareness have more difficulty applying knowledge of complex vowel graphemes compared to complex consonant graphemes when decoding unfamiliar words. Conclusions: Implications of the methodology and results are discussed in light of future research.
Sandra Gillam (Utah State University)Jason Anthony; Jeffrey Williams; Lillian Duran; Rachel Aghara; Paul Swank; Mike Assel; Susan Landry - Spanish phonological awareness: Dimensionality and sequence of development during the preschool and kindergarten years
Purpose: This study examined the dimensionality and sequence of emergence of Spanish phonological awareness (PA) skills in 3- to 5-year-old children. Method: A 3 by 4 factorial design crossed word structure (word, syllable, phoneme) with task (blending multiple-choice, blending free-response, elision multiple-choice, elision free-response) to provide examination of children's performances on 12 phonological awareness skills. Over 1200 Spanish-speakers were assessed at the beginning and end of the school year. Results: Confirmatory factor analyses demonstrated a unifying latent ability explained much of children's performances on the PA tests. Item response theory analyses demonstrated that items varied in difficulty and how well they discriminated individual differences in latent PA. Item parameters were stable across item sets (rs=.75-.86) and time (rs=.60-1.00), and ability estimates were moderately stable across time (r=.64). Conclusions: Examination of item characteristic curves revealed that word structure had little effect on item difficulties but that assessment method had substantial effects on both item difficulties and item discriminations. Implications for assessment, early intervention, and cross-linguistic theories of phonological awareness will be discussed.
Darlene Godoy de Oliveira (Darlene Godoy de Oliveira)Ana Navas, Alessandra Seabra, Elizeu Macedo - Properties of eye movement and pupil diameter in adult good readers in a lexical decision task: effect of lexicality and frequency.
Purpose: Lexical decision tasks assess word recognition providing measures of access and level of lexical development. Moreover, obtaining measurements of eye movements and pupil diameter provides consistent psychophysiological parameters about cognitive engagement. This study examined the effect of lexicality and frequency on eye movements and pupil diameter in a lexical decision task. Method: Participated 20 university good readers (mean age 22.3). On task, the psycholinguistic characteristics of frequency, regularity and length were distributed between the words. Pseudowords were created with changes of letters from words. The stimuli were presented on a monitor attached to an eye-tracking equipment (Tobii ®). Participants pressed buttons as quickly when they judged the stimuli, after presentation for 2 seconds. Results: Repeated measures ANOVA indicated significant effect of lexicality on number of correct answers (F [1,19] = 11.150, p = 0.003), with greater accuracy for pseudowords, and number of fixations (F [1,19] = 21.494 , p = 0.000), with pseudowords less fixed. For variable frequency, there was significant effect on number of correct answers (F [1,19] = 25.264, p = 0.000) for high frequency words. The low frequency words had higher reaction time (F [1,19] =9.255, p = 0.007) and pupil diameter in inspection of word (F [1,19] = 6.150, p = 0.023) and during the reaction time(F [1,19] = 13.576, p = 0.002). Conclusions: The pattern found for the lexicality indicated greater ease of semantic processing, while effects of frequency showed that accumulated experience with reading facilitates the processing by the measures used in study.
Clara Gomes ()Araceli Valle; Sylvia Defior - Vocabulary training to improve reading comprehension: A randomized trial with Spanish-speaking children
Purpose: Vocabulary knowledge is critical for development of reading comprehension. Additionally, children from families with low social economic status are at a higher risk of developing vocabulary deficit. Our goal was to examine the efficacy of two types of vocabulary training both based on oral language to ameliorate children's vocabulary knowledge and consequently to foster the development of reading comprehension abilities in primary school years. Methods: 89 third graders from socially and economically disadvantaged school areas in Spain participated in this 2-month language intervention. Children were randomly assigned to either one of the target intervention groups or to a control group. The intervention groups consisted of (1) direct instruction of definitions of words and (2) instruction of words in context that included teaching of context strategies for deriving meaning of unknown words. In the control group, books were read aloud to the children without any explicit teaching of vocabulary. Children were assessed before and immediately after the delivery of the intervention. A third assessment point is planned for May 2011 to check for long-term effects of the intervention. Results: Based on pilot data and the literature for English-speaking children, we expected vocabulary knowledge for the taught words to improve in both definition and context intervention groups from pre to post-tests. Nevertheless, a better general vocabulary knowledge (non-taught words) was expected for the context group in which strategies were taught. Results from immediate and delayed post-tests will be discussed. Relevance: This study utilizes randomized controlled trial testing with a Spanish population.
John Goodrich (Department of Psychology, Florida State University); Christopher Lonigan; JoAnn Farver - Do oral language skills impact the relationship between first and second language phonological awareness skills in Spanish-speaking English language learners? An evaluation of the lexical restructuring model.
Purpose: The Lexical Restructuring Model (LRM; Metsala & Walley, 1998) states that as children's vocabularies increase they develop a more refined representation of the sounds of which those words are comprised, and in turn become more sensitive to the detection of phonemes. In addition, many studies have demonstrated that among English language learners (ELLs) there exists a strong positive relationship between first and second language phonological awareness skills. This study examined whether or not the relationship between children's first and second language phonological awareness skills is moderated by their oral language skills. Method: In this study, 466 children classified as Spanish-speaking ELLs were administered measures of early literacy in both Spanish and English as part of a larger study, including measures of expressive vocabulary, blending, and elision skills. Results: Analyses indicated that Spanish expressive vocabulary skills moderated the relationships between both Spanish and English elision and blending skills, and English expressive vocabulary skills moderated the relationship between English and Spanish elision skills. However, English expressive vocabulary skills did not moderate the relationship between English and Spanish blending skills. Probes of these interactions indicated that, for all interactions, as expressive vocabulary skills increased, the strength of the relationship between first and second language phonological awareness skills increased. Conclusions: This evidence supported the LRM for development of phonological awareness for ELLs, demonstrating that there may be a need for an intervention for poor vocabulary skills to precede an intervention for poor phonological awareness skills.
Amanda Goodwin (Peabody College, Vanderbilt University) - The unique role of morphological awareness compared to phonological recoding in predicting reading and vocabulary achievement for Spanish-speaking English language learners
Purpose: The current study extends knowledge of morphological awareness beyond existing understanding by presenting a new measurement model of morphological awareness and examining the particularly salient role of morphology for English language learners (ELLs).This study examined unique contributions of fourth grade morphological awareness and phonological recoding to fourth grade oral vocabulary and fifth grade word reading, reading comprehension, and reading vocabulary. Method: For 197 Spanish-speaking ELLs, structural equation modeling was used to separate morphological awareness from phonological and orthographic demands present in opaque morphological relationships by creating a latent variable stemming from the shared variance of four morphological tasks with different levels of morphological transparency. In addition, 3 structural models were investigated. Results: Findings indicated that when controlling for phonological recoding, morphological awareness made a significant and meaningful contribution to passage comprehension and reading vocabulary, but not word decoding, with oral vocabulary acting as a mediator. The study also showed that phonological recoding was a significant predictor of each reading outcome when morphological awareness was not included as a predictor, but only significantly predicted word reading when controlling for morphological awareness. Conclusions: Understanding how morphological awareness and phonological recoding uniquely contribute to oral vocabulary, word reading, reading comprehension, and reading vocabulary can help researchers design interventions to improve achievement. By including morphological awareness in the model, the interpretation of the role of phonological recoding changes, suggesting more of a need for morphological instruction. Significance of these findings to research and educational settings are discussed.
Alexandra Gottardo (Wilfrid Laurier University)Adrian Pasquarella - Reading comprehension in adolescents: Back to basics
Reading comprehension is crucial for school success, especially in adolescents. However, most recent work has examined strategies assumed to facilitate reading comprehension in adolescents (Graesser, 2007; McNamara, 2007). In contrast, research examining reading comprehension in younger children has focused on building models based on relevant cognitive-linguistic subprocesses. The current analyses examine variables related to performance on standardized measures of reading comprehension based on the simple view of reading. Fifty-five English-speaking adolescents were tested on several reading and language measures (mean 15 years, S.D. 1.1 yr). Analyses were conducted including two measures of reading comprehension, the Gates-MacGinitie and the passage comprehension of the Woodcock Language Proficiency Battery. Although word reading was significantly related to reading comprehension when entered in the second step of a hierarchical regression analysis (Beta = .559, p < .001), it was no longer significantly related to reading comprehension when measures of vocabulary and grammatical awareness were entered. The measure of vocabulary knowledge based on the knowledge of synonyms and antonyms was significantly related to reading comprehension on both measures (Betas =.374 & .47, p < .05) when decoding and other measures of vocabulary (PPVT-III, EOWPVT) were entered. Morphological knowledge was also related to reading comprehension performance on the Gates-MacGinitie when decoding and other language measures were entered (Beta = .253, p < .05). Therefore, decoding is not strongly related to reading comprehension as was found with younger skilled readers (Catts, Hogan & Adolf, 2005). Additionally, the results show that research examining models of reading comprehension in adolescents should continue to inform practice.
Amy Grant (); Amy Grant; Alexandra Gottardo; - The role of genre and activity preferences in early reading development: Are there gender differences?
Recently published, local policy research has focused on "gender differences" in reading development based on the results of large-scale international studies (e.g., Council of Ministers of Education, 2007), and how early differences in genre preference may be related to differential development of reading skills (Ontario Ministry of Education, 2004, 2009). Empirical research, however, has found mixed results as to whether these differences primarily exist early in development (Halpern, 1997; Soderman et al., 1997), and disappear later in the school system (Feingold, 1988; Hogrebe, Nist, & Newman, 1985), or do not exist at all (Harper & Pelletier, 2008; White, 2007). The current study investigated the role of genre and activity preferences and their relation to reading skills within a sample of Grade 1 readers (N = 42, 26 females). Preliminary analyses revealed no gender differences on both standardized and unstandardized measures of vocabulary, decoding, phonological awareness, rapid naming, and print exposure (all p › .054, most p › .120). Additionally, there were no significant differences in the percentage of fiction (t = .84, ns) or non-fiction (t = -1.25, ns) books boys and girls read during the duration of the school year, or on an adapted measure of Stanovich and West's (1989) activity preference questionnaire (all p › .17) which included new categories such as videogames, and trading cards in addition to the traditional preference categories of reading and television. There was also no significant relation between the genre children reported reading, and their reading skills, nor between children's activity preferences and their genre preferences. These results show that with Grade 1 students, there is evidence against notions regarding what boys like to read, and how they are performing in school.
Katherine Green (Georgia State University)Gary Bingham, Nicole Patton-Terry, Kizzie Albritton - Examining the relative importance of children's language, literacy, and social skills to invented spelling
Purpose This study was designed to examine the relative contribution of children's oral language, print awareness, letter knowledge, and social skills to their development of invented spelling. Method Two hundred forty seven African American and Latino four-year-old children participated in this study. Children's early language and literacy development was assessed at the beginning and end of prekindergarten on the following measures: The Peabody Picture Vocabulary Test, 4th Edition (PPVT-4: Dunn and Dunn, 2007) the Test of Preschool Early Literacy (TOPEL: Lonigan, Wagner, Torgesen, & Rashotte, 2007), Rhyme Awareness Task (PALS-PK: Invernizzi, Justice, Landum, & Booker, 2004), Name writing (PALS-PK: Invernizzi et al., 2004 ), Invented spelling (PALS-K; Invernizzi, Sullivan, & Meier, 2001). Parents completed an evaluation of their child's social skills (Fantuzzo, Mendez, & Tighe, 1998). Results Multiple regression analyses were used to explore the relation between children's oral and written language (IVs) competencies in the fall of prekindergarten (pre-k) and their invented spelling (DV) at the end of pre-k. Results reveal positive associations between invented spelling and children's vocabulary (β = .27, p < .001), letter knowledge (β = .42, p < .001), and social skills rating (β = .19, p < .05), but not their print awareness or phonological awareness. Conclusions Because children's invented spelling skills generally emerge before children are established readers many believe that both oral and print related skills are important to its development (Gentry & Gillet, 1993; Molfese et al., 2006; Ouellette & Sénéchal, 2008). This study adds to existing research on children's development of writing skills by highlighting the role that children's social development skills, most likely through their ability to stay on task, play in children's invented spellings.
Luana Greulich (Andrews University)Stephanie Al Otaiba;Jessica Folsom;Jane Meadows - Where Do We Go From Here? A Literature Review of Response to Intervention
Purpose of Review The purpose of this literature review was to answer three research questions related to RtI. First, we described criteria researchers used to identify students in need of intervention and to what extent researchers determined the quality of Tier 1. Second, we examined the nature of intervention (intensity and duration) that students received at Tiers 2 and 3, and whether students moved between tiers. Third, we examined how RtI was defined and what proportion of students did and did not respond to intervention. Methods Electronic and hand-searches were completed and six criteria were met. First, studies were published in peer-reviewed journals after 2000. Second, studies implemented at least Tier 2 intervention or Tier 3 intervention. Third, studies included an appropriate control group. Fourth, studies were limited to those that included samples from grades K-3. Fifth, studies were included only if they reported which measures were used for progress monitoring. Last, studies must have reported students' responsiveness to the intervention. Results The results are varied across the 17 studies, due to the variety of definitions, measures,and methods that were used to determine eligibility and responsiveness. Therefore, we identified communalities across studies such as intensity and duration, but it was impossible to compare responsiveness across studies due to the significant differences in the study designs and curriculums used. Conclusion Finally, methodological issues are discussed associated with determining non-responders, as well as strengthens and weaknesses of RtI, which inform the direction of further research and provide valuable information on implication to practice of RtI models.
Vibeke Grover (Institute for educational research, University of Oslo);Veslemøy Rydland - Peer text discussion and reading comprehension
Purpose This study examines under which conditions peer group discussions of science texts with multiple and divergent perspectives promote students' text comprehension. More specifically, the study addresses whether students' prior topic knowledge as well as peer group composition impact the extent to which group discussions of main text ideas promote text comprehension. Method 410 fifth-grade students (202 boys) in 21 multiethnic classrooms in Norway participated. The students read multiple texts (a narrative text, an editorial letter and a text-book chapter) addressing the science topic of global warming before participating in group discussions on main text ideas with peers. Reading comprehension was assessed prior to (pre-comprehension) and subsequent to (post-comprehension) the peer discussion, using different, but equivalent researcher-developed comprehension tests. Before text exposure, students' prior topic knowledge was assessed. Results The students whose prior topic knowledge was below the 25th percentile, scored significantly better on post-comprehension compared to pre-comprehension while the other percentile groups did not raise their post-comprehension scores. In fact, students with prior topic knowledge above the 75th percentile, scored significantly worse on post-comprehension. Moreover, group competence composition (prior topic knowledge and pre-comprehension) explained some variance on students' post-comprehension. Conclusion Students' prior topic knowledge as well as group competence composition impacted the extent to which the students' comprehension benefitted from peer group discussions in this specific reading task involving multiple texts with divergent perspectives. Educational implications concerning facilitation of classroom support for reading comprehension will be emphasized in the discussion.
Paula Guardia (University of Cambridge); Usha Goswami - Neighbourhood Density Effects on Short-term memory, Rapid Automatised Naming and Reading Skills in Spanish.
Purpose. Accurate phonological representations are crucial for efficient phonological and reading skills. Neighbourhood Density (ND), a lexical factor, refers to the number of similar-sounding words stored in long term memory, and which provides more complex pattern of influence on the restructuring of phonological representations than vocabulary. Theoretically, words from Dense neighbourhoods are more accurately represented than words from Sparse neighbourhoods because Dense words must be distinguished from a large number of other very similar-sounding words (Goswami, 2001). More efficiency in access and retrieval is expected for Dense neighbourhood. The purpose was to explore the effect of ND on STM and RAN tasks and to their prediction of reading in Spanish. Method. 94 non-readers (mean age 5.7) were followed up to first grade. ND was experimentally manipulated in STM and RAN tasks (adapted from Thomson, 2004). Reading (accuracy, speed, comprehension) was measured at the end of first grade. Results/conclusions. For STM, children recalled words from Dense neighbourhoods more accurately. For RAN, children named words from Sparse neighbourhoods faster (p<0.01). Similar results were found in English (Thomson, 2004). Despite the advantages on speed retrieval for words from Sparse neighbourhoods, the results showed that STM and RAN Dense made bigger contributions to explain reading than Sparse. The better the children performed in Dense tasks, the better they performed in reading. STM Dense explained reading accuracy. RAN Dense explained all reading measures. The results, pioneering in Spanish, suggested that STM/RAN Dense seem to be good proxy measures of the quality of the mental lexicon and possible loci for intervention.
Ying Guo (Children); Joan Kaderavek; Laura Justice - A standardized tool for assessing the level of children's orientation to book reading: The C.O.B (Children's Orientation to Book Reading Rating Scale)
Purpose: The purpose of this study was to examine the psychometric characteristics of the Children's Orientation to Book Reading Rating Scale (C.O.B), an observational tool designed to characterize children's level of orientation (i.e., interest, engagement, and attention) during adult-child shared book reading. Two research aims were addressed: (1) to document the reliability and validity of C.O.B via computation of inter-rater reliability and correlation between the C.O.B. and parent-report of children's literacy interest, and (2) to document the predictive role of children's engagement on children's early literacy skills. Method: Participants included 46 preschoolers and 17 teachers. Researchers used the C.O.B. to code children's level of engagement during videotaped, whole-class shared-reading sessions collected from each teacher. Parent-reported measures included children's attention and effortful control skills. Children early literacy measures included print concept knowledge, letter knowledge and emergent reading skills. Results: Results showed that inter-rater agreement was high with interclass correlation coefficients of .86. C.O.B is significantly correlated with parent-reported children's attention (r = .42) and effortful control (r = .34). Regression analysis demonstrated that after controlling for children's SES and language ability, the C.O.B was a significant predictor of children's letter knowledge (β = .33, p = .02) and trended towards predicting children's overall emergent reading skills (β = .31, p = .08). Conclusion: The C.O.B. demonstrated adequate psychometric characteristics. This tool provides researcher and applied professionals a means of assessing children's orientation to literacy and can be used to explore the impact of engagement on literacy learning.
Stephanie Guthrie (University of Michigan)Frederick Morrison - Kindergarten classroom instruction: What happens and what matters for literacy growth?
Purpose A great deal of evidence suggests that effective literacy instruction is differentiated based on child knowledge (e.g., child by instruction interactions). In addition, recent policy initiatives emphasize the importance of literacy instruction in kindergarten. This paper explores instruction during a typical day in kindergarten classrooms, and reveals the types of literacy instruction that predict reading achievement in a sample of average- to high-ability students. Method 68 kindergarten classrooms were observed 3 times per year, and instruction was coded for activity type and management. Codes were summed according to subject (e.g., literacy, non-instructional, math), activity (e.g., code/meaning-focused, basic number skills), grouping (whole class or small group), and management (teacher or child). 158 participating children's phonological awareness, passage comprehension, and decoding skills were also assessed. Results Instructional activities changed in duration and frequency over the course of the kindergarten year, with the majority of time focused on literacy and non-instructional tasks. Within literacy, meaning-focused activities were the most common. HLM analyses were conducted, with each model controlling for initial scores and phonological awareness. Teacher-managed meaning-focused instruction was a significant predictor of both outcomes (βdecoding = .0057; βpassage comp = .0093); however, teacher-managed code-focused instruction was not significantly related to kindergarteners' reading development (βdecoding = -.0008; βpassage comp = -.0006). Discussion These findings provide additional information about the focus of kindergarten instruction. Moreover, the results support previous research regarding child by instruction interactions, and particularly the best instructional practices for average and above-average readers.
Shanna Hagan-Burke (Texas A&M University)Michael Coyne - The effects and interactions of student, teacher, and setting variables on Kindergarteners' response to early reading intervention
Purpose: This study examined the separate influences of student, teacher, and setting factors on the reading achievement of at-risk kindergarteners and investigated whether the type of supplemental reading intervention (i.e., explicit, code-based reading instruction or typical school intervention) moderated the influences of student, teacher, and setting factors on end-of-kindergarten reading performance. Method and Results: Students in the experimental condition received explicit, systematic, code-focused instruction; students in the comparison condition received their schools' typical intervention practices. Using multilevel structural equation models, we (a) investigated the separate influences of students' entry-level language and literacy skills, teacher knowledge and skills, and setting characteristics; and (b) examined whether type of reading intervention moderated their influences. Among student variables, entry-level alphabet knowledge and rapid object naming (RON) were statistically significant predictors of end-of-kindergarten outcomes, but RON's influence on phonemic outcomes was stronger in the comparison condition. In the teacher model, higher instructional quality scores predicted better decoding outcomes in the comparison condition, but had no statistically significant influence on decoding outcomes in the intervention group. In the setting model, intervention delivered via pull-out (versus in-class) and smaller instructional group sizes benefitted phonemic outcomes in both conditions. Conclusions: Findings suggest that student, teacher, and instructional variables all influenced kindergarteners' response to beginning reading interventions. These findings underscore the complexity of understanding the effects of early reading interventions and highlight the need for future research to unpack and make sense of this complexity in ways that provide support to those who must make ongoing decisions about how to best provide and adjust intervention supports to children experiencing reading difficulties.
Corinne Haigh (Bishop's University); Robert Savage; Caroline Erdos; and Fred Genesee - The role of phoneme and onset-rime awareness in second language reading acquisition
Purpose: This study investigated the link between phoneme and onset-rime awareness and reading outcomes in children learning to read in a second language. We sought to: 1) contrast the predictive validity of well-matched phoneme and onset-rime blending tasks in biliteracy development, 2) examine the specific link between phonological awareness and reading and of the role of letter knowledge both within and across languages. Method: Phoneme and onset-rime awareness, word identification, and letter knowledge tasks were administered in English and French in the spring of kindergarten to English-dominant children in early French Immersion programs (n = 98). A battery of tests of reading achievement in English (word and pseudoword decoding, and word and pseudoword reading fluency) and French (word and pseudoword decoding, word and pseudoword reading fluency, speed, accuracy, and fluency of reading in connected text, and reading comprehension) were administered at the end of grade 2. Results: Regression analyses indicated that English phoneme manipulation was a significant predictor of both English and French reading outcomes after controlling for kindergarten knowledge of letter names and word identification. French onset-rime knowledge measured in kindergarten accounted for significant variance for French reading outcome measures. Conclusions: Results support the existence of a strong and probably causal link between English phoneme manipulation in kindergarten and both English and French reading outcomes in grade 2. Practically, these results provide information about what phonological awareness measures can be used in kindergarten to predict later reading outcomes for children learning to read in a second language.
Stephen Hamilton (University of California, Davis)Jane Oakhill; Alan Garnham - Automatic Gender Stereotyping, an ERP Investigation
Purpose To investigate neurological and behavioral responses to violations of definitional and stereotypical representations of gender Method Event-related brain potentials (ERPs) and reaction times (RTs) were recorded as participants took part in a gender-priming paradigm. Participants had to classify target pronouns as male or female after they were exposed to gender related primes. Primes were related to gender definitionally (e.g. mother, father), or stereotypically (e.g. nurse, doctor). Results When stereotype primes were paired with gender incongruent target pronouns (e.g. nurse-he) the pronoun elicited a negative-going wave with features similar to an ERP component associated with semantic and pragmatic anomalies (the N400). When definitional primes were paired with incongruent pronouns (e.g. father-she) the pronoun elicited a positive-going wave with features in common with an ERP component related to syntactic anomalies (the P600). RTs also varied as a function of prime and target congruency, with incongruent prime-target parings evoking longer response latencies than congruent prime-target pairings. Conclusions The results are taken as evidence that single-word stimuli are able to invoke robust representations of gender, even when those representations are based upon stereotypical assumptions. The proposal that this effect occurs as a consequence of automatic processing (e.g. Banaji & Hardin, 1996) is also discussed.
Brenda Hannon (Department of Psychology) - A New Measure for Assessing the Contributions of Higher-level Component Processes to A new tool for assessing the contributions of higher-level cognitive processes to oral comprehension in pre-readers.
Purpose: The present study reports the development of a theoretically-motivated measure that provides estimates of a pre-reader's ability to recall text, to make text-based inferences, to access knowledge from long-term memory, and to integrate this accessed knowledge with new information from the text. Methods: The new pre-reader task is based on the adult versions developed by Potts and Peterson (1985) and Hannon and Daneman (2001), but it uses pictures and auditory text to make it more suitable for children 4-6 years of age. Children completed the pre-reader component processes task as well as measures of comprehension, word decoding, working memory, and fluid intelligence. A number of regression analyses were completed in order to determine the relative predictive powers of the pre-reader task versus word decoding, working memory, and fluid intelligence. Results: The results showed that the pre-reader task accounted for more than 50% of the variance in comprehension performance. In fact, the pre-reader task was a better predictor of comprehension performance than were measures of either working memory or word decoding skills. Finally, the pre-reader task was a very good predictor of performance on a composite measure of fluid intelligence. Conclusions: The pre-reader component processes task is a very good predictor of the contributions of higher-level cognitive processes to performance on measures of comprehension and fluid intelligence.
Lindsay Harris (University of Pittsburgh); Michal Balass; Charles Perfetti - Detecting readers' orthographic and semantic processes with error-related negativities (ERNs)
Purpose: Recent ERP research has indicated that the error-related negativity (ERN) is sensitive to errors committed during simple linguistic tasks involving speech perception and lexical decision. The waveform's robustness to errors in high-level language tasks, such as access to word form and meaning has not been tested. We designed two studies to determine whether complex orthographic and semantic knowledge are stable enough to elicit the ERN. Methods: In Study 1, ERPs to correct and incorrect responses were recorded during a time-pressured, forced-choice spelling judgment task in 24 adults. In Study 2, ERPs to correct and incorrect responses were recorded during a time-pressured, forced-choice semantic categorization task in 28 adults. Data was collected using a 128-channel Geodesic sensor net and analyzed using Net Station software. Results: In both studies, ERNs were reliably observed between 30ms and 80ms post-response for both error types (false alarms and misses). D-prime sensitivity measures revealed a relationship between ERN magnitude and readers' response criterion, with increased stimulus discrimination resulting in a greater ERN effect. Conclusions: The ERN can be elicited in a high-level language processing task involving word reading, and its magnitude is associated with respondents' form and meaning discrimination confidence. The results suggest that the stability of an individual's word-form or semantic knowledge can be gauged by the amplitude of the ERN waveform component, making it a potentially useful tool to examine learning.
Sara Hart (Department of Psychology, Florida Center for Reading Research)Christopher Schatschneider; Jeanette Taylor - Developmental genetic and environmental influences on oral reading fluency
Purpose. The present study examined the development of genetic and environmental influences on oral reading fluency (ORF) from first to fifth grade. Method. Participants were drawn from the Florida Twin Project in Reading, a longitudinal sample of 4300 twins from Florida. Data were collected on each child as part of Reading First and entered into a state-wide system known as the Florida's Progress Monitoring and Reporting Network. These analyses are based ORF from DIBELS. Results. Development was measured using the fall assessment point across five school years. Results from latent growth curve modeling suggested significant genetic and child specific environmental overlap among the intercept centered at first grade, slope and quadratic latent factors. Moreover, beyond those influences, there was significant shared environmental and nonshared overlap between the slope and quadratic factors. As a next step, the more sensitive simplex model was used, which models novel genetic and environmental influences at each time point ( "innovations") as well as the "transmission" of genetic and environmental influences from one time point to the next. Results suggested that while there was significant genetic and environmental transmission across all time points, differential innovation was found. Novel genetic influences were measured for first, second and third grade only. Shared environmental innovation occurred only in the first grade, and nonshared environmental influences occurred at every grade. Discussion. These results highlight the developmental process of genetic and environmental influences, particularly identifying the environmental effects on growth which appears to be a product of entrance into formalized education.
Hengameh Hassan-Yari (Queen)John R. Kirby; S. Hélène Deacon - Morphological priming, morphological awareness, and reading ability
Purpose We examined the effects of reading ability and morphological awareness on Deacon et al.'s (2010) Fragment Completion priming task, in which participants are shown five types of prime (either the same word, an inflected word, a derived word, an orthographic control, or no prime), and then shown word fragments (e.g., p _ _ t) and asked to fill in the blanks to make real words. We hypothesized that morphological priming in Fragment Completion would be related to morphological awareness and reading ability. Method 147 children were tested in Grades 3 and 4. In Grade 3 they received measures of IQ (Block Design, Vocabulary), phonological awareness (PA: Elision, Word Blending), Orthographic Processing (OP: Orthographic Choice, Wordlikeness, WordChains), morphological awareness (Word Analogy, Sentence Completion), and reading (Word Identification, TOWRE, Passage Comprehension). In Grade 4 they completed the Fragment Completion task. Results Priming occurred (means: Identity=2.23, Inflected=1.76, Derived=1.69, OrthoControl=1.30, and NoPrime=.98). Regression analyses indicated that priming scores were predicted significantly by reading ability or morphological awareness, whether IQ, PA, or OP were controlled or not. Conclusion Contrary to the results of McCutcheon et al. (2009) with a lexical decision task, the present results with a spelling task show that morphological priming is related to both reading ability and morphological awareness. We discuss the implications for the measurement of morphological processing and awareness, and the consequences for theories of reading.
Michael Hebert (Vanderbilt University)Steve Graham - The impact of writing assessment: A meta-analysis
Purpose: Despite its importance as a critical skill, many students still experience difficulty with reading (Lee, Gregg, & Donahue, 2007). Writing is an often overlooked avenue for improving reading skill. Reading Next (Biancarosa & Snow, 2004) identified writing as a key element of good literacy instruction. Writing to Read (Graham & Hebert, 2010) confirmed that students comprehend text better if they write about it, and become better readers as a result of writing instruction. To take advantage of these benefits, students need to be provided high quality writing instruction. A common assumption is that teacher, peer, and self-evaluation of writing progress will make students better writers. As the National Commission on Writing (2003) noted, assessment has "an important role to play in the improvement of the teaching and learning of writing" (p. 21). Method: This review presents findings from meta-analyses of experimental and quasi-experimental studies designed to answer the following questions: 1) Does overall writing performance improve when assessment is included as part of classroom instruction? 2) Are there specific factors that influence the effects of writing assessment? Results and Conclusions: Results showed that students' overall writing performance is improved when teachers monitor student progress; students are taught how to assess their own progress; or teachers provide students with feedback on their writing. Additionally, several factors of writing assessments were found to influence the effects of holistic writing quality scores, including technology use during the assessment; typed versus handwritten texts; legibility; spelling; grammar; length; and scorer familiarity with the writer.
Lindsay Heggie (Queen's University)Lindsay Heggie; Lesly Wade-Woolley; Tosca Burtenshaw - Adults' sensitivity to derivational suffixes in the reading of morphologically complex words
Purpose. Most multisyllabic words in English are also morphologically complex. Research has shown that readers use morphological information to guide them in reading these words (Jarmulowicz 2007; Kuo & Anderson, 2006). Derivational morphology, through Level 1 (stress-neutral; PUnish-PUnishment) and Level 2 (stress-shifting; STAble-staBIlity) suffixes, provides valuable information for reading multisyllabic words. Previous work has shown that derivational suffixes guide readers' application of stress (e.g., Clin, Wade-Woolley, & Heggie 2009). In Wade-Woolley (2007), adults appear inexplicably more sensitive to Level 2 than Level 1 suffixes. But here, readings of stems and derivations were divided by intervening tasks; performance may therefore simply reflect a lack of memory as to where the stress was placed on the stems when derivations were read. Method. Adult participants (n=56) read lists of pseudoword stems with stress-neutral and stress-shifting suffixes. Half read words in stem-derivation pairs, and half read all stems immediately followed by all corresponding derivations. We expected that accounting for distance effects would improve adults' reading of stress-neutral derivations such that performance on Level 1 and Level 2 words would not be significantly different. Results. Counter to our expectations, and in keeping with previous findings, participants more accurately interpreted Level 2 suffixes than Level 1, regardless of presentation (stem-derivation pairs or all stems, all derivations). Conclusion. Level 1 suffixes appear not to share as much information, or perhaps not as saliently, as Level 2 suffixes. No stress shift is unlikely to be the default state for suffixation, as it is obeyed less consistently. Implications for the reading of morphologically complex words, as well as future research directions, will be discussed.
Elfrieda H. Hiebert (University of California, Santa Cruz);Guy Trainin; Kathleen Wilson - Comprehension and reading rates across extended grade-appropriate texts
Purpose: For the first time in standards documents, the Common Core State Standards (CCSS) has designated grade-levels for text. These levels are ambitious and reflect increasingly more complex text features and content. The CCSS grade-level exemplars also reflect the expectation that students-even second graders-have the ability to read extended texts. Research on students' persistence while reading silently is limited. Method: This study examined the comprehension and reading rates of 42 second- and fourth-graders with grade-level texts. The texts were on a similar informational topic (G2=early homes; G4=temporary homes) but reflected age-appropriate features for length (G2=600 words; G4=1,000 words) and lexiles (G2=430, G4=890). Texts were divided into four sections, each followed by four questions. Computer administration allowed the examination of rate and comprehension for different sections. Results: Thirty percent of second graders and 21% of fourth graders were designated as noncomprehenders since they failed to attain 70% comprehension. Second-grade comprehenders had lower reading rates than fourth graders (G2=183; G4=198) but similar comprehension (G2=77%; G4=73%). For both grades, comprehenders had consistent rates and comprehension across segments 1 to 3 but rates increased and comprehension decreased for segment 4. Conclusions: The pattern on the final segment suggests that students' reading persistence may be influenced by factors in addition to text features. Slower readers speeded up when peers in the computer lab had completed the task. This influence can be eliminated by individual assessments. However, the manner in which various factors influence students' persistence in classroom reading tasks and tests requires attention.
David Hill (University of Pittsburgh)Seth King; Dr. Christopher Lemons - Critical components of elementary RTI models: an examination of coherence and fidelity in tiers 1 & 2.
Purpose Response to Intervention (RTI) is premised on the idea that increasingly intensive layers of instruction will reduce the number of children who fail to make appropriate academic gains. The purpose of this review was to examine two critical components of such models: coherence and fidelity. A systematic literature review was conducted to determine how researchers conceptualize and document coherence between Tiers 1 and 2 and to examine the methods for and importance of monitoring fidelity of implementation. Method A systematic literature search was conducted using PsychINFO and ERIC. Studies included in the review were empirical studies in which supplemental reading instruction was provided to elementary students in an RTI framework. Sixty-seven studies met inclusion criteria. Studies were coded for features of coherence and fidelity. Additionally, the relationship between student outcomes (indicated by effect size) and coherence and fidelity were examined. Results The review of literature indicated that while researchers often espouse the value of making explicit connections between Tiers 1 and 2 and of closely monitoring fidelity of implementation, the prevalence of reporting these two critical components is lacking across published studies. Further, the examination of student outcomes related to these components revealed that both may have important impacts regarding the interpretation of findings. Conclusions This review highlights a need for researchers to better document the coherence between tiers of instruction and fidelity of implementation in studies in which RTI models are evaluated. Consequences for not doing so and future directions for research will be discussed.
Connie Suk Han Ho (The University of Hong Kong)Pui-sze Yeung, Yau-kai Wong, David Wai-ock Chan, Kevin Kien-hoa Chung, & Lap-yan Lo - Early indicators of persistent reading difficulties among Chinese readers in junior elementary grades
Purpose The present study aimed at identifying the indicators of persistent reading difficulties among Chinese readers in junior elementary grades. Method A battery of cognitive-linguistic and reading measures were administered to three groups of Chinese children with different reading trajectories ("persistent poor readers", "improved poor readers" and "normal readers") from Grade 1 to the beginning of Grade 4 in a 3-year longitudinal study. The three groups were classified according to their performance in a standardized Chinese word reading test in Grade 1 and Grade 4. Results Results of ANOVA on the reading-related measures in Grades 1 and 2 among the three groups revealed that rapid naming was the most important identifier of early word reading difficulty. Additional difficulties in other domains, including orthographic skills, morphological awareness, and syntactic skills, were needed for reading problems to persist. Chinese persistent poor readers did not differ significantly from normally-achieving readers on measures of oral language skills, phonological awareness, or phonological memory. Conclusions The present findings show that poor rapid naming is an early indicator of reading difficulties in Chinese and this may help developing early identification tool for reading difficulty in Chinese. The present findings are also in line with the suggestion about the predominant role of the 'semantic pathway' in the 'triangle model' in reading Chinese words in contrast with the less significant role of the "phonological pathway". In terms of educational implications, intervention programmes may include training of orthographic, morphological, and syntactic skills to enhance word reading and reading comprehension of young Chinese poor readers.
Andrew Holliman (Coventry University)Clare Wood; Gareth Williams - Screening for potential reading difficulties: a new test of speech rhythm sensitivity
Purpose: The last 12 years have seen the development of a literature which has shown that speech rhythm sensitivity is implicated in successful reading acquisition. However, recent evidence suggests that speech rhythm is not a unitary concept; indeed, the different components of speech rhythm (i.e., stress, intonation, and timing) assessed at different levels (i.e., word-, phrase-, and sentence-level) may be related to reading development in different ways. In this study, a new measure of speech rhythm sensitivity was developed, which assesses all aspects of this skill, and its relationship with reading was explored. Method: One-hundred five- to 7-year-old English-speaking children completed a newly developed speech rhythm sensitivity assessment along with some control/cognitive assessments (i.e., phonological awareness, vocabulary, and non-verbal IQ) and a battery of reading assessments (i.e., word reading, fluency, comprehension, and spelling). Results: The results are pending. Conclusions: The findings from this study will inform current models (and theories) of literacy development and make a significant contribution to the literature. The task itself may ultimately form the basis of an early identification or screening measure for young children at risk of reading difficulties.
Joni Holmes (Northumbria University)Susan Gathercole; Darren Dunning - Can working memory training improve children's reading and language skills?
Purpose: Evidence-based computerised 'brain-training' programs offer huge potential for overcoming the cognitive deficits associated with reading difficulties. We have demonstrated that regular and sustained training specifically on working memory tasks improves working memory in children with disorders of memory and attention, in most cases taking scores from below-average to age-appropriate levels. The purpose of this project was to assess whether any gains in working memory following training transfer to improvements in reading. Method: Three groups of children were assessed on measures of working memory and reading before and after 6-8 weeks of working memory training. These consisted of 26 typically-developing children (9/10yrs), 22 children with dyslexia (8-10yrs) and 15 children with language difficulties (5-10yrs). A control group of 19 typically-developing children (9/10yrs) were assessed on the same measures before and after 6 weeks of normal schooling. Results: Training led to significant improvements in non-trained working memory tasks for all groups. There was no immediate impact of training on reading in the typically developing or language difficulty groups. Both single word reading and reading comprehension improved significantly following training for the dyslexic group. There were no changes pre- to post-test for the control group. Conclusions: Working memory training consistently boosts children's working memory and might offer a relatively short but intensive educational intervention for overcoming the poor reading skills of children with dyslexia. However, the inconsistent effects of training on reading skills across subpopulations warrants further rigorous investigation before this approach can be advocated as an effective remedial approach.
Jane Hornickel (Auditory Neuroscience Lab www.brainvolts.northwestern.edu, Northwestern University, USA)Steven Zecker; Nina Kraus - Neural changes following auditory training in children with developmental language disorders
Purpose: The impact of auditory training on the nervous system is well established in animal models (Fritz et al., 2003). Auditory training can engender gains in reading-related skills for children (Moore et al., 2005; Tallal et al., 1996), but the neural underpinnings of these improvements are poorly understood. Method: Children with exceptionalities underwent auditory training with commercially available training games (Earobics or Fast ForWord) or classroom assistive listening devices (FM system). Speech-evoked electrophysiological responses from the auditory brainstem and cortex and measures of reading were assessed before and after training. Results: Learning impaired children and children with autism spectrum disorders showed enhanced brainstem and cortical responses to speech after playing Earobics or Fast ForWord, respectively (Hayes et al., 2003; Russo et al., 2005; Russo et al., 2010), particularly for speech presented in noise. This suggests that auditory training can improve nervous system resistance to the deleterious effects of background noise. Preliminary results suggest classroom FM system use also alters auditory function in children with reading disorders resulting in reduced variability in subcortical response timing and an enhanced ability by the nervous system to utilize stimulus regularities. FM system use also appears to benefit reading ability. Conclusions: Children with developmental language disorders, including reading impairments and autism spectrum disorders, show improved neural responses to speech after auditory training. Auditory training can benefit auditory neural function and associated auditory-based language skills, thus providing effective remediation by targeting biological correlates of communication.
RaMonda Horton-Ikard (School Communication Sciences and Disorders Florida State University) ;Kenn Apel; Danielle Brimo; Elizabeth B. Wilson-Fowler - Examining the relationship between kindergartners home literacy experiences and narrative language abilities
Purpose: Research indicates that home literacy activities influences children's oral language development related to grammar and vocabulary. However, there are relatively few studies which have examined the role that home literacy experiences play in shaping discourse language skills, such as narrative ability. The current study examined if children's reported home literacy experiences related to their discourse performance using a narrative retell task. Method: Participants were thirty-eight typically developing kindergarteners (M=72 mos.). A narrative sample was collected from each child using a wordless picture storybook. Each child's parent was asked to complete the Home Literacy Experience Survey (Evans, Levy, & Jared, 2001) which describes children's emergent literacy experiences across six categories. Narrative samples were transcribed using the Systematic Analysis of Language Transcripts (SALT, 2010) and scored using the Narrative Scoring Scheme (NSS- Heilman, Miller, Nockerts, & Dunway, 2010). The samples were examined for overall NSS scores, mean length of t-unit (MLTU), and number of different words (NDW). Descriptive statistics and correlation coefficients were used to examine the relationship between home literacy categories and narrative language performance. Results: The average MLTU produced was 6.99 (SD=1.35), while the NDW was 82.42 (SD=33.8). The mean NSS score achieved by the participants was 15.34 (SD=3.82). Findings revealed a positive correlation between certain aspects of the NSS (macrostructure) and the survey Practice reading/writing category. The findings a negative correlation between survey Beginning Print/Book Activities survey category and MLTU. Conclusion: Children's home literacy experiences differentially mediate their narrative language development for macro and micro-structures.
Suzanne Houston (UCLA); Tami Katzir; Elizabeth Sowell; Carly Rosso; Eric Kan; Ariel Starr; Guinivere Rodriguez - Relationship of rapid letter naming and phonological awareness to brain thickness in typically developing children
Purpose: There is debate in the field whether rapid naming makes unique contribution to reading, or whether it can be subsumed under phonological processing. The goal of this study was to examine whether the rapid naming and phonological awareness of typically developing children made independent contribution to specific brain thickness of areas previously identified in functional studies as involved in these process. Methods: 43 typically developing children were grouped by reading comprehension ability (poor, average and high) went through cognitive and reading related assessments as well as MRI structural imaging. Results of the hierarchical regression analysis revealed that when predicting Left Inferior and Superior Partial Thickness there were main effects for: rapid naming as well as an interaction effect with comprehension group. No Effect was found for a blending task. When predicting brain thickness of left and right parsopercularis in the frontal areas, main effects were found across the board, however only the interaction of comprehension group and blending, and not rapid letter naming, was found significant. Conclusion: Taken together these results suggest a unique dissociation of the interaction of reading level and rapid naming vs. phonological awareness in predicting brain thickness in frontal and partial areas in different points of reading development.
Chung-Hui Hsuan (Asia University, Taiwan)Yin-Mong Ling, Pei-yin shan, wan-Juin Lai, Jia-Huei shi - The role of morphological awareness in Chinese reading from preschool to grade 3
Purpose: Morphological awareness has been found important in Chinese reading through various assessments at different ages. These assessments were based on the awareness of morphemes in compounding vocabulary, such as when discriminating morphemes embedded in different vocabularies or requiring children to produce new vocabularies. This study aims to investigate (1) the development of morphological awareness among Chinese children not only in compounding morphemes but also in derivation and inflection morphemes, such as classifiers (e.g., 個/ge/), location morphemes (e.g., 上/shang/, 'upper'), or word-forming affix (e.g., -化/hua/, '-lize'); and (2) the relationship between morphological awareness and Chinese reading from preschool to grade 3. Method: 150 children from the second year of kindergarten to grade 3 (30 participants were recruited from each grade) will be assessed by a set of morphological awareness tests, receptive oral vocabulary knowledge, character reading, and reading comprehension during December 2010 and January 2011. Results: Data will be analyzed in February 2011 by correlation at each grade. Conclusion: The results will allow us to investigate (1) the development of morphological awareness among Chinese children from preschool to grade 3; (2) the relationship between morphological awareness and reading through the trajectory of reading development under controlled oral vocabulary knowledge in children.
Anh Hua ()Janice Keenan - Assessing comprehension with retellings and comprehension questions: Why do they differ?
Purpose: A surprising finding in Keenan et al.'s (2008) comparison of reading comprehension tests was that performance varied even on the same test depending on whether comprehension was assessed with questions versus retellings; the correlation between the two was only .41. This study explores reasons for the modest relationship. What makes certain ideas likely to be recalled? Are they different than ideas assessed by questions? How often does a child recall ideas relevant to a question but get the question wrong? Or vice versa? Does it depend on whether the question is literal or inferential? Does it differ for children with comprehension deficit (CD)? Method: Retellings and question accuracy were compared for 44 children (22 CD, 22 Controls matched for age and word decoding) on Qualitative Reading Inventory passages. Literal and inferential questions were compared both on question accuracy and on whether information related to the questions was in the retellings. Relevance ratings were obtained to map each question onto passage idea units so that recalls could be scored not just for total amount but also for ideas underlying each question. Results: Although total amount recalled correlated with question accuracy only r= .45 (n=44), preliminary analyses show that restricting scoring to relevant ideas increased r to .63. Interestingly, subjects recalled more idea units related to inferential questions despite their superior performance on the literal questions. Conclusion: This comparison of retellings and questions expands our understanding of comprehension assessment and of how CD children may construct mental models differently from Controls.
Laura Hume (Florida State University, Florida Center for Reading Research); Darcey Sims; Christopher J. Lonigan - How Preschoolers' Motivation for Literacy and ADHD Characteristics Relate to Early Literacy Skills
Purpose - This study examined how ADHD characteristics and motivation for literacy affect early literacy skills in preschoolers. Both ADHD characteristics and motivation for literacy have been shown to relate to early literacy skills and ADHD characteristics have been linked to motivation, but researchers have yet to examine how ADHD and motivation for literacy combine to affect early literacy skills in young children. ADHD characteristics may influence literacy skills by lowering motivation for literacy. Children with higher levels of inattention and hyperactivity may dislike typical reading related activities requiring children to stay focused, resist distractions, and sit still for an extended period of time. This dislike could lead to less engagement in literacy related activities and less early literacy skill development. Examining the contributors of early literacy skill development in preschoolers may help develop ways of preventing long term reading and academic difficulties. Method - Multiple regression analyses were completed using data from 165 preschool children. Children were assessed on early literacy skills, inattention, and hyperactivity. Teachers rated ADHD characteristic behaviors and parents reported children's motivation for literacy. Results - Both inattention and motivation for literacy contributed unique variance to early literacy skills. Motivation for literacy did not significantly interact with ADHD characteristics, but motivation for literacy partially mediated the effect of ADHD characteristics on early literacy skills. Conclusions - These results suggest both ADHD characteristics and motivation for literacy uniquely effect the development of early literacy skills and ADHD's effect on early literacy skills is, in part, due to its effect on motivation for literacy.
Jane Hurry (Institute of Education, Uni of London);Jane Hurry; Eirini Flouri; Kathy Sylva - Literacy difficulties and emotional and behaviour disorders: causes and consequences
Purpose: There is a well-established association between literacy difficulties (LD) and emotional and behaviour disorders (EBD). However, as there is a scarcity of experimental studies, directions of causality are not clear. This study investigates causal relationships between LD and EBD, and implications for intervention. Specifically we test the hypotheses that ;1) LD cause or exacerbate externalising or internalising problems; 2) EBD cause or exacerbate LD, and; 3) the presence of EBD interfere with the effectiveness of reading intervention. Method: In a quasi-experimental study, 342 six-year-olds in the bottom 20% of readers were followed up immediately post intervention (first ,follow-up) and four years later (second follow-up). In the first year of the study, 89 received a phonological intervention (Phono), 81 received Reading Recovery (RR), and 172 in the control group received standard tuition. Results: RR (but not Phono) significantly improved children's literacy at first follow-up (es = .80), but resultant literacy gains had no effect on EBD. This suggests that LD do not cause EBD. Neither intervention had long term effects on literacy or behaviour. Hyperactivity interfered with literacy gains resulting from the less intensive Phono, but not the more intensive RR. Conduct disorder and hyperactivity had negative effects on literacy progress, supporting propositions that these disorders exacerbate children's LD. Conclusions: LD do not appear to cause or exacerbate EBD, but there is some evidence that conduct disorders and hyperactivity exacerbate LD. This may due to EBD interfering with the effectiveness of instruction, suggesting that co-morbidity should inform teachers' instructional decisions.
Mutsuo Ijuin (Tokyo Metropolitan Institute of Gerontology); Taeko N. Wydell - Can a connectionist model account for the hypothesis of granularity and transparency? A computer simulation model for reading Japanese Kanji and Kana in developmental dyslexics
Purpose : The Hypothesis of Granularity and Transparency (Wydell& Butterworth, 1999) argues that any orthography where the letter-sound correspondence is transparent will not produce a high incidence of phonological dyslexia (transparency). The hypothesis also argues that even when this correspondence is opaque, if the smallest orthographic unit representing a speech sound is large, there will not be a high incidence of phonological dyslexia (granularity). The present study investigated the granularity aspect of the hypothesis using a connectionist network to read Japanese logographic Kanji and syllabic Kana. Method : Two connectionist networks were trained to read single-character Kanji and Kana, with transparency controlled at 100% and granularity manipulated between the two scripts (Kanji granularity is larger than that of Kana). In this study, the granularity size was the only factor to discriminate Kanji from Kana. The intact network first learned about the phonological information of characters before any orthographic information was introduced, and subsequently the network was trained to read characters aloud. The phonological dyslexic network was simulated by impairing the representation of phonological information before training the network to read. Results : The intact network learned to name Kanji faster than Kana and learning Kanji was more robust to phonological lesions than learning Kana in the phonological dyslexic network. Conclusion : The connectionist networks simulated a significant advantage of learning to read Kanji over Kana as predicted by the Hypothesis of Granularity and Transparency. This result suggests that the unit size of granularity is likely to play a critical role in reading acquisition.
Marcia Invernizzi ()Francis Huang, Karen Ford - Revisiting the "gift of time": Age and early literacy skill acquisition
Purpose Age at kindergarten entry has been of interest to parents and policy makers and the practice of voluntarily holding back a child until he or she is older ("redshirting") has received much attention in the popular press. Upon entering kindergarten, older children, have been found to perform better in reading skills. Given that redshirting if more prevalent with White males from more affluent families, this places children from economically challenged and minority families at a disadvantage, who may not be able to afford to keep their children out of school for an additional year. We investigated if the effects of age persisted over time. Method We used a longitudinal sample of over 400 students (from kindergarten to the second grade) who come from both low performing and high poverty schools in a Mid-Atlantic state. Using the kindergarten entry cut-off date (October 1) in the state and a child's birthdate, we created two groups comprised of the oldest and youngest children whose age difference in kindergarten is maximally separated by 11 months. Results Using hierarchical linear modeling, we find that in the fall of kindergarten, older children had higher scores on emergent literacy measures (ES=0.30), while controlling for school- and student-level variables (gender, economic status, and race/ethnicity). However, latent growth curve analysis revealed that, over time, the initial difference in scores is narrowed and ceased to be statistically significant. Conclusions Findings indicate that while age may initially contribute to higher literacy scores, over time, younger children are not put at a disadvantage.
Alice Jackson (University of Maryland); Donald Bolger - Neural correlates of word learning from context
Purpose - Most new vocabulary is acquired through incidental learning from discourse context (Nagy & Herman, 1987). The context variability hypothesis (Bolger et al., 2008) states that varying contexts will support better abstraction of a word's core meaning features than contexts that do not vary, and further states that definitions are a specific type of context consisting solely of core meaning features. This study examined the neural correlates of learning novel words from high- and low-variability or limited contexts and definitions. Method: Participants were taught the meanings of sixty novel pseudowords through exposure to either sentence contexts or dictionary definitions. Following training, participants generated free-response definitions of the trained pseudowords and made multiple-choice judgments of cloze sentences. Event-related potentials (ERP) were collected as participants read congruent and incongruent sentences that included the trained pseudowords. Results: In the multiple choice cloze task, participants were most accurate in selecting words trained via varied contexts relative to limited contexts including definitions. In the ERP congruency judgment task, while there was no effect of context variation for the 'incongruent' sentences, while 'congruent' sentence judgments showed effects of word training type. Participants were more accurate on sentences that used words trained via varied contexts than limited contexts or definitions. Based on preliminary ERP data, an N400 effect is expected to the words trained via varied contexts versus the limited contexts and definitions. Conclusions: These results support the context variability hypothesis by demonstrating better word learning through highly variable contexts than low-variability contexts or definitions.
Linda Jarmulowicz (University of Memphis) - Stress and syllable errors in derived word production
Reading ability relies on various levels of the phonological system. Both segmental and suprasegmental information must be learned, stored, and eventually made explicit. We know that phonological awareness plays a critical role in reading development, and as children mature, awareness of morphological structure becomes important in comprehension. These two pieces are particularly important in orthographies like English that code for both phonemes and morphemes. Prosody, at the word, phrase, and sentence level, is emerging as another area important in reading. The current study evaluates the types of errors made in the production of derived words with stress changes by children with dyslexia compared to 3rd and 5th grade typical readers. Results indicate that children with dyslexia have difficulty shifting stress in this elicited production task, and that they have a different error profile than younger readers.
Christine Javier (Wilfrid Laurier University); Alexandra Gottardo; Esther Geva - The nature of language development in English as a Second Language (ESL) learners
Purpose: Changes in language proficiency may occur when a bilingual's environment shifts and he/she must adjust his/her linguistic abilities in order to adapt to the new environment (Grosjean, 1998). Since Canada is home to many cultures, many immigrant families enroll their children in school every year. The purpose of this study was to compare the degree of language development in a group of English as a Second Language (ESL) learners from one grade to another. Language proficiency in both their native language (L1) and English (L2) were assessed. Method: Participants were ESL students in Grade 1 who were recruited from several elementary schools in Southwestern Ontario. Participants were tested on their L1 and L2 language proficiency in Grades 1 and 2 using versions of the Peabody Picture Vocabulary Test - III (PPVT-III), the Test of Word Reading Efficiency (TOWRE) (Torgesen, Wagner, & Rashotte, 1999), and the Woodcock Language Proficiency Battery-Revised (Woodcock & Muñoz-Sandoval, 1995). Results: Results indicated that on all of the language measures, performance in Grade 1 was highly correlated with performance in Grade 2 (all paired samples correlations between .715 and .824). Paired samples t-tests revealed that participants' raw scores on all language measures in both the L1 and L2 increased from Grade 1 to Grade 2 (all p-values < 0.001). Conclusion: These findings suggest that despite schooling exclusively in English, participants showed significant growth in both languages, due to continued exposure to participants' L1 at home. Both theoretical and practical implications for ESL learners are discussed.
Juan E. Jiménez (University of La Laguna)Juan E. Jiménez; Eduardo García; Cristina Rodríguez - Analyzing the role of complexity of syllable structure and the effects of task differences in Spanish dyslexic teenagers
Purpose: The primary purpose of the study reported here was to explore the effects of the complexity of syllable structure and the effects of task differences in the explanation of deficit in phonological awareness in Spanish dyslexic teenagers. Method: A sample of 78 subjects was selected and organized into three different groups: 28 reading-disabled children, 25 normal readers matched in age with the former, and 25 younger normal readers at the same reading level as those with reading disabilities. We administered phonological awareness tasks (i.e., isolation vs. deletion) which included items with different complexity of syllable structure (i.e., CV vs. CCV). Results: A (3 × 2 × 2) Group (RD vs. normal readers matched in age vs. younger normal readers) × Task Differences (isolation vs. deletion) × Complexity of Syllable Structure (CV vs. CCV) mixed ANOVA were performed on the number of correct responses as a dependent variable and they were calculated separately across participants and items. The results showed a significant interaction between the complexity of syllable structure and task differences. The deletion task revealed the phonological deficit when a CCV structure was presented where the operation that is required is the analysis of phonemes composing cluster onset.
Clinton Johns (Haskins Laboratories); Julie Van Dyke; James Magnuson; Kenneth Pugh; Einar Mencl; David Braze; Whit Tabor; Donald Shankweile - Online memory retrieval during reading is affected by phonological stability
Purpose: Phonological skill is related to the development of reading skill. This study tests the hypothesis that low-quality output from early processing (e.g., unstable phonological representations) limits the efficiency of later, higher-level processes (e.g., syntactic analysis). We propose that the mechanism of retrieval, required for processing complex syntactic constructions, provides the link between these processing levels. Thus, online sentence processing will be affected by how to-be-retrieved phonological representations are encoded. Method: Participants were non-college-bound young adults (age 16-24; current N=11, planned N=16). Event-related potentials were recorded as participants read sentences varying according to retrieval difficulty and phonological stability. Sentence syntax was either easy (subject-relative clauses, SR) or difficult (object-relative clauses, OR). ERPs were time-locked to the main verb, which elicits retrieval of its argument (subject noun). Stability of subject nouns (e.g., maid) was manipulated via fast-priming with either phonologically consistent (raid) or conflicting (said) primes. Consistent primes are meant to enhance phonological encoding; conflicting primes create interference. Results: Preliminary N400 analyses indicate that phonological stability affected retrieval. The classic SR-OR difference was observed in conflicting-prime conditions, where enhanced negativity in the ERPs indicates impaired integration in OR conditions. However, when subjects were preceded by phonologically consistent primes, no such difference manifested. Consistent-prime OR conditions resulted in a mid-frontal positive shift at the main verb, similar to previous ERP findings associating such positive shifts with item retrieval on the basis of familiarity. Conclusions: Results support the phonological bottleneck hypothesis, indicating that high quality phonological encodings facilitate retrieval during sentence comprehension.
Kathleen Jubenville (); Rosemary Lever; Monique Sénéchal; Stephanie Pagan - Writing stories: An analysis of component skills
Purpose. The goal of this research was to test two alternative accounts of individual differences in reading and writing. The first was a general-literacy view that posits that the same higher-level skills underlie both reading comprehension and written composition. In contrast, the domain-specific view posits that components skills are task specific. Method. 100 Grade 4 children were assessed on reading comprehension and written composition. Children composed a story based on a picture-book. Composition quality was a composite measure that included story structure (i.e., story grammars), story sophistication (i.e., the number of ideas), and language sophistication (i.e., the number of morphemes). General higher-level skills were inference ability and comprehension monitoring. Specific writing higher-level skills included online planning and revising. Basic literacy control variables were word reading and spelling, and other controls were expressive vocabulary, working memory, and non-verbal intelligence. Results. Fixed-order regression analyses were conducted to test the two views of literacy. The results for reading comprehension were consistent with the literature, in that inference ability and comprehension monitoring were unique predictors after controlling for basic reading skills and other controls. In contrast, inference ability and comprehension monitoring did not account for a significant portion of the variance in composition quality. For composition, it was the specific skills of online planning and revising that explained unique variance after entering controls and basic writing skills. Conclusions. The findings revealed that different higher-level skills explained individual differences in reading comprehension and written composition. These findings are consistent with a domain-specific view of literacy.
Laura Justice (Ohio State University);Sara Hart - Preschoolers' learning-related behaviors: contributions to literacy development during classroom-based intervention
Purpose: The purpose of this study was to examine the contribution of preschool children's learning-related behaviors and competencies (viz., literacy engagement, inhibitory control) to their responsiveness to classroom-based literacy intervention. Method: In this multi-site randomized controlled trial involving 84 preschool classrooms, teachers were randomly assigned to implement a print-focused book-reading program or a control program for the academic year. Children's literacy skills were assessed at four time points over three years to assess longitudinal impacts of the intervention; learning-related behaviors were collected during the intervention itself. Results: Three-way interactions (literacy engagement, inhibitory control, treatment status in preschool) with reading achievement in grade one as the outcome indicated heightened intervention impacts for children with positive engagement and higher inhibitory control during treatment receipt. Conclusions: The way in which children's learning-related behaviors and competencies may interact with various intervention approaches is an important consideration when implementing classroom-based literacy interventions.
Holger Juul (Centre for Reading Research, University of Copenhagen); Holger Juul; Carsten Elbro; Mads Poulsen - From accurate to fluent word reading - in how many days? A new approach to early reading development.
Purpose: In an ongoing Danish study of beginning word reading, we attempt to clarify what the predictors of early reading development predict: accuracy or speed? Letter knowledge and phoneme awareness are hypothesized to predict accurate reading, because they promote understanding of the alphabetic principle ('cracking the code'). However, for the subsequent development from 'accurate reader' to 'fluent reader', other factors, e.g. RAN, are expected to be more predictive. Method: A new methodological approach is employed to separate accuracy development from speed/fluency development. Through frequent testing of children from the beginning of reading instruction (N=183), we estimate the number of days in development until fixed accuracy and fluency criteria are reached for short regular words. Accuracy: at least 70 pct. correct. Fluency: at least 40 words correct per minute. Results: Consistent with the hypothesis, preschool letter knowledge and phoneme awareness accounted for about 35 pct. variance in developmental time before the accuracy criterion was reached. This equals the predictive power of these measures in studies of reading skills at a fixed time (e.g., variance at the end of Grade 1). The predictive powers of these and other preschool measures will also be reported for the subsequent development of fluent reading. At present, however, some participants have yet to reach the fluency criterion. Conclusions: 'Days to criterion'-measures allow us to divide reading development into meaningful phases and to explore the cognitive underpinnings of each phase separately. The advantages of this approach will be discussed and compared to growth curve analysis.
Ann Kaiser (Vanderbilt University) - Discussant
Our discussant will share perspectives on some of the challenges in applying RTI in early childhood settings. She will critically analyze the results of these studies, highlighting their contributions and the needs for future research. She also will discuss the viability of the proposed RTI model for preschool settings.
Patricia Kantor (Florida State University)Richard K. Wagner - Modeling the Growth of Writing: A Three-Year Longitudinal Study
Purpose: Given the almost limitless number of variables that exist for analyzing writing samples, it is difficult to identify key elements for instruction, remediation, and assessment. One possibility is to apply factor analysis techniques to identify key factors. The current study will investigate the relation of key writing factors in a longitudinal sample of 1st through 3rd graders. Method: Beginning in 1st grade, 318 children participated in a longitudinal study of literacy skills that included a yearly assessment of writing. Coh-Metrix was utilized to identify variables in the writing samples. Confirmatory factor analysis was used to determine key factors and latent growth modeling was used to explore the developmental structure of writing. Results: A preliminary analysis that involved researcher coding of samples of writing from a study of 1st and 4th graders identified a five factor model that best described the writing samples: macro-organization, complexity, productivity, spelling and punctuation, and handwriting fluency. Large differences were found between 1st and 4th graders in the productivity and handwriting fluency constructs (Wagner et al., in press). Additionally, when samples were analyzed using Coh-metrix, which has the benefit of not relying on human judgment, large differences continued to be found for variables that reflected productivity, as well as Flesch readability indices. Conclusions: The results of the study will extend prior research by looking at the growth in factors of writing during the early stages of writing acquisition. Additionally, using computerized analyses such as Coh-Metrix, provides the ability to examine a greater number of variables with high reliability.
Arjette Karemaker (University of Oxford) ;Nicola Pitchford; Claire O'Malley - Assessing the effectiveness of a whole-word multimedia reading intervention designed to support development of early literacy skills
Purpose: Two studies investigating the effectiveness of a whole-word multimedia software intervention in supporting developing literacy skills in children aged 5-6 years are discussed. The main question in this presentation is: is the multimedia software program Oxford Reading Tree (ORT) for Clicker effective in supporting typically- developing and struggling readers compared to traditional teaching methods using printed texts from the same scheme? Sub-questions relate to: 1) Does multimedia software support different reading processes? And, 2) Do the multimedia features built into ORT for Clicker support children's developing development of literacy skills? Method: Class-based intervention studies were conducted with typically developing children (Study I) and children that were struggling with learning to read (Study II). A within group design was adopted; each child was given each of the two interventions, and the order of intervention was counterbalanced across the groups. Each intervention was integrated into the literacy hour over five consecutive days. Measures of written word recognition, written word naming, phonological awareness and attitudes towards computers were taken before and after each intervention. Results: Collectively, results showed heightened performance gains across a broad range of basic literacy skills after intervention with the whole-word multimedia software compared to traditional pedagogical practices. In addition, attitudes of both teachers and poor beginner readers were more positive and less negative towards ICT after using the multimedia software. Implications for theory and practice are discussed. Conclusions: Reading intervention with whole-word multimedia software can be an effective classroom aid in supporting early literacy skills.
Devin Kearns ()Douglas Fuchs, Donald L. Compton, Lynn S. Fuchs, Karla K. Stuebing, & Samuel A. Patton III - Describing the Cognitive Correlates of Reading Disability
Purpose: The present study examined cognitive correlates of reading disability (RD) as a function of comorbidity to determine whether these groups had different cognitive profiles. Method: Students with RD vs. RD+MD (math disability) vs. RD+ADHD (attention-deficit/hyperactivity disorder) and RD+MD+ADHD (n = 212), as well as students typical achievement (TA, n =51) were tested on a large reading and cognitive battery. Profile analysis was conducted to examine how the cognitive and reading profiles of these groups were alike and different. Results: The profile analysis revealed that, across subtypes, students with RD were particularly weak in phonological awareness and oral expression, even after accounting for the overall lower level of students with RD. This group, however, had relatively strong in visuospatial short-term memory relative to peers with TA, after accounting for level. Among the RD subtypes, those with RD-only were less impaired on cognitive measures but the subtypes did not have different patterns of strength and weakness among themselves. Posthoc power analysis suggested that, given the sample size and the number of comparisons, the absence subtype differences could not be attributed to low power. Conclusions: The finding that students with RD have weak phonological processing and oral expression after accounting for level corroborates findings from prior studies. The finding that students with RD-only, RD+MD, RD+ADHD, and RD+MD+ADHD did not differ on a range of cognitive measures is, however, surprising and suggests that the factors in all three disabilities are related.
Janice Keenan (University of Denver, Boulder)Richard Olson, Brian Byrne & Stefan Samuelsson - Preschool predictors of listening comprehension and reading comprehension
Purpose: How well can reading skill be forecasted from preschool behavior? The present study examined how children's grade 4 reading comprehension, listening comprehension, and word decoding was predicted by their preschool performance on vocabulary, verbal memory, grammar/morphology, print knowledge, phonological awareness, and rapid naming. We also explored genetic and environmental influences on the relations between preschool and grade 4 latent traits. Method: 925 twins were tested both in preschool and at the end of grade 4 using at least two measures per construct to allow for phenotypic and behavior-genetic modeling of latent traits. Results: Preschool measures of vocabulary, verbal memory, and grammar/morphology strongly predicted listening comprehension at the end of grade 4, suggesting that comprehension problems can be detected early. Although all preschool measures significantly predicted independent variance in grade 4 reading comprehension, with the strongest contributions coming from phonological awareness and vocabulary, the variance explained in reading comprehension by the preschool measures was less than in listening comprehension. This difference reflects decoding, which was the least well predicted in grade 4 from our preschool measures. Behavior-genetic modeling revealed a high level of shared genetic influences between grade 4 listening comprehension and its primary preschool predictors, and between grade 4 decoding and its primary preschool predictors. Taken together, the preschool predictors of listening comprehension and decoding accounted for all of the genetic influence on Grade 4 reading comprehension. Conclusions: Reading comprehension, and especially listening comprehension, at the end of grade 4 can be predicted by children's skill profiles in preschool.
Cornelia A.T. Kegel (Leiden University);Adriana G. Bus - Differential susceptibility in early literacy instruction through computer games: An example with the Dopamine D4 Receptor gene (DRD4) as moderator
Purpose: This study is the first randomized controlled trial testing whether genetic predispositions may influence how susceptible children are for a computerized early literacy intervention. The Dopamine D4 receptor gene (DRD4) has been associated with attentional mechanisms and in line with the differential susceptibility paradigm we advanced the proposition that children with the less efficient long variant of DRD4 (7-repeat allele) would be more susceptible to both (a) adverse effects of poorly designed programs, and (b) beneficial effects of an optimal training. Method: The intervention is an Intelligent Tutoring System designed to promote alphabetic knowledge in young children. 182 kindergartners were divided over three groups: one receiving the program with all available features, one completing the same computer assignments but without oral feedback to children's responses, and one control group. Results: Carriers of the long variant of the DRD4 gene profited most from the computer program with positive feedback and performed at the lowest level of early literacy skills in the absence of such feedback whereas the carriers of the short DRD4 variant showed a flat profile. Conclusions: Our findings suggest that educational interventions may have modest effects because only a sub-sample depends on the quality of an intervention. Less susceptible children adapt to most learning environments without performing too well or too bad.
Nenagh Kemp (School of Psychology, University of Tasmania); Brett Kessler; Rebecca Treiman - Adults' spelling of doubled consonants in pseudowords
Purpose: The question of when to double or extend consonant spellings (e.g., the l in bailiff, billet) is an aspect of English spelling that causes difficulty even for adult spellers. If adults simply memorize the need (or not) for doubling consonants in individual words, their consonant doubling in pseudowords should be random. However, if adults learn probabilistic spelling patterns from written language, then their consonant doubling in pseudowords should be more systematic. Method: University students (N= 42) wrote 144 pseudowords to dictation. For the consonants usually spelled f, l, k, and ch, English vocabulary statistics favor doubled or extended spellings (ff, ll, ck, tch) after a single vowel, especially in monosyllabic words (e.g., sell, stitch). We analyzed whether participants followed these patterns by varying vowel length (short, long) and number of syllables (one, two, three), and also tested whether familiarization with pseudowords enhanced the use of surrounding letter context in spelling. Results: Participants doubled/extended consonants more after single vowels, although more in one- than in two- and three-syllable pseudowords. The number of letters that participants used to spell the vowel was a stronger determiner of consonant doubling than was the phonological length of the vowel. Pseudoword familiarization had a limited effect on spelling. Conclusions: This study provides evidence that adults make use of preceding vowel context to decide whether to double/extend consonants, and that this sensitivity extends even to multisyllabic pseudowords. Further, graphic factors seem to affect the extent of such context-conditioned spelling more than phonological factors.
Sarah Kershaw (Florida State University, Florida Center for Reading Research); Chris Schatschneider; Amy Barth; David Francis; Barbara Foorman; Yaacov Petscher; Jack Stenner; Carl Swartz - Genre differences in text cohesion that predict reading comprehension and fluency
Purpose A growing body of research suggests that individual features of text, such as referential cohesion, causal cohesion, conceptual cohesion, and word frequency, affect reading comprehension and fluency. However, little is known about how combinations of these text features affect a student's ability to comprehend various genres of text. Thus, the current study seeks to utilize text features derived from Coh-Metrix to (a) investigate genre differences in the factor structure of 725 texts, and (b) determine how these factors predict reading comprehension and fluency by combining data from three samples of students from grades 3-10. Method The first set of analyses included an exploratory factor analysis (EFA) by genre to investigate genre differences in the factor structure of text features such as referential cohesion, causal cohesion, conceptual cohesion, and word frequency derived from Coh-Metrix. The second set of analyses included structural equation modeling to determine how these factors predict reading comprehension and fluency. Results Results from EFA analyses by genre revealed three factors (referential cohesion, causal cohesion, and syntactic similarity) consisting of 9 variables that accounted for roughly 75% of the total variance. Variables loading on all three factors remained the same across genre with the exception of the proportion of pronouns to noun phrases. Results from structural equation modeling are consistent with the literature that show referential cohesion to be a significant predictor of reading comprehension. Conclusions These results suggest that factor structure varies based on genre and that features of text play a significant role in predicting reading comprehension and fluency.
Michael Kieffer (Teachers College, Columbia University);Nonie K. Lesaux - Who benefits from instruction to improve morphological awareness? A multivariate approach to treatment by student interactions
Purpose: This study investigates whether and how the effects of instruction on morphological awareness differ by students' prior skills. It draws on data from a cluster-randomized controlled field trial of a 20-week multi-component academic language intervention in sixth-grade classrooms. Rather than testing whether treatment effects differ by scores on one pretest measure, this study uses a new approach to examine whether students' multivariate profile of reading and language skills predict differential responsiveness to the intervention. Method: Eighty classes in 14 urban middle schools were randomly assigned to treatment or control (n = 1738). Students completed measures of vocabulary, reading, and morphological awareness at pretest and posttest. Latent class analysis (LCA) was used to determine the number and nature of unique multivariate skill profiles at pretest, which were then tested in subsequent multilevel models as moderators of treatment effects. Results: Preliminary results indicate at least four distinct skill profiles demonstrated at pretest, and that treatment effects on morphological awareness differ by profile. Treatment effects were strongest for students demonstrating a profile of below average vocabulary, low morphological awareness, and low-average decoding (d = .41; p < .001). Treatment effects were weaker for students demonstrating average language skills combined with adequate decoding (d = .14; p = .009), and non-significant for students demonstrating low language skills combined with substantially impaired decoding (p = .332). Conclusions: Results suggest that morphology instruction may be most beneficial as a tier-two intervention for older struggling readers with under-developed language skills, but without substantial impairments in decoding.
David Kilpatrick (State University of New York, College at Cortland)Leslie Cole -first author; David Kilpatrick, second author - Storage of Rime-Based Sub-Word Orthographic Units in Children and Adults: Evidence from a Pseudo-Rime Task
Purpose The consolidated alphabetic phase of Linnea Ehri's theory of sight-word acquisition suggests that sub-word orthographic letter strings are stored in a unitized fashion in long-term memory (LTM). Some research paradigms have provided support for this notion while others have not. The present studies make use of a different research paradigm based on pseudorimes, which are pronounceable rimes that do not function as rimes in English (e.g., enk in menk). Method Three studies, the first with college students (n = 30), and the others with second graders (n = 27; 26) and fifth graders (n = 28; 27) exposed participants to a timed task of reading nonsense words from multiple lists. Some lists contained words built upon genuine rimes (e.g., ank in mank) and others upon pseudo-rimes (e.g., enk in menk). In Study 2, children also read a third list built from real word based rime units (e.g., ant in mant). Results Both children and adult participants read significantly more words from lists based on real rimes than pseudo rimes. Second graders showed the pattern of real word rimes > real rimes > pseudo rimes in Study 2 while the fifth graders showed real word rimes = real rimes. The third study, which held phonology of pronunciation constant (e.g., -irst vs. -erst), showed similar results. Conclusions This research paradigm provided evidence of consolidation of orthographic rime units in LTM in children and adults. Study 2 provided cross-sectional development in this consolidation process.
Sunjung Kim (University of Central Arkansas)Rebecca Wiseheart; Linda Lombardino - Diagnosing dyslexia: the dual route versus the double deficit model
Purpose The purpose of this study was to determine whether the Dual Route Hypothesis (DRH) or the Double Deficit Hypothesis (DDH) best accounts for the profiles observed in an archival database of children who were diagnosed with dyslexia. Methods Archival data were analyzed for 125 6-12 year old children who were diagnosed with dyslexia. To test the DRH, we categorized children as having surface, deep, and combined dyslexia. To test the DDH, we categorized these same children as having a phonological decoding deficit only, a rapid naming deficit only, or combined deficit. Results We used strict criteria for classifying the children as having reading profiles representative of dyslexia based on a DRH and the DDH. Only 36% of the children were classified as dyslexic using the DRH when we used traditional untimed measures of reading. However, when we used timed measures of reading instead, the percentage of children classified under DRH rose to 60%. Alternatively, when we used criteria for the DDH based on traditional phonological processing composite scores, 66% of the children were classified as having dyslexia. This number rose to 71% when we used subtest scores instead. Conclusions Findings from using both classification systems revealed the importance of using timed test measures of reading, along with specific subtest scores for identifying dyslexia in children who span a wide age range and who vary in the severity of their reading disability. Implications relevant for defining dyslexia for both research and psycho-educational purposes will be addressed.
James Kim (Harvard University)Carol Booth Olson, Robin Scarcella - Enhancing the interpretive reading and analytical writing of mainstreamed english learners in secondary school: results from a randomized field trial using a cognitive strategies approach
We conducted a multi-site cluster randomized controlled trial of a cognitive strategies approach to teaching text-based analytical writing for mainstreamed Latino English Language learners (ELLs) in 9 middle schools and 6 high schools. In the Pathway Project intervention, teachers are trained to use a pretest on-demand writing assessment to improve text-based analytical writing instruction for mainstreamed Latino ELLs who are able to participate in regular English classes. The intervention draws on well documented instructional frameworks for teaching mainstreamed ELLs. Such frameworks emphasize the merits of a cognitive strategies approach that supports these learners' English language development. Pathway teachers participated in 46 hours of training and learned how to apply cognitive strategies by using an on-demand writing assessment to help students understand, interpret, and write analytical essays about literature. 103 English teachers were stratified by school and grade and then randomly assigned to the Pathway Project professional development intervention or control group. Students taught by Pathway teachers scored .54 standard deviations higher than control teachers on an assessment of analytical writing. In addition, Pathway teachers who were trained for two years had a larger impact on analytical writing (d = .72) than Pathway teachers who were trained for only one year (d = .17). Although there was no significant impact on reading comprehension scores, the effect sizes favored year 2 Pathway teachers over control teachers (d = .08). Implementation data indicated that the largest differences between Pathway and control teachers were on measures of writing rather than reading instruction.
Young-Suk Kim (Florida State University)Stephanie Al Otaiba; Kristen McMaster; David Parker; Jessica Sidler Folsom; Luana Greulich; Aaron Rouby; Kadie Ann Bourgoin - Measuring quality of writing for beginning writers
Purpose - The purposes of this study were (1) to examine how to measure the quality of writing and (2) to examine relations among various aspects of writing using qualitative and quantitative scoring for beginning writers. Method - Participants for this study were 521 first graders from 8 schools and 35 different classrooms. The mean age of participants was 6.2 years (SD = .38). There were 233 females and 289 males from a range of SES. Children wrote on a prompt, "One day when I got home from school,…" Qualitative coding used an adapted version of Six Traits Plus and was on a likert-like scale ranging from 1-5. The categories of qualitative coding included (1) ideas, (2) organization, (3) word choice, (4) sentence fluency (grammar), (5) spelling, (6) mechanics, and (7) handwriting. Quantitative scoring included Curriculum-Based Measures (CBM), and writing fluency. Curriculum-Based Measures (CBM) included Correct Word Sequences (CWS), Incorrect Word Sequences (IWS), and Number of T-units. Writing fluency measures included number of words, sentences, and ideas. Results - Preliminary findings suggest that adequate reliability is obtained on the qualitative coding on all the seven categories. Various components of qualitative score were closely related. Children's performance on qualitative measures was positively related to the majority of quantitative CBM and writing fluency measures. Conclusions - It appears that qualitative scoring, although it has multiple components, has strong internal consistency. Qualitative scoring is also related to many aspects of quantitative scoring of writing for beginning writers.
John Kirby (Queen's University, Kingston, Ontario)Bozena White - Levels of representation in reading comprehension
Purpose: Our purpose was to examine levels of representation (Kintsch, 1998) in high school students' reading comprehension. We (a) investigated whether these levels can be reliably distinguished in commonly used tests, (b) tested the effects of approaches to learning, and (c) determined whether the difficulties of "poor comprehenders" are confined to specific levels of comprehension. Method: Participants were 141 Grade 10 students. Reading comprehension measures were the Canadian Tests of Basic Skills (Nelson, 1998), Ontario Secondary School Literacy Test, and Summary Writing (Kirby & Pedwell, 1991). Other measures included the Study Process Questionnaire (Biggs et al., 2001), nonverbal ability, vocabulary, working memory, and word reading accuracy. Results: (a) Items from the three reading comprehension measures were classified reliably into the three levels of details, main ideas, and themes. (b) Hierarchical regression analyses indicated that the deep approach to learning was a significant predictor only of theme-level comprehension, after controlling nonverbal ability, vocabulary, working memory, and word reading ability. (c) Poor comprehenders, identified by regression, performed worse at all three levels of comprehension; these differences persisted after controlling vocabulary, working memory, and the deep approach separately. Conclusions: These results indicate that levels of representation can be identified in measures of reading comprehension and suggest that deeper levels should be assessed systematically and conceptions of learning addressed in interventions. The performance of the poor comprehenders suggests that semantic/inferential difficulties occur as the lowest levels of cognitive representation are being established and prevent the higher levels from being established.
Tijs Kleemans (Radboud University Nijmegen, The Netherlands); Eliane Segers; Ludo Verhoeven - Cognitive and linguistic precursors of early numeracy in children with specific language impairment: the role of rapid naming
Purpose: Research on the impact of language on numeracy and arithmetical skills suggest that children with Specific Language Impairment (SLI) could be at risk in developing insufficient numerical (i.e. counting, simple addition) representations. Moreover, given that these children also show problems in rapid naming, a slower unification of verbally stored numerical representations (i.e. counting words; arithmetic factual knowledge) may have significant repercussions for their arithmetical skills. Therefore, rapid naming may be a unique factor in explaining why these children have difficulties in processing verbal numerical tasks. This was further investigated in the present study. Methods: Sixty- two children with SLI and 111 children from mainstream education were tested on cognitive (i.e. general intelligence, working memory, rapid naming), linguistic (i.e. phonological awareness, grammatical ability) and numerical (i.e. early numeracy, addition, subtraction) capacities. The children were longitudinally followed in the second year of kindergarten and first grade. Results: When predicting numerical skills we found a mediating role of phonological awareness and grammatical ability for both SLI and mainstream. However, when controlling for both cognitive and linguistic precursors, rapid naming appeared to be a unique factor in explaining the numerical capacities of the children with SLI. Conclusions: The results of the present study suggest that children with SLI also have problems in developing numerical representations. As a clinical implication, classroom teachers should be made aware of the fact that intervention based on compensative strategies in the early years of childhood may be the only way to prevent further delays in arithmetic.
In Yeong Ko (University of Maryland, College Park)Min Wang - Processing of compound words by adult Korean-English bilinguals
Purpose: The purpose of the present study is to investigate how Korean-English bilinguals process compound words in both Korean and English. The research questions are: (1) When Korean-English bilinguals process Korean and English compound words, what information is used to segment compound words into their constituents? (2) What is the relative contribution of the first and second constituent primes to lexical decision for Korean and English compound words? Method: Two masked priming experiments were conducted with 80 adult Korean-English bilinguals in South Korea. In both experiments, within-language prime-target pairs (Korean-Korean for Experiment 1 and English-English for Experiment 2), co-varying morphological decomposability, semantic and form relatedness were presented. In Experiment 1, 2 (condition: +M+S+O vs. -M-S+O) x 4 (prime type: first constituent vs. second constituent vs. first unrelated vs. second unrelated) design was employed. In Experiment 2, 4 (condition: +M+S+O vs. +M-S+O vs. -M-S+O vs. -M+S-O) x 4 (prime type) design was employed. The four types of English prime-target pairs and their examples are: (1) morphologically decomposable, semantically transparent and orthographically overlapped (+M+S+O, e.g., key - keyhole), (2) morphologically decomposable, semantically opaque and orthographically overlapped (+M-S+O, e.g., dead - deadline), (3) only orthographically overlapped (-M-S+O, e.g., pump - pumpkin), (4) only semantically related (-M+S-O, e.g., frigid -cold). Results: In Experiment 1, mean RT showed a constituent priming effect in +M+S+O condition, but no priming effect in -M-S+O. The priming effect of the first constituent was not different from that of the second constituent. In Experiment 2, mean RT showed no significant priming effect in all conditions. Conclusions: Morphological information plays an independent role irrelevant to the form information when Korean-English bilinguals decompose compound words into their individual constituent morphemes in their L1 (Korean). However, morphological decomposition is not relied upon in their L2 (English) processing. The results also suggest an equivalent role of first and second constituents in Korean compound word processing.
Tuire Koponen (University of Jyväskylä, Finland); Paula Salmi; Minna Torppa; Kaisa Aunola; Timo Ahonen; Jari-Erik Nurmi; Marja-Kristiina Lerkkanen; Anna-Maija Poikkeus - RAN and counting as predictors of reading and arithmetic calculation fluency: two population based samples
Purpose: This presentation reviews the results from two longitudinal sub studies of RAN and counting as predictors of reading and calculation fluency among two population based samples. Study 1 examined to what extent children's cognitive abilities predicted their single-digit and procedural calculation skills, and the covariance of these with reading. Study 2 will examine to what extent children's performance in counting and rapid serial naming (RAN) predict fluency of arithmetic calculation and reading at school age among a population based sample. The effects of phonological awareness and memory will be controlled. Methods: Study 1: The children (N=178) were first examined in kindergarten according to their basic numerical skills, linguistic skills and visual attention, and later in the 4th grade according their calculation and reading skills. Study 2: The children (N=1874) were first examined in kindergarten according to their basic numerical skills, linguistic skills and other cognitive skills later in the 2nd and 3rd grade according their calculation and reading skills. Results: Study 1: The results showed that single-digit and procedural calculation, as well as their covariance with reading, were predicted by unique cognitive factors. In addition to counting ability, single-digit calculation, as well as its covariation with text reading, was predicted by naming speed. In contrast, procedural calculation was predicted by number concept skill and mother's education after controlling for single-digit calculation. Study 2: Results will be analyzed. Conclusions: The role of RAN and counting as predictors of reading and calculation fluency will be discussed during the presentation.
Ofra Korat (Bar-Ilan University) ;Iris Levin - E-book enhancing vocabulary and comprehension: The effects of adult's vocabulary support, static and dynamic dictionaries
Purpose: We investigated the effects on word meaning acquisition and story comprehension of e-book reading, with three facilitators: adults' support, dynamic visual support or static visual support. Method: Participants were 145 Israeli Hebrew speaking preschoolers (aged 4 to 6) from middle SES neighborhoods. The sample was assigned to four groups, each reading an e-book, four times, with: (1) adults' vocabulary support (2) dynamic visual vocabulary support, (3) static visual vocabulary support, or (4) without vocabulary support. Children were randomly assigned to groups, within each of 6 classrooms. A pretest-posttest design was used to measure gains in receptive and expressive word meaning and story comprehension. Results: On all measures children's progress was dependent on group, with adult's support appearing as most effective, dynamic support as second, static support as third, and no support as least effective. The gains however differed significantly only between some of the groups, but always between the group supported by the adult and the control group with no support. Conclusions: E-books can be used effectively to facilitate word meaning acquisition, which enhances story comprehension. The contribution of e-book for these purposes can be improved by using adult's support and dynamic visual support.
Anthony Koutsoftas (Seton Hall University)Pradyumn Srivastava - Contributions of reading and oral language ability on the quality of writing in children with and without language learning disabilities
Purpose: Writing is a complex process that is preceded by oral language (OL) ability. Previous research has shown that children demonstrate writing skills even without explicit reading instruction (Berninger, 2009) and that writing skills continue to develop simultaneously with OL (Beers & Nagy, 2008). Thus the relationship among the three language processes (reading, writing, and OL) is not clear. The purpose of this research is to investigate the contribution of OL and reading ability (RA) on writing skills of children with and without language learning disability (LLD). Method: A subset of data from a larger study examining reading fluency and written discourse in fourth and fifth grade children with LLD and typical development (TD) was used to address the study questions. Fifty fourth and fifth grade children (25 with LLD and 25 with TD) provided one narrative and one expository writing sample in response to a prompt. Samples were scored using a commonly employed writing rubric. Results: Four separate stepwise multiple regressions, one for each genre and group, were used to answer the research questions. Three out of the four regressions were significant: Narrative-TD, F (2, 24) = .69, p = .51, Narrative-LLD, F (1, 24) = 5.95, p = .02, Expository-TD, F (1, 24) = 4.33, p = .05, Expository-LLD, F (1, 23) = 10.13, p < .01. Conclusions: Significant amounts of variance were accounted for by the standardized OL or RA measure in three of the four regressions (Adjusted R2 = .17, .12, .28, respectively). OL predicted writing for children in the LLD group while RA predicted expository writing in the TD group.
Emiko Koyama (Ontario Institute for Studies in Education)Emiko Koyama; Esther Geva; Guanglei Hong - Summer reading loss: The impact of ESL and reading disability status.
Purpose: Educators and parents often voice concerns about the possible negative impact of summer vacation on student learning. While meta-analyses have shown the impact of socioeconomic status on summer learning loss, not much is known about how children's characteristics such as speaking English as a Second Language (ESL) and having a reading disability affects summer learning loss. The present study investigates whether children who speak English as a Second Language and children at risk for Reading Disabilities have a slower growth rate in reading during the summer months between Grades 1 and 2 compared to their not-at-risk peers. Method: The current study utilizes piecewise regression to model reading growth during Grades 1 and 2 and during the summer break. The sample consisted of 216 elementary school children (156 ESL students and 54 students at risk for reading disability) residing in a metropolitan city in Canada. Wide Range Achievement Test reading subtest was used as outcome. Results: The results from the unconditional model showed that the reading growth during the summer break was slower than that during the Grades 1 and 2 academic years, but not negative. The hypothesis that ESL students and students at risk for Reading Disabilities would experience a larger summer loss was not supported. Conclusions: When removing the effect of socioeconomic status, summer reading loss seems to affect individuals uniformly, irrespective of whether they are ESL or at risk for reading disability. The impact of home literacy and other summertime activities will be further examined and discussed.
Anuenue Kukona (University of Connecticut & Haskins Laboratories); David Braze; James S. Magnuson; W. Einar Mencl; Kenneth R. Pugh; Whitney Tabor; Julie A. Van Dyke; Donald Shankweiler - Reader skill differences in rapid compositional semantic processing
Purpose: We investigated individual differences in skilled and unskilled readers' responses to interference from "locally coherent" (but globally incoherent) word-level structure in the visual world paradigm, which presents listeners with spoken sentences about a visual scene. Magnuson et al. (2010) found that higher-skilled readers (via PPVT) were more susceptible to interference from locally coherent rhyme structure: skilled individuals fixated images of rhyme competitors (e.g., speaker, when hearing "beaker") more than their less-skilled counterparts. Method: Our sample included urban young adults (N=48) with a wide range of skills. Listeners heard 16 sentences like "The boy will eat the white cake," while viewing picture sets containing objects like white cake, brown cake, white car, and brown car. Results: During the adjective ("white"), listeners anticipatorily fixated the white cake most, based on compositional semantic information from the verb plus adjective. Listeners also fixated the locally coherent white car reliably more than brown car, based on local (but not global: cars are inedible) adjective information. Critically, this local coherence effect was modulated by reading skill: we found a reliable negative correlation between fixation proportions to white car (minus brown car) and reading comprehension (PIAT-R sentence comprehension; and a similar pattern with PPVT). Conclusions: Skilled readers were less susceptible to interference from locally coherent word-level (adjective) structure. Our results suggest that poor readers may have difficulty forming multi-word, compositional (semantic) structure because they are distracted by local word information, which may only be further compounded in print.
Victor Kuperman (McMaster University);Julie A. Van Dyke - Strategies of morphological processing depend on individual verbal skills
Purpose: Computational models postulate multiple parallel routes of morphological processing (Baayen & Schreuder, 2000; Kuperman et al., 2008). We explore how individual verbal skills affect the balance between whole-word and morphemic routes. We identify the skills that influence recognition of suffixed words and/or their morphemes and examine interactions with lexical properties of both. Method: 71 non-college-bound adolescent speakers of US English took 18 skill tests, including phonological awareness, working memory, RAN and comprehension and read 240 sentences, while their eye-movements were recorded. For 80 suffixed words, we fitted linear mixed-models to eye-movement measures, with test scores as predictors. Results: Word identification and comprehension tests interacted with whole-word frequency (trucker) such that the frequency effect on reading times was stronger for poor scorers than better ones. PIAT-R and word/nonword segmentation tests showed interactions with base word frequency (truck), such that a higher-frequency base came with shorter reading times for poor scorers, a weak effect for mid-range scorers, and a noticeable inflation of reading times for high-scorers. Conclusions: Evidence supports both morphemic and whole-word routes, yet their engagement varies by skill: good readers suffer from competition from morphologically related words more than mid-range readers, while bad readers receive a recognition boost from base words. We conclude that readers strategically adjust weights of different sources of morphological information, depending on the quality of lexical representation (Perfetti, 2007). Predictivity of segmentation tasks also suggests the involvement of a general cognitive analytical skill in recognizing complex words, which was not observed when reading simple words.
Lior Laks (Tel-Aviv University); Ruth Berman - Modality-driven distinctions between spoken and written narratives in Jordanian Arabic
Purpose: The purpose of the study is to shed light on the linguistic manifestations of Arabic diglossia by studying the linguistic differences between Spoken and Written Arabic. Method: The talk examines differences between narratives elicited from adult native speakers of Jordanian Arabic in both writing and speech, based on a 7-minute film entitled "Quest" in which an imaginary clay figure searches for water in four mythical landscapes. Analyses concern both language-general differences between spoken and written texts such as text length and lexical and syntactic density (Berman & Ravid, 2009; Chafe, 2004; Nir & Berman 2010) and Arabic-specific distinctions in inflection, derivational morphology, morpho-syntax, and semantics. Results: The study shows that clearly defined diagnostic measures can be devised to characterize the well-known phenomenon of Arabic diglossia (Ferguson 1959), as reflected in spoken and written discourse respectively, by means of carefully controlled elicitation methods that allow for direct comparability between linguistic expressions in the two modalities. Preliminary analysis reveals differences in the use of passive verbs, action nominals, syntactic case marking and the expression of iterativity. Conclusions: The study demonstrates grammatical differences between spoken and written usage in terms of the expressive strategies adopted by native speakers in the course of text construction. Such differences indicate interesting avenues for further corpus-based research in Arabic dialects as well applications for language pedagogy for teaching Standard Arabic to native speakers as well as for teaching colloquial and standard Arabic as a foreign language.
Susan Lambrecht Smith ();Allan B. Smith; Jenny R. Roberts; John L. Locke - Language profiles in children with reading disability from infancy to grade school
Purpose: A variety of speech and language deficits associated with reading disability (RD) are apparent before the age of formal reading instruction. This study follows the early speech and language development of 7 children with RD retrospectively from infancy. Results are discussed in terms of strengths and weaknesses in individual risk profiles. Method: Participants were selected from a group of 58 children participating in a larger longitudinal study. Children enrolled at or before 24 months of age and identified as having RD at school age were selected for inclusion. Phonological, lexical and syntactic production were assessed using speech samples collected from informal play sessions from 15 to 30 months. Vocabulary growth was also assessed using the MacArthur CDI. Measures of pre-reading, including phonological awareness, alphabetic knowledge and rapid naming were administered at age 5. Results: Three of the seven participants demonstrated profiles consistent with broad expressive language delays, including delays in vocabulary, syntax and productive phonology beginning in infancy. While the other participants did not evidence overt global language delays, two showed mixed weaknesses in early productive phonology. All participants performed poorly on measures of phonological awareness in preschool. Conclusions: Results of the present collective case study indicate the route to RD may vary as a function of strengths and weaknesses present within the language domain. These differences are not trivial, as they have significant implications for early identification of children at-risk for RD, and the manner in which therapeutic intervention should be tailored for these individuals.
Karin Landerl (University of Graz, Austria) - Typical and atypical development of basic numerical skills and associations with reading and spelling
Purpose: Comorbidity of deficits in reading and arithmetic has repeatedly been demonstrated but the underlying factors are not very well understood. A central mechanism that has recently been identified to underlie arithmetic development are basic numerical skills. A number of simple tasks (e.g., symbolic and non-symbolic number comparison, dot counting, number line) and experimental effects (e.g., distance or size-congruity effect) are informative on how children represent numbers in their cognitive system. The main aim of the current study is the longitudinal investigation of basic numerical development in children with typical and atypically poor arithmetic skills and whether or not the cognitive representation of number is associated with reading attainment. Methods: At the end of 1st grade, 150 children were screened for either age-adequate or untypically poor arithmetic skills on a standardized battery assessing mental calculation (addition, subtraction, multiplication). Over the next three school years, a computerized battery of basic numerical skills was given five times (beginning and end of grades 2 and 3 and beginning of grade 4). Reading and spelling were assessed in grades 2 and 3 and a working-memory battery was given in grade 3. Results: There are substantial correlations (between .45 and .59, p <.001) between arithmetic and literacy measures at both assessment points. Detailed data analysis of the associations between literacy measures and basic numerical skills is currently under way. Conclusions: Findings will help to specify in what way cognitive skills underlying reading and spelling development impact on the development of the cognitive representation of number.
Karla Lassonde (Minnesota State University, Mankato); Edward J. O'Brien - Male-biased occupations and the limits of gender-fair language: When firefighter no longer means male
Purpose - A series of experiments were designed to test whether gender-neutral language (e.g., firefighter), used in place of male-biased language (e.g., fireman), actually eliminates the bias using an implicit measure called the contradiction paradigm (Albrecht & O'Brien, 1993; O'Brien & Albrecht, 1992). A major benefit of this measure is that it taps into memory processes that naturally occur during reading and does not require an explicit response. Method - Eighty participants read a series of twenty-four narrative passages in each of four conditions. In each passage, a target occupation was introduced using either male-biased or gender-neutral nouns. A target sentence followed defining the gender of the occupation character using either the pronoun "he" or "she." The time it took participants to read the target sentence was recorded. Results - In Experiment 1, participants read target sentences defining the character as a woman (e.g., "she") more slowly following passages that contained male-biased nouns and gender-neutral nouns. In Experiment 2, an additional sentence was added to the text that defined the occupational character as a woman. This information eliminated the disruption in reading found in Experiment 1. Conclusions - The results suggest that gender-neutral language can moderate against activation of gender stereotypes but only when combined with an explicit mention that the target occupation character is female. Currently we are assessing whether the addition of an adjective (e.g., careful/strong) used to describe the occupation in text will either reinforce the gender stereotype or will work to eliminate any male bias when combined with gender-neutral language.
Mark Lauterbach (CUNY Graduate Center); Francis Tabone; Sara Martinez; Michael Termini - How a little calculus can help special education teachers understand reading tests better: the development of a teacher friendly growth metric for standardized reading test data
Purpose To develop an accurate and easily interpretable metric that captures current ability and growth in reading skill for moderately to severely delayed readers across a wide spectrum of reading levels for both instructional planning and research purposes. Method The population was 67 special education students between the ages of 7.5 and 21 with an average full scale I.Q. of 70 who attended a private special education school. Reading data from the Group Reading and Assessment and Diagnostic Evaluation (GRADE) was collected over a three-year period. Logarithmic curves were fit to the students' total test w-ability scores for each student and compared to the expected growth by comparing the derivatives at corresponding points between the student's growth curve and the w-ability curve provided in the test manual. The scores were standardized in terms of the expected gain for that particular point on the curve. Results The result is a score centered at zero (normal development rate) and extends in terms of the proportion of w-ability expected at that particular point for a year's growth (-.5 = 50% below and .5 = 50% above), that is strongly related to I.Q. and severity of disability but not related to absolute score. Conclusion This has proved useful for teachers who work with students with a wide range of reading abilities/disabilities for both understanding the nonlinear growth expectations in reading and instructional goal setting for individual students. Additionally, it is useful for program analysis where a majority of students are well out of their age/grade norms.
Amy Lederberg (Georgia State University);Amy R. Lederberg;Carol M. Connor; Elizabeth M. Miller; Susan R. Easterbrooks - Teaching deaf and hard-of-hearing prekindergarteners early literacy skills
Purpose-This paper will present preliminary evidence for the efficacy of a newly-created curriculum called Foundations for Literacy for promoting the development of alphabetic knowledge, phonological awareness, and vocabulary among deaf and hard-of-hearing (DHH) prekindergarteners. Method-A battery of standardized tests was administered to intervention (n=20) and comparison DHH children (n=39) who were eligible for the intervention but did not receive it. Intervention children were taught an hour a day by research teachers for the school year. Average age of both groups was similar (M age =53 and 54). Results-A MANCOVA (with pretest raw scores as a covariate) showed that intervention children made significantly greater gains than did comparison children on alphabet knowledge and phonological awareness skills, F (3, 53) = 6.99, p < 0001. Intervention children's phonological awareness standard scores on the TOPEL-PA subtest increased from 84 to 95; while comparison children changed from 89 to 86. MANCOVAs that examined changes in vocabulary (WJVocab, EOWPVT, PPVT) did not show an effect of intervention. Follow-up MANOVAs showed a main effect of time for vocabulary, with standard scores increasing, on average, 5 points for both the PPVT (83 to 88) and the EOWPVT (81 to 86) indicating that children were generally showing greater gains in vocabulary than might be expected based on their age but that Foundations was not more effective than the instruction received by the comparison children. Conclusion--These results suggest that DHH children have the potential for developing age-appropriate skills that are foundational for reading.
Julia Lee (Universiti Malaysia Sarawak); Stephanie Al Otaiba - Socioeconomic and Gender Group Differences in Early Literacy Skills: A Multiple Group Confirmatory Factor Analysis
This poster presents the findings of SES and gender group differences in early literacy skills among kindergarteners. Students (n = 462) attended 10 schools serving a diverse socioeconomic population in a mid-sized southern city. Students' mean age in fall was 5.16 (SD = .30). About 53.5% of students were male. The present study is based on archival data of a larger longitudinal School-based Classification and Prevention of Learning Disabilities study. Using multiple group confirmatory factor analysis, we examined measurement invariance of the early literacy constructs (i.e., alphabet knowledge, phonological awareness, and spelling) across SES-gender groups. We found that there was measurement invariance (i.e., no construct bias) of the early literacy constructs across boys and girls who received free or reduced price lunch (FRL), a proxy for SES, and those who did not receive FRL. Furthermore, the findings suggest that the poverty gap and gender gap in early literacy skills at the early stages of schooling are more nuanced than previously found (e.g., Chatterji, 2006; West, Denton, & Germino-Hausken, 2000). It is not simply the case that all boys perform more poorly than girls. Rather, the gender gap is moderated by SES; poverty defines how well young boys and girls perform on early literacy skills. The standardized effect sizes of the latent means suggest that boys and girls who are eligible for meal subsidies are particularly weak in all three constructs. These findings suggest that it is important for high-quality early childhood programs to be provided at home and in preschools.
Erica Lembke (University of Missouri) - Screening and progress monitoring in early elementary using transcription measures
Purpose. The purpose of these studies was to examine technical adequacy and technical features of slopes produced from Curriculum-Based Measures (CBM) of beginning writing. Research questions included: What is the reliability and validity of word and sentence copying and dictation tasks? What are the growth rates of these measures when administered weekly and are growth rates significant? Method. Study 1--concurrent criterion validity of early writing transcription measures was examined (n=15 grade 2 students). Students completed 3 minute measures of word and sentence copying and dictation, along with other holistic and atomistic tasks that served as the criterion measures. The measures were scored for Words Written (WW), Words Spelled Correctly (WSC), Correct Word Sequences (CWS) and Correct minus Incorrect Word Sequences (C-IWS). Study 2--(n=23 grade 1 students), the same copying and dictation measures were utilized and scored weekly for 8 weeks. Results. Dictation measures yielded strong validity correlations with the atomistic criteria, particularly when scored using CLS, WSC, and CWS (r > .80 for 22 out of 32 correlations across all scoring methods). Alternate-form reliability was strongest for Word Copying and Sentence Dictation (r = .71 to .98), and criterion validity with the TEWL (Hresko et al., 1996) was strongest for Word Dictation (r = .52 to .61). Using HLGM, growth for all tasks was significant and weekly growth rates were greatest for the word-level measures (Copying and Dictation). Conclusions. These studies demonstrate that transcription measures may be used as static or growth measures for beginning writers.
Christopher Lemons (Peabody College of Vanderbilt University); Amanda Kloo; Naomi Zigmond - The 2% test: Who is crying over spilled milk?
Purpose: An alternate assessment based on modified academic achievement standards (AA-MAS) or 2% test was offered as one testing option within the NCLB accountability framework for students in Pennsylvania during the 2009-2010 school year. The purpose of this study was to: a) examine which students teachers placed into the AA-MAS, b) determine how differing groups of students performed on the assessment, and c) gain an understanding of the decision processes Individualized Education Plan (IEP) teams used in making the test participation decisions. Method: First, a descriptive analysis will be presented in which the characteristics of students placed into the AA-MAS will be provided (e.g., disability category, previous performance category on the state assessment). Second, a summary of characteristics of those children who were able to obtain a performance of 'proficient' on the AA-MAS will be provided. Third, the results of interviews with staff from four school districts will be presented. In each district, staff interpreted the state guidance differently, resulting in a qualitatively different population of students being placed in the AA-MAS. Results: Results indicate that the AA-MAS in the state of Pennsylvania remains a rigorous, challenging assessment of grade-level content and that prior performance on the general assessment is a significant predictor of performance on the AA-MAS. However, there is significant confusion between school staff regarding the target population for the test. Conclusions: Results indicate that the state guidance may be misaligned with the developed test and that schools that follow the guidance strictly may be placing a group of students into the assessment who are not likely to obtain proficiency on the AA-MAS. Suggestions for improving this testing option will be provided.
Arne Lervag (University of Oslo)Monica Melby-Lervåg - Oral Language Skills Moderate Nonword Repetition Skills in Children with Dyslexia: A Meta-analysis of the role of Nonword Repetition Skills in Dyslexia.
Purpose: A number of studies have shown that children with dyslexia experience problems with nonword repetition (e.g., Goulandris, Snowling & Walker, 2000; Messbauer & de Jong, 2003). There is, however, disagreement regarding the magnitude of the nonword repetition deficit (e.g., Boada & Pennington, 2006 versus Marshall, Snowling & Bailey 2001). The current study seeks to explain these differences between studies using a meta-analysis, and examines studies that compare children with dyslexia with chronological age matched controls (CA) and reading level controls (RL). Of primarily concern is whether differences in phonological awareness, verbal short-term memory skills and/or oral language skills are able to explain the variation in nonword repetition skills found between studies. Method: A meta-analysis reviewing studies that have focused on the relationship between dyslexia and nonword repetition was conducted. Results: The results show that children with dyslexia have poorer nonword repetition skills when compared to both chronological-age and reading-level controls. However, the severity of the nonword repetition problem varies significantly between studies, and the most important predictor of this variability is oral language skills. Conclusion: We argue that the sizable explanatory value of oral language skills reflects sample variation in the rate of comorbidity between dyslexia and specific language disorders (SLI). Thus, samples with combined dyslexia and specific language disorders demonstrate the most severe nonword repetition problems.
Paul Leseman (Department of Education, Utrecht University);Lotte F. Henrichs; Hans Cohen de Lara; Kees Broekhof - Promoting academic language use in kindergarten science activities: a randomized controlled intervention study
Purpose This study examined how teachers can promote 5-year-olds' ability to verbalize their intuitive non-verbal understanding of physics phenomena. Even though children sometimes have surprisingly adequate scientific intuitions, sharing their insights with peers and teachers can be challenging, yet seems necessary for further conceptual development. We propose that an early command of 'academic language' (AL) skills enables children to succeed better in such sharing. Method 59 teachers and 230 children were observed twice (winter and late spring) during playful science lessons about light reflection and air pressure. After the first observation, thirty teachers were randomly assigned to a training session in which they learned about AL and were made aware of the many affordances of early science lessons to promote AL in 5-year-olds. A second observation was used to determine training effects. Results Trained teachers and children, guided by trained teachers, show increased lexical richness of the conversations in science activities compared to the no-treatment control group. Teachers and children in the experimental group made much more use of specific 'technical' words (air pressure, reflection) and general academic words (explore, discover, conclude). Utterances in the trained group were on average longer and contained more causal, temporal and contrastive connectives. Trained teachers engaged children more often in extended scientific reasoning rather than simple labeling or describing. Conclusion. Promoting teachers' awareness of the affordances science lessons offer for academic language learning has positive effects on the quality of teacher-child discourse, facilitates children's sharing of their scientific intuitions, and stimulates their scientific reasoning.
Kyle Levesque (Dalhousie University); Hélène Deacon - Do children see 'break' in breakable more than in breakage? Influence of base morpheme frequency and suffix productivity on children's derived word reading
Morphological awareness has been shown to be a robust contributor to reading development independent of phonological awareness and orthographic knowledge. Moreover, investigations in this area have contributed to the assertion that developing readers are sensitive to the morphemic structure of derived words. In particular, the frequency of the base morpheme (e.g., 'bold' in boldly) appears to have a privileged role in children's word reading beyond that of family size (number of morphological relatives of a base) and family frequency (summed frequency of morphologically related words). However, a recent study involving adult readers has raised the concern that base frequency effects may be mediated by suffix productivity (form-meaning consistency across derived words; Ford, Davis, & Marslen-Wilson, 2010). Therefore, it is important to consider the potential impact of suffix productivity on children's word reading in relation to other morphemic factors. The purpose of this study is to better understand the influence of morphemic structure on children's derived word reading. Using a correlational design, we examine the independent contributions of family size, family frequency, base frequency, and suffix productivity, while controlling for variables such as number of phonemes, semantic relatedness, and derived word surface frequency. No published study to date has investigated the combination of these factors using item-based regression analyses to help clarify their influence on word reading. To our knowledge, this is the first attempt to contrast the role of family size, family frequency, base frequency, and suffix productivity in a sample of school-aged children.
Chuchu Li (University of Maryland, College Park)；Candise Lin; Min Wang - The processing of tonal information in reading Chinese characters among young children
*Purpose: To examine 1) Do children represent tonal information in visual character recognition when it is not represented; 2) Does tone processing differ between reading Pinyin and Chinese characters; 3) What is the developmental pattern in processing tonal information? *Method: Sixty students in total will be recruited from Grades 2, 4 and 5 in an elementary school in Beijing, with 20 in each grade level. Two naming experiments will be conducted using priming paradigm. In the first naming task, primes and targets are both Chinese characters. There are four prime conditions: 1) T+S+ (the prime and target share the same Tone and syllable Segment); 2) T+S- (the prime and target share the same Tone but differ in syllable Segment); 3) T-S+; and 4) T-S-. In the second naming task, primes are Pinyin and targets are Chinese characters. We decided to focus on two of the aforementioned conditions: T+S+ and T-S+. *Results: Data collection and analysis will be completed by May, 2011. We hypothesize that if tonal information is represented and processed, there will be a greater priming effect in the T+S+ condition compared to T-S+ and a greater priming effect in the T+S- compared to T-S-. This difference may be greater in younger children. Pinyin primes may have a greater priming effect than the character primes. *Conclusions: The significant tone priming effect will suggest that children activate tonal information during visual character recognition even when it is not represented in the written characters.
Miao Li ()John Kirby; Wei Zhao - An examination of the Simple View of Reading Model in Bilingual Children
Purpose. We investigated whether the Simple View of Reading (SVR) model is applicable to Chinese English-immersion students and whether the SVR model should include vocabulary and strategy components. Method. Participants were 246 Grade 8 students in English-immersion programs in China. English measures were Woodcock Listening Comprehension (LC) test, Test of Word Reading Efficiency (D), Gates-MacGinitie Vocabulary subtest, Reading Strategies (modelled after Cromley, 2007), and Gates-MacGinitie Reading Comprehension (RC) subtest. Chinese measures were Woodcock Listening Comprehension test (LC, Chinese version), Chinese Character Reading (D). Results. Data collection has just been completed. Preliminary results indicate that English LC and D significantly predict English RC. The model still holds after controlling for Chinese LC and D. English vocabulary and strategy explained additional variance in English RC after Chinese LC and D, and English LC and D alternatively. Structural equation modeling will be used to test the revised model and to estimate the direct and indirect effects of each predictor on RC. Conclusions. The SVR model can be applied to students whose English is a second language (L2) and it still holds after considering L1 factors. L2 vocabulary and strategy contribute to the L2 RC over and above the effects of both L1 LC and D and L2 LC and D. Implications for theories of L2 reading development will be discussed.
Meng Feng Li (Department of Educational Psychology and Counseling, National Taiwan Normal University); Meng-Feng Li; Yi-Fen Su; Chi-Ching Chuang - The development of character-identification strategies for Chinese regular and irregular compound characters among Chinese beginning readers
Purpose - The aim of this study was to investigate the cues used by Chinese beginning readers to identify regular and irregular compound characters. Among modern Chinese characters, about 82 percent of the characters are semantic-phonetic compound characters, which contain two parts: a semantic radical, which provides information about meaning, and a phonetic radical, which carries information about pronunciation. According to the consistency of pronunciation between compound characters and their phonetic radicals, regular and irregular compound characters are defined. This study attempted to examine the relative role of sound, visual similarity, phonetic radical, and semantic radical in regular and irregular character identification for Chinese beginning readers. Method - Ninety nine average readers were recruited from first, second, and third grades respectively. A task developed by Khomsi (1985) was administered to know which cues were used by children to identify regular and irregular compound characters. A 3×2×7 (grade level× regularity × stimulus type) mixed-factor designs were adopted. Results - The three-way interaction effect was significant both across subject means and across item means. The post-hoc comparisons revealed that phonetic radicals were the most important cues for first and second graders to identify regular characters, but not irregular characters. Conclusion - Most first graders and some of the second graders tended to use phonetic radicals, not merely sound, as cues to identify regular characters. However, for irregular characters, because the pronunciation between characters and their phonetic radicals are inconsistent, beginning readers rarely adopted phonetic radicals as cues for character recognition.
Chi-Shun Lien (National Chung Cheng University) - Children's moral theme comprehension: the processes of inference generation during and after reading moral stories
Purposes Young children are usually unable to extract the superordinate themes from moral stories, whereas they are capable of understanding the protagonist's intention and consequences of action. The purpose of this study was to examine whether readers at different age levels were able to make accurate inference and construct appropriate mental representation of theme during and after reading moral stories. Method Participants were 20 fourth graders, 20 sixth graders and 15 college students, recruited from Chiayi, Taiwan. Each participant was tested individually and read two moral stories promoting two different values (i.e., responsibility and caring). Participants were presented a moral story sentence by sentence and asked them to think aloud for probing the on-line process of reading. After reading, two off-line measures were administered: participants were asked to recall the story and then to answer the factual questions, inferential questions and identify the story theme. Participants' responses were scored on the basis of a scoring key containing a set of categories of inference. All responses were scored twice by two experienced raters and disagreements were resolved after discussion. Results For the on-line measure, a significant developmental trend was found in think-aloud protocols. College students generated much more inferences than 4th and 6th graders. Most importantly, college students produced more accurate global inferences, whereas young children tended to make more local inferences. In addition, younger children produced more communicative messages (e.g., self-questionings, suggestions, comments, etc.) and inaccurate inferences than older participants. For the off-line measure, college students performed significantly better than 4th and 6th graders on all measures. The 6th graders outperformed 4th graders on the measure of theme identification. However, there was no difference between 4th graders and 6th graders on the free-recall, and the factual and inferential questions. Conclusion The findings showed that young children were less able to generate global inferences and unable to construct a global coherent mental representation for moral story. They could not capture the underlying value in the story. For implication, it is important for educators to teach reading strategies, which encourage young children to generate global inference during and after reading.
Candise Lin (University of Maryland, College Park); Min Wang; Yi Xu - Stress Perception among Mandarin-English Bilinguals, Korean-English Bilinguals, and English Monolinguals
Background We compared stress sensitivity among Mandarin-English (ME) and Korean-English (KE) bilinguals, and English monolinguals. Lexical stress exists in Mandarin and English, not in Korean. When learning English stress, ME may have an advantage if their knowledge of L1 stress transfers. KE may experience difficulty because they have no experience with stress. Methods Participants were 20 ME, 10 KE, and 20 English speakers. The sequence recall task had two minimal pairs of nonwords: a phoneme contrast (/'kupi/-/'kuti/) and a stress contrast (/'mipa/-/mi'pa/). For both contrasts, participants learned to associate the first nonword with the key "1" and the second with "2." Participants recalled sequences of 2-6 nonwords by pressing the keys (press 121 for /kupi-kuti-kupi/). To control for non-linguistic rhythmic sensitivity, an additional sequence recall task was administered using a minimal pair of drumbeats which resemble the acoustic properties of the stress contrast. The second experiment was lexical decision which consists of real words and nonwords that differ in stress location (ladder: /'lædər/-/lə'dær/) Results ME performed better than English speakers in rhythmic sensitivity (p=.004). On the stress contrast, ME are better than KE (p=.031) and they are marginally better than English speakers (p=.086). No significant group difference was found for the phoneme contrast (p=.113). Data collection is underway for the lexical decision task. Conclusions ME have higher stress sensitivity than KE as a result of transfer of stress knowledge from Chinese. ME's higher non-linguistic rhythmic sensitivity may facilitate their stress sensitivity. More data for the KE group will be collected.
Pei-Ying Lin (University of Saskatchewan)Yu-Cheng Lin - The Visual-Spatial Orthographic Skill in Chinese and Bilingual Children
*Purpose - This study examines Chinese and Chinese-English bilingual children's visual-spatial orthographic skill. *Method - Two groups of Grade 6 students in Taiwan and Toronto were participated in the study (52 Chinese and 35 Chinese-English bilingual students). They received a computer-based Chinese character structure task, which reflects the seven traditional Chinese character structures (Horizontal structure type 1 and 2, Vertical structure type 1 and 2, P-shape, L-shape, and Enclosure) and three levels of stroke density (high, medium, low levels). The accuracy and response time of each item were recorded by E-Prime and analyzed with Rasch item response model. *Results - (1) Vertical, Enclosure, as well as P- and L-shapes are easier than Horizontal structures for Chinese children, whereas Horizontal and Enclosure structures are easier than Vertical, P- and L-shapes for bilingual children. (2) Chinese children processed Vertical type 2 and Enclosure structures faster than other structures. Bilingual children also processed Enclosure structure faster than other structures. However, Chinese children processed Horizontal type 2 structure significantly slower than bilingual students. (3) Accuracy and response time were significantly correlated for bilingual children. (4) No significant effect of stroke density was found between and within the groups. *Conclusions - Chinese and bilingual children's accuracy and processing speed of visual-spatial orthographic skill are influenced by the commonly used reading pattern and orientation (Chinese: vertical; English: horizontal). Bilingual children with slow processing speed are likely to have poor visual-spatial orthographic skill. Finally, character structures play a more important role than stroke density for the children.
Yu-Cheng Lin (University of Texas at El Paso)Pei-Ying Lin - The Effects of Word Spaces on Reading Efficiency of Chinese Normal and Poor Comprehenders
Purpose: This study investigated the effects of word spaces on the reading efficiency in Chinese normal and poor comprehenders (PCs). Method: Participants were 36 fifth-grade normal readers and 22 PCs. Half of normal readers and PCs were randomly assigned to the word-spaced condition as experimental groups (normal: n = 18, PCs: n =11). The other half was assigned to the unspaced condition as control groups (normal: n = 18, PCs: n = 11). Experimental and control group were matched for age, word recognition, nonverbal IQ, vocabulary, and socioeconomic status. The word-spaced texts read by the experimental group were inserted with word spaces, whereas the unspaced condition was ordinary texts without any word spaces. All participants received six 45-minute reading training sessions. In each training session, students read the word-spaced or word-unspaced stories (depending on the condition they were assigned to) and were given comprehension questions by the instructor after the stories. Twelve short stories with an average length of 212 words were designed and randomly divided into three versions, four stories in each. The three versions of stories were presented to participants on computers with E-prime to measure children's reading efficiency such as the reading speed and comprehension (before, during, and after the trainings at three different time points). Results: Word spaces decreased the normal readers' reading speed but increased PCs' reading speed after the reading training. Moreover, word spaces did not benefit the comprehension of normal readers and PCs. Conclusions: Educators should provide word-spaced texts to Chinese PCs.
Kristen Lin (); Donald Bolger - Representation of Letters and Words in the Brain
Purpose - To determine the neural correlates of integration between orthography and phonology in visual word processing. Method - Adults ages 18-30 were tested using a masked priming lexical decision task where priming effects were observed for both word and pseudoword stimuli. The study was divided into 2 parts: a behavioral experiment manipulating prime duration and an fMRI experiment. In the first experiment, masked priming included 8 conditions; non-matched pairs (fish-tree), identical pairs (goat-goat), phonologically consistent pairs (tank-sank), and phonologically inconsistent pairs (hint-pint) of either pseudowords or words. Ninety participants were divided evenly into 3 groups and masked primes were presented at 17, 34, or 50ms. In the second experiment, 17 participants were given a similar task in the fMRI scanner. Results - In experiment one, the 17ms trials showed that subjects responded significantly faster to words than to pseudowords, but there was no difference between conditions. At 34ms and 50ms, subjects responded significantly faster to trials with phonologically consistent primes than to inconsistent primes and with a trend towards a similar pattern in pseudowords. For the fMRI experiment, inconsistent words more than consistent words significantly activated the inferior parietal lobe (IPL) and the inferior frontal gyrus (IFG). Conclusions - Behavioral results indicate facilitation by consistent words and inhibition by inconsistent words, especially at the 34 and 50ms presentations. fMRI results show the IPL, an area indicated in integration and extracting regularities within a word, and the IFG, an area involved with more demanding reading tasks. This also corroborates our behavioral results.
Sven Lindberg (DIPF / IDeA)Telse Nagler; Marcus Hasselhorn - Individual Differences of Elementary School Children's Calculation and Reading Strategies
Purpose: The purpose of this study is to examine elementary school children's reading and calculating strategy behavior to identify specific and overlapping strength and deficits across the domains by using microgenetic analysis. Method: Overall the sample consists of 86 1st grade students. All children were confronted with a set of 30 word cards consisting of regular and irregular words and pseudowords and a set of 32 calculation tasks of different difficulty to evoke the usage of several distinct strategies. The words and calculating tasks were presented individually according to migrogenetic trial-by-trial procedure during which a task related question is asked after each item. Each session was taped on video. Results: The analysis of the strategy frequencies revealed that all children used different and multiple strategies and the applied strategy behaviors differ among the varied word and calculating tasks groups. Repeated measures analysis of variance (ANOVA) showed that the use of retrieval and elaborated strategies increase with the skill level. Moreover, skilled children in both domains used strategies more effective and made fewer mistakes whereas less skilled children used primarily unelaborated strategies and made more mistakes. Conclusion: The use of the microgenetic analysis to examine children's reading and calculating skills seems to be very promising. Children's strategy behavior can be described in more detail and individual differences between different skilled students can be discovered. The peculiarities of the strategy behavior of less skilled children may help to understand their shortcomings and can be used as the foundation of specific training.
Robin Litt (Oxford University)Kate Watkins; Kate Nation - Paired associate learning deficits in dyslexia: a reflection of association type or verbal task demands?
Purpose - Previous research has established a relationship between reading and Paired Associate Learning (PAL). However, the specific nature of this relationship is not well understood. We investigated the hypothesis that children with dyslexia have specific impairments in crossmodal (visual-verbal, verbal-visual), but not unimodal (verbal-verbal, visual-visual) PAL. Method - Forty-five children (15 dyslexic, 15 CA controls, 15 RA controls) aged 8-11 were matched for nonverbal intelligence and tested across four conditions (visual-verbal, verbal-verbal, visual-visual, verbal-visual). Stimuli comprised novel abstract symbols and nonwords. On day one, participants completed a computerized PAL task, consisting of two presentation trials and five test trials with feedback. The next day, delayed recall and yes/no recognition tasks were administered. Learning rate, accuracy, and reaction time were recorded. This procedure was repeated for all conditions. Data were analyzed using mixed factorial ANOVAs and multiple regression. Results -Children with dyslexia performed as well as CA controls in the nonverbal condition (visual-visual), but significantly lower in all conditions with a verbal component . Performance was similar to RA controls. Contrary to the hypothesis, performance was not selectively impaired in crossmodal PAL. However, the finding of impaired verbal-visual PAL, which required no verbal output, suggests that the demands of a verbal response cannot explain the findings. Conclusions -This study raises important questions as to whether previously reported PAL deficits in dyslexia are due to crossmodal PAL or verbal learning impairments. While the answer is not entirely clear, the results generate interesting hypotheses for the future.
Yongguang Liu ()Ling-po Shiu - Fast orthographic learning in Chinese and its relationship to phonological decoding
Purpose: The present paper investigated fast orthographic learning (FOL) in Chinese children. FOL means that orthographic representation of a word can be learned incidentally after minimal exposure. First, we examined whether there is FOL using a procedure modeled after Share (1999) (Experiment 1). Second, the effect of phonological decoding on FOL was examined by manipulation of articulatory suppression (Experiment 2). Method: In Experiment 1, thirty second graders encountered each of 5 pseudoword targets six times in the context of reading stories. Then the participants completed two posttests: an orthographic choice test (containing alternate homophonic spellings of the targets) and a spelling test, both immediately (Time 1) and after 3 days (Time 2). In Experiment 2, twenty-five third graders learned the five targets through a lexical decision procedure (Group 1) and another twenty-five through a naming procedure (Group 2). Group 1 vocalized unrelated sounds during target exposure to minimize phonological decoding. The two groups were matched by their scores of school Chinese tests. Posttests were the same as in Experiment 1. Results: In Experiment 1, the targets were chosen correctly in the orthographic choice test more often than chance level. Specifically, the targets were chosen more often, spelled more accurately than their alternate homophonic spellings. In Experiment 2, Group 1 performed as well as Group 2 on all posttests (except spelling in Time 2). Conclusions: FOL was found with Chinese children. The effect of phonological decoding on orthographic learning needs further clarification.
Jessica Logan (The Ohio State University) - Growth in reading from Kindergarten to Grade 6: Genetic and Environmental Influences
Purpose - The present study is the first to examine the genetic and environmental influences on growth in reading skills from kindergarten through early adolescence. Method: Participants were drawn from the Western Reserve Reading Project, a study of 314 twin pairs based in Ohio. Twins were assessed via six annual home visits from kindergarten through Grade 6. Assessments included word identification, non-word decoding, reading comprehension, and rapid naming. Analyses were conducted using latent growth curve modeling. Variance of the intercept, linear, and quadratic slopes was partitioned into variance due to genes, shared environment, and non-shared environment. Results: Results suggested evidence for quadratic growth in children's reading skills from kindergarten through Grade 6. Specifically, higher intercepts were related to slower growth, and faster deceleration of growth. Significant genetic and environmental influences were observed for growth in word-reading, decoding, and rapid naming. Significant environmental effects on growth were also observed for reading comprehension. For all four outcomes, genetic influences on the intercept and slope were unique to each other. However the same environmental influences were shared between the intercept, slope, and quadratic slope. Conclusions: Results suggested that although genetic effects are present, growth in reading skills is influenced by the environment. Further, this growth may be malleable through environmental influences, such as school environment, family environment, or intervention.
Linda J. Lombardino (University of Florida);Lombardino Linda; Tibi Sana; Ho Yiting - Literacy skills in second grade Arabic speaking children: A comparison of children from three demographic areas in the West Bank
Purpose The study was designed to (1) investigate differences in performance between children who live in different demographic areas within the West Bank and (2) provide a preliminary database by which to identify young Arabic children who are at risk for reading disorders. Methods: One hundred-fifty Arabic speaking children between 7:0 and 7:6 years of age and in the beginning of their second grade year in school were tested. Measures were developed to assess the following skills: reading-related knowledge (RRK) (e.g. letter knowledge, concept of word, invented spelling), phonological awareness (PA), rapid naming (RAN), and word reading (WR). Children were selected from three demographic areas in West Bank (city, village & camp). Results: Strong correlations between RRK measures (letter knowledge, concept of word, invented) and PA tasks were found. Moderate correlations were found between RAN and PA subtests and between RAN and RRK tasks. Most importantly, significant differences were found between the performance of children from the city and those from the village on the RRK and PA measures. Twenty-two percent of the children performed one standard deviation below the mean on the RRK tasks combined and 16% performed one standard deviation below the mean on the PA tasks. Conclusions: Reasons for the disparity in performance among children in different demographic areas will be discussed along with recommendations for implementing prevention and early intervention programs to reduce the large number of children, especially in the villages, who are failing to acquire the necessary skills for becoming skilled readers.
Debra Long (University of California, Davis)Stephen Hamilton - As The Plot Thickens: Interactions between Text- and Reader-Characteristics in the Processing of Narrative Texts
Purpose To investigate relations among text properties, reader characteristics and sentence reading times (RTs) using Hierarchical Linear Modeling (HLM), a statistical technique designed for the analysis of nested data. Method Data that varied from sentence to sentence were considered at the first level of the model, and data that varied from person to person were considered at the second level of the model. At level one, variance in RTs was estimated as a function of length, word frequency, new argument nouns, and semantic overlap. In addition, the properties of a target sentence (sentence n) were used to estimate variance in RTs for sentence n+1 (spillover effects). At the second level, coefficients derived from the level-one analyses were used as criterion variables upon which individual difference variables were regressed. These variables were Nelson Denny Vocabulary, ND comprehension, working-memory span, and decoding skill. Results A span-by-decoding interaction at the intercept showed that poor decoders took longer to read sentences than good decoders, unless they also had high working-memory scores. Spillover from new argument nouns varied as a function of individual differences: A 3-way, cross-level interaction between new argument nouns, decoding and span revealed that slow decoders experienced more spillover from new argument nouns, but this effect was attenuated by WM. Conclusions Our results show that the influence of text properties on reading vary substantially as a function of individual differences. They accord well with Perfetti's (1985) verbal efficiency theory, which asserts that decoding problems affect integration processes by monopolizing working-memory resources.
Christopher J. Lonigan (Florida State University; Florida Center for Reading Research);Beth M. Phillips - Does preschool children's self-regulation moderate the effectiveness of language and literacy instruction? Evidence from a randomized evaluation of a Tier II intervention
Purpose: To determine if young children's self-regulation affected positive impacts of an early literacy intervention in the context of a Response to Instruction model of intervention. Method: Assessments of children's language and literacy skills were conducted at the start, middle, and end of the preschool year. Preschool children were identified as failing to make adequate progress in language and early literacy domains (i.e., phonological awareness, print knowledge) in the context of an effective classroom curriculum. Based on the mid-year assessment, 184 children who scored at or below the 20th percentile on a variety of language and early literacy measures were deemed eligible, and were randomized to a Tier II intervention condition or a no-additional-intervention control. Children randomized to the Tier II intervention received small-group instruction for 10 to 20 minutes per day for 12 weeks. Self-regulation was assessed using the Head-Toes-Knees-Shoulders task and teachers' reports. Results: Children receiving the Tier II intervention significantly outscored children in the control group at the end of the preschool year. Generally, measures of self-regulation were associated with growth in skills; however, in all but a few cases, self-regulation did not moderate the impact of the intervention on children's language and early literacy skills. Conclusions: The results of this study fail to provide a broad level of support for the hypothesis that children's self-regulation has a direct effect on the benefit of instructional activities. The nature of the intervention (i.e., small-group, focused, time-limited), however, may be partially responsible for the absence of a moderating effect.
Susan Loveall (University of Alabama); Frances Conners - Testing the self teaching hypothesis: phonological recoding vs. no-exposure
Purpose - The self-teaching hypothesis suggests that orthographic learning is accomplished through phonological recoding. In most self-teaching studies, children recode nonwords in text and later complete spelling and multiple choice tests of the nonwords. Correctly spelling the nonwords or choosing them over pseudohomophones is considered support of the self-teaching hypothesis. However, it may be possible to succeed on these tests using prior orthographic knowledge. For a more stringent test of the self-teaching hypothesis, our 2-part study compared recoding, repetition, and no-exposure conditions. Method - In Experiment 1, 19 participants completed learning tasks under two conditions - (1) phonologically recode nonword targets and (2) repeat pronunciations of nonword targets. Three days later, they completed spelling and orthographic choice tests. In Experiment 2, 27 participants completed the spelling and orthographic choice tests without exposure to the nonwords. Results and Discussion -Experiment 1 participants chose the previously recoded nonwords more often than pseudo-homophones t(18) = 7.16, p < .001. Also, they performed better in the recoding condition than in the repeating condition on both spelling, F (1, 18) = 64.14, p < .001, and orthographic choice F (1, 18) = 3.58, p = .075. Experiment 2 analyses showed that participants in the recoding group performed better than participants in the no-exposure group on both spelling, F (1, 43) = 8.82, p < .025, and orthographic choice, F (1, 43) = 15.23, p < .025. Results are consistent with the self-teaching hypothesis and rule out the explanation that participants' prior orthographic learning determined their orthographic learning performance.
Maureen W. Lovett (Learning Disabilities Research Program) - Interventions for children and adolescents with reading disabilities: Variability in reading growth
Wei-Pai Lu (National Changhua University of Education)Wei-Pai Lu - Is discrepancy a fake phenomenon for children with reading disabilities ? Findings from a nationwide longitudinal dataset
Purpose The goal of this study is to extend the previous two-group classification hypothesis studies and reexamine the phenomenon of discrepancy among students with a reading disability. I asked the following research questions: (a) Do students with a reading disability label differ from poor readers without a disability label with respect to the ecological factors? (b) Is reading disability a unique entity? If it is, do the profiles of poor readers with and without a disability label fit "simple view of reading formula"? Method The sample is drawn from the Early Childhood Longitudinal Study Kindergarten Class of 1998-99 (ECLS-K). I identified three groups: average students, students with reading disabilities, and poor readers without a disability, and I compare groups at two time points: third grade and kindergarten. A technique of multivariate analysis was used to examine the group-differences and a multivariate post hoc technique was also used to examine the contribution of each variable to resulting effects. Results The results from the multivariate model provide evidence that students with reading disabilities differ from their poor reading peers. The most two important variables to respectively separate these two groups at kindergarten and at third grade are the general knowledge variable and the science knowledge variable. All reading disable subgroups achieved higher than poor reading subgroups in these two variables, regardless their SES levels. Comparing to garden-variety poor readers, children with labeled reading disabilities have uneven cognitive profiles that cannot be explained by their extrinsic background.
Yang Luo (OISE/University of Toronto)Xi Chen; Helene Deacon; Hong Li - Development of Chinese Orthographic Processing: A Cross-Cultural Perspective
Purpose This study investigated the development of orthographic processing in Chinese literacy acquisition, and whether and to what extent home print exposure, character reading, and phonological skills would predict Chinese orthographic processing. Furthermore, we carried out the investigation in China and Canada to examine the effect of linguistic context on these developmental and predictive patterns. Method We tested 94 kindergarteners and first graders in Mainland China and 91 Canadian-Chinese kindergarteners and first graders in Canada on Chinese orthographic processing (orthographic discrimination, radical position awareness and radical form awareness), phonological awareness, character reading, rapid digit naming at the beginning and the end of the academic year. Children's home print exposure and nonverbal abilities were measured at the pretest. Results Children in China developed Chinese orthographic processing skills faster and earlier than Canadian-children in Canada. Yet both groups demonstrated early differentiation between Chinese and alphabetic orthographies perhaps even before kindergarten, and developed rudimentary radical position awareness by the end of kindergarten. Radical form awareness, however, did not appear by the end of Grade 1 for the Canadian group, while grew dramatically during Grade 1 for the Chinese group. When predicting Chinese orthographic processing, character reading was a unique predictor in both groups, yet phonological awareness was predictive only in China, and home print exposure only in Canada. Conclusion Our findings suggested a developmental sequence for different aspects of Chinese orthographic processing, and the important influence of linguistic contexts, home print exposure, character knowledge, and phonological processing on the development of these skills.
Juan Luque (University of Malaga (Spain))Miguel López-Zamora; Carlos Álvarez - Beyond Decoding Deficit: Inhibitory Effect of Positional Syllable Frequency in Dyslexic Spanish Children
Given errors and lower processing time in reading pseudowords are considered major signs of dyslexia, there exists an implicit assumption: deficit is basically located at decoding phase. However, the inhibitory effect of positional syllable frequency might show up a complementary view. This effect consists on words composed of frequent syllables produce longer reaction times and more errors than words with less frequent syllables (Carreiras, Álvarez, & de Vega, 1993). The high-frequency syllables are shared by more words so extra- time is needed to deactivate the lexical neighbors. Therefore, a lexical decision task containing a classical PSF paradigm implies two processes at least, the activation (decoding phase) and the inhibition of lexical candidates. We hypothesized dyslexic readers must show a stronger inhibitory effect than normal readers because they are slower decoders but they could also be slower inhibitors. Dyslexics and control readers (7 & 9 yrs) received a lexical decision task, containing high/low frequency words by high/low syllable frequency words. Results showed an interaction between lexical and syllable frequency factors, and a strong inhibitory effect was found on low-frequency words condition. More interesting, while the third order interaction between lexical frequency, syllable frequency and reading level factors did not reach the significance, these two factors by dyslexics/controls factor was statistically significant. The inhibitory effect showed that dyslexics deficits are no restricted to the grapheme-phoneme conversion phase, but it is a wider phonological deficit as current theories have pointed out. Lack on inhibitory processes could cause memory overload and so comprehension difficulties.
Elizeu Coutinho Macedo (Elizeu Macedo); Maria José Cicero Oger Affonso; Anna Carolina Cassiano Barbosa; Carolina Mattar Julien de Toledo Piza - Writing skills in developmental dyslexia: the brazilian picture naming-writing test (PNWT1.1- Writing)
Purpose: Studies in dyslexia have given priority to reading assessment, even though difficulties with automatic letter writing and naming have been evidenced. Writing problems have been underdiagnosed, partially due to the lack of instruments available in Brazilian Portuguese. The Picture Naming-Writing Test (PNWT1.1-Writing) is a computerized instrument that allows the assessment of writing competence. Therefore, the purpose of this study was to analyze writing patterns of dyslexics in a naming-writing figures task. Method: Participants were fifteen children with dyslexia (DYS), of both sexes, with ages ranging from 8 to 12, from public and private schools in the State of São Paulo. These were matched to two control groups: Age Controls (AC) composed of 15 children with no reading problems, paired by gender, age and type of school; and Reading Controls (RC) composed of 15 younger readers, paired with dyslexics by gender, type of school and reading performance. Results: quantitative analyses indicated that the total score of the DYS in the PNWT1.1-Writing was significantly lower when compared to AC; no significant differences were observed in the total score of DYS when compared to RC; the RC group was slower than both DYS and AC. The analysis of types of error showed significant differences between the DYS and the two control groups in: grapheme-phoneme univocal correspondence, omission of segments and phoneme-grapheme correspondence regardless of rules, which indicate both phonological and orthographical difficulties. Conclusions: the analysis of types of error in the PNWT1.1-Writing provide greater understanding of difficulties faced by individuals with dyslexia and demonstrate their writing patterns differs from those observed in both age and reading controls.
Denise Magdales (University of Florida - Speech and Hearing Center); Rebecca Wiseheart; Linda J. Lombardino; Lori J.P. Altmann - Phonological awareness in simultaneous and sequential bilingual college students
Purpose: Phonological awareness (PA), a strong predictor of successful literacy acquisition, has not been studied well in bilingual speakers. Competence in the target language and cognitive ability may contribute to performance on phonological awareness tasks. We predicted that simultaneous bilinguals would perform like monolingual speakers, but sequential bilinguals would perform more poorly on complex phonological awareness tasks. Method: Three groups of students (age 18-30) participated: 26 monolingual English speakers, 21 bilingual Spanish-English speakers who acquired Spanish before English, and 13 bilingual Tagalog-English speakers who acquired both languages simultaneously. Participants completed tests of working memory, executive function and PA, plus a novel Pig Latin Task. In the Pig Latin Task, participants translated words into 2 "dialects" of Pig Latin, one that split onsets and one that kept onsets whole (e.g., reen-gay vs. een-gray). Results: Groups did not differ on complex working memory and executive function tasks. Sequential bilinguals performed more poorly on verbatim memory, blending, and RAN tasks, but not other PA tasks. On the Pig Latin task, Group interacted with Onset Type. Monolingual speakers performed significantly better with Split-Onsets than Whole-Onsets. Simultaneous bilinguals performed well with both. Sequential bilinguals performed poorer on both than other groups. Conclusions: Sequential bilinguals performed more poorly on several measures of phonological awareness as well as verbatim memory, all tasks that emphasize access to phonological forms. Simultaneous bilinguals had native-like phonological awareness. Differences in performance between bilingual groups may arise from less automatic access to the phonology of English.
Joe Magliano (Northern Illinois University); Amanda M. Durik: Janet K. Holt; - Assessing the Impact of Topic Interests on Comprehension Processes
Purpose: There is a growing body of research indicating that the level of text-topic interest affects performance on comprehension tests (both standardized and unstandardized tests). However, to date, most of this research has focused on outcome measures rather than the processes that support comprehension. The goal of the present research was to assess the extent to which topic interest and other cognitive skills (prior domain knowledge, prior topic knowledge, reading skill, working memory capacity) are related to inference and comprehension processes. Method: In Study 1, a group of participants read a set of science texts silently and sentence-reading times were collected. In Study 2, typed "think-aloud" protocols were collected to provide insight into what individuals were thinking about at a particular point in time while reading. Multilevel modeling was used in both studies to assess the extent that interest and other individual differences factors moderate sensitivity to text features that are correlated with comprehension processes (i.e., bridging and elaborative inferences). Results: The results indicated that readers with higher levels of interest process texts more slowly and demonstrate greater sensitivity to features that support the construction of a coherent mental model for a text. Moreover, there is some evidence that interest helps less-skilled readers become more engaged in the reading. Conclusions: These results provide some explanations for a growing body of literature that shows that text-topic interest affects performances on comprehension tests. Moreover, they suggest that topic interest could be used as a scaffold to promote comprehension.
Joyce Mak (); Esther Geva - English script recognition in ELL kindergarten prereaders
Purpose: What aspects of script are salient and meaningful to young ELLs? This study examined the emergence of English script recognition and its relationship to early English word reading in ELLs. Methods: "Mr. Cricket" assessed English script recognition by asking children to decide for each item whether "Mr. Cricket was reading English." The "scripts" included pictures, digits, non-Latin, and Latin "texts". The sample consisted of 129 ELLs (Mean = 69.71 months) from three different home language backgrounds (Chinese, Portuguese, and Spanish). Children were assessed on the Cricket Task in Kindergarten and on letter and word reading tasks in Kindergarten and Grade 1. Results: A Principal Component Analysis established two composites: Composite 1 was based on non-Latin "texts" (symbols, digits, Chinese). Composite 2 was based on the Latin alphabet (strings of letters, English pseudowords, Portuguese). Composite 1 was significantly correlated with letter naming in Kindergarten, r = .232, p = .01, but not with word reading due to a floor effect on the latter. Composite 2 did not correlate significantly with letter or word reading in Kindergarten. Neither composite was significantly correlated with word reading in Grade 1. Conclusion: Distinguishing crude visual features of the alphabet is essential. Realizing that symbols, digits, or Chinese are "not English" is an essential aspect of script awareness that is related to alphabet recognition. However, in Kindergarten, the ELLs have not been sensitized to finer aspects of the script and do not attend yet to details of the written word, which may require more reading ability.
Jeannette Mancilla-Martinez (University of California, Irvine)Nonie Lesaux - The Role of Sentence-Level Syntactic Knowledge on Fifth Grade Spanish-speaking Language Minority Learners' Reading Comprehension Outcomes
Purpose: This study investigates the role of English syntactic knowledge on the English reading comprehension achievement for a sample of fifth grade Spanish-speaking language minority (LM) learners in the U.S. Specifically, the contribution of sentence-level syntactic processing on reading comprehension outcomes is examined, thus advancing our understanding of processes that influence reading comprehension outcomes for this growing population of at-risk readers. Method: The investigation was conducted as part of a longitudinal study that includes 173 Spanish-speaking LM learners, followed from early childhood through early adolescence. For this investigation, data from the 5th grade year (wave 6) is featured. Given the multi-faceted nature of reading comprehension, the study employs standardized measures tapping component skills known to influence comprehension, including word-level reading and oral language, as well as syntactic knowledge. Results: Findings demonstrate that students' syntactic knowledge is more closely associated with their word-level reading skills than their oral language skills. In turn, this knowledge accounts for unique variance in students' reading comprehension outcomes, above and beyond the contributions of both word-level and oral language skills. Next, we will conduct an item analysis on the syntactic measure to inform instructional efforts aimed at improving students' syntactic knowledge and, ultimately, reading comprehension outcomes. Conclusions: This study contributes to a small research base focused on syntactic understanding to inform theory in this domain. In turn, these data on the processes that influence reading comprehension outcomes help inform our understanding of effective instruction to meet LM learners' needs.
Eva Marinus (Macquarie University/ ARC Centre of Excellence in Cognition and its Disorders (CCD)); Anne Castles - Noisy neighbors: interference in the orthographic lexicon
Purpose: In this study we aimed to further clarify the link between the development and tuning of orthographic representations and word reading performance. To this end we assessed interference of high-frequent neighbors in word-reading speed and accuracy. Research Questions: Are poor readers more susceptible to the interference of high-frequent neighbor words than typical readers? Are certain neighbors (e.g., rime vs. body) noisier than others? Method: 35 Grade-2 and 31 Grade-3 children (aged 7.14-9.55) from two primary schools in Sydney were tested on standardized word reading tasks (TOWRE and Castles & Coltheart 2), nonverbal reasoning (KBIT) and passive vocabulary (ACE). Interference of orthographic neighbors was measured with an experimental naming task in which children were asked to read aloud four-letter words with and without high-frequent neighbors. Speed, accuracy and actual responses were recorded. Results: All children responded less accurate and slower to words with multiple high-frequent neighbors. In addition, the speed cost for words with high-frequent neighbors was stronger for poor readers. Interestingly, the interfering effect only occurred when the high-frequent neighbors were from different types (rime, body and middle neighbors). Conclusion: High-frequent neighbors cause interference in the orthographic lexicon and this effect is stronger in poor readers. The latter can partly explain their problems in developing word reading speed and accuracy.
Sandra Martin-Chang (Concordia University);Kyle Levesque; Robin Grumet - Is 'list' the new four letter word? Cognitive processes in contextual and isolated-word reading
Purpose: Reading words in context elicits more conceptual processing than reading words in isolation. The opposite pattern also holds; isolated-word reading is more perceptually-driven than contextual reading (Martin-Chang et al., 2010). In the past, these data may have been used to promote the "specialness" of context. However, a transfer appropriate processing view argues that while conceptual processing is a natural consequence of reading connected text, it can also be induced in isolation. The current investigation tested this premise. Method: Sixty participants read 72 target words counterbalanced over three conditions (length, pleasantness, story). In two isolated-word reading conditions, participants circled a response on a 7-point scale rating either word 'length' or 'pleasantness'. In the third condition, words were read within a 'story'. Cognitive processes were assessed via a surprise explicit memory task (conceptual) and an implicit word stem-completion task (perceptual). Results and Discussion: Although the words in both the 'length' and 'pleasantness' conditions were read in isolation, the targets that were rated for pleasantness (M=9.23) were recalled at a much higher rate than those judged for length (M=2.97). Memory for words that were read within a meaningful story (M=4.43) fell between 'list' and 'pleasantness'. Explicit memory is commonly used as a proxy for conceptual processing, indicating that the words judged for pleasantness were processed more conceptually than those read in context. The opposite pattern of results was found in the implicit word stem task (length=10.53, story=9.20, pleasantness=9.27). The results are discussed in terms cognitive processing during conceptual and isolated-word reading.
Ana Marty (Florida State University, Florida Center for Reading Research)Ana H. Marty; Silvia Palenzuela; Christopher J. Lonigan (Florida Center for Reading Research, Florida State University); JoAnn M. Farver (University of Southern California); & Kimberly D. McDowell (Wichita State University) - Preschool children who are English Language Learners: Family, language, and regional factors associated with early literacy skills
Purpose: Children who are English-language learners (ELL) are at considerable risk for educational difficulties, particularly in reading. In the United States, Spanish-speaking children constitute the largest ELL subgroup and are the fastest growing school-age population. The Spanish spoken in the homes of these children is not a universal entity, however. Depending on the region of origin (i.e., Central and South America, Cuba, Puerto Rico), there are dialect variations in vocabulary, pronunciation of words, and differences in the vocabulary acquired during the preschool period. Moreover, children from these different language backgrounds are likely subject to different SES, socio-cultural, and school factors. The purpose of this study was to examine how heritage language and regional factors related to children's early literacy skills. Method: Approximately 500 3- to 5-year-old children completed measures of language, phonological awareness, and print knowledge in Spanish as a part of a larger multi-state study concerning the development of an early literacy assessment for Spanish-speaking ELL preschoolers. Children's parents completed a background information form that included language origin, family demographics, and children's exposure to Spanish in the home. Results: Analyses revealed that families with different heritage languages differed on a number of factors (e.g., language spoken in home, SES). Children's scores on measures of Print Knowledge, Elision, Blending, and Definitional Vocabulary differed based on the children's heritage language, SES, and degree of exposure to Spanish in the home. Conclusions: The results of this study indicate that there is substantial variability among preschoolers who are Spanish-speaking ELLs.
Julie Masterson (Missouri State University)Arthur Maerlender; Rene Friemoth Lee; Devery Larsen - Characterizing Spelling and Reading Fluency Development in Persistently Poor Readers
Purpose: Response to Intervention models require the use of measurement tools that have optimal sensitivity to progress both within and across grades. We used traditional and experimental metrics to compare spellings between children who were persistently poor readings and their typically developing peers across Grades 1 through 5. The students' performance on an experimental measure of reading fluency also was documented, and comparisons to their spelling profiles were made. Method: Participants included 38 students classified as persistently poor readers and 31 students who were typically developing. Students in Grades 1 and 2 spelled 50 words consisting of common consonant, consonant cluster, and vowel spelling patterns. Students in Grade 3 through 5 spelled an additional 34 with greater linguistic complexity. Accuracy was measured via the percent words correct (PWC) and the Spelling Sensitivity Score (SSS; Masterson & Apel, 2010). The SSS credits misspellings due to omissions, illegal misspellings, legal misspellings, and correct spellings differently. We also administered the DRFT (Maerlender, 2007), a one-minute task requiring students to silently read and respond to sentences gradually increasing in linguistic complexity. Results/Conclusions: Both the PWC yielded statistically significant differences between the upper and lower grades; however, only the SSS differentiated performance between the middle grades. The SSS also indicated some qualitative differences between the groups, including the protracted use of illegal misspellings by the PPR group. In the early grades, groups were similar in both the types of spellings used and reading fluency performance; however, as grade level increased the disparity between the PPR and typically developing students increased.
Catherine McBride-Chang (The Chinese University of Hong Kong) - The role of Chinese lexical compounding morphological awareness for early reading difficulties in Chinese (but not English) among Hong Kong Chinese children
Purpose: In this talk, I will highlight the role of morphological awareness in two longitudinal studies of Hong Kong Chinese children with reading difficulties. Method: Study 1 included children who were either at-risk for dyslexia by virtue of having a sibling with dyslexia or because of an early language delay at age 5 and subsequently tested for dyslexia at age 7. Study 2 looked retrospectively (i.e., across their previous performances at ages 5-9) at metalinguistic skills of children found to be consistently in the bottom 25% of a representative sample of Hong Kong Chinese children in word reading in Chinese, English, both, or neither across ages 8 and 9. In both studies, tasks of vocabulary knowledge, phonological awareness (including lexical tone sensitivity), rapid automatized naming (RAN), and lexical compounding morphological awareness (MA) were administered. Results: Across both studies, the most consistent correlates of subsequent word reading difficulties were early MA and RAN performance. Those who were poor in reading English only did not manifest MA difficulties; in fact, their MA performances were significantly better than were those who were poor in Chinese word reading only. Interestingly, those who were poor in reading both Chinese and English were consistently poorest only in RAN. Conclusion: Results underscore the importance of testing early morphological awareness in the form of lexical compounding for those who might be at-risk for reading difficulties in Chinese.
Samantha McCormick (Royal Holloway University of London); Kathy Rastle; Colin Davis; Linda Bayliss - A word learning approach to investigating orthographic effects on speech processing
Purpose : A training study using novel words investigated the impact of orthographic learning on spoken language processing. Previous research showing effects of orthographic consistency has suffered from a range of methodological problems limiting its persuasiveness. The use of novel materials allowed us to teach people identical phonological forms but separate written forms, thus overcoming these difficulties. Method: On Day 1, adult participants were taught twenty novel words words paired with a picture and were only introduced to their spelling on Day 2. The taught spelling for each word was either regular or irregular with respect to grapheme to phoneme consistency (CHESK vs CESK for /tʃɛsk/). Orthographic consistency was counterbalanced across participant groups. Production (on all days) and perception tasks (on Day 3) assessed the effects of orthographic learning on knowledge of the learned words. Results: There was no difference in picture naming times across groups of words before the introduction of written names but after orthographic learning, while naming times for all pictures decreased, naming times for pictures with regular and irregularly spelled names decreased at different rates. This advantage for pictures with regularly spelled names was maintained on Day 3 both in production and perception tasks. Conclusions: These results provide compelling evidence for the impact of orthographic learning on language, critically, on equivalent spoken forms. These results are discussed in terms of two theories suggesting that the recruitment of phonological information automatically activates orthographic knowledge or that orthographic learning itself shapes phonological representations.
Carol McDonald Connor (Florida State University) - Non-mainstream dialect use in second grade
Purpose: Previous research indicates that most minority students decrease use of non-mainstream American English (NMAE) and increase use of School English (SE) in kindergarten and first grade, with implications for their literacy skill gains. But what happens to students who have not increased their use of SE by second grade? Method: In this study, we examined second graders' (n=680 students, 40 classrooms, 8 schools) use of NMAE on the DELV-S Part 1. Schools ranged in percentage of children qualifying for the US free and reduced lunch program (FARL) from 4% to 96% with a mean of 44%. Fall NMAE was negatively associated with fall reading and language skills. Only 20% of students (n=102) used substantial amounts of NMAE, defined as using NMAE on at least 40% of items on the DELV-S. Results: Overall, HLM revealed that these students (n=102) demonstrated increasing using of SE from fall to spring but this varied significantly. The greater students' increase in SE, the greater were their gains in reading comprehension, controlling for fall NMAE, language and reading skills. Students were more likely to increase SE use when they attended more affluent schools. Fall reading and language skills did not predict change in NMAE use. Conclusions: Results of this study support the dialect shifting hypothesis; that the mismatch between NMAE and SE is not the reason for minority students' reading difficulties, per se. Rather; it is students' failure to shift from using NMAE to SE in academic contexts that is associated with weaker reading gains.
Sarah McGeown (Psychology Department, University of Hull); Rhona Johnston; Emma Medford - Reading instruction affects the cognitive skills supporting early reading development
Purpose: This study examined whether different cognitive skills were associated with early reading success when children were taught by different methods of reading instruction. Method: Seventy-nine children were taught either by an eclectic approach (n = 40) which included sight-word learning, guessing from context and analytic phonics, or by a synthetic phonics approach (n = 39), where children were taught solely to sequentially sound and blend letters to read unfamiliar words. Synthetic phonics was introduced to a later cohort at the school; therefore teacher, class, school and SES variables were constant for both methods of instruction. Children were tested prior to any reading instruction and post-tested six months later. Results: For the eclectic group, letter knowledge, vocabulary and rhyming skills predicted later reading ability, whereas for the synthetic phonics group, letter knowledge, phonemic awareness and memory span predicted later reading skill. At post-test, synthetic phonics taught children had better letter-sound knowledge, word reading and phoneme awareness. Conclusions: Reading instruction appears to affect the cognitive skills that support early word reading. Teaching children at the phoneme level to sequentially blend letters led to a greater dependence on short term memory as children needed to retain the sequence of sounds in memory to blend them together. Children taught via a predominantly whole word and context approach relied more heavily on large phonological units (i.e., rhyme) and vocabulary. Implications for our understanding of the skills supporting children's early reading development and the effect of reading instruction on children's emergent reading strategies are discussed.
Anita S McGinty (Center for the Advanced Study of Teaching and Learning, University of Virginia);Laura M. Justice; Tricia A. Zucker; Carolyn Gosse; Lori E. Skibbe - Shared-reading dynamics: Mothers' question-use and the verbal participation of children with SLI
Purpose: This study used sequential analysis to explore bi-directional and dynamic dependencies between mothers' question-use and children's verbal participation during shared reading, to thereby consider patterns of verbal attunement within the dyads. Method: The sample comprised mothers and their preschool-aged children with Specific Language Impairment (SLI; n = 14). Each mother and child extra-textual utterance was transcribed and coded. Mother utterances were coded as Questions or Other; in turn, questions were coded for Cognitive Challenge and Directiveness. Child utterances were coded for syntactic complexity (Minimal, Low, High). Results: Descriptive data show variability in mothers' question-use and some variability in the complexity of children's talk during shared reading; however, sequential data show a lack of attunement of these two behaviors. Mother's question-use did not facilitate higher levels of verbal participation by children. Further, the complexity of children's verbal participation did not influence the cognitive challenge and directiveness of mothers' question-use. Conclusions: The findings were contrary to hypotheses and collectively suggest unique and dynamic challenges within shared reading experiences involving mothers and their young children with SLI. Post hoc analyses explore the extent that patterns of dependency varied across individual dyads and by mother- and child- characteristics. Future directions for research are discussed.
Margaret McKeown (University of Pittsburgh)Isabel Beck - Comparing two approaches to vocabulary instruction for kindergartners
Purpose The study examined kindergarteners' word learning and comprehension effects of two approaches to vocabulary instruction. One approach comprised repeated readings of a storybook and explanations of word meanings; the other presented multiple contexts for the words and interactions aimed at deep processing. Method Materials were three storybooks, 10 Tier 2 words for each story, a friendly explanation and activities for each word. In Repetition instruction, each story was read three times and words were explained during each reading. Interaction condition provided one story reading, followed by explanation of word meanings and opportunities for interactions in novel contexts. Participants were students and teachers in 8 kindergarten classrooms from a working-class community. The design was within subject; each student experienced each kind of instruction. Targeted outcomes ranged from recognition of word meanings to comprehension (Context Integration and Listening Comprehension measures) and expression (Production Task), each assessed with experimenter-designed measures. Results Data analyses comprised linear mixed-effect model ANOVA's, with subject as random effect; story as a repeated measure. Repetition and Interactive approaches showed no difference for recognition of word meanings; interactive instruction better abled students to integrate words into contexts and to produce words associated with a picture. However, the listening comprehension measure showed no differences between conditions. Conclusions Implications relate to vocabulary instruction and assessment. Regarding instruction, results extend the consensus that effective instruction for higher order goals seems to require attention to deep processing. Regarding assessment, the study provided two novel assessments for kindergartners that tap higher-level language processes.
Kristen McMaster (University of Minnesota)David Parker; Xiaoqing Du - Monitoring first-graders' writing progress using sentence- and story-writing tasks
Purpose. The purpose of this study was to examine technical features of slopes produced from Curriculum-Based Measures (CBM) of beginning writing. Research questions included: Do sentence- and story-writing measures yield reliable and stable slopes? How many data points are needed to obtain reliable/stable slopes? Do sentence- and story-writing measures reflect growth in a short time period? Method. Eighty-five first graders completed weekly CBM prompts for 12 weeks. Then, the Test of Written Language-3 (Hammill & Larsen, 1996) was administered. Correlations among measures were calculated to determine alternate-form reliability and criterion validity. To determine reliability of slopes, we calculated correlations between slopes produced in odd and even weeks, as well as between slopes produced across incremental durations compared to the overall 12-week slopes. To determine stability, we examined standard errors of the slopes. To determine the duration needed to yield slopes that reflected growth, we calculated t-values, effect sizes, and the statistical power needed to detect slopes greater than zero. Results Alternate-form reliability and criterion validity of CBM scores were sufficient (rs > .70 and .50, respectively). Examinations of group- and individual-level slopes indicated 8 or 9 data points were needed to yield reliable/stable slopes. The Picture-Word task appeared more sensitive to growth, yielding slopes significantly greater than zero within 4 weeks. Conclusions. This study demonstrates that Picture-Word and Story prompts may be used to monitor progress of beginning writers. A critical next step is to determine the instructional utility of the measures.
Chris McNorgan (Developmental Cognitive Neuroscience Laboratory, Northwestern University);Chris McNorgan; Aubrey Alvarez; Annum Bhullar; Jessica Gayda; James R. Booth - fMRI activation in language network predicts future reading ability
Purpose We investigated whether brain activity was predictive of future reading ability, and if so, how this brain-behavior correlation informs models of reading ability. Method A longitudinal study followed 26 normally developing children ranging in age from 9 - 15 years who were initially assessed for reading skill and performed a rhyming judgment task using fMRI. Patterns of brain activation in this task predicted changes between initial and a follow-up assessment of reading skill administered up to 6 years later. Results Brain activity in areas typically active during imaging studies of reading was found to predict future reading ability, however the predictive ability of these areas depended on age. Increased activity in neural circuits associated with phonological recoding was predictive of greater gains in reading fluency in younger children, whereas increased activity in orthographic processing circuits was predictive of smaller gains in reading fluency for older children. Conclusions Interpreted within the context of a connectionist model of reading, these results suggest that developmental constraints on the cross-modal mapping system lead to age-dependent roles of the orthographic and phonologic processing system, leading to optimal and sub-optimal reading strategies at different ages: younger children who are more sensitive to higher-order phonological word characteristics make greater reading proficiency gains whereas older children who continue to focus on whole-word orthographic representations make smaller reading proficiency gains.
Emma Medford (University of Hull); Sarah McGeown - The effects of behaviour on the development of emergent literacy skills.
Purpose: Emergent literacy skills, such as phonological awareness, are associated with reading development. However, classroom behaviours have also been found to be associated with early reading success. The current study examined the influence of negative externalising and internalising behaviours on the development of children's emergent literacy skills during the first year of reading instruction. Method: Thirty two children (18 girls, mean age 4 years, 7 months) completed assessments of letter-sound knowledge, vocabulary, rhyme and phoneme awareness prior to reading instruction and half way through their first year. Class teachers completed the Strengths and Difficulties Questionnaire (a measure of children's behaviours and psychosocial problems) for each child. Correlations and regression analyses were used to examine the influence of negative externalising behaviours (hyperactivity and conduct problems) and negative internalising behaviours (emotional problems and peer relationship problems) on children's emergent literacy skills. Results: Externalising behaviours were negatively associated with post-test letter sound knowledge and phoneme awareness whereas internalising behaviours were not associated with emergent skills. Furthermore, externalising behaviours explained significant variance in post-test letter sound knowledge and post-test phoneme awareness after accounting for pre-test abilities, whereas internalising behaviours explained no significant additional variance. Conclusions: Children's negative externalising behaviours appear to influence the development of some emergent literacy skills. It may be that engagement in these behaviours is associated with less focus during literacy instruction. Interventions focused on improving early externalising behaviours may indirectly aid the acquisition of early literacy abilities. Future longitudinal research with a larger sample will investigate these conclusions further.
Chelsea Meenan (University of Denver); Janice M. Keenan - Differences in the Cognitive Profiles of Children Diagnosed by Common Reading Comprehension Measures
Purpose: Several studies demonstrate that tests of reading comprehension measure component skills differently (Cutting & Scarborough, 2006; Keenan, Betjemann, & Olson, 2008). The purpose of the present study is to explore the implications of these inconsistencies by examining how children identified as being in the high and low tails of the distribution of each test vary in their cognitive profiles. Method: 995 participants (8 - 18 years) in a behavioral genetic study of reading took the GORT, Qualitative Reading Inventory, Woodcock-Johnson Passage Comprehension, and the PIAT. Word decoding skill, IQ, ADHD symptomology, and working memory were compared for children below the 10th percentile and above the 90th percentile of each test. Results: In the low tail, differences between tests were found in word decoding and working memory. Low scorers on the PIAT and WJ had significantly poorer decoding and working memory. Among children above the 90th percentile, word decoding was lower on the QRI than the other measures, suggesting that decoding skill is less essential for good performance on this measure. IQ was also lower on the QRI than the WJ and PIAT, perhaps because of its high correlation with decoding skill. There were no differences across tests in ADHD ratings. Conclusions: Differences across tests are manifest in the cognitive profiles of children diagnosed as particularly good or poor in reading comprehension. This has implications for how tests are used in research and practice.
Monica Melby-Lervag (University of Oslo, Dep. of special needs education)Solveig-Alma Halaas Lyster Charles Hulme - Phonological skills and their causal role in learning to read: A meta-analytic review
Purpose: We report a systematic meta-analytic review of studies of the relationships among children's phonological skills and decoding. We focus on three key phonological skills that have been widely considered as possible causal foundations for learning to read: phoneme awareness, rhyme awareness and verbal short-term memory. Method: Our review included both extreme group studies comparing children with dyslexia to typically developing controls, and correlational studies. Our analysis identified 239 relevant papers and a total of 688 effect sizes were calculated. Results: Extreme group comparisons indicated that children with dyslexia show a large deficit on phoneme awareness in relation to typically developing children of the same age as well as in relation to typically developing children matched on reading level. There were reliably smaller group deficits on both rhyme awareness and verbal short-term memory for both studies comparing with chronological age matched controls and reading level controls. Analyses based on correlational studies showed that phoneme awareness was also the strongest correlate of individual differences in reading ability in such studies, and that this effect remained reliable after controlling for variations in either verbal short-term memory or rhyme awareness or both. Discussion: These findings provide strong support for the pivotal role of phoneme awareness as a predictor of individual differences in early reading development. We discuss whether such a relationship is truly a causal one, and the implications of research in this area for current approaches to intervention for children with reading difficulties.
Silvana Mengoni (Centre for Reading and Language, University of York); Hannah Nash; Charles Hulme - The interaction of phonological and orthographic learning in children with Down syndrome
Purpose: For children with Down Syndrome (DS), language, particularly phonology, is a relative weakness. Reading, however, is a strength and it has been proposed that reading practice benefits spoken language (Buckley, 1995) but there is little empirical evidence to support this. It has also been found that phonological skills play a lesser role in reading development than in typical development (Hulme et al., submitted). Two experiments were designed to further examine the relationship between phonology and reading in Down syndrome. Method: In both experiments children with DS aged between 7-17 years and a reading-matched typically developing (TD) group aged 5-7 years participated. Experiment 1: Children were taught the spoken names of ten novel objects, half of which were presented with the spellings of the names. Experiment 2: Children were taught the written forms of 12 nonwords, half of which received phonological pre-training. Results: Experiment 1: There was a significant advantage for learning the spoken names of pictures that were paired with their spellings during training. Performance in the two groups was equivalent and both groups benefitted equally from orthographic support. Experiment 2: Learning the written forms of the nonwords was facilitated in both groups by phonological pre-training. Conclusions: Children with DS can learn new spoken words and benefit from the presence of the written form of the word to the same degree as TD children. Furthermore, when learning the written form of a word, prior experience with the spoken form is beneficial for both TD children and children with DS.
Jeremy Miciak (University of Texas at Austin); Scott Baker; Doris Baker; Keith Smolkowski; Sylvia Linan-Thompson; Alejandra Mielke - Relation between instructional interactions and reading achievement in first grade bilingual classrooms
Purpose: This study examines teacher and student instructional interactions and the reading achievement of first grade Spanish-speaking students in the United States. Specifically, we will present results of our analysis of the relation between the mean rate of learning opportunities observed during Spanish reading instruction and student performance on oral reading fluency and comprehension in Spanish and English. Method Design: Randomized controlled trial with assignment at the school level Participants: 38 schools in Oregon and Texas; 1,299 students; 120 teachers Measures: Observation Measure: Student-Teacher Interaction and Context Observation - Reading (STICO-R; adapted from Smolkowski & Gunn, 2010). This measure documents the frequency of teacher behaviors (i.e., modeling and academic feedback) and student behaviors (group responses, individual responses, covert responses, and mistakes). Reading Achievement Measures. Aprenda: Lectura de Palabras, Lectura de Oraciones y Comprensión de Lectura (Harcourt, 2005); Stanford Achievement Test: Word Reading, Sentence Reading, and Reading Comprehension subtests (Harcourt, 2005); IDEL Fluidez en las Palabras sin Sentido and Fluidez en la Lectura Oral (Baker, Good, Knutson & Watson, 2007); DIBELS Nonsense Word Fluency and Oral Reading Fluency (Good & Kaminski, 2005). Data Analysis: Hierarchical linear modeling with two levels: students nested within schools. Results: In process. We hypothesize that increased mean rate of learning opportunities observed during reading instruction will be significantly associated with improved student reading achievement. Conclusions: Instructional quality, independent of the language of instruction, plays a significant role in the reading performance of bilingual children. Students attending classrooms where teachers provide clear models, and consistent opportunities to respond in group formats and individually, should have better reading outcomes than students in classrooms where there are fewer instructional interactions between teachers and students.
Maggie Middleton (The Ohio State University)Margaret E. Middleton; Stephen A. Petrill - The relationship between narrative comprehension and reading comprehension
Purpose This study examined the genetic and environmental contributions to the relationship between narrative comprehension and reading comprehension. Methods We examined this issue in a sample of 275 identical and same-sex fraternal twin pairs (MZ = 112, DZ = 163) recruited from the Western Reserve Reading Project. The McDonald's Story, The Shipwreck, and The Dragon Story from the Test of Narrative Language, Woodcock Johnson Passage Comprehension, and the Peabody Reading Comprehension subtest were assessed from tester-administered measures. Results Two latent factors of narrative and reading comprehension were formed using the language subtests and the comprehension subtests respectively. The variance of the latent factors was divided into estimates of additive genetic, shared environment, and nonshared environmental influences using Cholesky decomposition. Results suggest that heritability is statistically significant for Narrative Comprehension and Reading Comprehension in addition to significant unique nonshared effects (including error) for Reading Comprehension at waves 1, 2, and 3. The standardized path estimates of the latent factors suggest statistically significant genetic, shared, and nonshared environmental overlap at all waves respectively. Moreover, there is unique genetic influence on Reading Comprehension above and beyond Narrative Comprehension at wave 1 and 2. Conclusions These results suggest that the relationship between language comprehension and reading comprehension is influenced primarily by genetic overlap. Additionally, evidence for unique genetic influences on reading comprehension, above and beyond language, suggests that reading comprehension is not entirely accounted for by language skills.
Brett Miller (Eunice Kennedy Shriver National Institute of Child Health and Human Development) - Funding Opportunities for Literacy Research Available from the Eunice Kennedy Shriver National Institute of Child Health and Human Development
This poster will present current funding opportunities available from the Eunice Kennedy Shriver National Institutes of Child Health and Human Development (NICHD) for graduate students, post-doctoral fellows, and independent researchers interested in additional training, mentorship or research funding. NICHD has a long history of providing support for research and training initiatives to increase understanding of both normal and atypical development of reading and written language skills throughout the life course and the training and research opportunities highlighted will focus on scientific topics related to literacy and related learning disabilities. A program staff member will be available to answer general questions about NICHD funding or questions about funding for specific research projects or training opportunities for individuals from U.S. or non-U.S. institutions.
Marie Moore (University of Alabama); Susan Loveall; Frances Conners; Allyson Phillips - Individual differences in RAN and orthographic knowledge: Word-specific vs. general
Purpose The self-teaching hypothesis suggests that orthographic learning results from phonological recoding experience. However, rapid automatic naming (RAN) may also bear on orthographic learning, as various proposed explanations of RAN's relation to reading link it with orthographic processing. Yet, reported relations between RAN and orthographic measures are mixed. We explored (a) whether RAN contributes to orthographic knowledge beyond the contribution of phonological recoding, and (b) whether this depends on the exact measure of orthographic knowledge (word-specific vs. general). Method Forty-one 2nd and 3rd grade children (mean age = 8.68, mean IQ = 103.65) completed a battery of tests measuring reading-related skills including (1) phonological recoding (Word Attack subtest, WRMT-R); (2) word-specific orthographic knowledge (Orthographic Choice and Homophone Choice); (3) general orthographic knowledge (Orthographic Awareness); (4) RAN (Rapid Digit Naming and Rapid Letter Naming subtests, CTOPP); and (5) IQ (KBIT-2). Results and Conclusions Using two hierarchical regression models, we tested the effects of RAN on word-specific orthographic knowledge and general orthographic knowledge after accounting for phonological recoding and IQ. In the first model, RAN contributed significantly to word-specific orthographic knowledge, R2∆ = .16, F(3, 38) = 11.45, p < .001. However, in the second model, RAN did not contribute significantly to general orthographic knowledge, R2∆ = .03, F(3, 38) = 2.21, p = .10. Thus, the role of RAN in orthographic learning may depend on the type of orthographic knowledge learned. Further research could examine this relationship in longitudinal designs that allow for the measurement of word-specific and general orthographic learning.
Ann Morrison (Metro State College of Denver) - Effects of intervention delivered by Title 1 paraeducators on the emergent literacy skills of urban, at-risk kindergartners
Purpose: Many Title 1 schools utilize paraeducators as interventionists despite the fact that conditions for their effectiveness have not been well established. This paper will report the effects of a kindergarten Tier 3 program taught by paraeducator interventionists on at-risk kindergartenters' emergent literacy skills. Method: A single subject design was used to evaluate growth phonological awareness, letter naming, and letter sounds. Pre- and posttesting were used to evaluate growth in vocabulary, listening comprehension, phonological awareness, and print awareness. Ten low-achieving students and one paraeducator participated in the study. Weekly measures of phoneme segmentation, letter naming and letter sounds were measured over 16 weeks. Intervention was provided by a paraeducator who received two hours of training before the project began and weekly coaching from the Principal Investigator. Emergent literacy intervention was provided 25 minutes daily for ten weeks to groups of five students at a time. Content focused on phonological awareness, letter naming, letter sounds, print awareness, oral vocabulary, and comprehension. Results: Single subject analysis demonstrated significant student gains in letter naming, letter sounds, and phoneme segmentation. Pre- and post-test data indicated significant gains in initial sound fluency, print awareness, and vocabulary. Gains in comprehension were inconsistent. Conclusion: Results support a growing body of evidence regarding the conditions under which paraeducators can affect statistically significant gains in the emergent literacy skills of at-risk kindergarteners. This study provides knowledge to guide the continued development of the intervention. Continued examination of paraeducator involvement in the development of at-risk kindergartner's comprehension skills is warranted.
Frederick J. Morrison (University of Michigan) - Discussant contribution
Dr. Frederick Morrison will discuss the scientific and educational implications of the results of the four paper contributions.
Jack Mostow (Project LISTEN, Carnegie Mellon University); Sunayana Sitaram - Mining data from Project LISTEN's Reading Tutor to analyze development of children's oral reading prosody
Purpose: How does children's oral reading prosody develop over time? Miller & Schwanenflugel (2008) investigated this question by meticulously analyzing children's readings of a short text passage in grades 1 and 2 and comparing their prosody to fluent adult readings of the same passage, based on the observation that similarity to adult prosody is a good measure of children's oral reading expressiveness. Method: We implemented an automated, scaled-up version of this approach by training models of oral reading pitch, intensity, and duration on a corpus of narrations of 12408 sentences by 11 fluent adult narrators. We used these models to evaluate the oral reading prosody of 77693 sentences read by 205 children (mostly) 7-10 years old who used Project LISTEN's Reading Tutor during the 2005-2006 school year. Results: The models assess a child's prosodic contour for a given sentence by estimating the probability that an adult would generate it, based on syntactic and other features of the words. Aggregating such evaluations over multiple children and sentences makes it possible to pinpoint specific common prosodic deficits in children's oral reading. Conclusions: Mining this corpus provides an unprecedented opportunity to explore, at a large scale and fine grain size, the detailed prosodic characteristics of fluent adult reading that children lack most, learn fastest, and in what order. References Miller, J., & Schwanenflugel, P. J. (2008). A Longitudinal Study of the Development of Reading Prosody as a Dimension of Oral Reading Fluency in Early Elementary School Children. Reading Research Quarterly, 43(4), 336-354.
Kristen Munger (Syracuse University); Maria Murray; Benita Blachman - First grade teachers' knowledge and beliefs about reading: changes after participating in a phonics intervention study
Numerous studies and national reports indicate that both early and at-risk readers benefit from systematic and explicit phonics instruction. Nevertheless, not all teachers seem to be aware of these benefits. Some studies suggest that teachers may not be knowledgeable about basic phonics concepts, or using phonics may conflict with their underlying belief systems for how reading should be taught. Purpose: The purpose of the present study was to investigate the extent to which first grade teachers' knowledge and beliefs about reading changed after participating in a larger intervention study in which they implemented a phonics-based intervention and participated in monthly professional development sessions on phonics instruction. Method: At the end of the intervention, thirty-three teachers (18 treatment and 15 control) completed a questionnaire (adapted from Mather et al., 2001) that included "knowledge items" (e.g., how many speech sounds are in spoken words), as well as "belief items" (e.g., basic skills should never be taught in isolation). Differences in pretest and posttest scores were compared between treatment and control groups. Results: After participating in the larger intervention study, treatment teachers knew significantly more about phonics instruction compared to control teachers. There was also a significant difference between treatment and control teachers' beliefs about phonics instruction, with treatment teachers endorsing more items consistent with evidence-based teaching practices for early and struggling readers. Conclusions: This research suggests participating in an evidence-based reading intervention study can influence not only teachers' knowledge about phonics but also their beliefs about the teaching of reading.
Maria Murray (SUNY Oswego - Curriculum and Instruction Dept); Kristen Munger; Elfrieda H. Hiebert - An analysis of reading intervention texts: what types of words are presented to at-risk readers?
According to Hiebert and Martin (2003), each word in a passage presents an opportunity to either recognize a known word or decode an unknown one. Texts that at-risk readers encounter in commercially available interventions may or may not provide these opportunities. Little empirical research has been done on such programs. Purpose: The purpose of this study was to analyze texts from the popular first grade intervention program by Fountas and Pinnell called Leveled Literacy Intervention (LLI). Method: LLI words were coded for word repetition, decodability, and occurrence of singletons. Words in Scott Foresman's Reading Street (SF) decodable texts were analyzed for comparison purposes. Results: Thirteen percent of words in LLI texts are repeated 10 times or more, many only recognizable using picture cues (e.g., dolphin, cherries). Just 46% of repeated words are decodable according to the simplest syllable pattern (closed). SF text is less repetitive (10% of words repeated 10 times or more), with 73% of repeated words being closed syllables. Early LLI texts' decodability is considerably more difficult than SF, and they also contain a high percentage of words appearing only once (15%) compared to SF (6%). These singletons are less likely to be recognized by at-risk readers due to lack of repetition. Conclusions: Early LLI and SF texts approach word recognition differently by emphasizing either word repetition or decodability. Empirical evidence indicates that poor decoding ability characterizes at-risk readers. Closer examination of LLI intervention texts is needed to determine if they assist readers in learning to decode.
Telse Nagler (IDeA Center Frankfurt)Telse Nagler; Sven Lindberg; Marcus Hasselhorn - Accelerated reading in German elementary school children: An investigation of the acceleration phenomenon (AC) and its characteristics
Purpose: This study aims at the replication of the acceleration phenomenon (AC) in the German language firstly for normal and poor elementary school readers. Method: A sample of 41 third grade children (22 normal and 19 poor readers) was investigated using the acceleration paradigm: Subjects reading speed and comprehension measurements were determined firstly in a normal (Self-Paced 1) and subsequently in an accelerated reading condition (Fast-Paced). The performance was compared to a third condition in which the children read again in their normal reading speed (Self-Paced 2). The subjects answered questions referring to the sentences for reading comprehension measurements. All subjects were tested with the same word material to reveal acceleration benefits within and across groups. Results: Earlier AC studies showed that normal and poor readers profit from the acceleration on their respective level of performance. Comparable patterns were observable in this study: The reading impaired children showed significantly slower reading speed and significantly more comprehension errors compared to the control group. Both groups benefited from the acceleration in the Fast-Paced condition, resulting in enhanced comprehension scores and reading rate. The poor readers were able to improve their comprehension to the baseline of the controls. Conclusion: This study successfully replicated the AP in the German language, firstly for German elementary school aged children with and without reading difficulties. An important finding is the significant reading improvement of poor readers reaching the baseline performance of normal readers through accelerated reading. These results can serve as a foundation for possible training intervention.
Bill Nagy (Seattle Pacific University) - A discussion of symposium findings
Bill Nagy has agreed to play the role of discussant for this symposium.
Ana Luiza Navas (Faculdade de Ciencias Medicas da Santa Casa)Joana Pinto; Paula Delissa - Influence of different types of texts on reading fluency measures in a transparent orthography
Purpose: In transparent orthographies, short and frequent words are quickly read when presented in isolation. Therefore, it is important to estimate the influence of different types of texts in reading fluency assessments. The purpose of this study was to analyze pause durations, reading rate and accuracy in texts with different psycholinguistic characteristics. Method: We evaluated 90 subjects, 55 female and 35 male, with ages between 16 and 74 years old, with no sensory or cognitive deficits, and who considered themselves fluent readers. Participants were requested to perform an oral reading task, with four texts, with different psycholinguistic characteristics, and similar length. Results: Text 1 contained mostly short words, text 2 contained mostly long words, Text 3 contained simple syntactic structures and Text 4 complex syntactic structures. Oral readings were recorded by a digital recorder and durations of pauses were analyzed with Praat. We found longer pause durations for Text 2 then for Text 1, while Text 4 presented longer pause durations compared to Text 3. Furthermore, there was a difference on pause duration for word class. There was a higher reading rate for Text 1 then for Text 2, and for Text 3 in relation to Text 4, evidencing the interference of word length and syntactic complexity on reading rate. The mean percentage of reading accuracy was similar for all texts, demonstrating good accuracy scores among these adults.
Jessica Nelson (Carnegie Mellon University) - Individual differences in reading skills and experience are reflected in eye movements during reading
Purpose: The goal of this study was to characterize how individual differences in reading skills and experience are reflected in patterns of reading behavior, as measured by the eye movement record. Method: A factor analysis of a large database of questionnaire and reading test scores enabled identification of five major dimensions of individual variability: expertise (speed and experience), sublexical skills, accuracy focus (reading at a pace that enables high comprehension), learning/memory, and amount of casual reading done. Thirty-five adult participants with scores along each of these dimensions read paragraphs while their eye movements were monitored. Results: A mixed-effects linear regression analysis revealed that experienced readers demonstrated more efficient reading behavior than less-experienced readers, especially for low frequency words and words with high frequency neighbors. This pattern reduced the word frequency effect and increased the neighborhood frequency effect for these readers. Readers with good sublexical skills also showed more efficient reading behaviors than those with poorer sublexical skills including, specifically, shorter first fixation durations for high frequency words. Main effects of accuracy focus and interactions between expertise and accuracy focus were also found. Conclusions: The pattern of findings suggests that individual differences in skills, knowledge, and strategy are evident in the eye movement record. There are not only predictable main effects reflecting reading speed differences, but also interactions with lexical properties: Reading expertise especially strengthens knowledge of low frequency words, whereas sublexical skill benefits the rapid identification of high frequency words. The importance of reading practice and sublexical skills persists into adult reading.
Tom Nicholson (Centre of Excellence for Research on Children's Literacy, Massey University)Louise Turner - Slow but steady - the positive effects of a minimal intervention approach for reading difficulties
This paper presents a one-year longitudinal analysis of a reading intervention program that compared a code- and meaning-emphasis tutoring approach for struggling readers. Children completed a 2-week summer school just before the start of the school year that consisted of a daily 1-hour one-to-one lesson. They then received a single, weekly, 40-minute one-to-one lesson during the school year. A sample of pupils aged 6 to 10 years completed the study. Children completed assessments at the beginning and end of the summer, at mid-year, and at the end of the year. The tuition did not show clear effects until the end of the year. The code and meaning approaches both had positive effects but the meaning approach was more effective for reading in context.
Anne-Mette Veber Nielsen (University of Copenhagen - Department of Scandinavian Studies and Linguistics);Holger Juul - Amount of independent reading predicts gains in word reading fluency during first grade.
Purpose. Amount of reading outside school has been found to be related to word recognition ability and reading comprehension among students in middle school. But little is known about the effect of independent reading on early reading development. The aim of this study was to investigate the effect of amount of independent reading on word reading fluency during first grade. Method. Seven first grade classes (N=154) participated in the study. A range of preschool measures as well as measures of word reading fluency from five different testing points in first grade were available. Simple linear regression were used to estimate growth rates of word reading fluency. A method to register amount of independent reading was developed. 42 easy readers were distributed to each class. From Nov. until Jan. the teachers registered the lending of easy readers. This method made it possible to calculate amount of reading in number of letters. The difficulties in measuring beginning readers amount of independent reading will be discussed. Results. Hierarchical regression analyses revealed that amount of independent reading was a significant predictor of gains in word reading fluency from mid first grade to the end of first grade after early word reading fluency, letter knowledge, phonological awareness and RAN-digits were controlled for. Conclusions. It is important to find correlates to growth rates in reading. These findings suggest that independent reading contributes uniquely to gains in word reading fluency in first grade and stress the importance of being able to control for amount of independent reading.
Sean Noe (The Ohio State University); Howard Goldstein; Rhonda Tabbah - Characterizing language and literacy coverage through content analysis of preschool curricula
Purpose: Although several approaches have sought to identify evidence-based curricula (e.g., PCER Initiative, 2008-2009; What Works Clearinghouse, 2009), clear guidance for selecting high-quality curricula remains lacking. The early literacy domains of phonological awareness, print awareness, comprehension, vocabulary, and alphabet knowledge have been found to be strongly related to reading achievement. To help provide needed curricular selection guidance, this study examined the extent to which preschool curricula include these 5 early literacy domains. Method: Ten widely used curricula were reviewed. The content of each set of curriculum materials was analyzed to summarize the scope and sequence of each curriculum per the five domains. Results: Results revealed wide variability in the instructional content of preschool curricula. Analyses revealed curricula for educators who seek a balanced curriculum across the five domains. Other curricula emphasize a particular early literacy domain (e.g., OWL curriculum for vocabulary). For children who need additional instruction in phonological awareness, Imagine It! and Storytown would be good candidates for adoption. Others provide limited scope and sequence information. Conclusions: The results from this study provide researchers and educators with a framework for evaluating the extent to which preschool curricula cover the five domains found in the literature to be most strongly related to reading achievement. Concordance in scope and sequence across curricula also may inform content for Tier 2 and 3 curricula.
Mark Noordenbos (Radboud University Nijmegen);Eliane Segers; Willy Serniclaes; Holger Mitterer; Ludo Verhoeven - Neurological evidence of the allophonic mode of speech perception in children at-risk for dyslexia.
Purpose: There is evidence that individuals diagnosed with dyslexia are better in the discrimination between speech sounds of the same phoneme category (allophonic perception: Serniclaes et al., 2004; Bogliotti et al., 2008; Dufor et al., 2009). In a previous study we showed that Kindergarten children at-risk for dyslexia discriminated between sounds of the same phoneme category ), but that this allophonic perception was gone in first grade. This study was the first study to show the allophonic mode of perception in pre-reading children. In the present study the same children were tested in first grade using neurological measures to investigate the discrimination of speech sounds in the brain. Method: Children at-risk for dyslexia (n=33) and chronological-age control children (n=32) participated. Neurological data was collected in an odd-ball paradigm, using the Mismatch Negativity (MMN) to find out if children unconsciously hear differences between sounds. Results: A significant MMN was found for an allophonic mode of speech perception in the neurological data of the at-risk children, but not of the control children. Both groups showed an MMN for the discrimination between stimuli of different phoneme categories. Conclusions: The phonemic perception of at-risk children in first grade in the behavioural tasks might be explained by the onset of formal reading instruction, in which the emphasis is on decoding skills. However, the present data show that allophonic mode of perception still remains at an unconscious level in children at-risk for dyslexia. The implications for reading acquisition by these children will be discussed.
Jane Oakhill (University of Sussex)Alan Garnham; Anna-Marie Armstrong - Factors that influence the interpretation of gender stereotyped terms
Purpose: In previous studies we have shown that conflicts with stereotyped gender expectations (female surgeon, male nurse) result in processing costs. The aim of the present studies was to explore the circumstances under which stereotyped attributions of gender can be reinforced or attenuated during reading. Methods: The first study explored the effects of simply providing further (neutral) information about a protagonist introduced by a stereotyped term (with the intention of giving the reader time to develop a more specific representation of that protagonist). The second study explored the effects of attempting to reinforce or overturn the stereotype attribution by using gender-biased adjectives (e.g. nurse...talkative/fierce.....he). Both studies used reading times to assess the processing costs associated with gender stereotype conflicts. Results: The results suggest that stereotyped gender attributions can be moderated by other aspects of the text. In the first study they were strengthened simply by provision of further (neutral) information about the protagonist (there was a strong interaction between length of the text and whether a pronoun in the final sentence matched or mismatched the stereotype). In the second study, stereotype attributions could be overturned by the presence of a (stereotypically male or female) adjective, which presumably suggested to the reader that their initial stereotyped gender allocation was open to question. Reading times increased on phrases with stereotype-inconsistent adjectives, but in such conditions subsequent reading of a stereotype-inconsistent pronoun was facilitated. Conclusions: Gender stereotype inferences are compelling for readers, but their effects can be moderated by other attributes the text.
Natalie Olinghouse (University of Connecticut)Joshua Wilson - Vocabulary predictors of writing quality and genre elements
Purpose: While vocabulary is considered an essential component of effective writing, the relationship between word choice and quality is not well understood, especially across different written genres. The central aim of the study was to examine the stability of several vocabulary characteristics across genres, and the relationship between vocabulary variables and writing quality and genre elements. Method: One hundred fifth-grade students wrote three compositions: fictional story, persuasive, and informational. Participants had 15 minutes to respond to a picture or sentence prompt. All writing prompts used an outer space theme to control for background knowledge. Vocabulary variables included diversity, content vocabulary, high frequency words, verb concreteness, and noun descriptors. A holistic scale measured writing quality and the inclusion and quality of essential genre features represented the genre element score. Results: Stability of vocabulary variables across genres was minimal, with correlations ranging from -.21 to .37 across genres. Overall, the five vocabulary measures predicted 35% (story) to 60% (informational) of the variance in genre elements. Commonality analysis indicated that diversity and verb concreteness were the best predictors of story genre elements. Diversity and content vocabulary were the best predictors of persuasive genre elements, while content vocabulary and verb concreteness best predicted informational genre elements. Analysis for writing quality will be completed in the winter of 2011. Conclusions: Students use different vocabulary, and different vocabulary measures are more closely related to genre elements depending on the written genre. The results have implications for choosing the best vocabulary characteristics to assess and teach within different genres.
Ake Olofsson (Umeå University)Astrid Ahl; Karin Taube - Dyslexia and study skills in higher education: learning and teaching.
Purpose - to investigate how dyslexia affects university students' ability to benefit from higher education, to describe the strategies that students with these problems use, and to examine the higher education institutions' support for these students. Method - structured and semi-structured interviews, self-reports and testing of 36 university students and interviewes with 30 lecturers from 3 Swedish higher education institutions (teacher training and nursing). Results - The students with dyslexia were found to have problems with a number of skills and academic tasks, e.g. note taking and expressing ideas in writing. Many of them described that their difficulties were long-standing and had been experienced already in elementary school, but the nature of their problems often change over time. The students seemed to make use of resources available to them, including additional time for examinations, access to dyslexia tutors and support with information technology. Conclusions - The results suggest that there are significant knowledge gaps in the educational institutions regarding people who have dyslexia, especially for students not identified formally and explicitly in the compulsory school. Such students may have a very different educational background compared to the traditional group with dyslexia and may be in need for a special type of support. It is suggested that knowledge about students with dyslexia is of great importance for the design of higher education in order to cope with new groups of students
Richard Olson (University of Colorado at Boulder); William Coventry; Brian Byrne; Stefan Samuelsson - Dynamic and static assessment of phonological awareness in preschool: a behavior-genetic study
Purpose: The primary goals of the study were (1) to compare static and dynamic assessments of phonological awareness (PA) in preschool for their prediction of reading ability at the end of kindergarten, and (2) to explore the genetic and environmental etiologies of individual differences in children's performance on these tasks and their correlations. Method: 1988 twin children were tested in preschool with static and dynamic measures of PA and measures of letter knowledge (LK). At the end of kindergarten they were tested with the TOWRE measures of word and phonemic decoding efficiency. Results: After controlling for LK and static PA, dynamic PA accounted for a statistically significant 1% of the variance in the TOWRE composite. Genetic influences were completely shared for static and dynamic PA. These genetic influences were partly shared with LK, and only the influences shared with LK were significantly correlated with genetic influences on the TOWRE. Shared environment influences were largely shared across the measures, with some independence between static and dynamic PA, while non-shared environment influences were specific to each measure. Conclusions: The very minimal but statistically significant 1% unique contribution of dynamic PA at preschool to TOWRE at the end of kindergarten shows that it offers little advantage over the more commonly used static forms, though we acknowledge that dynamic PA might be more predictive in cases of preschool educational disadvantage. Planned analyses will reveal if dynamic PA might provide more unique prediction for word reading, comprehension, and spelling in the later grades.
Eduardo Onochie-Quintanilla (); Ian Simpson; Sylvia Defior - Phonological awareness and naming speed as predictors of reading fluency in Spanish
Purpose: There is great interest in understanding which factors are crucial in literacy acquisition. Most research to date has been carried out using English-speaking samples. However, English cannot be considered a model for other languages due to its opaqueness. It has been argued that RAN might be a more important factor, and that Phonological Awareness (PA) might lose predictive power, in transparent writing systems compared to opaque languages, like English. Few studies exploring this issue have been carried out in Spanish, a highly transparent writing system. Method: 190 native Spanish speakers from the city of Granada were monitored over a two year period. Phonological awareness, naming speed and IQ were assessed in the last year of kindergarten (mean age 5;8), the beginning of first grade (mean age 6;5) and the beginning of second grade (mean age 7;5). Results: Both PA and RAN proved to be independent predictors of reading fluency at the beginning of the first grade. However the predictive power of PA decreased at the beginning of second grade. In contrast, RAN uniquely explained a larger amount of variance at the beginning of second grade than it did at the beginning of first grade. Conclusions: These results agree with results obtained in studies carried out in other transparent writing systems. PA looses predictive power of reading fluency in writing systems with more regular grapheme-to-phoneme correspondence. In contrast RAN becomes a more effective tool for predicting individual differences in reading fluency compared to PA over a 2 year period.
Eric Oslund (Texas A&M University);Deborah C. Simmons; Oi-man Kwok; Shanna Hagan-Burke; Michael D. Coyne; Leslie Simmons - Predicting reading outcomes for at-risk kindergartners: An examination of curriculum-embedded measures
Purpose. We investigated the predictive validity of multi-component and individual component (phonemic and alphabetic) curriculum-embedded mastery checks on kindergarten outcomes. Method. Participants were 137 children who participated in a responsive, tier-two kindergarten reading intervention. The intervention consisted of 100 lessons of explicit and systematic code-based instruction delivered in groups of 3-5 students, 30 minutes per day. Curriculum-embedded measures were collected at the end of each program component (approximately every 4 weeks) and used to predict phonemic and alphabetic latent outcome variables. Results. Multi-level structural equation modeling indicated that multi-skill progress monitoring checks explained between 58% (by late October) and 86% (by early February) of the variance on the phonemic latent variable and between 36% and 57% on the decoding outcome. Predictors differed across time points. The strongest predictors for both phonemic and decoding outcomes at the respective four measurement points were (1) recognizing letter names, (2) producing letter-sound correspondences, (3) identifying the first letter of a word presented orally, and (4) whole-word segmentation. Conclusions. Findings indicate that kindergarten outcomes of children with early reading risk who participated in supplemental intervention can be reliably predicted by multi-skill mastery checks. Additionally, a parsimonious set of measures that changed over time reliably predicted both phonemic and decoding outcomes. Predictors reflect the developmental progression of reading skills with earlier predictors involving naming a single letter or letter-sound correspondence and later predictors involving segmentation of whole words. In addition, the magnitude of variance explained by the predictors as a set increased across time.
Gene Ouellette (Mount Allison University)Elissa McCarron; Sacha Nadeau; Talisa Tims - Oral vocabulary and reading skills: More than just good friends?
Purpose: To elucidate the nature of the role of oral vocabulary in word reading. This contentious issue was investigated by operationally defining distinct facets of the oral vocabulary system, and considering the combined and unique contributions of each to word reading. Method: The present study examined the connections between breadth of the oral lexicon (lexical phonology), depth of knowledge (lexical semantics), organization of the lexical-semantic system and decoding and visual word recognition. Students in grade 6 (n = 61) were assessed with standardized measures of phonemic awareness, vocabulary breadth, depth of knowledge, and lexical-semantic organization, along with timed measures of non-word decoding and real word reading. The inclusion of timed measures is noteworthy, as past research has focused on reading accuracy only and results have been equivocal. It was hypothesized that if semantics influenced word reading through facilitating activation of underlying representations, this would be more evidenced in increased reading rate rather than in raw accuracy. Results. All oral language measures were found to be well correlated with decoding and reading rate. Regression analyses revealed that even after controlling for phonemic awareness, vocabulary breadth and lexical-semantic organization predicted both decoding and word recognition proficiency. Phonemic awareness and lexical-semantic organization were the only unique predictors of reading rate when all measures were considered together. Conclusions: These novel results suggest that the role of oral vocabulary in reading skill acquisition extends beyond phonology and lexical breadth; organization or richness of the lexical-semantic system may facilitate the storage and/or retrieval of phonological and orthographic representations, resulting in better reading fluency.
Heeyoung Park (Speech, Language, and Hearing Sciences, University of Florida); Linda J. Lombardino; Katelyn A. DiPietro; Denise C. Magdales; Danielle Schoepski; Katherine E. Martin; Lori J. P. Altmann - Processing Speed in Young Adults with Developmental Dyslexia: A Domain-General or Domain-Specific Deficit?
Purpose: The finding that dyslexic readers have processing speed deficits has led researchers to question whether these speed deficits are domain-general or domain-specific, but to date, no clear evidence exists to address this issue. This study is designed to investigate whether processing speed deficits in college students with developmental dyslexia are restricted to speed of linguistic information processing only or also affect non-linguistic information processing. Method: Seventeen dyslexic students were compared with seventeen age-matched normal readers, all monolingual native speakers of English. The study had a 2 stimulus type by 2 response type design. Stimulus types included English letters as linguistic stimuli and Korean letters as non-linguistic stimuli. Participants responded to stimuli verbally (Say the number corresponding to a symbol) or via motor response (Press the button under the correct response). Results: All participants were slower responding to non-linguistic stimuli. Dyslexic readers performed significantly more slowly on all processing speed tasks than normal readers. There were no interactions between group and other variables. Covarying age and fluid intelligence had no effect on these results. Conclusions: These results suggest that dyslexic readers are significantly slower than age-matched normal readers on both linguistic and non-linguistic processing speed tasks regardless of response modalities. These results support the hypothesis that processing speed deficits in dyslexia are domain-general. Findings will be discussed relative to their applicability to reading instruction and other educational issues.
Rauno K. Parrila (University of Alberta)Jennifer Barber - Intact orthographic learning in dyslexia: Evidence from an eye movement study with continuous text
Purpose: Most existing studies on orthographic learning suggest that dyslexics need more exposures to new words to learn them. We presented normally reading adults and high-functioning adult dyslexics orthographically difficult novel words in context while recording their eye movements. We expected that the two groups would differ both in word recognition time (as indexed by first-pass gaze durations) and in orthographic learning (as indexed by the slope of the regression line fitted onto successive gaze durations on the same novel word). Method: Participants were 20 typically reading university students and 20 university students with phonological dyslexia. Their eye movements were recorded during the reading of non-fiction text in which the novel words were embedded. Results: Eye movement data indicated that while the typically reading participants had consistently longer gaze durations, no differences were observed in orthographic learning. Instead, both groups showed a rapid and parallel reduction in gaze durations from the first to the fourth exposure, followed by a period of more stable gaze durations from fifth up to 12th exposure to the novel word. Conclusions: These results indicate that at least high-functioning adult dyslexics can develop orthographic representations of novel words equally fast as their normally (and well) reading peers. The results also suggest phonological access as the likely reason for consistently slower first-pass gaze durations, and indicate that the interaction between phonological and orthographic processing requires further examination in the context of reading in connected text.
Nicole Patton Terry (Georgia State University) - Lexical quality among young chidren who speak Nonmainstream American English
Purpose: The lexical quality hypothesis posits that differences in the quality of word representations in the lexicon are related to differences in reading skill. This study explored lexical representations (LR) of beginning readers who speak nonmainstream American English (NMAE), as dialect differences may compromise the quality of the LR. Method: Children in pre-kindergarten through second grade were given experimental tasks that measure phonological and semantic components of the LR indirectly. Each task was modified to include dialect-sensitive items (different pronunciation (chest vs. chess) or meaning (ice vs. jewelry) between mainstream American English (MAE) and NMAE) and dialect-neutral items (same pronunciation or meaning in MAE and NMAE). These tasks included: cross-modal lexical decision (i.e., does this spoken word "match" this picture), picture naming precision (i.e., name the target and then say it "slowly and clearly"), and acceptability judgment (i.e., listen to various pronunciations of a target and tell me if each is "okay"). Results: Data analysis is on-going. Preliminary analyses suggest significant dialect group effects (e.g., heavy NMAE speakers were more likely to provide NMAE pronunciations of words, accept NMAE pronunciations of words, and match NMAE pronunciations to targets). However, they also chose MAE pronunciations and target matches frequently on all three tasks. Conclusions: Preliminary results suggest that young NMAE speakers' LR contain information about both MAE and NMAE-they know much about MAE forms that align closely with print as they are learning how to read. The theoretical and educational implications for dialect variation and reading achievement will be discussed.
Rufina Pearson (Universidad Catolica Argentina)Linda Siegel - Predictors of bilingual reading in Spanish speaking children
Purpose Spanish-speaking children (N=69) in bilingual (English/Spanish) classes participated in a longitudinal study (kindergarten to grade 5) whose main aim was to investigate the relationship between early measures of phonological awareness in Spanish and reading achievement in both languages at grade 5. Method Participants were administered measures of phonological awareness (PA) and letter identification (LI) in Spanish at kindergarten, reading measures in Spanish at grades 1 and 2, and reading and verbal measures at grade 5 in both English and Spanish. Regression analysis on reading measures at grade 5 and PA, word reading and LI measures at kindergarten and grade 1 and 2 were conducted. The results were analysed in terms of: predictors of reading, strength of the prediction, effect on reading comprehension and interaction effects between good and poor readers. Results Phonological awareness (PA) and letter identification measures in Spanish at age 5-7 were predictive of reading measures in both languages at grade 5 as shown by regression analysis. Also PA in Spanish was predictive of English reading comprehension (RC). Oral language skills were predictive of reading comprehension within language as found by previous studies. However we found a cross language effect from Spanish language skills to English reading comprehension. Finally, we found that poor Spanish readers had difficulties decoding in English, although they do not differ from good readers on RC. Conclusions These findings support the transfer of phonological awareness skills across alphabetic languages, and the importance of phonological training to enhance bilingual skills.
Felipe Pegado (INSERM U992-Neurospin FRANCE)Kimihiro Nakamura; Stanislas Dehaene; Laurent Cohen - Literacy breaks the symmetry of alphabetic visual objects
All primates, including humans, recognise images in a left-right invariant-way. This mirror-invariance is useful to recognise objects both from left or right perspectives, but this very competency has to be 'unlearned' for reading acquisition in order to correctly identify letters (e.g. to distinguish a 'b' from a 'd'). In a first study, we presented pairs of visual stimuli (faces, houses, tools, strings and falsefonts), whose left-right orientation was manipulated, to adult literates and illiterates. The task was to judge if the pairs were 'same' or 'different', regardless of orientation (identity task). The subjects were explicitly instructed to assign 'same' for mirror-inverted pairs. The results showed an important behavioural cost to respond 'same' in mirror-trials, proportional to the literacy level, but only for strings and falsefonts. A strong bias to respond 'different' in mirrored-strings was also observed in good readers but not in illiterates. In a second study (Neuroimage, in press), we used an fMRI priming paradigm to probe the neural discrimination of mirror-inverted pairs of stimuli in skilled readers. We demonstrate that the left occipito-temporal cortex, namely the Visual Word Form Area (VWFA) distinguishes the left-right orientation of single letters, and yet exhibits mirror invariance for simple matched pictures. These results clarify how letter shapes, after reading acquisition, escape the process of mirror invariance which is a basic property of the ventral visual shape recognition pathway.
Meghan Pendergast (Georgia State University)Gary Bingham, Nicole Patton-Terry, Kizzy Albritton - Examining the English and Spanish language and literacy growth of preschool dual-language learners
Purpose The purpose of this study was to examine the growth of dual language learners' (DLL) oral language growth in English and in Spanish. Method Participants. Ninety-three four-year-old children across ten Georgia early childhood centers participated in this research. Latino children in this study primarily came from low-income backgrounds where Spanish was the home language. Procedures. Children's early language and literacy development was assessed at the beginning and end of prekindergarten on the following measures: The Peabody Picture Vocabulary Test, 4th Edition (PPVT-4: Dunn and Dunn, 2007) and the Test de Vocabulario en Imagenes Peabody (TVIP: Dunn, Padilla, Lugo, & Dunn, 1986), the Test of Preschool Early Literacy (TOPEL: Lonigan, Wagner, Torgesen, & Rashotte, 2007), Narrative Assessment Protocol (NAP: Pence, Justice, & Gosse, 2007), and the Rhyme Awareness Task (PALS-PK: Invernizzi, Justice, Landum, & Booker, 2004). Results Repeated Measure ANOVAs were used to examine differences in students' fall and spring scores on both Spanish and English assessment. Utilizing standard scores, results reveal that DLL children evidenced significant positive growth on English measures of vocabulary, expressive language, narrative quality, and rhyme awareness. Findings also demonstrate that these children did not experience a significant change in their Spanish language acquisition, as measured through Spanish measures of vocabulary, expressive language skills, rhyme awareness, or narrative quality. Multiple regression analyses demonstrated associations between DLL language growth and their home and preschool experiences. Conclusion Results will be discussed in relation to previous research documenting the impact of home language and early childhood environments as well as research on cross language developmental parallels (Bravo-Valdivieso, 1995; Gonzalez & Valle, 2000; Hammer, Miccio, Wagstaff, 2003; Hammer, Lawrence, & Miccio, 2008; Lindsey, Manis, & Bailey, 2003).
Peng Peng (); Douglas Fuchs; Donald L. Compton; Devin M. Kearns; Eunsoo Cho; Lynn Fuchs; Kristen L. McMaster - Rethinking Fidelity-of-Treatment Implementation:Conceptual and Practical Implications Based on Scale-up PALS
Purpose This analysis is based on a data from a multi-site, longitudinal scaling up study of the 2nd-6th-grade PALS reading program. We aimed to determine whether all components of the PALS program were implemented by teachers with equal accuracy; and to explore relationships between the fidelity with which PALS components were conducted and (a) teachers' total fidelity score and (b) students reading improvement. Method We used a 198-item observation checklist to evaluate the fidelity with which 80 study teachers and students from Tennessee, Minnesota, and Texas implemented PALS. Analyses were based on the fidelity data of 80 PALS teachers from Tennessee, Minnesota, and Texas. Project staff identified 9 PALS components that reflected program content, the quality of intervention implementation, and participants' degree of engagement. Hierarchical Generalized Linear Modeling (HGLM) was used to investigate relations between PALS components and teachers' total fidelity score. Correlation and regression analyses were used to explore the relative importance of PALS components to students' reading improvement. Results HGLM analysis indicated that teachers had relative difficulty implementing Paragraph Shrinking, which requires students to construct main idea statements at the paragraph level. Paragraph Shrinking was the only component of the 9 PALS components that correlated with teachers' Total Fidelity score and students' reading improvement, and it explained 7% of variance in students' reading improvement. Conclusion PALS components were differentially difficult for teachers to conduct with fidelity; they correlated differently with teachers' total fidelity score; and they were differentially important in explaining students' reading gains across the 17-wk project period. Findings may help researchers think about how to identify the "active ingredient(s)" in their intervention programs.
Charles Perfetti (University of Pittsburgh)Suzanne Adlof - Are integrative comprehension processes the same in listening and reading?
Purpose: The simple view of reading assumes that comprehension processes are equivalent between written language and spoken language, once word identification processes are controlled. However, key differences between spoken and written discourse (e.g., rate of input, memory load, pragmatic constraints) place differential demands on cognitive resources. Considering these differences, we report a new study of adult spoken language processing, which tests whether implicit measures of spoken language comprehension can combine with measures of decoding and vocabulary to predict reading comprehension. Method:Implicit processes of word-to-to text integration are observable through evoked potentials (ERPs). We measured adults' ERP responses to critical words in spoken sentences. Across three experimental conditions, the critical word repeats, paraphrases, or is inferable from information in the previous context. Each was compared with a baseline condition. Results: Regression analysis based on 23 of a planned 40 subjects indicates that reduction of the N400 while listening to paraphrases was significantly related to standardized reading comprehension test scores, after controlling for decoding and vocabulary knowledge. Conclusions: These results demonstrate that an implicit brain-based indicator of language comprehension is associated with a paper and pencil reading comprehension task that is loaded with strategic and other variables beyond comprehension. We also will discuss results for inference and repetition conditions; e.g. word repetition did not produce an N400 reduction in listening, perhaps reflecting stronger constraints against word repetition in listening compared with priming influences in reading.
Dolores Perin (Teachers College Columbia University) - Literacy Intervention for Low-Achieving Postsecondary Students
This study measured the potential efficacy of an intervention for low-achieving postsecondary students. The intervention addressed written summarization, question-formulation, vocabulary development, persuasive writing, and responding to comprehension questions. Performance in two conditions, science contextualization and generic text, was compared to that of a business-as-usual comparison group. Participants were n=246 students in 16 developmental education (remedial) classes in three community colleges. Students were randomized to condition within 12 of the classrooms, with the other four serving as the comparison. Outcomes were measured on alternate forms of a researcher-designed measure, and transfer was measured using the Nelson-Denny Reading Test. Using hierarchical regression controlling for pretest, site, science knowledge, science interest, student demographics and academic background, on the summarization posttest, the science and generic conditions exceeded the comparison condition on the inclusion of main ideas, and word count (ES = 0.36-0.70). Further, students in the science condition included one-third of a standard deviation more main ideas and their accuracy scores were 0.33 SD higher than those of the generic text group (both p<.05). Additionally, the science group was four times more likely to copy from the source text than either the generic or comparison group, who did not differ from each other. The increase in copying may reflect change on metacognitive awareness over the intervention period. In conclusion, findings on a measure closely related to the intervention were positive, and benefits of contextualizing the intervention in science text were seen on two variables. However, gains did not transfer to the reading test.
Douglas Petersen (University of Wyoming); Ronald Gillam - Predicting reading ability for bilingual Hispanic children using dynamic assessment Predicting reading ability for bilingual Hispanic children using dynamic assessment
Purpose: This study investigated the predictive validity of a dynamic assessment designed to identify bilingual Hispanic children at risk for future reading difficulty. Method: During kindergarten, 63 bilingual Hispanic children completed a dynamic assessment nonsense-word decoding task that yielded pretest to posttest gain scores, residuum scores, response to decoding strategy scores, and temporally related working memory information. At the end of first grade, the same children completed word identification, decoding, and reading fluency tasks designed to represent the narrow view of reading. Results: The dynamic assessment measure pertaining to the response to reading strategy instruction measure yielded the highest classification accuracy, with sensitivity and specificity at or above 80% for all three formative criterion reading measures, including 100% sensitivity for two out of the three first-grade measures. Conclusions: The dynamic assessment of reading strategy surfaced as a parsimonious, valid means of predicting first-grade word-level reading ability for Hispanic, bilingual children.
Stephen Petrill (Department of Human Development and Family Science, Ohio State University) - Math Fluency is etiologically distinct from decoding fluency, untimed math, and untimed reading performance: Evidence from a twin study.
Purpose. The cognitive and educational psychology literatures suggest that math fluency, although correlated with untimed math measures, may also constitute an additional dimension of math performance. Using a quantitative genetic perspective, we examined the genetic and environmental etiology of math fluency; and whether math fluency was distinct from decoding fluency, as well as untimed reading and math skills. Method. We examined 314 pairs of school-aged twins from the Western Reserve Reading and Math Projects. Twins were assessed through a 90-minute home visit at approximately Age 10 and were reassessed in their homes approximately one year later. Measures of calculation fluency, reading fluency, untimed reading and untimed math were loaded onto latent reading and math factors at each age, then simultaneously decomposed using a multivariate genetic analysis. Results. Results of the study suggested that the shared environment and genetics influenced the covariance among math fluency, untimed math measures, and reading measures. However, roughly 2/3 of the variance in math fluency was independent from untimed math measures and reading, including reading fluency. A portion of this genetic variance was related to decoding fluency. However, the majority of this independent variance was due to genetic factors related solely to math fluency that were longitudinally stable across two measurement occasions. Conclusions. These results suggest that math fluency, although related to decoding fluency and other math measures, may also be a genetically-distinct dimension of mathematics performance. In contrast, shared environmental influences on math fluency are generalist; associated with untimed measures of math and reading performance.
Yaacov Petscher (Florida Center for Reading Research) - A comparison of methods for scoring multidimensional constructs unidimensionally in literacy research.
Purpose: The purpose of this investigation is to highlight a sequence of latent variable models which may be applied to scores representing multidimensional constructs. This paper will a) highlight traditional methods for assessing dimensionality in literacy research, including a strictly unidimensional model, an uncorrelated multidimensional model, a correlated multidimensional model; and a bifactor model. The bifactor model will be discussed as an alternative to common modeling approaches as a method to evaluate distortions which occur when multidimensional data are fit with unidimensional models, and as a method to score mulitidimensional constructs in a unidimensional framework. Method: Data on 8,545 students in grades 3-10 on the computer adaptive reading comprehension assessment from the Florida Assessments for Instruction in Reading (FAIR) were analyzed for this investigation. The four general latent variable models were estimated using Mplus software and were compared using the Akaike's and Bayesian Information Criteria, as well as the log likelihood statistic. In addition to the four general latent models, three forms of the bifactor model were tested, including the restricted, unrestricted, and partially restricted models Results: The results indicated that the reading comprehension data were multidimensional in nature, and that the general bifactor model provided the best fit to the reading comprehension data across all grades. Conclusions: The bifactor model serves as an important analytic approach to handle multidimensional data.
Beth Phillips (Department of Educational Psychology and Learning Systems, Florida State University and Florida Center for Reading Research) - A new classroom language environment scale: Initial validation and exploration
Purpose Accumulating findings suggest the importance of early language and vocabulary development for children's successful literacy acquisition. Whereas substantial attention has been devoted to the specific language interactions between parent and child that may promote language growth, less attention of this specific type has been devoted to teacher-child interactions. This presentation summarizes the development, reliability, initial validation, and teacher and classroom correlates (e.g., experience, education, auspice, ratio) of the Classroom Language Environment Observational Scales (CLEOS). This measure, unlike more typical, broad-coverage classroom environment scales, focuses exclusively on the language environment created by teaching staff with respect to language and vocabulary scaffolding, instruction, and incidental learning opportunities. Method One hundred twenty classrooms representing a range of public prekindergarten, Head Start, and private childcare auspices were observed with the new measure. Teachers completed a background survey summarizing education, experiences, and instructional goals and preferences. Descriptive, correlational, and predictive analyses explored the interrelations among structural and process features as measured by the CLEOS. Convergent validity with a widely used general classroom quality measure (the Teacher Behavior Rating Scale) also was investigated. Results Preliminary results support the reliability and validity of the CLEOS, as well as a varied pattern of relations with teacher and setting characteristics. Results indicate the wide range of language environments created by teachers, and suggest that teaching experience and expertise, rather than education credentials, may be related to higher quality. Conclusions Findings suggest a need for greater professional development attention to supporting teachers' incidental language scaffolding and explicit vocabulary instruction.
Linda M. Phillips (University of Alberta, Edmonton, Canada) - Critical issues between home and school in family literacy intervention programs
Purpose: The purpose is to argue that broad-based collaborations between families and educators are essential to children's literacy development and are too often hindered by ill-conceived negative opinion on both sides. Method: Monthly observations and interviews of 156 urban and rural families and teachers were conducted as part of a 5-year study. Classroom atmosphere, scaffolding, and fidelity of program implementation were measured using descriptive statistics and qualitative analyses. Results: 68% of parents wanted opportunities for their children to learn and 17% to improve their own literacy; 60% were surprised that teachers didn't change their teaching even though the children had been in a family literacy program. Families and teachers (95%) acknowledged the need to accept responsibility for developing a positive relationship. Conclusions: Treating parents as a collective group obscures differences that must be considered if productive collaborations are to occur. What is typical of families with certain cultural, linguistic and social backgrounds is important to learn but there is always variability within the groups in terms of beliefs about children's learning. Parents must understand the role and lives of teachers and vice-versa in order to maximize children's success at school. Early literacy development is a complex phenomenon widely acknowledged to be affected by many aspects of child development as well as the gap between home and school perceptions and practices. Home-school collaborations holds promise for bridging the gap between home and school perceptions and practices. There is no gain in blame.
Elayne Cristina Morais Pinheiro (Universidade Federal de São Paulo)Elayne Cristina Morais Pinheiro; Thaís Barbosa; Alan Baddeley; Ana Luiza Gomes Pinto Navas; Orlando Francisco Amodeo Bueno - Adaptation of a reading screening test to Brazilian Portuguese
PURPOSE: Pen and paper screening tests are essential instruments for teachers and professionals working with education. However, there are few reading screening tests in Brazilian Portuguese. The aim of this study was to adapt a reading screening test, The Reading Decision Test (RDT), to Brazilian Portuguese. This test ensures that children who are learning to read a regularly spelled language can be tested on in terms of decoding and comprehension. The RDT contains three subtests (letters, words, sentences), each containing two versions (Form 1 and 2), with increasing difficulty levels. This allows children to be tested on two occasions. Each task measures the ability to discriminate nonsense letters, words (pseudowords) and sentences, from real ones. METHOD: The test was applied to groups of children while they were in school. Children (n=230), aged from 6 to 10 years, from public schools in São Paulo, answered both versions of each subtest with a one week interval between assessments. RESULTS: Intraclass correlations (ICC) showed that both versions of each subtest were correlated: letters (ICC=0.68), words (ICC=0.90), sentences (ICC=0.90) (p<0.05). Kruskal-Wallis and Mann-Whitney tests showed that both versions of each subtest demonstrated school grade differences (p<0.01). Spearman's correlation showed that the RDT was significantly correlated to an existing Brazilian isolated words reading test (r=0.79, p<0.001). CONCLUSION: The adapted RDT proves to be a rapid and sensitive measure of reading during the early stages of reading development. The test was able to differentiate reading profiles of children on different school grades.
Aparna Pisupati ();Sheryl Rimrodt; Raj Stewart; Nicole Davis; Ken Pugh; James Pekar; Laurie Cutting - Neurobiological Correlates of Learning the Meanings of "Words"
Purpose: After completing a dynamic training paradigm (DTP) associating meanings to pseudowords (TP), typically developing adolescents (Controls) had activation to TP that was more similar to real words (rW) than novel pseudowords (NP). We administered the same DTP to adolescents with poor reading skills (RD) to determine if the effect was similar. Additionally, because DTP alternates two different conditions for associating meaning to pseudowords, we assessed the effect of the different conditions. Method: 39 participants (21 Controls, 18 RD, ages 10-14 years) completed the DTP before fMRI scanning. DTP was evenly split between two conditions: Context (sentential) and Isolation (with a picture), and every participant completed both conditions. The fMRI task presented 160 "word" stimuli (40 TP, 40 NP, 80 rW). SPM's were generated to analyze the group differences for different learning conditions and stimuli type. Results: RD and Controls showed activation to TP that was similar to activation for rW; however, RD showed greater activation in left middle temporal gyrus and insula than Controls for NP. RD also showed greater activation in right IFG for Isolation training. No significant differences were observed between RD and Controls for TP from Context training. Conclusions: Though Controls and RD show differential activation for NP, there were no differences for TP; thus, neurobiological correlates consistent with training-based learning were suggested in both Controls and RD. Additionally, evidence of a differential response in RD to Isolation condition could have implications for optimizing reading instruction design, and warrants further study.
Nicola Pitchford (University of Nottingham, UK);Maria Ktori; Walter van Heuven - Inhibitory effects of exterior letter frequency on visual word recognition: differential patterns across English and Greek
Purpose: Skilled readers of English (deep orthography) identify letters that appear frequently in the initial and final positions of words faster than letters that are uncommon in these positions. However, for skilled readers of Greek (transparent orthography) this relationship is restricted to the initial letter, suggesting that differences in orthographic structure influence the nature of letter position encoding. Method: English and Greek word and nonword stimuli were manipulated orthogonally for exterior letter frequency. For each orthography, a set of 80 word stimuli was created: 20 with high frequency exterior letters (e.g., ACTOR), 20 with high frequency initial but low frequency final letters (e.g., ALIEN), 20 with low frequency initial but high frequency final letters (e.g., BOXER), and 20 with low frequency exterior letters (e.g., BACON). A matched set of orthographically legal and pronounceable nonword stimuli was created. Age and education matched groups of monolingual English (N=24) and Greek (N=24) readers and a group of bilingual Greek‐English readers (N=24) took part in a lexical decision task. Results: Results revealed an inhibitory effect of letter frequency on lexical decisions. Both English and Greek readers were faster at recognising words and nonwords with low compared to high frequency initial letters, suggesting initial letter frequency may be encoded prelexically. In contrast, only English readers were faster at recognising words (and not nonwords) with low compared to high frequency final letters, suggesting this effect arises from lexical processes. Conclusion: These findings show that orthographic structure modulates exterior letter processing in visual word recognition.
Anne Plisson (Université de Montréal);Rachel Berthiaume; Daniel Daigle - Spelling ability: A comparison of dyslexic and deaf children
Dyslexic children and deaf children encounter great difficulties when learning to read and write (Colin et al., 2007; Snowling, 2006). For different reasons, both populations' delay is explained by a phonological deficit (Musselman, 2000; Ramus et al., 2003). More specifically, this phonological deficit impedes spelling development in alphabetic languages like French (Sprenger-Charolles & Colé, 2003). The main goal of this research is to investigate spelling ability in populations known for their phonological deficit and to compare them. 16 dyslexic children and 16 deaf children matched on age (mean age: 11.1) and on reading level were compared to 21 normal-developing children of the same age (CA) and to 24 younger normal-developing children of the same reading level (CR). All participants had to write a text. Texts were analysed at the graphemic level. Results indicate that 1) dyslexic and deaf participants write shorter texts and use shorter words than CA and CR, 2) dyslexic and deaf children' success rate is comparable to CR, but lower than CA, 3) dyslexic participants, CA and CR have the same pattern of errors (their errors are most of the time acceptable at a phonological level). Deaf participants' errors are phonologically acceptable half of the time only. Those results are discussed in terms of spelling processes used by the different populations. The question of the relative importance of non phonological processes is raised.
Hanne Poelmans (ExpORL, Dept. of Neurosciences, K.U.Leuven, Belgium)Heleen Luts; Bart Boets; Maaike Vandermosten; Pol Ghesquière; Jan Wouters - Auditory steady-state responses indicate reduced neural phase-locking to the phoneme-rate of speech in adults with dyslexia
Purpose: Spoken language processing depends strongly on the efficiency of the auditory system to process auditory features in speech. It is hypothesized that in dyslexia, deficient processing of rapidly-changing auditory information underlies a deficient development of phonological representations, causing literacy problems. Particularly, the ability to process amplitude (AM) and frequency (FM) modulations was found to be sensitive in the differentiation between normal reading and dyslexic individuals. Low-frequency modulations between 4 and 20Hz correspond to the processing rate of important phonological segments (syllables and phonemes respectively) in speech and therefore provide a bridge between low-level auditory and phonological processing. Method: Neural AM and FM processing was investigated in a group of normal reading and dyslexic adults by means of auditory steady-state responses (ASSRs). The ASSR is a neural oscillatory activity that is phase-locked to the frequency of the auditory stimulus. ASSRs were evoked by amplitude or frequency modulations at 4 and 20Hz. Furthermore, psychophysical tests of speech-in-noise perception, AM/FM detection and phonological awareness were assessed. Results: Preliminary analyses suggest that compared to normal readers, dyslexic subjects showed decreased phase-locking to the 20Hz modulations, whereas no group differences were found for 4Hz. Furthermore, a relation between neural phase-locking to 20Hz modulations and psychophysical tests seem to be present. Conclusion: This study introduced a new technique in dyslexia research to show a decreased ability of the neural system of dyslexic adults to phase-lock to the phoneme-rate in speech. This insufficiency may influence the phonological encoding of spoken language in individuals with dyslexia.
Sharolyn Pollard-Durodola (University of Denver, Child, Family, and School Psychology);Jorge E. Gonzalez; Deborah C. Simmons; Aaron B. Taylor; Matthew J. Davis; Melissa Fogerty;Leslie Simmons - An observational study of teacher-child conversations around shared book reading: what happens between once upon a time and happily ever after
Abstract Purpose This study examines the relation between teachers' instructional emphasis and distribution of instructional time around shared-book reading with child receptive and expressive vocabulary outcomes. We hypothesized that inferencing through cognitively demanding questioning using textual and extra-textual talk would predict children's language outcomes. Research questions: 1. How does instruction before, during and after shared book reading activities relate to preschooler's vocabulary on standardized measures of vocabulary? 2. How are differing types of teacher instructional activities related to preschooler's vocabulary on standardized measures of vocabulary? Method Participants included 99 preschool disadvantaged children along with 13 preschool teachers participating in a larger study. Observations were conducted using an observational system titled "Multi-Option Observational System for Experimental Studies (MOOSES) Standardized measures of expressive and receptive vocabulary were used. Data were obtained from 39 video-taped observations. Observers coded teacher behaviors according to three foci: a) the distribution of time (before, during, after reading texts), b) the instructional emphasis (i.e., labeling, describing, associating, etc.), and c) the instructional focus of each event [i.e., target vocabulary word or comprehension/concept knowledge]. Results Findings indicated that the duration of teachers making vocabulary and comprehension associations significantly predicted posttest PPVT scores while time after reading was a significant predictor of posttest EOWPVT scores. Frequency of teacher associating vocabulary was also a significant positive predictor. Conclusions Findings from this study indicate that instruction that emphasizes higher cognitive skills such as making associations impact student receptive and expressive vocabulary learning. Implications for practice will be discussed.
Tatiana Cury Pollo (Universidade Federal de São João Del Rei)Cláudia Cardoso-Martins; Rebecca Treiman; Brett Kessler - A longitudinal evaluation of the syllabic spelling hypothesis in Portuguese
PURPOSE This study of Portuguese-speaking children investigated the prevalence of spellings in which symbols putatively represent entire syllables. Such spellings are a hallmark of Ferreiro and Teberosky's (1982) constructivist theory of spelling development, which has influenced early literacy education in many countries. METHOD Participants were 76 children in Brazil. We maximized the chance of observing children at a syllabic stage of development by testing them every 3 months from the age of 4 years to approximately 6 years. At each time, children spelled 12 words ranging from 1 to 4 syllables. Children were classified as syllabic at a test date if the number of symbols they wrote for a word corresponded to the number of its syllables significantly (at .05 level) more frequently than predicted by a permutation test. Children were classified as phonological if the letters they wrote were plausible spellings of the word's phonemes significantly more frequently than predicted by a Monte Carlo test. RESULTS Of the 454 testing sessions during which a child could have been classified as a syllabic speller, only 24 (.053) were so classified. This number is not significantly higher than expected by chance. Furthermore, in contrast to Ferreiro's claim that syllabic spellings are common among children who don't yet spell alphabetically, only 2 of these observations involved children classified as nonphonological spellers. CONCLUSIONS The results fail to support the influential idea of a syllabic stage in the development of spelling in learners of alphabetic writing systems.
María Porta (Unidad de Psicología Evolutiva y Educacional) - Beginning literacy: the effects of a classroom-based linguistic training program and its relationships with students´ home and classroom literacy environment
PURPOSE: Using a pretest-posttest comparison-group design, this 20-week study assessed the effects of a phonological awareness/letter training program (PALTP) on literacy achievement in 113 at-risk Argentinean children: 62 kindergarteners (M=5.2 years) and 51 first graders (M=6.4 years). METHOD: The PALTP was administered to the experimental group (EG; n=55) and focused on: phoneme segmentation; rhyme categorization; initial sound categorization; sound blending; and letter-sound knowledge. 34 thrice-weekly sessions of 30 minutes each were administered by the teacher to the entire classroom. One pre-test and two post-test measures of phonological awareness (PA) abilities, writing, and word reading were obtained, as well as a measure of reading comprehension. Level of parental education, home literacy environment, teacher beliefs and the literary environment of the classroom were all evaluated. RESULTS: Significant interaction effects between the subjects' age and the administration of PALTP were found: For kindergarteners, levels of PA, word reading, and comprehension for the EG were greater than those of the CG. Although no significant differences were observed between the first graders' groups, pair-wise comparison analysis showed greater pretest-posttest gains for the EG than for the CG on word-reading measures. Regression analysis for both age groups showed letter-sound knowledge and PA were equally predictive of word reading and teacher beliefs were predictive of PA. For kindergarteners, home literacy environment was predictive of PA and reading ability. Parental educational level did not predict reading achievement. CONCLUSIONS: PA training programs that include sound/letter matching activities help children understand the alphabetic principle and promote reading acquisition.
Mads Poulsen (University of Copenhagen) - Lexical access and letter access are involved in different aspects of reading in Grade 5
Purpose: This study investigated the effects of lexical access speed and letter access speed on reading fluency and reading comprehension. We hypothesized that 1) letter access speed would correlate with reading fluency but not comprehension, while 2) lexical access speed would influence reading comprehension. For readers who are struggling with recoding, most of the reading effort is probably tied up with recoding, leaving little to be explained by lexical access. Therefore we expected that 3) lexical access speed would primarily predict reading fluency for readers who were no longer struggling with recoding. Method: 85 Grade 5 students completed tasks of reading accuracy, fluency, and comprehension. In addition they were tested on isolated letter naming and isolated picture naming tasks as measures of letter and lexical access speed. All items in both naming tasks were unique. Parallel serial rapid automatized naming measures with letters and objects were also taken. Results: In the full sample letter access speed, but not lexical access speed, accounted for unique variance in reading fluency. The reverse was true for reading comprehension. Half the sample were at-ceiling recoders (95% accuracy). In this subset sample, both letter access and lexical access accounted for unique variance in reading fluency. The pattern of effects for lexical access did not change by controlling for serial rapid naming (RAN). Conclusions: The results suggest that letter access and lexical access are important for different aspects of reading. In addition, the influence of lexical access on reading appears to change with reading development.
Daisy Powell (Institute of Education, University of Reading); Lynette Chesson - Is Rapid Automatized Naming (RAN) a cause or consequence of reading skill?
Purpose: A full understanding of the causal nature of the RAN-reading link remains elusive. Our approach was to assess RAN and related cognitive skills in unusually young, English non-readers (mean age: 3; 10), to establish a) the cognitive correlates of RAN and b) how RAN and its correlates relate to later literacy. Based on previous research, we assessed both phonological and non-phonological (e.g. visual, executive) factors thought to underlie the RAN-reading link. Method: 43 children (mean age 3:10) were first assessed (T1) while in the nursery (pre-school) year of UK schools, and reassessed nearly a year later (T2), 3 months after the onset of literacy instruction. Children were non-readers at T1 (BAS single word reading score = 0). T1 measures included RAN (colour and object naming), verbal and non-verbal abilities, letter knowledge, visual processing, phonological awareness (PA), phonological memory (PM), and executive functions (EF). T2 measures: single word reading; letter knowledge. Results: Interestingly, the strongest concurrent correlate of non-alphanumeric RAN was letter-knowledge. RAN was also related to both phonological (PA and PM) and visual processing, consistent with our previous research. Preliminary regression analyses indicated that alongside letter knowledge, PM at T1 independently predicted T2 literacy measures, whereas surprisingly PA, EF and vocabulary did not. Conclusions: Both phonological and non-phonological factors contributed to RAN in these non-readers. However, early RAN and PA skills appeared less strongly predictive of T2 decoding skills than letter-knowledge and PM, perhaps due to reliance on sounding-out and blending strategies at this early stage of literacy acquisition.
Kelly Powell-Smith (Dynamic Measurements Group); Ruth Kaminski - Examining the effects of Tier 3 early literacy and language interventions in preschool
Purpose: Curricular options to provide Tier 3 support for children with delays in early language and literacy development have not been well conceptualized for preschools. This paper will report the effects of a Tier 3 curriculum that engages children in intensive early language and literacy intervention. Method: A multiple-baseline design across subjects was used in two studies, one for early literacy and one for language. Nine low-performing children for each intervention were identified for participation. During baseline, children received typical classroom instruction. Intervention began for a group of three children when their data were stable. After experimental effects were established for the first group, intervention was initiated for a second group, and later a third group. A follow-up phase examined skill maintenance. Mastery-monitoring measures were administered twice per week in baseline and once per week during intervention. Intervention was conducted for ~10 minutes three times per week. The early literacy intervention focused on letter names, phonemic awareness, and sound-letter correspondence. The language intervention focused on core vocabulary and use of core vocabulary in simple sentences. Results: Most of the children demonstrated robust experimental effects on multiple measures of language and literacy skill acquisition. Results revealed more and less effective components of instruction. Conclusions: Results provide knowledge that will inform practice and further refinement of the interventions under development. The study provides preliminary research support for the efficacy of a Tier 3 curriculum for preschool children needing intensive literacy and language support.
Sarah Priebe (University of Texas Health Sciences Center at Houston);Janice Keenan - Knowledge effects on word identification
Purpose : While the prevailing wisdom is that accurate word identification leads to better comprehension (cf. Pinnell et al., 1995), there is also the possibility that the relationship is bidirectional- perhaps better comprehension leads to more accurate word identification. The purpose of this study is to extend previous research on the effects of prior domain knowledge on word identification (Priebe, Keenan, and Miller, 2010) by comparing the same reader across two passages: one where they have prior knowledge, and one where they do not. By doing so, we can be assured that prior knowledge effects we find are not the result of greater world knowledge that might come with having prior knowledge of a topic. Method: 26 4th-grade children read aloud two passages of the QRI (Qualitative Reading Inventory-3; Leslie & Caldwell, 2001). All participants had no prior knowledge of the first passage, and prior knowledge of the second. These readings were transcribed, and errors were tabulated and coded in terms of their graphic and semantic similarity to target words. Results : When children had prior knowledge of the passage, they read the passage more accurately (F(1,24) =7.17, p <.05). In addition, the errors that they made on the prior knowledge passage were more semantically similar (F (1,24) =5.72, p <.05). Conclusions: The results from this study indicate that prior knowledge effects on word identification are not solely due to differences in world knowledge. In addition, this study has implications for assessment and intervention.
Athanassios Protopapas (University of Athens); Despoina Moirou; Artemis Markatou; Eleni Vlahou - Effects of practice on the relation of reading to color naming and interference
Purpose: We examined the effects of practice on color naming of neutral and interfering stimuli. Word reading is correlated with color naming and with Stroop interference, in which color words are printed in a different color ink which must be named. It is often assumed that interference arises from lack of automaticity of color naming and might decrease with naming practice. Method: 58 adults and 66 children (Grades 4-5) formed 4 groups and practiced for 3 days on 192 stimuli (144 for children) each day. The "interference" group practiced color naming with incongruent color words; the "color" group practiced color naming with strings of Xs; the "word" group practiced reading aloud color words; the "control" group did not practice. The days before and after practice each participant read color words and named colors of neutral stimuli and incongruent color words. Reading and inhibition tests were also included. Results: Interference practice led to decreased interference for both adults and children. Color practice did not reduce color naming time and did not affect interference. Word practice reduced children's (but not adults') interference. Interference changes could not be attributed to reduced impulsivity or improved cognitive control. Conclusions: Findings are inconsistent with lack of color naming practice being a major cause of interference. Rather, word reading seems to be the limiting factor, because practice at an age when reading is not yet fully fluent leads to reduced interference. Studies of interference help us understand the development of word reading efficiency and associated cognitive control.
Cynthia Puranik (University of Pittsburgh)Stephanie Al Otaiba; Kristen McMaster; David Parker, Jessica Sidler Folsom; Luana Greulich; Aaron Rouby; and Kadie Ann Bourgoin - Using the levels of language framework to examine writing in beginning writers
Purpose - The purpose of this study was to examine how to assess beginning writing at various language levels: text, sentence, and word levels. Research questions included: Are there dissociations in the writing of beginning writers at various language levels? Is performance at the various levels of language comparable? Method - Participants for this study were 521 first graders from 8 schools and 35 different classrooms. The mean age of participants was 6.2 years (SD = .38). There were 233 females and 289 males from a range of SES. Using Curriculum-Based Measures (CBM), writing samples at the word, sentence, and text level were collected. These included a measure of spelling, letter writing fluency, sentence writing, and composition. Samples were coded to yield the following: Words Written (WW), Correct Word Sequences (CWS), Incorrect Word Sequences (IWS), and Number of T-units. In addition, children's letter writing fluency scores were calculated and spelling was analyzed. Results -Children were divided into good, average, and poor writers based on the WW score. Group performance was compared at the text, sentence, and word levels. Preliminary findings suggest there was a wide range of performance with one child being able to write only one word to another child writing 99 words. Conclusions - We anticipate, based on preliminary data analysis that among beginning writers that word sentence and text level performance are moderately related. Implications of final analysis are expected to have implications for writing instruction. For example, do beginning writers need writing instruction and support at various language levels?
David Purpura (University of Illinois at Urbana-Champaign); Laura Hume; Darcey Sims; Christopher J. Lonigan - Emergent literacy skills and early numeracy skills in preschool: the value of including emergent literacy skills in the prediction of early numeracy skills development
Purpose: The purpose of the current study is to examine if emergent literacy skills in preschool uniquely predict mathematics skills one year later while controlling for initial early numeracy skills. Method: Sixty-nine children were assessed in the spring of their first or second preschool year on the Preschool Early Numeracy Skills (PENS) test, the TOPEL, and the Stanford-Binet Copying test. These children were then assessed a year later (40 of the children were still in preschool and 29 were in kindergarten) on the PENS test and the Applied Problems and Calculation subtests from the Woodcock-Johnson III Tests of Achievement. Three stepwise regressions were conducted using Time 2 PENS, Applied Problems, and Calculation as the dependent variables. Age, Time 1 PENS scores, and the Copying test were included in Step 1 of the regression and the three subtests of the TOPEL (Print Knowledge, Vocabulary, and Phonological Awareness) were included in Step 2. Results: Both Print Knowledge and Vocabulary accounted for unique variance in the prediction of Time 2 PENS scores and Applied Problems scores while controlling for Time 1 PENS scores, Age, and Copying. Phonological awareness did not uniquely predict any of the mathematics domains. None of the emergent literacy skills accounted for unique variance in the prediction of the Calculation score. Conclusions: The findings of this study help to identify an important link between emergent literacy skills and early numeracy skills and indicate that emergent literacy skills play a unique role in the development of important early numeracy skills.
Pauline Quemart (Centre de Recherche sur la Cognition et l'Apprentissage)Severine Casalis - Impact of the exposure to members of a morphological family on the development of orthographic representations for base morphemes: a self-teaching study
Purpose - We conducted a self-teaching study (Share, 1995, 1999) in order to investigate whether young readers benefit from successful decoding of members of a morphological family to develop orthographic representations for base morphemes. Method - French third and fifth graders had to read three short stories, in which we included a pseudoword target (a new animal: a "fembrot"; a new tree: a "maunard" and a new musical instrument: a "vaitris"). We created two versions for each of the three stories. In the first version, the pseudoword target appeared six times, but only one time as such and the five other times embedded in a suffixed pseudoword (fembrotier; fembrotet, fembroterie…). In the second version, the pseudoword target appeared six times as such and was never associated to a suffix. Results - We examined the probability to spell correctly the target word according to the condition (no decoding of the pseudoword; decoding of the pseudoword alone; decoding of the pseudoword in its suffixed forms) and to the grade. In each grade, the rate of correct spelling was lower when the participants had not been exposed to the base than when they had been exposed to it. In addition, the rate of orthographic learning was higher when the participants had successfully decoded the base alone than when it was embedded in suffixed forms. Conclusion - As early as third grade, phonological recoding is critical to the acquisition of orthographic representations for base morphemes, and this orthographic learning occurs even when bases are embedded in suffixed forms.
Jamie Quinn (Florida State University); Richard Wagner - Sex differences in dyslexia and the accuracy of its diagnosis
Purpose: The purpose of this study was to explore sex differences in dyslexia, to discover if these differences are due to ascertainment bias, and to determine the accuracy of its diagnosis. Methods: Gender and odds ratios were calculated for 491,103 beginning 2nd grade students using four definitions of reading disability at four levels of severity, the bottom 30th, 15th, 5th, and 3rd percentiles. We used two IQ-discrepancy definitions and two low-achievement definitions for reading disability. We conducted three tests to assess school identification for ascertainment bias. Finally, to analyze the accuracy of identification procedures, we calculated sensitivity and specificity rates for school and research-based identification procedures. Results: Sex differences in dyslexia increased with increasing severity of the reading problem, peaking at a ratio of 2.4:1 for a broad measure of reading connected text for meaning and at a ratio of 1.6:1 for a narrow measure of decoding nonsense words. Accuracy of school-based identification was poor overall and worse for girls than boys: Only 1 out of 4 boys and 1 out of 7 girls with dyslexia were school identified as learning disabled. Conclusions: Sex differences in dyslexia are attributable largely to male vulnerability rather than ascertainment bias, based on the results of three tests of ascertainment bias and the fact that sex differences were found when all participants were assessed without ascertainment. The rates of identification by the schools are dismal; at best the schools identified only 20% of the students identified as reading disabled by research-based criteria.
Ralph Radach (University of Wuppertal);Jessica Hill; Christian Vorstius; Albrecht Inhoff - Word shape does not contribute to word processing in sentence reading
Purpose: Since the beginning of reading research, the shape of words in the visual periphery has often been credited to be an influential factor in the acquisition of linguistic information (Huey, 1908). Empirical work on this topic has so far been restricted to single word recognition paradigms, and some results may have been compromised by possible confounds with other types of word form and letter features. The present work examines the role of word shape and letter discriminability in two sentence reading experiments using the boundary technique. Method: The first experiment manipulated mask similarity, with either similar or dissimilar letters, and whether word shape information was provided or denied, while controlling for lexical frequency (high) and word shape frequency (rare). A second experiment was designed to provide optimal conditions for parafoveal word shape processing. To this end a contextual manipulation was introduced, with either high or normal contextual constraint, while controlling for lexical frequency (high), word shape frequency (rare), and mask similarity (dissimilar). Results: Experiment 1 indicated a small but significant effect of letter similarity on parafoveal processing, such that gaze duration on the target word was longer for dissimilar masks. No effect was found for word shape. Experiment 2 demonstrated significant preview and predictability effects; however, again there was no evidence for any role of word shape. Conclusions: Our findings suggest that word shape information, although available, is not used in early word processing during sentence reading. This is in contrast to earlier research, in which we found a substantial processing advantage of mandatory noun capitalization in German. We attribute this apparent contradiction to differences in the diagnostic value of visual word form features, strongly suggesting that word shape is not an important part of 'holistic' word information.
Gloria Ramirez (Thompson Rivers University);Li-Jen Kuo; Kelsie Baab; Ying Li; Tae-Jin Kim; Pamela Bollinger - The development of morphological and syntactic awareness among young second language learners: A study of children in Spanish-English two-way immersion bilingual program
Purpose: This study investigated the impact of early experience in a second language on children's development of morphological and syntactic awareness - two aspects of metalinguistic awareness that become progressively more important beyond the initial phase of literacy development (Kuo & Anderson, 2008). The study extended the scope of previous research by using measures that took into account the structural similarities and differences between bilinguals' two languages. Method: The study was conducted at two elementary schools that offered both general education and Spanish-English dual-language programs. Participants included 100 fourth graders with one of the following language profiles: 1) Monolingual English-speaking children in the general education program (Eng-GenEd.); 2) English-speaking children in the dual-language program (Eng-Dual, English-speaking); and 3) Spanish-speaking children in the dual-language program (Spn-Dual). Participants were assessed on researcher-developed measures of vocabulary, morphological awareness and syntactic awareness in Spanish and English. Results: Results show that Eng-Dual scored lower than Spn-Dual on Spanish vocabulary and syntactic measures. Despite receiving less instruction in English, Eng-Dual performed similarly to Eng-GenEd on all morphosyntactic measures in English. Furthermore, Eng-Dual outperformed Eng-GenEd on the measure of English vocabulary that involved Spanish-English cognates. Spn-Dual scored lower than Eng-GenEd and Eng-Dual on morphological awareness in English, but performed similarly to their English-speaking peers on syntactic awareness and measures of vocabulary that involved Spanish-English cognates and morphologically complex words. Conclusions: Findings from the study have implications for two major theories of bilingualism; cross-language transfer and cognitive flexibility, as well as for policies regarding language programs.
Kathy Rastle (Department of Psychology, Royal Holloway University of London); Matthew H. Davis; Marjolein Merkx - The acquisition of morphological knowledge investigated through artificial language learning
Purpose. Morphological knowledge plays a key role in word recognition and provides the primary means for lexical productivity. This research investigated how this knowledge is acquired using an artificial language paradigm in which participants learned novel suffixes (e.g. -nept) presented in novel word contexts (e.g. sleepnept). We sought to determine whether participants could acquire representations of the novel suffixes that were sufficiently robust to generalise to untrained exemplars (e.g. sailnept), and to characterise the time course of this generalisation process. Method. Undergraduate participants were trained on eight novel suffixes presented in the context of familiar stems (e.g. sleepnept, teachnept). Participants were tested immediately after training or two days after training on a variety of speeded and non-speeded tasks including a speeded auditory repetition test that assessed performance on trained items as well as items comprising an untrained stem and trained novel suffix (e.g., sailnept). Results. Results of the auditory repetition task showed an advantage for trained items immediately after training, as these were repeated significantly faster than items with trained stems but untrained suffixes. Critically, this advantage generalised to untrained items comprising the novel suffixes (e.g., sailnept) but only in the group of participants tested two days after training. Conclusions. This study demonstrates that participants can acquire morphological representations in a laboratory context that are sufficiently robust to generalise to the interpretation of novel exemplars. It also suggests that this process of linguistic generalisation may require a period of offline overnight consolidation consistent with complementary systems theories of lexical learning.
Anabelle Reta Sánchez (Florida State University); Yaacov Petscher; Laura B. Lang - Examining the growth of reading for migrant students in Florida
The purpose of this investigation was to study trajectories of literacy for Florida migrant students in grades 4, 8, and 10 using measures of reading comprehension, spelling, and text reading efficiency. We tested the extent to which migrant students differed in level and slope on the identified measures frm a group of non-migrant English Language Learner students, and a comparison group which reflected the demographics of students in Florida. A total of 3,122 migrant students were identified for the selected grades in Florida during 2009-2010, and were compared to the reference using a series of multilevel growth models. Scores of reading comprehension, spelling, and text reading efficiency were drawn from the Florida Assessments for Instruction in Reading. Developmental ability scores were used to measure growth in reading comprehension and spelling, while adjusted fluency scores were utilized for text reading efficiency in order to account for form effects across multiple passages. Migrant students significantly differed frm the referent groups in their initial ability scores across all literacy measures across all three grade levels; however, only minor differences were estimated in slope. Namely, though migrant students' developmental trajectories were positive, they did not differ from the comparison groups in reading comprehension or spelling. Slope differences in text reading efficiency were apparent in favor of migrant students in grades 4 and 8. Migrant students' growth during the year fails to close the gap observed at the beginning of the school year. Implications for intervention strategies will be presented.
Patricia Riddell (University of Reading, UK);Taeko Wydell; Rachel Pye; Rachael Sperring - The effects of letter rotation on lexical decision in English and Japanese Adults
Purpose: The Psycholinguistic Grain Size Theory suggests that readers of transparent orthographies read words at the level of the grapheme/phoneme while readers of deep orthographies read several graphemes simultaneously (Goswami and Zeigler, 2005). In contrast, Wydell and Butterworth (1999) have suggested that both granularity and transparency are important when reading text. We compared these theories by rotating the individual characters in words by varying amounts. This disrupts the appearance of groups of letters more than individual letters. We predicted that, if only grain size was important then Kana would be more affected than English which would be more affected than Kanji, while if both grain size and transparency were important, we predicted that English would be more affected than Kanji which would be more affected than Kana. Method: Fifty-two English adults (mean age: 36 years) and 54 Japanese adults (mean age: 43 years) completed Raven's matrices, a single word reading task and 3 lexical decision tasks (0°, 60° and 160° random letter rotation). Results: Results revealed that there was an interaction between language and level of rotation on reaction time after accounting for differences in baseline reading ability and age (F(4,314)=27.497, p<0.001). Post hoc testing showed that there was a significantly greater effect of rotation on Kana than on English and Kanji, and that English was more affected than Kanji. Conclusion: Our results support the Psycholinguistic grain size theory suggesting that grain size rather than transparency and granularity are important in defining differences in reading ability across languages with different orthographies.
Mabel Rivera (University of Houston)David Francis; Kristy Santi - Teaching Literacy to English Language Learners: What Have We Learned?
Purpose Over the last decade the study and validation of effective ways to teach literacy to English Language Learners has increased. The purpose of this study was to collect and disseminate knowledge of effective reading and writing assessment and interventions for ELLs in grades K through 12. Method We examined evidence of enhanced instruction in elements of literacy for ELLs performing below grade level. This information was first culled from a variety of sources, including major literature reviews and single studies, and then summarized in a literature review. Results The findings from this review indicate that assessment plays a crucial role in the educational experience of all ELLs. Assessments inform the process of educational placement, which in turn influences instruction and academic outcomes. ELLs need comprehensive, early, explicit, intensive interventions that reflect a close match between their source(s) of difficulty and the intervention. In addition, high-quality teacher preparation and professional development programs on teaching ELLs should address theoretical knowledge and pedagogical methods on second language acquisition. Conclusions Meeting the needs of ELLs requires a multi-dimensional strategic approach, with significant attention to three areas: 1) assessment, 2) intervention programs or models, and 3) professional development. The findings of this study demonstrate the significant positive impact that effective reading and writing interventions may produce, particularly for ELLs with low performance outcomes. Finally, teachers with strong content knowledge and specialized training have a positive impact on student performance.
Greg Roberts ()Audrey J. Leroux; Sharon R. Vaughn; Jack Fletcher - Response to intervention in middle school: longitudinal results from a tiered reading intervention with struggling adolescent readers
Purpose This study is part of a larger 3-year NICHD-funded initiative that addresses effective instruction in a tiered-intervention framework. To date, there is little research on the longitudinal impact of well-implemented response-to-intervention (RTI) models, and there is no multi-year, single cohort research on RTI's potential in middle school settings, despite the fact that RTI is increasingly prevalent in US schools, particularly in settings where "turn around" is a priority. This study considers the empirical basis for the emerging RTI-related policies and practices. Method A group of 1009 at-risk 6th graders were randomly assigned to a treatment or a business-as-usual condition prior to enrolling in 6th grade (610 in treatment and 399 in comparison). Treatment was an increasingly intense intervention provided daily to students not responding to less intense intervention provided in the prior school year. Non-responding students and a random sample of responders in both conditions were evaluated at pretest and posttest each of the three years. Missing data were assumed missing at random. Group-specific trends were estimated and compared using multi-group, latent-variable growth modeling. Parallel processing models were also estimated to test the hypothesis that treatment decreases students' reliance on word-level skills to comprehend text. Results Intercept, slope, and shape (quadratic) were estimated for the total group and within each condition. Across 3 years, the treatment group outperformed the comparison on a measure of reading comprehension when slope and shape were jointly constrained (p<.01). On sight word reading, slope for the treatment group was statistically significantly greater than slope in the comparison group (p<.05). Also, achievement in word reading (i.e., slope) was more predictive of comprehension slope in the comparison than in the treatment (.01). Conclusions An intensive 3-year treatment increased middle school students' word reading skills and reading comprehension. Treatment also "uncoupled" students' reliance on word-level skills for comprehending connected text, due presumably to the comprehension strategy-based nature of the intervention.
Sarah Robins (Washington University in St. Louis, Philosophy-Neuroscience-Psychology Program); Rebecca Treiman - Parent-child conversations as a source of early letter knowledge
Purpose - In research presented at SSSR 2010, we studied informal conversations between parents and their preschool children to see whether they talk differently about letters and pictures. Here we examine talk about letters, asking about various features of these conversations and how they change across the preschool years. Method - We examined transcripts from 32 corpora of conversations between U.S. English-speaking parents and children between ages 1;0 (years; months) and 5; 0 (N = 533), available in the CHILDES database. Features of the conversations were analyzed using a multilevel model. Results - Conversations included many questions about letters. Parents asked more questions than children did, although children began to catch up around age 4;0 (p < .001). These questions focused on letters' names and shapes. Even with older children, there were few questions about letters' sounds. Letter names were often uttered in sequences, a clue that they belong to the same category. While the use of sequences increased as children grew older (p < .001), the proportion of alphabetic sequences declined (p < .001). That is, parents and children transitioned from talking about letters as part of the alphabet to letters as part of words. Prior to 3;0, most of the letters associated with words were associated with the child's name (p < .001). Conclusions Informal conversations both provide a source of and reflect children's letter knowledge. The nature of these conversations helps to explain why children are more knowledgeable about some aspects of letters than others.
Astrid Rodriguez (CUNY Graduate Center)Linnea Ehri - Reading Text in Two languages Facilitates Reading Comprehension in Lower Proficiency ESL Students
Purpose. Research on cross-linguistic influences on reading comprehension among language-minority adults is scant. This study was conducted with Spanish-speaking students in community college ESL classes. Of interest was whether students would comprehend a text better if they read it first in Spanish (L1) followed by English (L2) than if they read it twice only in English. A second question was whether the contribution of L1 reading to L2 reading comprehension would be greater for students with lower than higher English proficiency. Method. Two levels of English proficiency (Level 2, very limited, and Level 3 limited proficiency) were distinguished using the Woodcock-Munoz Language Survey-Revised (2005). Students at each level (N=40) were randomly assigned the Spanish/English or English/English text reading conditions. Students read the text aloud. Reading comprehension was then assessed with free and prompted recall and cloze tasks. Results. Students with lower English proficiency comprehended the text significantly better when they had read it in Spanish before English than when they had read it twice in English. No advantage was observed for students with higher English proficiency. Think alouds revealed that thoughts were commonly expressed in both Spanish and English as all groups read the English text. Conclusions. Results suggest that memory representations of text are not language-specific for ESL students and that meaning transfers from L1 to L2 independently of surface forms. Results are consistent with Cummins' (1979) linguistic interdependence hypothesis and also with his linguistic threshold hypothesis regarding the conditions supporting the contribution of L1 to processing in L2.
Cristina Rodríguez (Researcher)Cristina Rodríguez; Juan E. Jiménez - The role of phonological awareness and verbal working memory on sentence comprehension in Spanish children with and without reading disabilities.
Purpose: The main purpose of this research was to analyze the effects of phonological awareness and verbal working memory on written sentence comprehension in Spanish normally achieving readers and dyslexic children from a cross sectional study. Different studies have shown the relevance of verbal working memory (VWM) as a critical factor in the explanation of sentence comprehension performance (Just & Carpenter, 1992 ; Wiseheart, Altmann, Park & Lombardino, 2009). However, the syntactic processing deficit in children with dyslexia could be also explained by other phonological skills such as phonological awareness (PA). Method: A sample of 178 children was selected from second to sixth grade. Participants were classified into two different groups: 89 normally achieving readers (CA) and 89 reading disabled children (RD). Groups were compared in the performance of sentence comprehension that require reassignment of thematic roles (v.gr. passives sentences) using the sentence-picture-matching paradigm. Three different designs were carried out: 1º) A Group (CA,RD) x Grade (2nd to 6th ) analysis of variance in accuracy on sentence comprehension task; 2º) An ANCOVA Group (CA, RD) x LEVEL (2nd to 6th) using VWM as a covariate; and 3º) An ANCOVA Group (CA,RD) x LEVEL (2nd to 6th) using PA as a covariate. Results: Main effects of group and grade were found in the first design. In the second design, grade differences disappeared in sentence comprehension task after VWM was controlled. Finally, main effects for group and level disappeared when PA was a covariate. Conclusion: The poor performance of RD group on sentence comprehension task is better explained by underlying deficits in PA than in VWM.
Kathleen Roskos (John Carroll University) ;Karen Burstein - What does e-book pedagogy look like? A formative study of an early literacy instructional model
Purpose. The role of the e-book in early literacy instruction remains ill defined, even as e-books are proliferating and lab-based research points to their potential. This study proposes an instructional model that integrates the e-book into the preschool curriculum. Its primary objective is to formatively assess model components for functionality (purpose) and usability (ease of use). Grounded in instructional design and early literacy research, the model addresses 4 components of e-book pedagogy: (i) e-book quality; (ii) environment; (iii) engagement; (iv) instruction. Method. Using a formative research approach, the model was implemented in 8 Early Reading First classrooms with small groups (n=3); four ERF classrooms served as controls. Each component was successively collaboratively implemented, assessed and modified over a 4-week period, followed by 2 weeks of adjusting each component. Assessments included: (1) a research-based survey tool of e-book quality; (2) physical environment design criteria applied to video/photo samples; (3) videotaped interactive behaviors at-screen + child eye-gaze at-screen + pre/post child inhibitory behaviors for engagement; and (4) video observations of fidelity to a science-based procedure, pacing, teacher-child language relevant to instructional goals. Data were analyzed using mixed methods. Results/Conclusion. Model functionality was improved, including an e-book quality survey tool; environmental specifications; at screen teacher protocols; e-book shared reading procedures. An important research task is not only to understand how e-books impact early literacy learning, but also how to use them well in teaching. This study defines and describes an e-book model design with potential to enhance the overall literacy-learning environment for young children.
Maureen Ruby (Eastern CT State University); Kristine Mika - Assessing teacher knowledge of literacy: Foundation for professional development
Purpose: Teachers don't always know what they don't know about literacy instruction, resulting in misguided professional development (PD). This overview of research in progress presents an approach to inventorying teachers' declarative and procedural literacy knowledge through use of a Teacher Knowledge Inventory (TKI) and providing objective evidence to teachers and administrators for planning PD to meet identified teacher needs at the district, school, and specific grade levels. Method: Participants include over 225 K-8 classroom teachers from three districts (rural, urban, and suburban) serving diverse populations. An inventory of 100 multiple choice items similar to the California, Connecticut, and Massachusetts reading assessments for teacher certification was created and administered to teachers in districts that requested to take the inventory. Data were analyzed descriptively and presented to administrators and teachers. District use of results (PD, curriculum and assessment revision, RTI) are described. Results: Initial results reveal significant deficiencies in teacher knowledge of literacy. Teacher response to results was an overwhelming "demand" for PD based on results. Districts also reevaluated student outcomes from state and district assessments using the lens of the results of the TKI and adopted a new K-2 CORE reading program, intervention programs for at risk 6-8 readers, and district level assessment and progress monitoring systems. Data from state reading assessments indicate positive impact of on reading scores. Conclusion: Data suggest inservice teachers lack literacy knowledge and skills tested on many states' initial certification assessments, and identified deficiencies can be addressed by targeted professional development.
Susie Russak (Beit Berl Academic College, Israel)Janina Kahn-Horwitz - Trajectories in EAL spelling development: comparisons between good and poor spellers
The present study examined the developmental trajectory of spelling knowledge in English (EAL) among 229 5th, 8th and 10th grade Hebrew L1 speakers in order to glean insight into the effects of exposure and practice with the English orthography on spelling proficiency. We specifically explored the developmental differences between good and poor spellers in relation to the acquisition of novel phonemes (ð, æ, ɛ, h), and orthographic conventions (silent e, consonantal clusters). Spelling development was examined in Hebrew L1 and English EAL. The Hebrew measures used for the study were standardized spelling, orthographic and phonological choice measures. Experimental English measures included real and pseudo word spelling, orthographic and phonological word choice and the standardized WRAT Spelling measure. ANOVAs showed that the differences between poor and good spellers were only significant in the 5th grade, except for the case of the silent e where there were significant differences in the 8th grade as well. The developmental profile within groups showed that for the poor spellers only the difference between scores in 5th and 8th grade for cc, ð, ɛ, were significant. The phoneme æ, proved to be difficult for all spellers until 10th grade On the other hand, the initial h proved difficult to spell for the poor spellers but not for the good spellers in 5th grade. Our findings show that most of the novel phonemes are accurately spelled after 6 years of English study. However, the novel phoneme æ remains a challenge regardless of ability and years of study.
Amir Sadeghi (Islamic Azad University, Damavand Branch)John Everatt; Brigid McNeill - Phonological and Orthographic Processing in Persian Reading Comprehension
This talk will present the findings of a study investigating the cognitive-linguistic predictors of reading comprehension levels of Persian speaking children. The Persian orthography, unlike English, is written from right to left. It is cursive and most of the letters change their shape when connecting to letters on one or both sides. The orthography also has the feature of using marks to represent sounds within the language. These marks are not always included in written text, particularly in passages targeted at more experienced readers. Over 200 school-children in Iran from grades 2 to 5 were given measures of text reading involving Cloze completion or passages followed by comprehension questions. Ability levels on these measures were compared to scores on additional measures of language competence, phonological ability, orthographic processing and speed of processing. Analyses indicated that Persian reading comprehension levels were predicted by measures of linguistic process and decoding ability, with the latter being predicted by phonological and orthographic processing skills. The findings will be discussed in terms of the application of the simple model of reading to Persian, which has an orthography with varying transparency between letters and sounds. Cross-language comparisons will be made with English, the language from which the model was derived, and Arabic, which shares the same orthography. (words = 211)
Elinor Saiegh-Haddad (Bar-Ilan University) - Linguistic processing deficits in Arabic reading disability
Purpose: The purpose of the study is to investigate the phenotype of reading disability in Arabic by examining the cognitive and linguistic processing (phonological and morphological) skills, as well as the word-level reading and spelling abilities of Arabic native speaking reading disabled versus normally developing children in grades 1-4. Method: A total of 160 children in grades 1 through 4 were tested on a range of phonological, morphological, and word-level reading and spelling tasks. Screening for reading disability was based on Low Achievement Models. Results: The results revealed significant deficits among reading disabled children on all phonological and morphological processing tasks, with a clearer deficit in the phonological than in the morphological domain, and with no particular deficit in the morphological processing of root versus word-pattern. We also found a clear word decoding deficit among the reading disabled children on all word-level reading tasks (voweled, unvoweled, isolated, and within text), but with the reading disabled showing a significant difference on the reading of text-level voweled versus unvoweled words but not on the reading of voweled versus unvoweled words in isolation. Spelling of words and of pseudo words, in particular, also appeared clearly deficient among reading disabled as against normally developing readers. Conclusions: The results are interpreted as reflecting universal deficits as well as the role of Arabic-specific linguistic structure (phonological and morpho-orthographic) in the manifestation of dyslexia in Arabic.
Paula Salmi (University of Jyväskylä, Finland); Tuire Koponen; Kenneth Eklund; Tuija Aro; Riikka Mononen; Pekka Räsänen; Timo Ahonen - RAN and counting as predictors of reading and arithmetic calculation fluency: children with language impairment and a sample with high prevalence of dyslexic readers
Purpose: This presentation reviews the aims and results from two different sub studies of RAN and counting as predictors of reading and calculation fluency. Study 1 examined the influence of language impairment on calculation fluency and how cognitive factors explain the variance among SLI children. Study 2 examined to what extent children's performance in counting and rapid serial naming (RAN) predicted fluency of arithmetic calculation and reading at 2nd and 3rd grade among a sample with high prevalence of dyslexic readers. The effects of phonological awareness and half of the variance in verbal short-term memory were controlled. Methods: Study 1: the performance of 9 to 11-years old children with SLI (N=29) was compared to that of typically developing children along a continuum ranging from pre-school to third grade (N= 20, 47, 40 and 33). Study2: Participants were 200 children from the Jyväskylä Longitudinal Study of Dyslexia (JLD). The Sructural Equation modelling (SEM) approach was applied. Results: Study 1: Most of the children with SLI (69%) had difficulties in calculating fluently. The subgroup analyses revealed that calculation fluency was connected to naming speed. Study 2: The results showed counting as a strong predictor of both later calculation and reading fluency. Similarly, RAN was equally associated, both with later calculation and reading fluency. These associations remained, even after controlling for phonological awareness. The association between counting and reading was stronger in a sample including dyslexic readers. Conclusions. Both counting and RAN could work as early predictors of reading and calculation fluency.
Marwa Sarsor (Tel-Aviv University); Iris Levin - Letter knowledge in a Diglossic language: Learning of letter knowledge among Arab kindergartners in Israel
Purpose: Arabic letters have two characteristics - two names and 3-4 shapes per letter. These complicating characteristics affect reading acquisition. We studied with which of the letter name systems children are more familiar with, which is easier to learn, and how does learning affect alphabetic awareness, i.e., reporting letters that start/end oral words. We studied also the relative familiarity with and learning of different letter shapes. Method: Kindergartners were divided into four groups according to the stimuli they were trained on: letter names (colloquial or literary) and letter shapes (for beginning or for the end of a word in a cursive form). Each child was trained to name four letters s/he was unfamiliar with. Training took place in two sessions of a dyadic structured game (experimenter and child), lasting 20-25 minutes. The child was exposed to one name on one shape of each letter. Children were pretested and post-tested on letter naming and on alphabetic awareness. Results: On pretest, some children were familiar mainly with colloquial names and others with literary names, but overall familiarity with the systems was indistinguishable. Pretest-posttest comparisons showed that children were as successful in learning colloquial and literary names. The same trends appeared on letter shapes. On alphabetic awareness, children tended to use colloquial names. Conclusions: Arabic children can be taught to name letters from an early age in the two name systems and on all shapes. Learning varieties of names and shapes is useful, for literacy teaching-learning at the early ages in kindergarten and school.
Robert Savage (Department of Educational and Counselling Psychology)Louise Deault; Maria Di Stasio - The relationship between classroom environments and growth in attainment in Grade 1.
Purpose - Pressley and colleagues (2001) claim that growth in literacy in elementary classrooms is well-mapped by their indices of Atmosphere Instruction Management and Engagement (AIMS). We thus explored the validity of the AIMS observation schedule in predicting a) teacher self-descriptions of their teaching b) observations of teaching, c) student perceptions of their classrooms and d) growth in literacy during Grade 1. Method - An AIMS interview was completed by 29 first grade teachers. Researchers observed these classrooms. Student perceptions were gathered using an adapted Classroom Environment Scale (CES, Frederickson et al., 1999) evaluating Involvement, Affiliation, Rule Clarity, Task Orientation, and Order and Organization in the classroom. Results - Analysis using Hierarchal Linear Modeling (HLM) showed first that all the data displayed variation at the classroom level. HLM was then used to assess all other associations. Students' perceptions of classroom involvement, affiliation, and of order and organization, were predicted by teacher's perceptions of AIMS classroom management, and student engagement sub-scales. Growth in listening comprehension ability was predicted by all four AIMS subscales. Student perceptions of teacher support in the classroom were linked to reading comprehension attainment measures. Conclusions - Some scales of the AIMS (Atmosphere and Management) predict growth in some measures of literacy (listening comprehension). Triangulation across observation interviews and student surveys was robust. More work is needed to develop and evaluate scales assessing word reading and reading comprehension. Student perceptions of their own classrooms seem to be important predictors of adult observations and attainment as early as Grade 1.
L. Brook Sawyer (Lehigh University);Laura Justice; Ying Guo; Katherine Glenn-Applegate; Stephen Petrill - Relation between home literacy environment and print knowledge of preschool children with language impairment
Purpose The specific aims were two-fold. The first aim was to determine the unique contributions of the home literacy environment (HLE) on print knowledge, preschool children with language impairment (LI), while the second aim was to determine whether child characteristics (i.e., language ability or behavior ratings) moderated this relation. Method The sample consisted of 49 preschool children with LI. Seven conceptualizations of HLE were used as predictors of children's print knowledge and were measured via parent report and observation: storybook exposure, literacy teaching, child's print motivation and the HOME measure total score and three subscales containing literacy-related items (i.e., Learning Materials, Quality of Language Stimulation, and Quality of Academic Stimulation). Moderators included parental behavior ratings and children's language ability. Results Four findings emerged from multiple regression analyses. The first finding was children's print motivation was the only HLE conceptualization that uniquely predicted children's print knowledge. The second and third findings revealed that language ability moderated the relation between (a) storybook exposure and (b) quality of academic stimulation, whereby higher levels of print knowledge were associated with higher quality academic stimulation or increased storybook exposure for children with higher language ability. The fourth finding showed that social skills was also a moderator, with higher print knowledge associated with higher quality language stimulation for children with higher social skills. Conclusions Results indicate differences in the relation between HLE and print knowledge for children with LI versus typical children. Additionally, children's characteristics play an important role.
Alfred Schabmann (University of Cologne)Barbara Maria Schmidt - Simulating reading instruction - the prognostic validity of PA and RAN in a shallow ortography
Purpose We tested the hypothesis that Phonological Awareness (PA) is only a "kick-off" for learning to read in regular orthographies and is no longer a valid predictor for reading after some kind of reading instruction has started. On the contrary we hypothesized that Rapid Automatized Naming (RAN) remains a predictor even after children were exposed to print. For spelling it was assumed that PA would be a more valid predictor even after some initial reading instruction. Method RAN and PA was measured on a sample of N=80 Austrian preschoolers. A subsequent simulated 2-hour "reading instruction" was given: Children were instructed to map 4 letter-like signs to 4 phonemes and read simple words that were constructed of the 4 phonemes. At the end of the simulation, decoding was tested using similar words. Reading and spelling in grade 1 were tested using lists of words and non words. Data were analyzed using Structural Equation Modelling. Results Overall the short simulation was a better predictor for reading than PA was. Both direct as well as indirect effects (mediated by the simulation) on reading were found for RAN, but only indirect effects were found for PA. For spelling no effects were found for the simulation, but a weak direct effect for PA and a stronger effect for RAN. Conclusions In shallow orthographies, PA is of limited prognostic validity both for reading and spelling. The idea to simulate the reading instruction process as an additional strategy to predict reading problems is promising.
Christopher Schatschneider (Florida State University); Laura Justice; Yaacov Petscher - Influence of peers' skill levels on children's language growth within preschool classrooms
Purpose: Two studies examined the extent to which peers' skill levels are associated with children's language growth within targeted-enrollment pre-K classrooms (study 1) and in early childhood special education (ECSE) classrooms (study 2). Methods: The language skills of 338 children from low-income households in 49 pre-K classrooms (study 1) and a similar number of children from ECSE classrooms (study 2) were directly assessed in fall and spring of the academic year. In each, a measure of peer language skill was created to represent the mean language score of all children in a classroom at the fall assessment point. Mixed models were used to predict children's spring language skills from peer language skills when controlling for initial status, and also to examine interactions between peer language skills and initial status. Results: Study findings showed that peer effects exert significant effects within early childhood classrooms, and appear to be most influential to children with low language skills. Conclusions: Peer effects appear to warrant serious consideration for their potentially positive and negative benefits to young children's language development.
Barbara R. Schirmer ()Laura Schaffer - Effectiveness of the Guided Reading Approach in improving reading achievement: Longitudinal investigation with students who are deaf
Purpose The purpose of the study was to investigate the effectiveness of the Guided Reading Approach on the reading achievement of elementary and middle school level deaf students during four years of intervention. Method The design was single subject experimental research. In addition, qualitative analysis of observations of instruction was conducted to determine fidelity to the Guided Reading protocols. The participants included an intact group of students who are deaf in grades 1-4 at the outset of the study and their classroom teachers at a school for the deaf. Of the 19 students who participated in year one, only one student moved during the 4 years of the study. The independent variable was the Guided Reading protocols. The dependent variable was reading achievement level as measured by Running Records, which were conducted monthly. Results In line with the use of intra-subject comparisons and visual analysis of data in single subject experimental design, Running Records data were graphed for visual analysis. Qualitative observational data were coded using the four steps of the Guided Reading Approach as a priori categories for the content analysis. Conclusions Findings indicated that almost 40% of the deaf students attained grade level proficiency and 67% of the students gained four years of progress in reading ability, although all of the students were reading well below grade level at the outset of the intervention. For most of the students, progress was inhibited by summer regressions; as these regressions lessened, progress increased.
Naomi Schneider (The Ohio State University); Howard Goldstein - Examining language and literacy in preschools for response to intervention foundations
Purpose: The purpose of this study was to examine the proportion of children who started the year before kindergarten with early literacy risk (i.e., identified as needing Tier 2 or 3 intervention) and whether a broad cross-section of prekindergarten settings were implementing high quality, evidence-based early literacy instruction, a necessary foundation for RTI. Method: In a 4-site descriptive study of 735 pre-kindergarteners from 68 classrooms representing a number of program types, children were assessed with a number of early literacy and language measures. Results: Many children appear to be entering the preschool year with significant delays in language and early literacy and may need Tier 2 or 3 intervention. Children qualifying for Tiers 2 and 3 at Fall were significantly lower (normatively) than Tier 1 children. These children made progress by Spring but did not close the gap with children identified in Tier 1. Literacy-focused instruction was infrequent, but associated with increased child engagement, provided promising targets for intervention. Conclusions: Observations of preschool programs revealed many classrooms are weak in Instructional Support, especially the dimensions most critical for supporting early literacy. Although exposure to Tier 1 (universal) early literacy curricula is keeping most children from losing ground in their pre-kindergarten year, most 4-year olds with early literacy risk are not narrowing the gap in preschool necessary to improve readiness for reading development.
Kirsten Schraeyen (Thomas More Antwerp)Astrid Geudens; Dominiek Sandra; Pol Ghesquière - Differences in the nature of dyslexic and non-dyslexic children's phonological representations: New perspectives from nonword repetion errors.
Purpose: Recent research with young dyslexic adults indicated that the ability to encode phoneme position might be a crucial factor when describing the quality of phonological representations. The results from a nonword repetition task showed that dyslexics didn't differ from matched controls at the level of phoneme retention but performed significantly worse on the retention of the phonemes' serial position within the syllable (Geudens, Schraeyen & Sandra, SSSR 2010). As masses of research data highlight the importance of high-quality phonological representations already in early reading development, we will report the results of a study in which the NWR test was administered to young children. Method: NWR (Scheltinga, 2003) data of 50 Dutch-speaking dyslexic and 50 non-dyslexic children (mean age 9;0) will be analysed with respect to the retention of both the phonemes themselves and their position, using logistic and linear mixed effects models for correct responses and error analyses. Results and conclusions: If our results with young adults can be replicated - a significant difference between dyslexics and non-dyslexics on the retention of phonemes' serial position within the syllable - we will have shown that memory for serial position is a key factor in explaining the difference between the quality of dyslexics' and non-dyslexics' phonological representations. This position factor may also be explained in terms of phonological processes (e.g. phonotactic development). In contrast to young adults, children may also show a phoneme retention effect in addition to a position effect as phoneme representations of young children may still be in development. Theoretical implications will be discussed.
Lucja Segal-Seiden (OISE/UT)Esther Geva - Development of contextual language in L1 and L2 children between grades 4 and 6
* Purpose: Investigating development of written language between grades 4 and 6 in children who are native speakers of English (L1) and children for whom English is a second language (L2). * Method: This was a longitudinal study, matched pair design, involving 44 L1 and 44 L2 (total n=88) normally developing children. Children were matched on language status and nonverbal cognitive ability (Raven Test score). Study included analysis of the stories written by the same children as part of TOWL-3 test in grades 4 and 6 in response to the picture stimulus. This poster focuses on results from TOWL-3 test, Subtest 7, "Contextual Language ", which tests the familiarity with linguistic structures in English. Data analysis was carried through Repeated Measures ANOVA with language group (L1/L2) and cognitive ability (Low/High) as the independent variables. * Results: The interactions of grade and cognitive ability, and grade and language were significant. High cognitive ability children significantly outperformed low cognitive ability children in each group. Specific tasks, eg., naming objects shown in picture and producing fewer run-on sentences were performed significantly better by the L2 children than by the L1. Total score for contextual language was significantly higher for high cognitive ability L2 children than for other groups. * Conclusions: Overall, in mastering contextual language rules in English story writing it was not the language status but the cognitive ability that determined performance, with high cognitive ability children outperforming low cognitive ability children. The high cognitive ability L2 children outperformed the other groups.
Eliane Segers (Radboud University Nijmegen, The Netherlands); Judith ter Vrugte; Ludo Verhoeven - Working memory, reading and mathematics: an information processing approach.
Purpose: we investigated the automatization of different types of mathematical operations in relation to automatization of word decoding, using task-switch experiments. According to the three-component model of DeHaene, not all types of mathematical operations are automatized, and simple additions and multiplications have an underlying verbal factor, whereas more difficult additions and subtractions do not. We investigated whether switch-costs between different operations would evidence this three-component model, and how executive functioning and word decoding ability would influence this relationship. We expected higher word decoding levels to facilitate task switches between operations with a verbal basis, and working memory capacity to be an underlying factor. Methods: Forty-two children from third grade and 27 children from fifth grade performed several tasks, measuring non-verbal intelligence, one-minute reading, digit-span backwards, and one-minute mathematics. During the experiments, they were assessed with 30 sums each time in five one-operant conditions (adding <10, adding >10, subtraction <10, subtraction >10, multiplication), and ten switch-conditions (e.g. adding<10 and subtraction <10). Both reaction times and correctness were measured, and task-switch costs were calculated. Results: We found that changes from complex to automatized operations led to a reduction in costs in comparison to baseline. Contrary to DeHaene's model, subtraction seemed to have a similar verbal basis as addition (but see Butterworth). Furthermore, word decoding and mathematical operations were found to be differentially related to working memory. Conclusions: Commonalities and differences can be found in the relation between working memory, reading, and mathematics. Theoretical and practical implications will be discussed during the presentation.
Mark S. Seidenberg (University of Wisconsin-Madison) - Dialect differences and decoding: is there a connection?
Purpose: Many African American children speak a dialect of English that differs from the mainstream dialect (MAE) emphasized in most schools. We examined whether knowledge of alternative dialects affects generating pronunciations from print, an important component of early reading. Method: We used a computational model of how children learn spelling-sound correspondences to examine effects of dialect knowledge. The models allow examination of the impact of dialect independent of other factors. In the mismatch condition, the model was taught a vocabulary of spoken word forms in AAE and then trained to pronounce written words with MAE pronunciations. Results were compared to a "match" condition in which the same dialect was used in both speech and reading phases. We also conducted a behavioral experiment examining factors that affect AAE speakers' reading aloud. Results: Learning was slower in the mismatch condition for "contrastive" words that have different pronunciations in the two dialects. This also occurred when the model was first trained on MAE pronunciations and then had to read aloud using AAE. Thus the effect is due to the existence of differing pronunciations, not AAE itself. Behavioral results indicate that knowledge of alternative pronunciations affects word naming latencies for adults, with larger effects for readers who use more AAE. Conclusions: The differing pronunciations of some words in MAE and AAE complicate the task of learning spelling-sound correspondences. Many bidialectal children may fall behind because the task of learning to read is more difficult but they are given the same time to reach developmental milestones.
Ben Seipel (University of Minnesota); Sarah Carlson; Kristen McMaster - Assessing discourse comprehension during reading using MOCCA
Purpose: The purpose of this study was to examine the psychometric properties of the MOCCA reading comprehension assessment, a cloze and multiple-choice hybrid assessment tool designed to help determine profiles for struggling comprehenders. MOCCA is designed as a cognitive diagnostic test to differentiate between two types of struggling comprehenders: comprehenders who paraphrase too frequently and comprehenders who generate invalid inferences too frequently. Method: A total of 232 students from the Grades 3-5 participated in group test administration, which included the MOCCA and the CBM Maze task; of these, 98 students completed individual testing including the Woodcock Johnson-III Word Attack and Word Identification subtests, a reading span task to test working memory, and a think-aloud task. Student scores from the CALT reading comprehension test and DIBELS Oral Reading Fluency were provided by the participating district. Results: Cronbach's alpha was calculated for all 40 items for all 232 students for all response types. Reliabilities for 'Correct' and 'Summary' response types were α = .927 and α = .846, respectively. These alpha rates demonstrate that the test items for these two response types have good internal reliability. MOCCA was also significantly correlated with the CALT (r=.58), CBM Maze (r=.325), DIBELS (r=.289 ), Word Attack (r=.252), and Word Identification (r=.37). Conclusions: With additional development, MOCCA will help teachers efficiently identify students who struggle with comprehension in different ways for the purpose of differentiating instruction .
Francisca Serrano (University of Granada)Sylvia Defior; Carmen Mª Hernandez - RFI: Reading Fluency Intervention program in Spanish dyslexic and poor readers
Purpose. A reading fluency intervention program for dyslexic and reading disabled children in Spanish is presented. Fluency in reading is considered a stepping stone in reading development, mainly due to the relationship with reading comprehension (Fuchs et al., 2005). Moreover, reading fluency difficulties in dyslexia are well documented, especially in transparent languages like Spanish (Serrano & Defior, 2009). Method. The program combines both repeated reading (Wexler, 2010) and accelerated reading (Breznitz, 2006) approaches. It provides a structured and sequential training at syllable, word and text reading levels. The training aims to improve reading automatization skills, starting at sublexical level (especially salient in Spanish) and to facilitate a more fluent and efficient word recognition, and thus fluent text reading. Additionally, it includes phonological skills training. Participants were 15 dyslexic and reading disabled children. A pretest-postest design was carried out. Assessment included a wide battery of measures: accuracy and fluency at word, pseudoword and text level reading; and comprehension, spelling, phonological awareness, orthography, prosody, vocabulary, and basic cognitive skills (visual perception, memory and intelligence). Results. Pretest-postest analysis shows a clear improvement in fluency measures at word, pseudoword and text levels. Results also indicate improvements in reading comprehension. Conclusions. The development of intervention measures for improving reading fluency is highly relevant both in typically developing and reading disabled populations. This is especially true in Spanish orthography, in which so far there are not scientific-based programs focused in reading fluency. Therefore, the relevance and practical implications of this intervention program is discussed.
Rose Sevcik (Department of Psychology Georgia State University); Jan Frijters; Robin Morris; Maureen Lovett; Maryanne Wolf - Remedial reading outcomes for children who vary according to specific language impairment status
Purpose: Current models for intensive reading interventions integrate multiple phonological and cognitive components, delivered within a language-intensive metacognitive structure. Specific language impairment (SLI) is known to be comorbid within reading disabled RD samples at rates of 14-20% (Catts et al., 2005). It remains an open question whether comorbid RD/SLI receive equal benefit from such intervention. Methods: Each of the 336 RD participants participated in at least 70 hours of intensive reading intervention with a core phonological analysis and blending component, along with metacognitive components focusing on vocabulary, reading strategies, fluency and/or comprehension skills (see Lovett et al., 2009). The presence of SLI among 23% of the sample was determined with scores < 1 SD on three of four language measures (two CELF, NEPSY, and PPVT-IV). Results: Outcomes prior, during, and immediately post-intervention formed the basis for slope estimates of reading outcomes. Lack of a statistically significant growth parameter for each participant was used as the primary metric of nonresponse (evaluated at α = .05). Growth rates 1.5 times this criterion (< 2.94 SE) defined good response. There were equal proportions of good- and nonresponse across SLI status (6-16% nonresponder; 70 to 52% good responder). When the outcome was full information from individual slopes, an overall effect for SLI status was observed (F(3, 330) = 2.92, p = .034) driven by greater gains in nonword decoding. Conclusions: A structured language-based intervention is equally effective with children who have SLI, with some evidence that intensive reading intervention is compensatory for some reading outcomes.
Dana Shafman (); Esther Geva - The contribution of morphological skills to reading comprehension in English (L1) and Hebrew (L2) young learners
Purpose: This study examined the role of morphological awareness (MA) in an "expanded" Simple View of Reading (SVR; Gough & Tunmer, 1986) framework. The framework was examined from a cross-linguistic perspective. Method: Grade 1 children with English as the home language (L1) attending a partial Hebrew Immersion program participated. Phonological awareness, rapid naming, word and pseudoword reading, oral language proficiency (vocabulary & syntax), and three separate MA tasks (inflections, derivations, compounds) were the predictors. Reading comprehension (RC) in both English (L1) and Hebrew (L2) were the dependent variables. Regression analyses examined whether MA measures explain unique variance in RC when entered following the other predictors. Analyses were conducted within language and cross-linguistically. Results: Within language analyses revealed that different aspects of MA explain unique variance in RC in each language. Derivational skills were highly significant in English, while compound knowledge was most important for Hebrew RC. English derivational morphology predicted Hebrew RC. Conclusions: The findings suggest that the specific aspects of oral language that contribute to RC are not identical in English (the L1) and Hebrew (the L2). These differences may be related to typological differences in morphological structures and the relative command of each language. The results suggest that a more nuanced SVR that takes into consideration typological differences and language status is needed. While it appears that morphological skills contribute to RC over and above word reading and vocabulary, the nature of the implicated morphological skills vary as a function of typological differences between the languages under consideration.
MIchal Shany (Haifa University);Tami Katzir;Ana Goldina; Esther Geva; - Inferring meanings of novel words from context in L2: The contribution of cognitive, linguistic, and reading abilities in L1 (Russian) and L2 (Hebrew)
Purpose: The objective of the current study was to examine the contribution of cognitive ability, memory, and language and reading skills in the mother tongue (L1) and second language (L2), to the ability to infer meanings of novel words from written context in L2. Method: Participants: 53 high school seniors, all native speakers of Russian that immigrated to Israel from countries in the former Soviet Union, and spent three years in Israel in intensive Hebrew immersion settings. Research tools: Participants were administered parallel cognitive ability, memory, and language and reading skills (decoding and comprehension) in both Russian and Hebrew. Following the Cain and colleagues (2003) task, participants were asked to read stories and infer the meanings of novel words. Results: Regression analyses revealed that vocabulary, reading comprehension and word reading accuracy in Hebrew (L2) and reading rate in Russian (L1) explained significant unique percentage of the variance on the new word-learning task. Results also revealed a significant advantage for girls with respect to the word-learning task. Conclusions: The results of this study indicate within and across language contributions to novel word learning in L2. The ability to infer novel word meanings from written context in L2 was explained concurrently by language and reading skills in L2, as well as by reading and language skills in L1. Findings from this study point to the continuing role that basic reading skills in both L1 and L2, and higher order, more complex skills in the L2 such as reading comprehension play in immigrant adolescents' L2 vocabulary development.
Laura Shapiro (Aston University, UK); Julia Carroll; Jonathan Solity - Predicting the acquisition of decoding and sight word reading over the first four years of school
Purpose We examined the influence of phonological awareness (PA) and other key skills on children's ability to learn a decoding strategy and on their subsequent word reading skills over the first four years of school. Method We measured print knowledge, PA, speech production, verbal short-term memory (VSTM), auditory, visual and motor skills in 444 children beginning school in the UK. We followed up these children's word reading and decoding progress at the end of each school year up until the fourth year of school. Results Confirmatory factor analyses of the baseline data suggested that PA, VSTM, speech production and auditory skills were separable constructs. Analyses of the outcomes at the end of the first year of school suggested that PA was the strongest direct predictor of early decoding, although auditory and VSTM skills were highly associated with decoding. In contrast, Print Knowledge provided the only direct link to early word reading, once correlations with all other baseline skills were modeled. Longer term analyses suggested that although the relationship between early PA, auditory and VSTM skills and decoding was maintained, the only direct predictor of word reading or decoding was the child's earlier performance on these tasks. Conclusions We conclude that early PA skills are critical for children to develop a decoding strategy, but once this strategy has been grasped, the main influence on a child's later reading is their previous success in reading.
Rebecca Silverman (University of Maryland); Patrick Proctor - The relationship between vocabulary and comprehension instruction and growth in reading for monolingual and bilingual children in third through fifth grade
Purpose: While there is research on reading instruction in early elementary school classrooms (e.g., Connor, Piasta, & Fishman, 2009; Juel & Minden-Cupp, 2001; Silverman & Crandell, 2010), there is limited research on reading instruction in upper elementary classrooms (except see e.g., Baker, Gersten, & Haager, 2006). The purpose of the study is to evaluate the relationship between vocabulary and comprehension instruction and reading comprehension outcomes among monolingual and bilingual students in upper elementary school. Methods: Observations at three points throughout a school year were conducted in 30 third through fifth grade classrooms. Assessments were administered to students in these classrooms at the beginning and end of the year. Assessments included the Woodcock-Munoz Language Survey, the Clinical Evaluation of Language Fundamentals, and the Gates MacGinitie Reading Test. Analyses: Observations were coded for frequency of research-based practices of vocabulary and comprehension instruction. Latent variables of vocabulary and comprehension were created from assessment data. Hierarchical Linear Modeling was used to analyze the relationship between observation codes and change in latent variables across the school year. Results: Preliminary analyses suggest that vocabulary and comprehension instruction was limited and had little impact on student growth in comprehension over an academic year. There was little differentiation of instruction for monolingual and bilingual students or for children with higher and lower ability. Conclusions: Findings from this study will inform professional development and curriculum and intervention design by providing an indication of what instructional variables are related to reading outcomes for students in upper elementary school.
Ian Simpson (); Eduardo Onochie-Quintanilla; Sylvia Defior; Betty Mousikou; Naymé Salas; Markéta Caravolas - Early letter knowledge in Spanish differs from early letter knowledge in English
Purpose: Many studies have looked at young children's knowledge of letter names and letter sounds in English. No Similar studies exist for Spanish. This longitudinal study examined English and Spanish children's letter knowledge to determine if the patterns found in English also occur in Spanish. Method: 190 children from southern Spain (mean age at commencement of study 6;7, range 6;1-7;1, 86 females) and 188 children from northern England (mean age at commencement of study 5;0, range 5;6-5;7, 90 females) took part in a series of letter naming and letter sound production tasks over a 2 year period. Results: Factors predicting task success for both groups included letter frequency and whether the letter appeared in the child's own name. An analysis of error type revealed that visual and phonological similarity led to letter confusion in both groups. While English children demonstrated superior letter sound knowledge, compared to letter name knowledge in all evaluations the Spanish children demonstrated letter name superiority. In English, children often perform better on letter names that begin with the sound of the letter (such as "B) than they do on other letters (such as "F"). Somewhat surprisingly we found the reverse pattern in Spanish with letter name beginning with the sound of letter eliciting more errors. Conclusions: While some differences between the two groups can be attributed to different teaching methods in each country, we argue that high-level differences in the two languages may account for some of the different performance patterns found between the two groups.
Darcey Sims (Florida State University/ Florida Center for Reading Research); Christopher Lonigan - The impact of early literacy intervention on the relation between behavior problems and emergent literacy skills in preschoolers
Purpose: Growing evidence suggests that behavior problems are commonly associated with early reading skills in preschoolers. Research has shown that interventions targeted at only one problem area (reading or behavior) do not significantly improve the other problem area. However, little is known regarding whether reading intervention influences the strength of the relation between these constructs. Given research showing that reading interventions improve reading but not overall behavior problems, it was hypothesized that intervention would moderate the relation between these constructs such that this relation would be weaker for children who had received a literacy intervention. Method: Measures of emergent literacy skills were completed by 268 3- to 5-year-olds (170 had received academic intervention as part of a longitudinal randomized-control-trial and 98 were in the control condition). Teachers also rated the children's prosocial, internalizing, and externalizing behaviors. Results: Across both groups, prosocial and externalizing behaviors were more often associated with emergent literacy than were internalizing behaviors. Multiple regressions predicting emergent literacy skills consistently demonstrated significant main effects for condition and child behaviors. However, in no case did the relation between emergent literacy skills and behaviors differ as a function of intervention. Conclusions: Results support the association between behavior problems and early reading skills. However, the strengths of these relations were not significantly different for children who had and had not received literacy intervention. This suggests that despite the impact of early reading interventions on improving early reading skills, they do not significantly attenuate the relations between behavior problems and early reading.
Lori E. Skibbe (Human Development and Family Studies, Michigan State University);Samantha L. Worzalla; Annemarie H. Hindman; Dorit Aram; Frederick J. Morrison - Longitudinal relations between maternal writing support and preschoolers' language and literacy skills
Purpose. The current study documented amounts and types of maternal writing support over a one year period. In addition, we investigated relations between maternal support and children's early literacy skills. Method. Eighty-seven mothers and their preschool-aged children were videotaped while they wrote an invitation for a pretend birthday party during two consecutive summers. Maternal writing support during this activity was coded at the level of the letter to document mothers' print (letter formation) and graphophoneme (letter-sound correspondence) support. In addition, children's literacy skills were assessed using the following subscales of the Woodcock-Johnson: letter-word identification, passage comprehension, and picture vocabulary. Results. Results indicate that, although there was variability in the types and amount of graphophoneme and print support that mothers provided to their children, mothers were most likely to rely on only a couple of strategies in both years, which were considered to be relatively low in quality. For letters which children received support, mothers provided significantly higher quality print and graphophoneme support in year two than in year one. Graphophoneme support in year one predicted children's literacy skills in year two, but did not predict their vocabulary scores. Print support was not related to any child skills in the present study. Conclusion. Results document the variability in maternal teaching strategies and describe how mothers change their support over a one-year period. They suggest that support focused on letter-sound correspondence may facilitate children's literacy development more than support primarily focused on helping children to write out letters.
Sara Smith (Doctoral Student, University of Oxford) - Developing receptive and productive tests of verb + noun collocational knowledge for use with young learners
* Purpose - Assessment of vocabulary in reading comprehension has tended to emphasize counting the number of words an individual knows, or vocabulary breadth. However, in order to process these lexical items effectively the learner must also have vocabulary depth. One area of vocabulary depth of increasing interest is the role multi-word phrases that express meaning as a whole, collocations, play in language acquisition and processing. While it has been long established through corpus linguistic evidence that such multi-word items are ubiquitous in naturally-occurring discourse (Sinclair, 1991), and that such items can negatively affect the reading comprehension of adult L2 learners of English (Bishop, 2004) at present the nature of collocational knowledge in young learners and its possible correlates with literacy remain relatively under-explored, largely because there are no available measures appropriate for use with children. * Method - The current study details the creation and validation of a receptive and a productive measure of verb + noun collocation knowledge for children between 7 and 10. This poster reports the findings from test administrations with approximately 80 British native and non-native English speakers. * Results - The tests yield reliable scores and show evidence of internal consistency and concurrent validity. The tests discriminate well between learners of different ages, as well as between native and non-native speakers. * Conclusions - Scores correlate highly with other measures of vocabulary indicating a relationship between vocabulary size and collocation knowledge. Possible future applications include exploring correlations between collocational knowledge and reading comprehension.
Brooke Soden (Ohio State University, Dept of Psychology)Jeanette Taylor; Chris Schatschneider - Evidence for common etiological influences on early literacy skills in kindergarten
Purpose The current study used a genetically sensitive twin study design to investigate the degree of overlap among etiological influences of print awareness, phonological awareness, and decoding during kindergarten when students were experiencing their initial formal reading instruction. Method Participants were 2,504 kindergarten twins in the Florida Twin Project on Reading. A multivariate Cholesky decomposition model was fit to the data on three decoding-related measures, Letter Naming Fluency, Phoneme Segmentation Fluency, and Nonsense Word Fluency. Results Genetic, shared environmental, and nonshared environmental factors were significant for all three reading skill phenotypes. The majority of genetic and shared environmental influence on decoding was due to common factors that included print awareness and phonological awareness. Notably, there was no additional genetic influence common to phonological awareness and decoding beyond that which was also shared with print awareness, but there was additional shared environmental influence common to just these two skills. Conclusions The bulk of genetic and environmental influences on decoding are common to the related skills print awareness and phonological awareness that begin early in the reading acquisition process prior to reading. That is, the same factors that are acting on the pre-reading skills print awareness and phonological awareness are also acting on later decoding. These findings underline that it is especially important that we understand the sources of variation common to these skills in order to recognize and identify prereaders who may become struggling readers.
Pernilla Soderberg (Linnaeus University)Ake Olofsson - Gavle University and Umea University - Spelling development during the first year of school
Purpose: To provide a detailed description of the development of spelling during the first year of schooling, and a comparison of the development of decoding and spelling. Method: One year longitudinal study of the first year of reading development in 35 seven year old Swedish children. Word decoding and letter knowledge was assessed with a computer based procedure and spelling was measured with paper and pencil tasks. The word material consisted of; (a) very familiar (content and functor) mono syllable simple words; (b) complex words which occur with high frequency in the early stages of children's reading schemes; (c) simple one-syllable pseudowords and (d) two-syllable pseudowords. Assessment of letter knowledge and phonological awareness was done prior to school entry. Results: The children had problems spelling words containing orthographic complexities, bi-syllabic nonwords were a bit easier to spell. The best result was found for simple words and for mono syllabic nonwords. Simple words and mono syllabic nonwords were relatively easier to spell than to read. On the contrary, complex words and bi-syllabic words were easier to read than to spell. Conclusions: When learning to read and spell, in the Swedish orthography, the beginning readers frequently encounter simple and short (mono syllabic) words. This may give the impression that spelling is easier than reading. This impression is reinforced by the phenomenon that it is often easier to interpret a misspelled word than a misread one. The pedagogical implications of these findings are discussed.
Fruzsina Soltész (Centre for Neuroscience in Education, department of Experimental psychology, University of Cambridge, UK)Usha Goswami; Denes Szucs; Victoria Leong - Entrainment of oscillatory activity to rhythmic acoustic streams in developmental dyslexia
Purpose: Children with developmental dyslexia have difficulties in auditory rhythmic processing and in rhythmic entrainment [1,2]. Previous studies with monkeys [3] and with humans [4] have shown that neuronal oscillations in the low frequencies entrain (phase-lock) to the structure of an attended rhythmic stimulus stream. Here we studied an analogous phenomenon in dyslexia. Method: Adults with DD and control subjects listened to a rhythmic auditory stream of sounds at a rate of 500Hz, which is the approximate rate of stressed syllables in speech. Deviants requiring response (15%, white noise) served to maintain participants' attention to the stimulus stream. EEG data was collected with a 128-channel Electrical Geodesics system. The prestimulus phase of Delta (0.5-3Hz) and of Alpha (8-12Hz) oscillations and their relation to the reaction time (RT) was analysed. Results: 1) A strong expectancy-related phase-locking was found prior to the stimulus in both groups in the Delta frequency oscillation. 2) The prestimulus Delta phase significantly predicted RT in the control group, but this correlation was completely absent in DD. 3) Meanwhile, the prestimulus phase of Alpha oscillations predicted RT only in DD. Conclusions: Both DD and controls showed expectancy effects in the Delta range, locked to rhythmic acoustic stimuli. However, the different oscillatory predictors of RT in DD suggest that DD rely on different brain mechanisms when attending a rhythmic stream of sounds. This finding may explain DD's weaker performance in auditory processing tasks, and may play a key role in phonological development via syllabic parsing and representation [5]. 1. Goswami et al, 2002, PNAS 2. Thomson & Goswami, 2008, J. Physiol. 3. Lakatos et al., 2008, Science 4. Stefanics et al., 2010, J. Neurosci. 5. Goswami, 2010, TICS
Erin Sparks (); S. Helene Deacon - Orthographic and morphological cues to English word stress: Linking cue sensitivity to reading ability
Purpose: Certain letter patterns are strongly associated with stress placement in English words (e.g., -age with trochaic stress; Arciuli & Cupples, 2006); research suggests that adults are sensitive to these stress cues. The present study investigates whether this sensitivity differs according to whether a cue is orthographic or morphological in nature, and examines links to reading ability. Method: We will recruit 100 adults with a range of reading backgrounds. Participants will complete a lexical decision task; its items include disyllabic words grouped into triplets that end in the same letter pattern. These endings represent either 1)an orthographic cue, consistent with the word's stress pattern (COURage), 2)an orthographic cue, inconsistent with stress pattern (corSAGE), or 3)a morphological cue, consistent with stress pattern (SHORTage). Participants will also complete measures of word reading (Woodcock, 1998) and morphological awareness. Results: We will compare reaction times and error rates for 1)consistent and inconsistent orthographic cues, and for 2)consistent morphological and orthographic cues. We expect to see effects primarily in error rates. Specifically, we expect fewer errors when an orthographic cue is consistent than when inconsistent, fewer errors for morphological cues than orthographic cues, and that the magnitude of the latter effect will be related to morphological awareness. We also expect better readers to be more sensitive to stress cues in all conditions. Conclusions: This study clarifies the roles of orthographic and morphological stress cues when processing multisyllabic words, and for the first time, links sensitivity to these cues to a wide range of reading abilities.
Deborah Speece (VCU School of Education)Rebecca Silverman, Kristen Ritchey, Lisa Pericola Case - Prediction and Stability of Reading Status Across the Elementary School Years
Purpose Our purpose is to provide a longitudinal analysis of predictors of reading status (at risk, not at risk) and stability of status in two cohorts of first and fourth children who were followed annually for three years. Method An unselected sample of 257 first grade and 230 fourth grade children participated. In fall of the first year, all children were administered reading and language measures; teacher ratings also were collected. Each spring children received a battery of reading and related assessments. Reading status was defined by factor scores derived from the spring batteries. Logistic regression and probability analyses will be used to address the prediction and stability questions. Results In first grade the most efficient screen included two word fluency measures and teachers' ratings of the number of reading problems. This model explained 46% of the variance in status and produced an AUC index of .96. In fourth grade the most efficient screen included reading comprehension, word reading fluency, and teacher ratings of reading problems, explaining 46% of the variance with an AUC value of .90. In neither study was growth an important contributor, contrary to our expectations. The longitudinal and stability analyses will be conducted next semester Discussion The contribution of teacher ratings is a novel finding in the screening literature so it will be of interest to see if it holds up longitudinally. Additionally, determining the probability of reading problems across years may clarify disparate findings in the literature in which probabilities range from .88 to .50.
Trina Spencer (The Ohio State University) - Assessment of Story Comprehension (ASC): A preliminary investigation of reliability and validity
Purpose: The purpose of this study was to examine preliminary evidence of reliability and validity of the Assessment of Story Comprehension (ASC) for preschoolers. These data are used to develop and refine the instrument. Method: The ASC was designed to allow repeated measurement of young children's comprehension of stories for the purpose of monitoring progress. Administration of the ASC involves children listening to a novel short story and answering factual and inferential questions about the story. In addition, children are asked to infer the meaning of a novel vocabulary word embedded in each story. Development of the ASC followed a logical and systematic content analysis process. After the initial development, a correlational research design was employed with 36 children attending public preschool programs. Each participant was administered three of the nine forms of the ASC as well as three comparison measures to investigate convergent and divergent evidence of construct validity. Correlational coefficients were computed for parallel form reliability, inter-rater agreement estimates were calculated between two independent scorers on 30% of the ASCs, and validity coefficients were computed between the ASC and comparison measures (i.e., PPVT, CELF, and the PLAI). Results: Moderate to strong correlations (.45-.67) were found for parallel form reliability. Scoring agreement between raters was 98% (range 90-100%). We found moderate to strong correlations with comparison measures, PLAI (.68), CELF (.42) and the PPVT (.51). Conclusions: The preliminary evidence of reliability and validity indicate that the ASC is a promising assessment instrument and further development and validation are warranted.
Elizabeth Spencer (The Ohio State University); Howard Goldstein; Sean Noe - Investigations of Tier 2 embedded storybook interventions for preschool children
The purpose of this study was to examine the effects of embedded storybook interventions to teach language and early literacy skills to preschool children identified as at risk of reading difficulties. Method: Preschool children (n = 18) participated in embedded storybook interventions designed to teach phonological awareness and alphabet knowledge skills or vocabulary and comprehension skills. Intervention was administered in 'listening centers.' Children listened repeatedly to prerecorded scripts under headphones in small groups at least three times per week. Scripts and accompanying story books were designed for the study and included instruction and practice on language and literacy tasks. All interventions were embedded in the story. The text of the story was interrupted by the narrator and children were invited to play a 'word game.' Each intervention included a teaching activity that provided children with multiple opportunities to respond to the targeted skill. Results: Children demonstrated an increase in participation and correct responses during the storybook interventions. Single subject designs show that children learned many of the targeted skills. Conclusions: Results provide evidence of learning of language and early literacy skills associated with implementation of embedded storybook interventions. The listening centers used in the current study have the potential to be used in everyday classroom settings. The centers require no special equipment and children can participate independently with minimal adult monitoring.
carrie stacey (university of north carolina at greensboro)Alan Kamhi - The relationship between text messaging, literacy skills and conventional spelling in students with and without reading disabilities
Purpose: To compare the impact of text messaging on conventional English spelling and reading skills in typically developing (TD) students and students with reading disabilities (RD). Method: Participants were 20 TD students and 20 students with RD (age range 14-19). The spelling portion from Wide Range of Achievement Test-3 was used to assess conventional spelling abilities. Three subtests were used from the Woodcock Reading Mastery Test to assess literacy skills. In addition, ten scenarios were used to obtain samples of textisms. Elicited text messages were scored for types of textisms used and the ratio of textisms to total words. All participants answered a questionnaire about their textism history and current practices. Results: Preliminary results indicated that students with RD had significantly fewer textisms than TD students. TD students exhibited a full range of textisms and many instances of creative spellings. For example, one student adds a "q" to words with a "g" as a signature letter (e.g., "I got" is "igqot." Other examples include "ne" for "any", "jhust" for "just" and "trin" for "trying". Students with RD were more likely to use conventional spellings and simpler forms of textisms than TD students. Conclusions: Consistent with previous studies, textisms do not indicate that students are poor spellers. Quite the contrary, the best spellers, TD students, exhibited more textisms and more creative spellings than students with RD who were poorer spellers. Educators might consider using textisms to promote conventional spelling abilities.
Eva Staels (Eva Staels);Wim Van den Broeck - Orthographic learning and the role of text-to-speech software in Dutch disabled readers.
Purpose The objective of our research was to explore whether orthographic learning occurs among poor Dutch readers while reading in silence. In addition, we explored for the first time the eventual positive effect of text-to-speech software on the ability of poor readers to acquire orthographic knowledge. Both research goals were investigated by replicating Share's (1999) self-teaching paradigm. According to the self-teaching hypothesis, phonological recoding of printed words into their spoken counterparts acts as a self-teaching mechanism enabling readers to acquire orthographic representations independently. Method The participants, 65 fourth- and fifth-grade disabled readers (mean age 9 years and 11 months), were asked to read eight stories containing embedded homophonic pseudoword targets (e.g., Blot/Blod) with or without the support of text-to-speech software. The degree of orthographic learning was assessed 3 or 7 days later. Results and discussion The results supported the self-teaching hypothesis by demonstrating that target spellings were correctly identified more often, named more quickly, and spelled more accurately than alternate homophone spellings. However, no remedial effect of text-to-speech software was demonstrated. In contrast with Ziegler and Goswami's (2006) psycholinguistic grain size theory, claiming that readers of transparent orthographies as Dutch remain reliant on serial phonological recoding, our results support the hypothesis of several researchers that all readers, even poor readers of transparent orthographies, develop connections between orthography and phonology at the whole word level (de Jong, 2006; Frost, 2006; Paulesu, 2006; Wimmer, 2006).
Rhona Stainthorp (Institute of Education, University of Reading)Diana Hughes - Stories of success: precocious readers as young adults.
Purpose: This is a report of a long term follow-up study designed to investigate the educational progress, attitudes to reading and reading behaviours of a group of young adults who were precocious readers before they started school (Stainthorp & Hughes, 1995, 1997, 2000, 2004). Studies of precocious readers are limited in number and there have been no published reports of the progress made by such people into adulthood. Method: The participants in a study of precocious reading ability were contacted to take part in a long term follow-up study. Eight people responded and these form the basis of a series of case studies. The participants are now 21 years old and in higher education. Tests of reading, spelling phonological awareness, and writing were administered. Additionally they were interviewed about educational achievements; attitudes to reading and reading habits; memories of the literacy teaching they received; and memories of taking part in the original studies. Results: This group of young adults have positive memories of taking part in the original project; they are a group of high achieving people who continue to value reading and writing. Their adult performance is however typical of their peers. Their phonological awareness, which was a significant characteristic of their early success is now less unusual. Conclusions: The small group who took part in the follow-up are illustrative but not necessarily typical. Nevertheless, their experiences provide a unique insight into the effects of success in achieving high levels of literacy skill at an unexpectedly early age.
Laura Steacy (Vanderbilt University); Jennifer K. Gilbert; Donald L. Compton; Eunsoo Cho - A look below the surface: An examination of word reading and decoding preferences in emerging readers
Purpose Prior to formal reading instruction, some students demonstrate preference for real word recognition (WR) over nonword decoding (NWD) while others display the opposite pattern. The purpose of this study is to: (a) explore the stability of these preferences from Grade 1 to Grade 2, and (b) explore the reading outcomes of students who demonstrate a preference (WR_P and NWD_P) and those who do not (NP). Method We used a longitudinal sample of 526 children representing the entire distribution of reading skills in Grade 1, with oversampling for poor readers. Students were given measures of WR and NWD at the beginning of Grade 1, end of Grade 1, and end of Grade 2. At each time point, two scores were calculated, representing preference for WR and NWD (a 10 point standard score difference was used to define preference groups). Two hidden Markov chain models were estimated to assess stability of preferences across time. After students were classified into 8 groups based on transition status, reading outcomes at the end of Grade 2 were compared. Results At the end of Grade 2, 8 of the 88 students who showed initial WR_P and 8 of the 24 students who demonstrated initial NWD_P displayed NP. Significant group differences were detected on measures of WR, NWD, and comprehension at the end of Grade 2. Conclusion Preliminary results suggest that WR_P and NWD_P are malleable by instruction in that the majority of students who initially displayed a preference demonstrate NP at the end of Grade 2. Among the 8 transition groups, students with stable profiles had the lowest second-grade WR and comprehension scores.
Dana Strait (Northwestern University): Jane Hornickel; Nina Kraus - Neural enhancement of regularities in ongoing speech underlies reading and music abilities: implications for reading remediation
PURPOSE Neural insensitivity to regularities in ongoing speech has been recently linked to child reading disability1. We do not know how auditory expertise, e.g., musical skill, relates to the neural processing of speech regularities, despite the fact that music and reading abilities are known to co-vary. We aimed to define relationships between musical skill and literacy-related aspects of neural function. We hypothesized that, in combination with auditory cognitive abilities, neural sensitivity to speech regularities provides a common biological mechanism underlying the development of music and reading abilities. METHOD We assessed auditory cognitive abilities (working memory and attention), music aptitude, reading abilities, and auditory brainstem enhancement of speech regularities in 42 school-aged children with a wide range of reading ability. Neural sensitivity to speech regularities was assessed by recording brainstem responses to the same speech sound presented in predictable and variable contexts. RESULTS Through correlation analyses and structural equation modeling, we reveal that neural enhancement to predictably-occurring speech underlies music and reading abilities in children. Relationships between music and speech processing are specifically driven by performance on musical rhythm tasks, underscoring the importance of rhythmic regularity for both language and music functions. CONCLUSIONS These data indicate common brain mechanisms underlying reading and music abilities that relate to how the nervous system enhances regularities in auditory input. Definition of common cognitive and neural mechanisms for music and reading skills supports the usefulness of music for promoting child literacy, with the potential to improve the efficacy of reading remediation. 1. Chandrasekaran B, Hornickel J, Skoe E, Nicol T & Kraus N (2009) Context-dependent encoding in the human auditory brainstem relates to hearing speech in noise: implications for developmental dyslexia. Neuron 64(3):311-9.
James Street (); Nicole Davis; Sabrina Benedict; Heather Harris; Laurie E. Cutting - Reader-text interactions and oral reading fluency
Purpose: Identifying characteristics of children's text is important for better understanding reading success/failure. One line of investigation is to understand more objectively the differences between texts on various quantifiable dimensions. Establishing methods to determine equality of texts, as well as how to manipulate them on certain dimensions, could ultimately contribute to understanding more about reader-text interactions and reading comprehension assessment. Methods: We created nine baseline passages that were comparable across 23 Coh-Metrix measures. From these passages, we produced four manipulated passages. Two passages were more difficult to decode, but word frequency was kept constant (DEC). Two passages had lower word frequency, but decoding difficulty was kept constant (VOC). Twenty adults read aloud five baseline passages, as well as the DEC and VOC manipulated passages. Data were analyzed to determine if differential Words Correct Per Minute (WCPM) existed between the five baseline and manipulated (DEC and VOC ) passages. Results: Results of a repeated measures ANOVA revealed that baseline passages were not significantly different from each other in WCPM; however, the DEC and VOC passages showed slower WCPM than all baseline passages. In particular, DEC passages showed slower WCPM than all five baseline passages (all p<.05). VOC showed a similar trend; however, traditional levels of statistical significance (p<.05) were reached only for two of the baseline passages. Conclusions: Findings suggest that passages can be manipulated on various dimensions (decoding, vocabulary), which may have implications for understanding both text structure and how to design optimal assessments to target specific reading-related skills.
Philippa Struthers (University of Otago); Elizabeth Schaughency; Samuel Clark - Home school communication, parent support strategies and reading development in the first year of school.
Purpose: How do parent-teacher communications about children's home reading, and parent strategies during home reading relate with reading development in the first school year? Method: Participants were 96 children (age 5-6) and primary caregivers from three New Zealand primary schools. Parent and teacher responses in home-reading logs (HRL) over ten school weeks were coded . Parents completed questionnaires regarding home reading practices and strategies used during child-parent reading. Indicators of child's reading development included researcher-administered measures from DIBELS and AIMsWeb and school-used measures including reading book level (BL), and teacher identification of "at-risk" children. Results: Preliminary analysis conducted on a subsample of 49 participants indicated all parents endorsed child home-reading. Child-parent reading frequency was not correlated with children's reading proficiency. HRL responses of "at-risk children" included more comments on reading development and problems. Parent reading support strategy use correlated negatively and significantly with children's BL (helping story comprehension, -.39, story content, -.47, drawing attention to words, -.44, and supplying the word, -.44). Child independent reading correlated positively and significantly with researcher-administered measures (.30 - .46), and school BL (.32). Conclusions: Preliminary analysis suggests greater communication between home and school and greater parental support during reading for children who are struggling with literacy development. Greater reading independence and less parental support during reading are shown for children who are achieving literacy milestones. Findings suggest parents and teachers may be scaffolding support in response to children's reading needs, consistent with emerging international research (Evans, Mansell & Shaw, 2006; Silinskas et al., 2010).
H. Lee Swanson (University of California-Riverside); Olga Jerman - The predictive validity of dynamic assessment, working memory, and reading comprehension in children with and without reading disabilities - a three year longitudinal study
Purpose: This longitudinal study determined (a) whether dynamic assessment (DA) of working memory (WM) was related to growth on measures of reading and vocabulary and (b) whether subgroups of children with reading disabilities (RD) (children with RD-only, children with both reading and arithmetic deficits, and low verbal IQ readers) and skilled readers varied in performance as a function of dynamic testing. Method: A battery of memory and reading measures was administered to 80 children (Mean age=11.0 yrs) across three testing waves spaced one year apart. WM tasks were presented under initial, gain, and maintenance testing conditions. Results: The results indicated that children with RD and skilled readers both benefited from DA procedures, but DA procedures did not statistically differentiate between the subgroups of children with RD. Additional results from a growth curve analysis were: (1) maintenance testing conditions were significant moderators of comprehension and vocabulary growth, whereas probe scores and gain testing conditions were significant moderators of nonword fluency growth, (2) WM performance was statistically comparable among subgroups of children with RD, but inferior to skilled readers across all testing situations, and (3) significant ability group (skilled vs. RD) x dynamic testing interactions emerged when predicting comprehension and vocabulary, suggesting that skilled reader's performance was more likely to be moderated by dynamic testing conditions than children with RD. Conclusion: Overall, the results suggested that DA added important variance in predicting later reading performance.
Marcin Szwed (Institute of Psychophysiology, Department of Psychology, Jagiellonian University); E. Qiao; L. Cohen; S. Dehaene - Expertise for fast and parallel reading in early and intermediate visual cortex: a cross-cultural fMRI study with French and Chinese readers.
Purpose: Which levels of the visual system are modified through acquisition of literacy? The visual system is hierarchical: simple feature analysis is done in early areas which then converge on high-level areas responding to complex stimuli e.g. faces. Several studies have shown that reading skills are linked to changes in high-level visual system (Visual Word Form Area). Here we investigated whether reading also relies on perceptual learning within earlier visual areas (perceptual learning is a form of implicit learning that involves improvement in sensory discrimination by repeated exposure to stimuli). Methods: in an fMRI study with French and Chinese adults, we matched French and Chinese written words and line-drawings of objects for their low-level visual features, and designed control images by scrambling procedures that keep local features intact. Results: Greater responses to words relative to controls, but not to objects relative to controls were found not only in high level visual areas, but also in early and intermediate visual areas V1-V8. Moreover, these responses to words were found only in French reading French (peak in areas V1/V2), and Chinese reading Chinese (peak in areas V4/V8). No enhanced V1-V8 responses were found for Chinese stimuli in French subjects or for French stimuli in Chinese subjects, even though Chinese subjects could partially recognize French words (67% accuracy). Conclusions: Early and intermediate visual activations reflect fast and parallel detection of character shapes, the hallmark of mature, expert reading. The impact of literacy on the brain is thus more widespread than previously thought.
Nadia Taibah (King Abdulaziz University, Jeddah, Saudi Arabia) - The effects of an intervention program on reading and pre-reading skills among at-risk Arabic speaking kindergartners
Purpose: This pilot study was to assess the potential for theoretically based intervention program in Arabic for reading named "I can read". The main objective was to enhance number of combination skills among children with concurrent risk for reading disability. The program aimed to promote five different areas in reading skills: phonological awareness (PA), phonology, morphology, vocabulary, and fluency skills. Method: Kindergartners were assigned into two groups, experimental and control group. Each group was divided into three levels, at risk, moderate, and advance abilities. The children at the experimental group received 50 sessions in small group over 14 weeks; each session lasted 45 minutes. The control group received government curriculum reading program based on phonics instruction. The current program was designed on small group instruction and each child within each group received activities based on his abilities. All children were pre-tested and post-tested on PA, letter naming and sight word reading, morphological skills, receptive and expressive vocabulary, and letter naming and word reading fluency. Results: Using quasi-experimental design controlling for SES, language exposure, and IQ skills, the acquisition effects were assessed. The results indicated that children in experimental group achieved higher scores than control group in PA, word reading, morphological, and vocabulary skills within all levels with higher effect on children with at risk. The effect on fluency skills was the least with at risk group compared with control group. Conclusions: Results provide the basis for additional work with larger samples with some modifications to enhance other reading skills such as fluency.
Novell Tani (Florida State University/FCRR)Carol Connor - The association of teacher perceptions with second graders' behavior and academic achievement: examining race and gender differences
Purpose: The achievement gap, particularly for African American boys, is a serious but perplexing educational and policy issue. The present study examined the effect of teacher's perceptions on student's behavior and academic achievement in second grade as well as potential racial or gender differences in these perceptions. . Method: We examined second graders' (n=560 students, 40 classrooms, 8 schools) self-regulation skills directly using the Head-Toes-Knees-Shoulders Task (HTKS) and indirectly using teacher report. We also assessed reading and vocabulary skills in the fall and spring and conducted classroom observations. Schools ranged in numbers of students receiving free and reduced lunch (FARL) from 4% to 96% with an average of 44%. Results: Preliminary results show teachers' perceptions of students' social and reading skills over and above their skills at the beginning of the school year. Moreover, there was a significant difference between African American boys and girls with regards to academic achievement on the WJ-Picture Vocabulary, Letter/Word Identification and Passage Comprehension assessments. Additionally, teachers at higher poverty schools were more likely to perceive greater behavior problems than the direct measure of self-regulation would tend to imply than did teachers at more affluent schools, which included fewer African American students. Conclusions: Results suggest that, in general, teachers' perceive African American boys to have more severe behavior problems than the direct measure of self-regulation might suggest would be the case and that; in general, teacher perceptions predict achievement even taking into account other factors. Implications for policy and practice will be discussed.
M. Luisa Tarczynski-Bowles (University of Coventry);Clare Wood; Andrew Holliman; Janet Vousden; Gareth Williams - A task validation study investigating school aged children's speech rhythm sensitivity and its relation to reading ability
Purpose: The central aim of this study is to determine whether a new task, namely the Lexical Judgement Task (LJT), can be used to assess children's speech rhythm sensitivity and whether performance on it is linked to literacy in the ways predicted by current models. Method: 44 children, aged between 5.2 and 9.4 years have been assessed. Children were assessed individually on several measures, e.g. for their phonological awareness (using Spoonerism subtest from PhAB), their phonological memory (Rapid Naming subtest from CTOPP), their reading ability (WRAT4) and general ability using the WASI vocabulary and matrix reasoning subtests. To assess their speech rhythm sensitivity the DEEdee task and the new Lexical Judgement Task were administered. Based on standardised reading scores, the children where allocated to one of two groups: a "Reading Delayed" group (reading scores of 85 or below) or a "Typical Readers" group (scores above 86). Results: Data collection is ongoing but the results are expected to show that children in the "typical reader" group will show better prosodic skills than the delayed group and the LJT will be able to predict significant variance in reading ability after general ability and phonological awareness have been accounted for. Conclusion: The results will aid the understanding of lexical stress sensitivity in relation to reading ability and provide insight into the way these are linked.
Agnes Tellings (Radboud University Nijmegen)Lex Bouts - Dutch third to sixth grade children's attribution of meaning to written pseudowords with and without affixes
Purpose: to investigate how elementary school readers of different ages attribute meaning to pseudowords, with and without affixes, with and without pseudomorphemes. Method: In Study 1, 176 grade 3-6 Dutch children did a written, individual lexical decision task and after completion were questioned about the pseudowords they had not crossed out (and apparently had considered to be words). In Study 2, 50 grade 3 Dutch children did an oral plus written, individual lexical decision task and after each item were questioned about the pseudowords they had conceived to be words. Both statistical (i.e., with respect to the influence of frequency, neigborhood size, and affix status)and qualitative analyses were performed but the qualitative analyses are the main subject of the presentation. Results: Two different forms of meaning attribution in general were found(i.e., concept-label errors and new concept-label constructions), and also two different forms of meaning attribution to the affix of affixed pseudowords (i.e., correct or nearly correct meaning attribution versus interpretation of the affix as an intensifier). Age was a factor in the division of the type of meaning attributions in both studies. Conclusions: Children develop not only quantitatively but also qualitatively in the way they attribute meaning to unknown written letter strings. A tentative conclusion is, that the development of affix knowledge goes through different stages, from the general to the specific, that is, from not recognizing affixes as affixes, to conceiving affixes as intensifiers, to learning the different, correct meanings of affixes.
Shane Templeton (University of Nevada, Reno);David Smith; Bob Ives - The nature of students' tacit and explicit knowledge of morphology in the context of a developmental model of word knowledge
Purpose - Few studies have addressed students' metalinguistic reflection on derivational morphological knowledge. The present investigation attempted to explore this aspect of 4th, 5th, and 6th grade students' morphological knowledge, in addition to their tacit understanding (Carlisle, 2000). The nature of these reflections was examined in relation to orthographic knowledge, vocabulary knowledge, and reading. Method - 54 students in grades 4, 5, and 6 were administered the Peabody Picture Vocabulary Test (2007); the state's criterion-referenced reading test; the Upper-Level Qualitative Spelling Inventory (Bear, Invernizzi, Templeton, & Johnston, 2008); the Test of Morphological Structure (Carlisle, 2000); and the Derivational Relatedness Interview, developed by the researchers to assess each student's ability to talk explicitly about and describe the meaning relationships among derivationally-related words. The interview was audiotaped. Although the Test of Morphological Structure is often treated in the literature as a measure of explicit morphological awareness, it does not require subjects to talk about the morphological relationships that it is exploring; the Derivational Relatedness Interview was designed to do so. Results - Intercorrelations among reading, vocabulary, orthographic knowledge, and morphology were all significant. The nature of explicit reflection on morphological relationships increased across grade levels. Conclusions - The ability to reflect metalinguistically on derivational morphological relationships increases with level of literacy development. This trend is being explored to the degree that it reflects a developmental model of literacy. Because of the significant intercorrelations among literacy measures, implications for efficient initial classroom assessment of students' vocabulary and morphological knowledge through the use of a well-constructed qualitative spelling inventory are discussed.
Danielle Thompson (Ascent Educational Consulting, LLC); Stuart Bernstein; Rachel Anderberg; Caitlin Orman; Cyrille Magne - Contributions of prosodic sensitivity and morphological awareness to word level reading: a perceptual task development study
Purpose: The primary goals were to create two auditory lexical decision tasks appropriate for imaging studies. Existing tasks of prosodic sensitivity and morphological awareness were adapted and studies examined unique word recognition variance accounted for by the tasks. Method: A correlational design was employed. There were 79 adult participants. Hierarchical regression was conducted with word reading as the dependent variable (Letter Word ID, WWJ-III). The independent variables were vocabulary (WWJ-III), word attack (WWJ-III), prosodic sensitivity, and morphological awareness. All tasks were examiner delivered except the lexical decision tasks, which were computer administered. Results: Prosodic sensitivity for words accounted for a significant 9.6% of the variance in word level reading. The contribution of prosodic sensitivity to stress shifts in real words was unique when controlling for vocabulary, word attack skills, and morphological awareness. Morphological awareness accounted for 4.7% of the variance in word level reading when controlling for other variables. The order of entry in the block designs made little change in the outcome. Conclusions: Outcomes are similar to previous investigations using non-computer based measures, suggesting the tasks may be feasible measures. Additionally, a preliminary child study (n = 14) supported the data with strong correlations for both tasks predicting word level outcomes. This is important for use in educational settings for early identification of missing implicit word knowledge as well as potential for use in imaging studies. Additional studies are in progress for adults (n = 50) and children (n = 130) and will be concluded for presentation.
Jenny Thomson (University of Sheffield); Carlo Cerruti; S. Lynneth Solis; Gottfried Schlaug - Translational & theoretical possibilities of transcranial direct current stimulation in reading research
Purpose In this study we examine the potential of using transcranial direct current stimulation (tDCS) to augment our understanding of the neural bases of reading and phonological awareness. tDCS modulates postsynaptic resting membrane potential - thereby altering the likelihood that another stimulus will cause the neuron to fire. The sustained effects of tDCS thus resemble the phenomenon of long-term potentiation. Here we asked how tDCS applied to left and right hemisphere temporo-parietal junction regions impact phonological processing and reading ability in the short-term. Method Thirty four typically-reading participants completed a series of phonological and literacy tasks both before and immediately after stimulation over two days. One day targeted the left and the other day the right hemisphere, counterbalanced for order effects across subjects and separated by at least one week. Half participants underwent excitatory (anodal) stimulation and half underwent inhibitory (cathodal) stimulation. Results Preliminary analysis of the data for anodal stimulation of the left hemisphere suggests an increased speed of phonological processing following stimulation. Conclusion The results of this study provide important preliminary data on the utility of tDCS as an investigative tool in reading research and a potentially remedial tool in dyslexia.
Sana Tibi (United Arab Emirates University, United Arab Emirates); Linda Siegel - Reading errors in Arabic: Implications for understanding the development of reading skills
Purpose: The aim of the present study was to measure the performance of elementary school students on oral reading fluency (speed and accuracy) in Arabic, and to determine the type of reading errors. Arabic has a clear divergence between the spoken forms of the language and the formal or written form, referred to as diglossia. For example, the phonology of formal written Arabic includes some sounds that may not be part of the spoken form of the language. Although formal written Arabic has a high rate of correspondence between its phonemes and its graphemes, reading in formal Arabic remains as a relatively difficult skill to develop. When children begin to learn to read in the standard form of the language, they are exposed to new phonemes. Method: The sample consisted of 90 male students randomly selected from elementary grades 1, 2 and 3 in a public school in the United Arab Emirates. All students were required to read a previously unseen vowelized Arabic reading passage with a time constraint of 60 seconds. Results: The results showed that the number of words read per minute increased with grade level. Analysis of the type of errors showed that errors included both phonological and morphological errors. The phonological errors appeared to be due, at least in part, to the diglossic nature of the language. The morphological errors were related to the morphological complexity of the language. Conclusions: This study contributes to an understanding of the development of reading accuracy and fluency in Arabic.
Elizabeth Tighe (Florida State University)Katherine Binder - Learning how to jittled/jittling/jittle words: an investigation of morphological accuracy and response times in low literate adults
Purpose: The purpose of this study was to assess accuracy and measure reaction times on morphologically complex words and control words in an oral reading passage and a word recognition task for low literate adults. Method: Fifty-eight adults enrolled in Adult Basic Education programs in Western Massachusetts were administered tasks measuring phonological awareness, morphological awareness, and general reading abilities. Additionally, participants were given an oral reading passage with inflected (-s, -ed, -ing) word endings and derived (phonological shift, stable, low and high frequency) words that were matched based on frequency with control words. A naming task was also administered with words matched to those in the passage. Accuracy and reaction times were measured for morphologically complex vs. control words in both the passage and the naming task. Phonological awareness and morphological awareness as predictors of reading comprehension were assessed utilizing the battery of literacy assessments. Results: Results indicated that participants were significantly more accurate with identifying control words as compared with morphologically complex words for both the oral reading passage and the naming task. Context was found to be an important factor - recognition was faster and more accurate when the words were found in context. Naming times also reflected that control words were read significantly faster than morphologically complex words. Finally, morphological awareness and phonological awareness were found to be significant predictors of reading comprehension. Conclusion: The results indicate that low literacy adults are in fact sensitive to morphological complexity and to contextual cues when identifying words.
Carolina Toledo Piza (Carolina T. Piza); Elizeu Coutinho Macedo; Thais Barbosa; Camila Cruz Rodrigues; Orlando Francisco A. Bueno - Reading and writing skills in children with dyslexia and good geaders
Purpose: Underpinning cognitive processes in reading and writing skills can help distinguish different profiles of disabilities. In this study the performance of dyslexic children was compared to two individually matched control groups in a Brazilian Reading and Writing Battery. Method: 28 dyslexic children (both genders), mean age 9,82 (±1,44) years, studying in public and private schools, were matched to: 1) Age Control Group (AC): composed of 26 good readers, mean age of 9,77 (±1,44) years, matched by age, sex, years of schooling and type of school. 2) Reading Control Group (RC): composed of 28 younger controls, with a mean age of 7,82 (±1,06) years, matched by sex, type of school and reading level. All groups were tested on tasks of the "BALE" battery, that evaluate sentence reading comprehension, oral sentence comprehension, spelling and reading of isolated words and pseudowords. Results indicated dyslexics obtained lower total scores and higher execution time, when compared to AC. No differences in total scores of reading tasks or oral comprehension tasks were observed in the RC and Dyslexia group. Nevertheless, dyslexics presented a slower reading speed, with a higher execution time and a lower score in spelling tasks, even when compared to younger controls. Detailed analysis of types of errors on reading tasks, demonstrated that dyslexics obtained lower scores in pseudowords reading, when compared to both groups. Conclusions: Findings suggest that overall scores of dyslexics were similar to those of younger readers. However, specific deficits in phonological and visual decoding, demonstrates both groups have different underpinning strategies.
Shelley Xiuli Tong (MARCS Auditory Laboratories, University of Western Sydney) - Toward a reconceptualization of morphological awareness in Chinese: Latent components of Chinese morphological awareness
Purpose: We examined latent components underlying Chinese morphological awareness development in Hong Kong Cantonese-speaking children. Method: Measures of fine-grained aspects of Chinese morphological awareness included homographic discrimination, morphological compounding, morphosyntactic completion tapping morphological awareness at the character, word and sentence level, respectively, and orthosemantic awareness, semantic category choice assessing morphological awareness at the subcharacter level. These measures were administrated to 199 kindergartners, 172 second graders and 165 fifth graders who are Cantonese-speaking children in Hong Kong. Results: Confirmatory factor analyses comparing alternative models of latent components of Chinese morphological awareness. The best-fitting model for the kindergartners is the two-factor model representing a division of morphological awareness in oral and written Chinese. In contrast, the best-fitting model for second graders is a two-factor model including morphemic and submorphemic awareness. However, the simplest one-factor model representing a unified morphological awareness showed the bet goodness-of fit index for fifth graders. Conclusion: There exists developmental difference in the components underlying Chinese morphological awareness. Across development, children become more able to process morphological awareness as a unidimension construct. This underscores the uniqueness of Chinese morphology and provides developmental evidence of change of morphological awareness constructs across age.
Xiuhong Tong (Psychology Department, The Chinese University of Hong Kong)Catherine McBride-Chang; Kevin K.H. Chung - Morphological structure processing in Chinese children with dyslexia
Purpose: The purpose of the present study was to investigate how Chinese children with dyslexia process Chinese compound word structure (referring here to position of the morpheme in a Chinese compound word). Method: Thirteen Hong Kong Chinese children from each group (dyslexic and age-matched control groups) were tested using an online lexical decision paradigm. Children were in grades 3 and 4. There were three conditions: the real condition, consisting of real words composed of two characters跑步 (running); the reversed condition, with the order of the two characters in the word reversed 洋海 , which comes from the real word 海洋 (ocean); and the random condition, with two real characters randomly combined to form arbitrary nonwords at the word level 情良. Results: Analyses of the online data showed that a main group effect was found in both reaction time and accuracy. Compared to the control group, the dyslexic group showed longer reaction times and lower accuracies in processing the reversed condition. Furthermore, for both groups, children performed worst in this condition, demonstrating that, even among typically developing children, it is difficult to judge whether or not a word is real when the two morphemes comprising it are reversed. Conclusion: Results suggest that children with dyslexia have more difficulty in processing morphological structure information; this may be because of a relatively weak morphological representation in the lexicon.
Wim Tops (Thomas More University College); Maaike Callens; Marc Brysbaert - Cognitive, metacognitive and personality profiles of postsecondary students with dyslexia
Purpose The goal of this study was the development of a theoretical framework for the guidance of students with dyslexia in higher education. There is indeed a lack of theoretical and empirical evidence in this domain, especially in non-English speaking countries. In this study we wanted to see whether the results for English speaking adults with dyslexia can be generalized to other alphabetical languages. Method 100 postsecondary students with dyslexia participated in this study, as well as 100 students without learning disabilities. We first focused on the cognitive functioning (IQ, memory, ...). We also investigated a wide range of reading and writing skills. Next, we administered personality and study strategies inventories to get better insights in the metacognitive profile of students with dyslexia. Finally, we followed both groups during their first year of undergraduate study. We compared the results of the personality and study strategies inventories of students with dyslexia who passed their first year with those who did not. Results We found that dyslexic students continue to have serious reading and writing problems, without significant differences in IQ and memory. Dyslexic students particularly fail in processing speed, not in accuracy. Students with dyslexia who passed their first year also had more metacognitive strategies than dyslexic students who didn't pass. Moreover, they had a specific personality profile, which was different from peer students without learning disabilities who passed. Conclusions Despite the differences in language and educational context, our findings are remarkably similar to those in English. This is good news, because it indicates that the profile is applicable to most alphabetical languages.
Minna Torppa (Jyvaskyla University, Department of Teacher Education);George Georhiou; Lerkkanen, Marja-Kristiina; Niemi, Pekka; Poikkeus, Anna-Maija; Siekkinen, Martti; Nurmi, Jari-Erik - Examining the simple view of reading (SVR) in an consistent orthography
Purpose: To present the examination of the Simple View of Reading, SVR (Gough & Tunmer, 1986) in a transparent orthography, Finnish. In transparent orthographies decoding accuracy lacks variance and fluency measures are applied. We examined the role of reading fluency in the SVR function and additionally the effects of classroom variation and task avoidance in a longitudinal dataset. We ask: 1) Does SVR model apply in Finnish? What are the associations of reading comprehension, listening comprehension and reading fluency ?, 2) Does classroom variation and task avoidance have an additional effect on reading comprehension, and 3) Does the SVR model apply for all children or are there subgroups with differing developmental paths and predictors of reading comprehension? Method: The dataset comprised of 2261 children whose reading comprehension was measured in grade 1, 2, and 3. Listening comprehension, reading fluency, and task avoidance were assessed in grades 1 and 2. Multilevel path modeling and multilevel mixture modeling (classroom as the between level) are applied. Results: On grades 1 and 2, both reading fluency and listening comprehension predicted reading comprehension. On grade 1 decoding fluency predicted reading comprehension more strongly than listening comprehension whereas on grade 2 the opposite pattern occurred. Grade 3 data and the effects of classroom variation and task avoidance are currently in analysis. Conclusions: Findings will add to the current knowledge of SVR in transparent languages, on the link between reading comprehension and reading fluency, and on the additional predictors of reading comprehension above SVR.
Daniela Traficante (Department of Psychology); Marco Marelli; Cristina Burani; Claudio Luzzatti - The role of frequency and grammatical category of base words in reading derived nouns: a study with primary-school skilled and poor readers
Recent reading models focusing on the role of morphemic units have addressed the issue of how orthographic representations become morphemically structured in the first steps of reading acquisition. Within this framework, the study of the development of reading abilities can be relevant. The present study investigated the role of base frequency in inducing access to morphemic structure information both in young skilled and poor readers, controlling for the grammatical category of the base word, in order to get indications about the lexical information exploited when reading derived words. Method. Twenty 4th and 5th grade poor readers and 40 skilled readers, matched by age, gender and cognitive development, participated in the study. Fifty nouns derived from noun bases and 50 nouns derived from verb bases were employed as experimental stimuli. Each item set was organized in two lists, matched on word frequency and contrasted on base frequency. Stimuli were presented randomly in a reading aloud task, along with four lists of control simple words matched to experimental stimuli for all relevant features. Results. When base frequency was high, nouns derived from noun bases were read by all children more correctly than simple nouns. As for nouns derived from verb bases, though, morphemic parsing did not exert any reliable effect. These results suggest that the facilitating effect of a high frequency base is reduced when its grammatical category is different than the category of the derived word. Data will be discussed in the light of the main computational reading models.
Rebecca Treiman (Washington University)Susan E. Stothard; Margaret J. Snowling - What makes some letter sounds easier to learn than others? Evidence from children in the UK and Canada
Purpose: Learning the sounds of letters is an important foundation for literacy. We studied letter-sound knowledge in two groups of English-speaking children: children from the UK, who learn the names of letters from an early age, often using lowercase letters, and children from Canada, where stress is placed on letters' conventional names and on uppercase letters. Method: We used mixed modeling to examine factors that may be associated with lowercase letter-sound knowledge, including whether a letter's sound is at the beginning of its name (i.e., acrophonicity), the frequency of the letter in books, whether the letter is at the beginning of a child's first name, whether the letter is the same shape as its uppercase counterpart, and the confusability of the shape with other lowercase letter shapes. We analyzed data from 182 UK reception year children and 149 Canadian kindergartners. Results: Children in both groups did better on the first letter of their forename than on other letters, and worse on visually confusable letters. Canadian children were influenced by acrophonicity, suggesting that they used their knowledge of letters' names to help learn the sounds, but UK children were not. Effects of frequency in book print were larger for UK children, and effects of case similarity were smaller. Conclusions: An understanding of letter learning requires knowledge of characteristics of letters as well as characteristics of learners. Our results show how characteristics of letters combine with cultural and educational practices to make some letter sounds easier and others harder to learn.
Chiung-hsien Tsai ()Yuhtsuen Tzeng - The development of Taiwanese children's on-line causal and anaphoric inferences
Making causal connections and using referring expressions plays a fundamental role in understanding discourse by linking successive utterances to form a coherent representation. The aim of this paper is to investigate the development of inferring causal and anaphoric coherence among students in Taiwan. Thirty-one grade fourth, 30 grade fifth and 32 grade sixth students participated. The experiment was implemented in Experiment Builder. Each participant read 36 experimental short texts and 18 fillers in randomized order. Each text consists of two sentences describing two protagonists engage in an event in the first sentence and a consequence or another event in the second. An interrogative question followed to probe their anaphoric or causal resolution. We manipulated causality (strong or weak) and anaphoric resolutions (overt or zero pronoun) in texts. Zero pronoun sentences are grammatical in Mandarin Chinese. All sentences were presented in the centre of a computer screen sequentially and participants press keys to proceed. There is a developmental trend, the higher the grade the shorter of their reading time and better accuracy rates. Importantly, the interaction between causality and anaphoric resolution was significant. For texts with anaphors, participants answer probe questions more accurately for sentences with high than low causal connections and their reading times were shorter for the second sentence which is critical for inference resolution. Zero pronoun conditions benefit from causality by shortening reading time. For texts with strong causality, there is longer reading time for text with anaphors but no differences in accuracy. Overt pronoun sentences speed up reading time but did not increase accuracy. Overall, causality plays a more important role than anaphoric connections and there is even a penalty for anaphoric connection if there are strong causal relations.
Sonja Ugen (Université du Luxembourg); Martin Brunner; Monique Reichert; Antoine Fischbach; Ulrich Keller; Christophe Dierendonck; Romain Martin - Reading comprehension strategies of biliterate students in German and French
This study examined the interaction of metacognitive, motivational and linguistic factors on reading comprehension of German-French biliterate students in Luxembourg where the educational system and the linguistically heterogeneous population contribute to a complex multilingual setting. About 6000 adolescent ninth-graders (who attended 8 years of German and 6 years of French classes) participated in this study. Students completed a standardized test on reading comprehension in German and in French, a knowledge test of efficient reading strategies in German and in French as well as a background questionnaire on several key individual characteristics (native language, preferred language, socio-economic status, academic self-concept and interest in German and in French). The analyses will focus on the students' reading strategies in relation to their reading comprehension within and across each language, controlling for background language (Germanic vs. Romance), preferred language and motivational aspects. As a conclusion, we will propose a summarizing model and discuss educational implications.
Joanna Uhry (Fordham University); Joanna Uhry; Laura Raynolds - Developmental spelling at three points in the kindergarten school year for L1 English- and L1 Spanish-speaking urban children
Purpose This study of emergent and beginning readers' developmental spellings examines phonemic representations in regard to possible influences (e.g., complexity of sound, position of articulation, and home language). It revisits our earlier work but with three rather than two times of kindergarten testing, and with a spelling-word list redesigned to offer more opportunities for contrast between L1-Spanish and L1-English spellers. Method Data were collected at three points in the school year from 95 kindergarten children in a Title I urban school. Measures included word identification, letter-name knowledge, phoneme segmentation, and invented spelling. Twenty dictated spelling words were scored developmentally using a system in which close articulatory proximity of substituted phonemes received higher scores than more distantly positioned phonemes. The list included consonants more apt to be substituted by L1 Spanish-speaking children than L1 English speakers (e.g., t, s for/th/; h, ch for /sh/; g for /k/, z for /s/; d for /t/; v for /f/ due to subtle phonemic differences in the two languages. Repeated measures ANOVAs were used for comparisons of substitution patterns between groups (L1 English and L1 Spanish), and across three times of testing. Results Consistent with our earlier work, English-Spanish contrasts in the target spelling patterns were more apparent early in the kindergarten year. Close analysis of changes in spelling patterns over time are examined. Conclusions Findings have implications for teaching as well as for developing research instruments that are sensitive to the phonological knowledge of emergent readers.
Catherine Ullman (Tufts University)Calvin Gidney - An analysis of the relationship between nonstandard dialect density and phonological awareness
Purpose: More than 50 years after Brown vs. Board of Eductiton, there remains an unexplained an inexcusable reading achievement gap between African-American and white students. One proposed partial explanation for this gap is the relative opacity of English orthography for those African-Amerian children who use a nonstandard dialect. African-American English (AAE) corresponds to English orthography less closely than does Standard American English (SAE). Our investigation focussed on whether the number of AAE features in a child's speech predicts that child's word reading and/or phonological awareness. Method: Participants were 28 monolingual African-American first through fourth graders. Data collection is ongoing. We collected a natural speech sample from each child. Morphosyntactic, Phonological, and total Dialect Density Measures (DDM) were calculated from each speech sample by dividing the number of observed phonological and morphosyntactic AAE features, by the number of morphemes, in that child's speech sample. Participants took the elision, segmentation, and blending subtests of the Comprehensive Test of Phonological Processing (CTOPP). Correlation and multiple regression analyses were conducted. Results: Correlation and regression analysess indicated significant relationships between the dialect density measures and blending performance, suggesting that children who have relatively more nonstandard dialect features in their speech have lower performance on the blending task. Conclusions: African-American children who use AAE may have more difficulty than their SAE-speaking peers in acquiring phonological awareness, and in learning to read novel words, and may require extra instruction in these areas.
Akira Uno (University of Tsukuba, Japan);Noriko Haruhara; Masato Kaneko; Noriko Awaya; Junko Kozuka; Takashi Gotoh - Visual memory deficits in a Japanese boy with development dyslexia: A case study
Purpose: The objective of the study was to ascertain which cognitive deficit caused the reading/writing difficulties exhibited by a Japanese speaking primary school boy aged 11, in particular, when reading Kanji nonwords, writing Kanji words, and reading/writing Roman alphabets (ROMANJI). Method: A right-handed boy aged 11 with developmental dyslexia (which was identified when he was 7 years old) was examined together with age-matched controls in order to ascertain the causes of his dyslexia, using (1) WISC-Ⅲ and RCPM (Raven's Colour Progressive Matrices); (2) STRAW (Screening Tests of Japanese Reading and Writing for Children) as reading and writing attainment tests; (3) PA (phonological awareness tests) and (4) MFFT(Matching Familiar Figure Test) and ROCFT (Rey-Osterrieth Complex Figure Test) as visual cognition and memory tests. Preliminary Results: The results showed the following: Intact Abilities: PIQ=106; VIQ=106; Full IQ=107; RCPM=30/36; Hiragana/Katakana Character/Word reading/writing; Kanji Word reading; MFFT. Impaired Abilities (compared to the age-matched normal controls): (i) Kanji Writing ability: below -1.5SD; (ii) Kanji Nonword Reading: below -1.5SD; (iii) PA Skills: his Z-score is below -2; (iv) ROCFT - copy: -1; immediate recall: -1.5; delayed recall: -1.5; (v) reading/writing Roman alphabets: below -2.0SD (16/26 correct). His reading/writing errors in Romanji are primarily visual errors (e.g., M-W; n-h; t-f; p-b; etc.), and this was the same for writing Kanji words and reading Kanji nonwords. Conclusion: These preliminary results suggest that the reading/writing disability exhibited by this boy is likely to be caused by a visual memory deficit rather than a visual perception deficit or PA deficit.
Sylviane Valdois (Universite Pierre Mondes, France);Marie Line Bosse - Visual attention span and reading acquisition: a causal relationship.
Purpose: The visual attention span (VAS) corresponds to the number of distinct visual elements which can be processed simultaneously in a multi-element display. We have demonstrated that a VAS disorder independently accounts for poor reading in dyslexic individuals (Bosse et al., 2007). More generally, we have shown that VAS abilities affect reading acquisition independently of phonological skills in typically developing children (Bosse & Valdois, 2009). This study examined the extent to which visual attention span (VAS) and phonological awareness (PA) skills, estimated in kindergarten, predict reading performance one year later. Method: We conducted a longitudinal study on 130 French-speaking children whose performance was first assessed in kindergarten (T1), then at the end of Grade 1 (T2). Tasks of verbal short term memory, letter name and letter sound knowledge, phonological awareness, VAS and reading were administered in kindergarten. They were repeated at the end of Grade 1 where text reading was further assessed. The data were analysed using multiple regressions and pathway analyses. Preliminary Results: The results showed the following: 1. Phonological skills at T1 are longitudinal predictors of reading performance one year later (at T2). 2. The VAS is a longitudinal predictor of reading at Time 2 independently of the children's phonological skills at T1. 3. Contribution of the VAS to reading performance (T2) remains after controlling for reading performance at T1. Conclusion: These results suggest that VAS and PA skills are independent predictors of learning to read, which is compatible with a causal relationship between VAS and reading acquisition.
Victor Van Daal (University of Stavanger);Vibeke Rønneberg; Malin Wass; - Orthographic depth of L1 and L2 learning: Not only reading and spelling, but also L2 vocabulary affected
Purpose: Learning to read and spell is harder in some languages than in others, because of differences in orthographic depth. This paper examines whether differences in orthographic depth of the first language (L1 ) have repercussions for the learning of a second language (L2). We looked at whether Danish children were slower at learning to read and spell in English than their Norwegian and Swedish counterparts. We hypothesized that, due to the less transparent orthography, Danish children would be impeded in orthographic learning and building up word-specific representations in L1 and in L2. Method: Participants were assessed on reading and spelling in L1 and in English with frequency-balanced cognate word lists and frequency-balanced word lists, respectively. L1 vocabulary was used as a covariate. Orthographic skills were assessed with a novel-word learning task. Visual memory was measured with a matrix test and a visual association test. Vocabulary was assessed with a task requiring to indicate whether they knew the meaning of printed non-words and words. Results: MANCOVA and SEM indicate that Danish readers of all three age bands (n = 48) were slower and less accurate in naming L1 and L2 words, and made more spelling errors in L1 and L2, than their Norwegian (n = 57) and Swedish (n = 63) counterparts. They progressed less well than the Norwegians and Swedes in the acquisition of English vocabulary. Conclusions: The results of the first and second waves of assessments are discussed within the self-teaching framework of learning to read and spell.
Kees P. van den Bos (University of Groningen)Barry de Groot;Margo Jansen - Development of word-reading speed differentially linked to the development of phonemic analysis skills and alphanumeric naming speed
In many traditional as well as current reading models it is assumed that, with age, word-reading speed increases because word-holistic or lexical processing mechanisms get dominance over word-analytical or sub lexical mechanisms. An indirect way of testing hypotheses based on this assumption is to predict differential developmental links with reading-related skills which tap these lexical and sub lexical mechanisms. In our research lexical and sub lexical reading-related skills are operationalized by alphanumeric naming tasks (RAN) and phonemic analysis tasks (or phoneme awareness, PA), respectively. The main hypothesis is that, with age, reading speed-RAN relationships increase, whereas reading speed-PA relationships decrease. In previous, cross-sectional research (Van den Bos, 2008) age levels of 8-year-old through 14-year-old children (total N = 927) were studied. Standardized tests of word-reading (WR), RAN, and PA were used. Hierarchical regression outcomes offered clear support for the hypothesis. The present research raises the same hypothesis, but in this study, a longitudinal design is employed. Method. Ninety 'random' readers from intact classes from 6 Dutch schools for regular education were followed from grade 3 to grade 6 (ages 9 through 12). During the four years of this study, children were tested at the end of each school year, with the same tests as mentioned above. Data were analyzed with Multi Level Analysis, which accommodates for auto-regression. Again, the results offer clear support for the developmental hypothesis. Contrasts with reading accuracy and spelling performance (for which different predictions than for reading speed apply) are discussed, as well as some implications for tests of the double deficit hypothesis of dyslexia (Wolf & Bowers, 1999).
Wim Van den Broeck (Vrije Universiteit Brussel);Astrid Geudens; Eva Staels - A new reliable method to study subtypes in developmental dyslexia.
In their seminal article Castles and Coltheart (1993) argued that the group of disabled readers may not be a homogenous group. They found evidence for the existence of two subtypes, phonological and surface dyslexics. These findings were, however, criticized on methodological grounds. It has been argued (e.g. Stanovich, Siegel, & Gottardo, 1997) that comparing disabled readers to a normative sample of children of the same age, as in Castles and Coltheart, may yield processing trade-offs between sublexical and lexical paths that depend on the overall level of word reading skill. To this end, reading-level match controls were used in a new series of subtyping studies. In this contribution we show that the regression method of Castles & Coltheart as well as the reading-level match method lack reliability, and additionally the latter method is shown to be biased. We propose an alternative method based on state trace analysis that is both reliable and unbiased. Method Disabled readers' individual state trace curves were based on the reading performance on word and nonword reading lists across several conditions differing in difficulty. Word difficulty was manipulated by varying the number of syllables, number of consonant clusters and body frequency. To determine subtypes parameters of individual state trace curves were compared to the overall distribution of parameters. Results and Discussion Individual state trace curves of 73 Dutch disabled readers were calculated. Although the majority of disabled readers could not be categorized within one or the other subtype, the disabled readers belonging to a subtype showed a symmetrical pattern (as many phonological subtypes as surface subtypes).
Verna A.C. Van der Kooy- Hofland (Leiden University);Adriana G. Bus - Differential susceptibility to early literacy intervention in children small for gestational age or late preterm: A randomized control trial
Purpose: Central question is whether intervention effects are moderated by mild perinatal adversities. The dominant paradigm is that children small for gestational age (SGA) or late preterm (LP) perform at the lowest level of literacy. The theory of differential susceptibility, however, predicts that those children may profit most from the optimal environment created by the computer-based literacy intervention. Method: Hundred five-year-old children with the lowest level of early literacy skills were randomized (ratio 2:1) to an Intelligent Tutoring System or a treated control group. According to the Perinatal Register in the Netherlands (PRN, 2010), 21 participants (intervention, n = 13; control, n = 8) were at birth small for gestational age (birth weight between the 2.5th - 10th percentile) or late preterm (LP). Results: The interaction between experimental group and SGA or LP was significant, (F (1, 13) = 8.11, p = .014) after controlling for parental education, verbal intelligence and regulatory skills. The SGA or LP group showed the highest score on posttests after the intervention, and the lowest scores in the control group. Conclusion: Child-related risk factors turn out to create greater susceptibility to positive environments. SGA or LP children are quick in acquiring elementary literacy skills when they participate in an enriched, computer-based literacy environment. However, SGA or LP children are most vulnerable to develop delays in early literacy skills without instruction. It will be discussed which factors might make SGA or LP children more sensitive to context, for better and for worse.
Maaike Vandermosten (Parenting and Special Education Research Group, K.U.Leuven, Belgium)Bart Boets; Heleen Luts; Hanne Poelmans; Jan Wouters; Pol Ghesquière - Neural correlates of phonological representations, speech perception and auditory temporal processing
Purpose: It is often assumed that speech perception and its auditory temporal correlates are impaired in persons with dyslexia, and that these problems will eventually result in poorly specified phonological representations. Here, we used two imaging techniques that are each particularly suited to invest the neural underpinnings of auditory temporal processing or the quality of phonological representations. Method: First, the neural correlates of auditory temporal processing are investigated with Diffusion Tensor Imaging (DTI). This technique enables the examination of the white matter tracts between different brain regions. As intact myelination of these white matter tracts is important for rapid conduction of action potentials, we hypothesize that a disturbance of myelination would be extremely detrimental for temporal auditory processing. Second, the neural quality of phonological representations is measured by means of multivariate pattern analysis brain decoding techniques. By analyzing the spatial profile of fMRI data we aim to quantify the distinctness and robustness of the phonological representations. We scanned 20 normal reading and 20 dyslexic adults using DTI and fMRI. During the fMRI-scan participants listened to syllables that differed either in stop consonants, vowel or both. In addition to scan-data, behavioral measures on auditory temporal processing and speech perception were administered to relate to the DTI-data. Results: Preliminary results of the DTI data indicate that adults with dyslexia have a decreased quality of white matter in the left superior longitudinal fasciculus, and that this bundle shows specific correlations with speech perception, but not with auditory temporal processing. The analyses of the fMRI-data are not finalised yet.
Sharon Vaughn (College of Education)Janette Klingner; Elizabeth Swanson; Alison Boardman; Michael Solis; Audrey J. Leroux; Greg Roberts; Sarojani S. Mohammed - The effects of collaborative strategic reading instruction on the reading comprehension of middle school students
Purpose This series of related studies is in its third year of four multi-site, multi-year randomized control trials designed to test the efficacy of an intervention designed to promote strategic use of reading comprehension practices within cooperative groups - collaborative strategic reading (CSR). During years 1 and 2, the researchers investigated CSR in English/language arts and reading intervention classes for 7th and 8th graders with reading difficulties attending relatively low SES schools. In year 3, the researchers focus on English language learners who are struggling readers. Method Participants were 7th and 8th grade students in English/Language Arts and Reading classes in Texas and Colorado. Students were randomly assigned to classes (61 in year 1, 48 in year 2) and then classes were randomly assigned to treatment or business as usual comparison groups. Treatment students received multi-component reading comprehension instruction that taught them to apply comprehension strategies in collaborative groups 2-3 times per week over a 26-week period. Reliable pretest and posttest measures were administered. Results Findings from year 1 indicated significant differences in favor of the treatment students on the Gates-MacGinitie Reading Comprehension Test but not on reading fluency. Findings from year 2 are preliminary, but indicate similar-sized effects as Year 1 for typical achieving students but not students with reading difficulties. Year 3 data are currently being collected. Conclusions Findings from year 1 suggest that CSR is a feasible and effective practice that can be readily integrated with positive impact. Findings from year 2 and preliminary findings from year 3 will be reported including the role of fidelity of implementation in the relationship between CSR and reading comprehension.
Sharon Vaughn (University of Texas at Austin) - Discussant
Dr. Vaughn will discuss the findings across the papers.
Anneli Veispak (PhD student at Katholieke Universiteit Leuven);Bart Boets; Pol Ghesquière - The "thing" that good braille readers have and poor braille readers don't: a relation to developmental dyslexia
Purpose A proportion of Braille readers have specific reading difficulties which cannot be easily explained. Developmental dyslexia is the term used in the sighted population in case of inability to develop age appropriate reading skills. The pan-modal temporal processing deficit has been assumed to be causal to both the phonological and literacy problems observed in dyslexia. Struggling Braille readers have been reported to have difficulties with phonological awareness and tactile perception, domains also affected in dyslexics. The aim of our study is to investigate the causal pattern between low level auditory processing and cognitive prerequisites for reading in individuals who are blind and read Braille. Methods The participants of the study are Braille readers (n= 35, age 9-25 years, normal intelligence, no gross deficiencies in audiology) and print readers (n= 35, age 9-25 years, normal intelligence, no gross deficiencies in audiology). The experiments used: (1) Reading-ability assessment (real-word, non-word, story-reading); (2) Speech perception (speech-in-noise perception tasks: words, sentences; categorical perception); (3) Phonological processing (phonological awareness: phoneme deletion, spoonerism; Verbal-short-term memory: non-word repetition, digit-span forward; Rapid automatized naming: digits, letters); (4) Tests for auditory processing (Gap, 2Hz frequency modulation and tone-in-noise detection tasks); (5) Tactile perception (Grating orientation task). Conclusions Comparing the performance of print and Braille readers, it becomes apparent that both groups outperform the other on curtain aspects. Due to the tactual and sequential nature of Braille the correlational pattern between the low level processing and cognitive prerequisites for reading is differently emphasized in blind compared to sighted participants.
Marian Verhallen (Leiden University); Adriana G. Bus - Effect of illustrations on young children's processing of story language
Purpose. Our aim was to test experimentally whether illustrations in children's books contribute to young preliterate children's understanding of story texts. We hypothesized that pictures affect processing of unfolding text thus contributing to language development. The alternative process would be that illustrations form an autonomous source of information that children will explore independent of the spoken text. Method. Subjects were twenty one 5-year olds from low and medium SES families. Children looked at illustrations from storybooks in three conditions: (1) no oral text, (2) oral text that matched the illustration, (3) oral text that did not match the illustration. The within-subject design entails a change in illustration from one condition to the other. To ensure that illustration and condition were not confounded we combined each condition with three different pictures. Where children looked in the illustrations was measured with a remote eye-tracker device. We scored the number and duration of eye fixations on visual details mentioned in the text (AOIs) as well as on visual details not mentioned in the text. Results. A repeated measures analysis showed that children fixated visual elements highlighted by text more often and longer. These visual details were fixated far less when children heard no text. When children heard a text that did not match the illustration the number of fixations was lower compared to hearing the correct text, their duration, however was not. Conclusions. Fixation of elements in illustrations that matches the text may enhance visualization of language and may thus provide a boost for children's language development.
Ludo Verhoeven (Radboud University Nijmegen)Jan van Leeuwe - Impact of listening comprehension on reading comprehension development
Purpose: In the "simple view of reading" proposed by Hoover and Gough (1990), reading comprehension is conceived as the product of word decoding and listening comprehension. It is claimed that listening comprehension or the linguistic processes involved in the comprehension of oral language strongly constrain the process of reading comprehension. In the present study, this theory will be challenged by means of empirical data across the primary grades. Method: Following a longitudinal design, the word decoding, listening comprehension and reading comprehension abilities of a representative sample of 2790 Dutch children throughout the primary grades were examined. By means of growth curve analyses and structural equation modeling, it was investigated to what extent children's reading comprehension could be explained from their word decoding and listening comprehension levels, while contrasting product versus additive models. Results: The data show lend support to the product model of the simple view. Listening comprehension turns out to have a much greater impact on reading comprehension than word decoding. Growth curve analysis shows that the developmental pattern of word decoding shows strong increments in the early grades whereas for listening comprehension such increments tend to occur in the later grades. Moreover, he regression of listening comprehension on reading comprehension increases as a function of word decoding ability. Conclusions: It is concluded that the product model of the simple view of reading only partly holds and that the impact of listening comprehension on reading comprehension varies across the grades and across word decoding ability groups.
Pamela Vincent (The Ohio State University)Stephen Petrill - Genetic and Environmental Influences on Reading and Math Growth
Purpose: Research on reading and mathematics ability has shown that the two skills are associated, especially when linguistic proficiency is required for math problems. This study examines whether genetic and environmental influences on growth in reading and mathematics skills overlap. Method: Participants were drawn from the Western Reserve Reading Project. 314 twin pairs were assessed once per-year in grades 4, 5, and 6. Latent growth curve modeling was conducted in Mx. Variance was partitioned to estimate shared vs. unique genetic and environmental influences on latent intercept and latent growth in timed and untimed reading and math skills. Model 1 examined growth in reading fluency and math fluency, model 2 examined reading comprehension and applied problems. Results: Significant genetic and environmental influences on growth were observed for both timed and untimed reading and mathematics measures. Growth in reading comprehension shared genetic variance with growth in applied math problems (r = .64), but shared environmental influences were not common. The opposite was shown for fluency measures, growth in reading fluency and math fluency shared common environmental influences (r = .50). In addition, math fluency demonstrated unique shared environmental influences on growth independent of reading fluency growth Conclusions: Results suggested common genetic variance for growth in untimed reading and math and common shared environmental variance for timed reading and math in grades 4-6. The overlap in shared environment for timed measures may suggest that school or home environment affects growth in both reading and mathematics skills, and that these influences differ for timed vs. untimed measures.
Lesly Wade-Woolley (Queen) - Discussing prosody and reading
The final paper will be a discussant's synthesis of the four papers presented by the other members of the symposium.
Richard Wagner (Florida Center for Reading Research, Florida State University); Mercedes Spencer; Andrea Muse - Dimensionality of morphological awareness
Purpose Morphological knowledge, which refers to a conscious awareness of or the ability to use the smallest units of meaning in a language, may be important in learning to read English. However, the underlying nature of this construct is not well understood. Method A battery including nine morphological knowledge measures was administered to a fourth grade sample. The participants in the study were 99 4th graders. In addition, two standardized vocabulary tests and three oral reading fluency tasks were included. Participants' scores on a statewide reading comprehension test were also obtained. A series of confirmatory factor analyses was conducted to explore the possibility that morphological knowledge may be divided into two or more subcategories. Additional models designed to quantify the relations between morphological knowledge, vocabulary, oral reading fluency, and reading comprehension were included. Results In all cases, a single-factor model of morphological knowledge was the preferred model. This indicates that the morphological tasks administered were all measuring the same construct, and that method effects were not playing a significant role in performance. The results of the analyses also showed that morphological knowledge and vocabulary are indistinguishable for fourth-grade students. When morphological knowledge and vocabulary were represented as a single latent variable, it accounted for a high and significant portion of the variance in reading comprehension. Conclusions When measured with multiple indicators, what typically are viewed as distinct constructs-morphological awareness and vocabulary knowledge-were actually indistinct, and highly related to reading comprehension. Implications will be discussed.
Rick Wagner (Florida State University) - Discussion of the presentations: Factors that affect reading acquisition
Dr. Wagner will be discussant, placing the findings about dialect and reading in the broader context provided by large-scale studies of factors that affect reading achievement.
Shinmin Wang (National Taiwan Normal University)Susan E. Gathercole - Executive functions deficits of children with reading difficulties: Domain-general or Domain-specific?
Previous research has suggested poor readers show impaired performance on working memory task. Recent findings now present a strong correlation between working memory and executive functioning tasks. It is plausible to examine the role of executive functions in poor readers. In the present study, three executive functioning components: working memory, inhibition (verbal v.s. nonverbal) and divided attention (indexed by dual task performance), were assessed in poor readers and typical readers. Results suggest poor readers showed domain-specific (verbal domain) damage across a range of executive functions. In working memory skills, deficits of poor readers were restricted to the verbal domain, and to some extent, stemmed from more basic verbal memory impairments. In inhibitory skills, poor readers showed deficits in higher-level inhibition skills, but only in verbal domain. In dual task performance, poor readers showed greater disruption in dual task than typical readers. The dual task effects appeared in the digit recall task. This recognition of the role of executive functions in poor readers may provide new directions of reading research and potentially new strategies for identification and intervention in poor readers.
Hua-Chen Wang (Department of Cognitive Science, Macquarie University /Centre of Excellence in Cognition and its Disorder (CCD)); Anne Castles; Lyndsey Nickels - Orthographic learning of regular and irregular words in skilled and less-skilled readers
Purpose: The self-teaching hypothesis (Share, 1995) proposes that orthographic learning takes place via phonological decoding. Decoding skill is thus considered to be critical for acquiring visual word representations. However, how decoding assists orthographic learning of irregular words remains unclear. This study looked at orthographic learning of regular and irregular words and explored how prior reading skills interact with the effects of word regularity. Method: Twenty-eight second and third grade children participated in the study. In order to simulate natural orthographic learning, the children were exposed to the sound and meaning of eight novel words prior to orthographic exposure for those words. After three days of pre-orthographic-exposure training, the participants were given the eight novel words in written form in short stories to read aloud. Half the words were assigned regular pronunciations and half irregular pronunciations. Subsequently, spelling and lexical decision tasks were used to assess the success of orthographic learning. Results: The participants were divided into skilled and less-skilled readers, based on their prior word reading ability. The results from a spelling task immediately after orthographic exposure revealed that the more-skilled readers learned regular words better than irregular words but there was no effect of regularity for the less-skilled readers. Conclusions: The findings support the view that orthographic learning is less effective when decoding can only be partial, using irregular novel words. Furthermore, skilled-readers are better able to capitalize on the regularities of English when acquiring orthographic representations than less-skilled readers.
Jeanne Wanzek (Florida Center for Reading Research at Florida State University)Stephanie Al Otaiba, Yaacov Petscher - Moderators of oral reading fluency for students with learning disabilities or social/emotional disorders receiving special education
Purpose: This study examines moderators of growth in oral reading fluency across 2nd and 3rd grade for students receiving special education services for learning disabilities or social/emotional disorders. Method and Results: Data were drawn from the Progress Monitoring and Reporting Network (PMRN) database. Students identified with a learning disability (LD) or a social/emotional disorder (SED) in grade 2 were selected and followed through grade 3. A total of 10,339 students with LD and 2,146 students with SED were selected. Oral reading fluency assessments were administered 3 times each year (fall, winter, spring) in grade 2 and 3. Hierarchical linear modeling was used to model student growth in grade 2 and 3 using a piecewise model. Gender, status as English language learner, and participation in the free and reduced lunch program were included as covariates in the models to evaluate potential moderating effects on student outcomes and growth in oral reading fluency at each grade level. Results indicated that males, ELLs, and students with FRL demonstrate lower outcomes than their counterparts in the general education population. However, in special education, gender predicted ORF outcomes for students with LD in both grades with males performing significantly better. Gender was not associated with outcomes for students with SED. ELL status and FRL status were also moderators of outcomes for students with LD and students with SED, with these students performing better in oral reading fluency than their peers receiving special education. Conclusions: While males, ELLs, and students with FRL typically have lower outcomes in oral reading fluency, special education may provide instruction that helps these students improve their outcomes relative to their peers receiving special education.
Julie Washington (Georgia State University) - Reading, writing, and oral language: influence and interaction of dialectal variation in school-aged children
Purpose: It has been hypothesized that the greater the mismatch between oral and written language the more difficulty children will experience with literacy skill development. However, when this hypothesis has been tested results have been mixed depending upon language features examined and the age of the children included. The interactions between dialect use and general language skill were explored for their influence on reading and writing skills within a sample of urban, school aged African American children. Further, results were examined relative to the participants' overall school performance. Method: Language and writing samples were collected from a cross-section of African American children from second through fifth grade using semi-structured elicitation procedures. In addition, standardized reading, phonological, and oral language measures were administered. Samples were transcribed and coded using Systematic Analysis of Language Transcripts (SALT). Coding of language and writing samples included standard language measures provided by SALT (e.g., numbers of different words, complex syntax analysis). In addition, samples were coded for features of African American English (AAE), and for density of dialect produced (DDM). Standardized measures were scored using published guidelines. Results and Conclusions: Data analysis is on-going. Preliminary analysis suggests that reading and writing performance are impacted by the interaction between a child's dialect use and general language skills. Specifically, trends suggest that even if children are moderate to heavy dialect users, the impact on their reading achievement may be offset by strong overall language skill, but writing will be more problematic. Educational and pedagogical implications will be discussed.
Barbara Wasik (Temple University);Annemarie Hindman - Teacher and Child Outcomes of an Intensive Language and Literacy Intervention
Purpose. This study examined the impact of an intensive language and literacy professional development (PD) intervention, entitled ExCELL (Exceptional Coaching for Early Language and Literacy) on Head Start teachers and children. In monthly group trainings and weekly individualized coaching sessions, teachers were taught strategies to increase children's opportunities to use language in purposeful ways and to build vocabulary. Method. In a randomized control trial study, all intervention (n=19) and control (n=11) teachers were pre- and post-tested on the CLASS and ELLCO and all children in their classrooms (n=358) were pre-and post-tested on the PPVT and PALS. Also, book reading activities were videotaped and coded for teacher-child language interactions. Results. HLM results indicated significant differences between intervention and control teachers on the ELLCO and the instruction quality subtests of the CLASS. Also, children in the intervention group performed significantly better than peers in the control group on the PPVT and the phonological sensitivity subtest of the PALS. Qualitative data from classroom book readings showed that children in the intervention classrooms used more language (e.g., asked more questions, made more comments) than children in the control group. There was no difference across groups in teachers' language, including the number of open-ended questions or number and types of comments. Conclusions. In sum, findings suggest that ExCELL PD increased the quality of teachers' classroom practices and, by extension, children's language use in the classroom. As a result, these practices increased children's vocabulary and phonological sensitivity skills, both essential precursors to reading.
Mi-young Webb (Georgia State University)Amy Lederberg - Measuring phonological awareness in DHH children: Evaluation of internal consistency, concurrent validity, and factor structure of phonological awareness skills
Purpose: With early identification and enhanced technology (e.g., digital hearing aids and cochlear implants), the majority of young deaf and hard-of-hearing (DHH) children now have access to spoken language and can develop phonological awareness (PA) skills (Easterbrooks et al. 2008.) The current study assessed the psychometric properties of two PA tests developed for hearing children when used with DHH children. Method: 102 DHH children (M age = 65.4 months; SD = 12.5 months), who perceived speech were assessed on the four tasks of Phonological Awareness Test (Robertson & Salter 1997) (rhyme, segmentation, initial phoneme isolation, blending) and TOPEL-Phonological Awareness (Elision and Blending ) in the fall and spring. Results: Item analyses indicated high internal consistencies (alpha coefficients > .90) and good discrimination power (item-total correlations > .35). Multiple criteria for factor retention indicated a one-factor solution underlying the measurement, accounting for 57.28% and 59.25% of the common variance for Fall Spring data. Comparisons with hearing samples suggests that, on average, DHH children scored in the low average range compared to hearing norms. TOPEL standard score for this sample was 84.12 (SD=19.87) compared to the validity sample of 103 (SD = 14). Conclusions: Consistent with research with hearing children (Anthony & Lonigan, 2004), this study suggests that PA tasks developed for hearing children tap a single underlying ability regardless of linguistic complexity when used with DHH children who have speech perception abilities. Importantly, these results suggest that tests developed for hearing children can be used to assess PA of DHH children.
Hsiao-Ling Weng (Teachers College, Columbia University) - Influence of explicitness and causal order on second grade readers' comprehension of causal relations in expository texts
Purpose - This study investigated the effects of clausal order (cause before effect vs. effect before cause), and explicitness (present vs. absent of causal connectives), on twenty good and twenty poor second graders' reading comprehension of expository texts. Method - Four versions of passages were composed: the implicit-normal clausal order version, the explicit-reverse causal order version, the implicit-reverse causal order version, and the explicit-normal causal order version. The numbers of idea units were comparable across passages and the readability level of all passages were suitable second grade readers. Comprehension was measured on both sentence-level (i.e., non casual questions/content questions and casual questions) and discourse level (i.e., summarization). Results - The results showed that explicitness, causal order, and reading ability are all significant factors in the second-graders' expository text comprehension. Expository texts with casual connective present and with causal order congruent with event order were relatively easy to comprehend for both good and poor readers. However, good readers performed significantly better than poor readers on the causal related questions and summarization but not on non-casual related questions. Conclusions - This research provides us with insights into the reasons young readers have difficulty understanding some of the causal relations but not all. Hence, with this knowledge, we can develop appropriate reading and instructional materials to help our students become better readers.
Katherine L. Wheat (Department of Psychology, University of York, UK)Piers L. Cornelissen; Peter C. Hansen - Exploring early activation of Broca's area during visual word recognition using magnetoencephalography
Purpose: To replicate early phonological priming at pars opercularis of left inferior frontal gyrus and precentral gyrus IFGpo/PCG within 100 ms of reading a word (cf. Wheat et al., 2010) and to test whether this finding extends to nonword targets. Method: MEG data were collected from 20 right-handed native-English-speakers, with no history of reading difficulty. Participants silently read words and nonwords preceded by masked primes; pseudohomophones (teech-TEACH, slynt-SLINT) and orthographic controls (telch-TEACH, slont-SLINT). No overt response was required except for catch trials. Results: The pseudohomophone-word condition showed significant (p<0.05) IFGpo/PCG activity at 0-200ms, with greater activity by 100-300ms. No IFGpo/PCG activity was seen for orthographic-word or pseudohomophone-nonword until 100-300ms. Conclusions: There are noticeably different progressions of activity between conditions that will be presented in more detail across overlapping whole brain analyses and in a ROI approach.
Erik Willcutt (Department of Psychology and Neuroscience, University of Colorado, Boulder) - Discussant
Purpose: In addition to commenting on the four papers, the discussant will also briefly describe results of simulations conducted to test the validity of different methods commonly used to test hypotheses regarding causes of comorbidity.
Joanna P. Williams (Columbia University);Lisa Pao; J.Grant Atkins; Rong Cheng; Jenny Kao; Jill Ordynans - Improving reading comprehension by teaching text structures to second graders
Purpose: to report on the development and evaluation of an intervention designed for second graders at risk for academic failur to teach reading comprehension embedded in social studies content. Method: The intervention (41 lessons, twice weekly) included instruction about the structure of three types of expository text (sequence, comparison, & cause/effect),emphasizing clue words, generic questions, graphic organizers, and the close analysis of well-structured examples of such texts. The intervention was compared to a program that focused on the same social studies content (homes, schools, and jobs of Native Ameraicans, colonists, and pioneers), which included no text structure training, and also to a no-instruction control. Seventeen teachers, randomly assigned to treatment, and 212 students participated. Results: Comprehension outcome measures assessed the ability to summarize, orally and in writing, short paragraphs containing novel social studies content organized according to the three structures, and the ability to answer comprehension questions about other similar paragraphs. On all of these measusres, the intervention group performed better than the other two groups, which did not differ. Content outcome measures assessed vocabulary and memory for social studies information. On these measures, the intervention and the content-only groups did not differ; both scored higher than the no isntruction group. Conclusion: These findings indicated that our text-structure intervention was effective in improving comprehension. Moreover, the comprehension gains of the intervention group did not come at the expense of the amount of social studies content acquired.
Rebecca Wiseheart (St. John's University);Rebecca Wiseheart; Linda Lombardino; Lori J.P. Altmann; Lauren Snyder - Differential effects of neighborhood density on digit and object naming speed
Purpose: A long standing hypothesis is that rapid automated naming (RAN) taps the integrity of a phonological access mechanism; however, this hypothesis has been difficult to prove. Because rime neighborhood density (RND) is a phonological variable that has been shown to have facilitative effects on other naming tasks, we predicted RAN speed would be faster for high RND stimuli, regardless of stimulus type (digits/object) or level of reading skill (developing/independent). Method: 18 Kindergarten and 21 3rd grade normal readers completed a battery of tests including cognitive and reading measures along with 4 experimental RAN tasks that varied in terms of phonological density (high/low) and stimulus type (digits/object) Results: RND exerted significant facilitation effects on RAN speed for objects but not digits in older but not younger children. Most importantly, a large difference in RAN speed across groups was accounted for by efficiency in sight word reading, even after age, orthographic knowledge, and processing speed were partialled out. Thus, while RAN's relationship to phonology was restricted to object naming, rapid naming of both stimulus types was predicted by the development of automatic word reading skill. Conclusions: Results offer strong support for the hypothesis that RAN's relationship to reading is largely independent of phonology. Implications for a non-phonological access mechanism common to both RAN and sight word recognition are discussed.
Maryanne Wolf (Tufts Unviersity); Mirit Barzillai; Robin Morris; Maureen Lovett - How our views of fluency influence intervention: From multi-component emphases to multi-component effects
Purpose: While the reading field has made impressive strides in designing and implementing interventions that successfully remediate phonological and decoding deficits, this progress has not been matched in the areas of reading fluency and comprehension, or semantic knowledge. To ameliorate the difficulties readers face in these areas, it is necessary to adopt an approach that identifies and addresses the range of linguistic skills necessary for fluent reading. We evaluated the success of one such multicomponential approach, the RAVE-O program, in addressing these more difficult to remediate areas in a population of struggling second and third grade readers. Methods & Results: We first use SEM techniques to illustrate how various linguistic processes, such as semantic and orthographic knowledge, are involved in the fluent reading of second and third grade struggling readers. We then present intervention results demonstrating the success of a multicomponential intervention approach in addressing the word and connected text level fluency weaknesses of this population while also improving performance in other areas critical for reading such as deep, flexible semantic knowledge. Conclusions: Results from this analysis attest to the importance of incorporating an emphasis on multiple-linguistic areas within reading interventions in order to address the word and text level fluency deficits and semantic weaknesses of struggling readers.
Julie Wolter (Utah State University)Heather Barger; Katherine Pike - Third grade dynamic assessment of morphological awareness and literacy achievement: A feasibility study
Purpose: The purpose of this feasibility study was to replicate and extend upon Larsen and Nippold's (2007) research to determine whether a dynamic assessment of third-grade morphological awareness is related to reading and spelling achievement,and whether third-grade children improve their performance on a dynamic morphological awareness assessment following systematic scaffolds. Method: Twenty three typically developing third-grade children were administered an adapted morphological awareness dynamic assessment task two times (second administration 2-3 days later) and a reading and spelling test battery. The morphological awareness dynamic assessment task consisted of words which contained high frequency base words and affixes (e.g., yellow, -ing) but resulted in low frequency derived words (e.g., yellowing). Children were asked to explicitly define the words, and if unsuccessful, were provided with a series of prompts which ranged from asking about the parts of words to a choosing from multiple-choice answers. Results: The dynamic assessment morphological awareness revealed a wide range of performance for third-graders and was significantly and moderately related to third-grade spelling, word reading fluency, and reading comprehension. Additionally, performance on the dynamic assessment morphological awareness task significantly increased from time one to time two. Conclusions: A third-grade dynamic morphological assessment may be a helpful assessment tool to predict elementary school literacy success. Moreover, scaffolds included in a dynamic assessment may prove to be a valuable way to not only increase morphological awareness performance, but also to provide treatment insights for implementing morphological awareness instruction for individuals struggling with this important skill.
Clare Wood (Coventry University, Psychology Department)Sam Waldron; Nenagh Kemp - The relationship between text messaging and young people's grammatical understanding
Purpose - Previous research has shown positive relationships between the use of 'textisms', spelling and reading ability in children, but has not considered the potential relationships between texting (SMS use) and grammatical understanding. The current study aims to investigate whether a relationship exists between texting behaviour and written and spoken grammar, and to determine the age at which any effects are most prominent. Method - The study's design is cross sectional, with three age groups; 8-11 years (n=100), 12-16 years (n=100) and 18-24 years (n=100). Tests include measures of grammatical skill (Test of Receptive Grammar (TROGII), a pseudoword orthographic choice task, and a grammatical translation task), IQ (WAIS), spelling ability (WRAT-IV), the occurrence of textisms (e.g., u for you) and grammatical violations (e.g. im fine, how is you?) from a two-day sample of participants' texts. Results - In this ongoing research, data will be examined via regression analyses to determine whther grammatical violations made in text messages are related to grammatical understanding, while controlling for individual differences in IQ and considering the potentially mediating role of spelling performance. Conclusions - The findings will address media speculation about the relationships which have been claimed to exist between grammatical understanding and use of grammatical violations in written language tasks.
Donna Wright (Dr. Donna Akilah M. Wright) - Word reading phase theory applied to a reading intervention with young children
Purpose: The aim of the study was to investigate the application of Ehri (2000)word reading phase theory as a guideline for teacher candidates to develop activities to improve the decoding ability of young children at risk for reading failure. Method: Ninety-four first, second and third graders were classified as pre-alphabetic, partial alphabetic, full alphabetic and consolidated alphabetic word phase readers using developmental spelling and decoding measures. Gaps in word reading ability were identified using measures of letter knowledge, consonant letter sound knowledge, vowel letter sound knowledge and multi-letter pattern knowledge. Teacher candidates developed intervention activities based on children's word reading phase classification and gaps in word reading ability. Teacher candidattes provided individual and small group tutoring. All participants were tested following the intervention using a Woodcock subtest: word identification to determine improvements in word reading ability. Results: A multivariate analysis was conducted. Findings indicated word identification ability and spelling development level significantly improved following the intervention. Skills in decoding unfamiliar words had limited improvement. However, there were significant interactions between grade and word identification ability and spelling development level and initial word reading phase. Conclusion: Ehri's word reading phase theory provided appropriate guidelines for prescribing literacy based activities, which improved the word reading skills of young children experiencing challenges decoding words.
Qiuping Wu (Department of Educational Psychology, The Chinese University of Hong Kong)Puiwan Cheng - Effects of Linguistic Unit and Cognitive Operation Complexity on the Relations between Phonological Awareness and Reading Achievement in Chinese: A Two-Level Hierarchical Meta-Analysis
Purpose: Although phonological awareness has been found to be a strong predictor for children's reading achievement across alphabetic languages, its role in Chinese, a morphosyllabic language, has not been consistently shown in the previous studies. The purpose of the present study was to synthesize research that reported the relationship between phonological awareness (PA) and Chinese word recognition (CWR) in Chinese and firstly attempt to explain the variability among studies in terms of linguistic unit (i.e., syllable, onset/rime) and cognitive operation complexity (such as isolation, blending, et al.). Method: A two-level hierarchical meta-analysis was employed to evaluate the strength of the association between PA and CWR, and identify the effects of moderators (e.g., linguistic unit, cognitive complexity, age, and so on). Correlation coefficients were first retrieved from the previous papers and then transferred into Fisher's z as effect size. Second, the variables related to subjects and tasks were coded as moderators. The final database for meta-analysis contained 38 studies with a total of 7293 participants. Results: The results indicated that PA was significantly correlated with Chinese word reading (r = .33). The additional moderator analysis showed that: 1) the synthesized effect size was the strongest in Taiwan (r = .47) of the three Chinese societies; 2) the effect size was consistent across age group; 3) with respect to the effect of linguistic unit, PA at the syllable level was more closely related to Chinese reading performance for young children in mainland China; 4) cognitive operation complexity had no significant moderator effect on PA-CWR associations. Conclusions: Taken together, phonological awareness is significantly related to Chinese reading across all age range, but phonological awareness of larger linguistic units may be more sensitive for predicting Chinese word reading performance for preschool children, especially in Mainland China. Implications for future research are discussed.
Taeko Wydell (Brunel University, UK);Liory Fern-Pollak; Sylviane Valdois - The effects of visual attention span (VAS) and phonological awareness (PA) skills on reading in English primary school children: a cross sectional study
Purpose: This study aimed to examine the extent to which visual attention span (VAS) and phonological awareness (PA) skills contribute to reading skills in English primary school children. Method: Using the Valdois's VAS (global/partial report tasks) paradigm as well as PA (Alliteration/rhyme fluency; Spoonerisms) and reading measures (BAS, regular/irregular words & nonwords) 40 YR-2 (aged 6-7) and 40 YR-4 (aged 8-9) English-speaking children were individually tested. The data were analysed using primarily multiple regression and factor analyses. Preliminary Results: The results showed the following: (1) YR-4 performance on all the tasks was in general better than that of YR-2. (2) PA and VAS predicted Reading accuracy in both YR-2 and YR-4, while there was no relationship between the two factors. (3) VAS predicted reading accuracy in the younger children to a greater extent than the older children. (4) PA significantly predicted Irregular WD reading accuracy in YR-2, while in YR-4, PA contributed to reading accuracy similarly regardless of regularity and lexical status. Conclusion: These preliminary results suggest that (1) VAS and PA skills are two dissociable elements of visual word recognition; (2) although VAS increases with reading experience, its contribution to reading accuracy decreases; (3) the influence of PA on reading accuracy is qualitatively different from YR-2 to YR-4, i.e., in the younger children PA skills are rudimental whereas in the older children these skills are more fine-tuned.
Wen Xiao ()Rebecca Treiman - Exploring iconicity in simple Chinese characters
PURPOSE Chinese characters derived from pictures, but to what extent do modern characters resemble pictures? The present study addressed this question empirically, the only previous attempt to do so having involved just 20 characters (Luk & Bialystok, 2005). METHOD All simple-structure characters from Shu et al. (2003)'s database were included in the study. Forty adults with no knowledge of Chinese saw an English word (or short phrase) together with 2 Chinese characters and were asked which character best matched the meaning of the English word. The 2 characters had similar number of strokes, and the wrong characters differed across participants. RESULTS The mean proportion of correct answers, an index of iconicity, was 53.6%. This was slightly but significantly higher than expected by chance. A binomial test showed that 45 of the 213 characters were guessed at a level higher than chance. In a regression model predicting the sequence in which characters are introduced in Chinese elementary-school textbooks, iconicity did not contribute significantly while number of strokes, frequency of occurrence, and grammatical category (content word vs. function word) did make significant contributions. CONCLUSIONS These results provide a systematic estimate of the degree to which simple-structure Chinese characters are guessable because of iconicity. The overall degree of iconicity is low, although some characters have a degree of iconicity. Contrary to the suggestion of Shu et al. (2003), more iconic characters don't appear to be taught earlier than less iconic characters once other factors are taken into account.
Ling-Yan Yang (School Psychology, College of Education, The University of Iowa)Yu Su, Jian-Peng Guo - The role of verbal-mediated visual processing skill in Chinese reading acquisition
Purpose: The purpose of the present study was to use recent advances in meta-analysis methodology in a review of the literature concerning the association of certain measures in visual processing skills and Chinese reading performance. Method: Seventy-seven primary studies were located through a systematic literature search from 1980 to January 2010. The inclusion and exclusion criteria resulted in 16 studies that contributed appropriate correlation coefficients across 32 independent samples with a total sample size of 2,376. Four components of the construct of visual processing skill were identified: visual perception, memory, speed, and verbal mediation. All analyses were computed using the Hunter-Schmidt Meta-Analysis Programs Package (Schmidt & Le, 2004). Results: The noteworthy finding of the present meta-analyses is that, in comparison to other visual component skills, verbal-mediated visual processing skill, that is, the ability to label visual stimuli with verbal names, has a large correlation effect size with Chinese reading acquisition, not only in the lower grades (<G2) where visual-verbal mediation accounts for 38% of variances ( =0.62, =0.07), but also in the higher grades (≥ G2) where it accounts for 50% of variances ( =0.71, =0.1). Conclusions: This finding is in line with Perfetti's (1986) theory regarding the universal cognitive factors in accounting for ability differences in reading: One is a general language comprehension factor, and the other is a name retrieval factor that operates when a symbol name is retrieved from memory in response to a visual input.
Hsiou-Wen Yang ()Yi-fen Su; Shin-feng Chen - Relation Between Chinese Literacy and Processing Speed Index Among Second- and Third- Grade Students in Taiwan.
Purpose- There are some research data show that poor readers tend to have lower scores on Processing Speed Index (PSI) in WISC III among English language users. This phenomenon is also found in Taiwan for students whose native language is Chinese. However, the relation between processing speed and Chinese literacy is unknown. Besides, there are other factors, such as phonological awareness and rapid naming speed which have been proved to be closely related to Chinese literacy. Therefore, the purpose of this research is to investigate the relation between Chinese literacy (Chinese character recognition and character writing), processing speed, phonological awareness and rapid naming in beginning readers. Method- A structural equation model was proposed in which the relationships among Chinese literacy, processing speed, and phonological awareness were explored. Pearson correlation coefficients and regression analysis were also adopted to investigate these variables. The research tests include Chinese character recognition test, Chinese writing test, Coding and Symbol Search subtests from WISC III (Processing Speed Index), the phonological awareness test, and rapid naming test. The subjects included 109 second and 114 third grade students in Taiwan. Results- The results indicate an acceptable model fitness in the proposed model: χ 2 =44.413, df=29, p=0.03; χ 2/df=1.53; GFI=0.958; NFI=0.917; RMSEA=0.05; RMSEA=0.05; SRMR=0.05. These indicate the model fits well to empirical data. Conclusion- In conclusion, processing speed, phonological awareness and rapid naming are predictors of Chinese literacy. However, naming is the strongest of the three. Implications for educational applications are also discussed.
Yi-Fen Yeh (National Taiwan Normal University)R. Malatesha Joshi, Erin McTigue - Morphological Knowledge of English Words among Chinese Secondary School Students
Purpose - In order to investigate the knowledge of compounding words in English, this study examined morpheme affixation of English derivational words and compound words in Chinese secondary school students. Method - English Derivational Awareness Task and Task of English Compounds were administered to 289 Chinese- students from grade 7, 8, and 9. An error analysis was conducted to examine the affixes and semantic values of the words in their responses. Results - Approximately 70% of the errors on derivational words and compound words showed their recognition of either all or partial morphemes inside the real or pseudo words. While students were able to add suffixes (e.g., -s, -ed, -er, -ist, -ship, -ly) to the root morphemes, they did not know where and how to use these affixes appropriately. Their low performance on derivational words was mainly attributed to insufficient knowledge of the suffixes and sentence structures. Additionally, a significantly high accuracy rate on real compound words, compared to derivational words, indicated that students' familiarity with compound words' structure exceeds their familiarity of derivational words. Conclusions - Chinese students performed better on compound words than derivational words in English, even though they were able to detect the morphemes within them. Enhancing ESL students' knowledge of derivational morphemes and sentence structures, as well as the inferential ability on the combination of morphemes would help teachers to increase students' vocabulary size and competence.
Astri Yulia ()R. Malatesha Joshi, PhD.; Syamsul Bahri Ys.; Zainab Allaith - Can Simple View of Reading (SVR) be applied to ESL undergraduate students?
Purpose According to the Simple View of Reading (SVR), reading comprehension (RC) is the product of decoding (D) and linguistic comprehensions (LC), which is generally measured through listening comprehension. However, there may be other factors, such as vocabulary, grammatical knowledge, first language proficiency, that may contribute to reading comprehension. The purpose of the present study was to investigate the relationship among listening comprehension (LC), grammatical knowledge (GK), and reading comprehension (RC) among university undergraduate students who speak English as a second language (ESL). Method Subtests of RC, LTC, and GK of TOEFL were administered to 209 undergraduates learning ESL in Indonesia. Additional information (i.e., length of time of English instruction received, first-language proficiency, language spoken at home) were also collected for further analyses. Results The results showed that R = .789 and the effect size was .622 (R2 = 62.2%). GK showed a slightly higher contribution (β= .471; rs2 = .842; p<.001) than LC (β = .403; rs2 = .785; p<.001). The variables were highly correlated: the correlation between LC and GK was .628; LC and RC was .699, and GK and RC was .724. For non-linguistic factors, only the length of time of English instruction showed significant effect on reading comprehension (R2 = .45; p<.001), others did not have significant effect. Conclusions Even though SVR has been a widely accepted model among monolinguals, it has not been explored widely among ESL students, especially among university students. The results show that SVR may have to be modified for ESL students.
Elena Zaretsky (College of Educatio and Human Development)Jelena Kuvac-Kraljevic - The role of orthography in early reading acquisition in children with SLI: Crosslinguistic study.
Purpose: The purpose of this study was to investigate early reading development among children with specific language impairment (SLI) as a factor of orthography (shallow vs. deep), examining the interaction between oral language- (lexicon, morphosyntax, semantics) and code-related (phonological awareness and verbal working memory) components considered to be influential in reading acquisition in view of the orthographic depth hypothesis (Katz & Frost, 1992). Method: 30 kindergartners (15 English- and 15 Croatian-speakers) were assessed on measures of PA, VWM, vocabulary, alphabet knowledge, decoding, sight word recognition and spelling. ANOVAs, regression and correlation analyses were utilized to identify the interaction between and predictive power of code- and oral language-related skills on early reading as a factor of orthography. Results: Children with SLI showed similar code- and oral language-related profiles, suggesting that core deficits are the same in this population regardless of the orthography. However, significant differences were found in the predictive nature of code- and oral language-related components on early decoding and encoding skills, similar to that of typically developing (TD) counterparts from the same linguistic backgrounds. Conclusion: The results confirm that while children with SLI from different linguistic backgrounds have similar language and cognitive profiles, the nature of the orthography dictates the use of reading pre-requisite in the early stages of reading acquisition. This finding supports previous research with typically developing children (Zaretsky, Kuvac, Core & Lancek, 2009) that exposure to languages with shallow orthography and rich phonological composition puts emphasis on different distribution of factors influencing early reading acquisition.
Jing Zhao (Texas A&M University); Yifen Yeh; L. Quentin Dixon; R. Malatesha Joshi; Erin McTigue - A cross-group and cross-linguistic study of the effect of morphological awareness on literacy skills in Chinese-English bilinguals
Purpose - To investigate the effects of morphological awareness (MA) in Chinese-English bilinguals, we examined the contribution of MA to literacy skills from two perspectives: 1) cross-group but within-language perspective with a group of Chinese and a group of U.S. English-speakers completing tasks tapping into word level knowledge in English; and 2) cross-linguistic and within-group perspective with Chinese-English bilinguals tackling word analyses in both English and Chinese. Method -In Study One, knowledge of Extract the Base (August et al., 2003) and Word Form Exercise (non-word) was measured in 166 3rd-grade monolingual U.S. children and in 339 Chinese-English 8th-grade bilingual children in mainland China. In Study Two, 289 students at grades 7, 8 and 9 in Taiwan were administered Chinese and English MA tests, including inflectional, derivational and compound words. Structural Equation Modeling was used for analyses. Results -Two measures in Study One loaded heavily on MA, which contributed significant variance in word spelling skills in both the U.S and the Chinese group. There was a stronger effect of MA in the monolingual group than in the Chinese-English bilingual group. Bilingual learners' Chinese MA system was predictive of their English MA. Multi-group analysis showed that students with lower Chinese MA developed their English MA more through their Chinese MA, but students with higher Chinese MA relied more on their cognitive skills. Conclusions -MA makes a unique and significant contribution to English spelling in Chinese-English bilinguals and English monolinguals. In addition, abundant compound words in Chinese facilitated the development of English MA.
Tricia Zucker (University of Texas Health Science Center at Houston); Sonia Cabell; Jill Pentimonti; Laura Justice - The Systematic Assessment of Book Reading (SABR): A methodological shift in assessing classroom-based shared reading
Purpose The Systematic Assessment of Book Reading (SABR) is a free tool designed to examine teacher-child interactions surrounding read-alouds. Researchers have called for methodological shifts in the measuring book-reading experiences to capture variations in behaviors (Hammett, van Kleeck, & Huberty, 2003; Yaden, 2003). We examined the SABR's psychometric properties and explored stability of teacher behaviors across the preschool year. Method Participants were 105 teachers working in 4-year-old preschool classrooms, serving low-income children. Teachers read aloud 30 researcher-provided texts. Teacher behaviors were coded in six videotaped sessions using the SABR, which contains 23 items coded in 15-second intervals. Results Confirmatory factor analysis revealed five SABR constructs: Language Development, Abstract Thinking, Print/Phonological Skills, Elaborations, Session Climate. The SABR demonstrated adequate inter-rater reliability, concurrent validity with a global classroom quality measure, and good predictive validity of children's skills. Specifically, hierarchical linear models indicated that the SABR predicted preschool gains in letter naming and expressive vocabulary. The SABR also predicted children's decoding and comprehension in grade 1. Preliminary stability analyses of teacher behaviors across time demonstrated sizable variability. Conclusions These findings emphasize the importance of measuring book-reading quality as a multidimensional factor. The SABR represents one new way to measure important characteristics of early experiences that are predictive of children's outcomes. Moreover, this research is the first in classroom settings to measure book-reading quality with multiple observations across a sustained period of time. This is an essential methodological shift if we are to accurately assess the contribution of adult reading behaviors to children's outcomes.