Click on the name of the first author and see the abstract, click on Top to go back.
Al Otaiba - Allen - Altmann - Anthony - Apel - Aram - Aram - August - Azimi - Babayigit - Baciu - Barnes - Barron - Barth - Bell - Bell - Berthiaume - Betjemann - Betts - Biancarosa - Blanchard - Boets - Bolger - Bosman - Boulware-Gooden - Bowers - Bowyer-Crane - Boyce - Branum-Martin - Brown - Burgess - Butler - Cabell - Cain - Calhoon - Canto - Carden - Carlisle - Carroll - Cartwright - Chen - Clarke - Cleuren - Cocksey - Coleman - Collins - Collins - Compton - Connor - Conrad - Cormier - Cornelissen - Corrêa - Corriveau - Coyne - Craig - Cromley - Cronin - Crowe - Cunningham - Cutting - Daigle - Davidson - Davis - Davis - Davis - de Jong - Deacon - Denton - Desrochers - Ebbers - Ediger - Ehri - Elleman - Eric - Evans - Farnia - Feng - Filippini - Foorman - Francis - Friend - Frishkoff - Frost - Frye - Fuchs - Fuchs - Fujise - Fulmer - Furnes - Gao - Gersten - Geva - Gilbert - Glasney - Godard - Gonzalez - González-Trujillo - Gooch - Gottardo - Goudey - Graham - Grant - Greenberg - Gregg - Guan - Guo - Guo - Gutierrez-Palma - Halliday - Hansen - Harris - Hayden - Hayes - Heggie - Henderson - Hindman - Hock - Hogan - Holliman - Hoskyn - Hosp - Howard - Hu - Hus - Ijalba - Inutsuka - Jackson - Jarmulowicz - Johnson - Johnson - Jones - Joshi - Justice - Kaefer - Kainz - Kalia - Kantor - Karemaker - Keenan - Kershaw - Kiefer - Kieffer - Killing - Kim - Kim - Korat - Kuo - Lambrecht Smith - Larkin - Lederberg - Lee - Lesaux - Leung - Levin - Li - Limbos - Lin - Lindo - Logan - Lombardino - Lonigan - Lord - Lovett - Lu - Luo - Luppescu - Lyytinen - MacArthur - Mahfoudhi - Malabonga - Mancilla-Martinez - Manis - Martin-Chang - Martinussen - Mason - McBride-Chang - McCann - McMaster - Meisinger - Mesmer - Miller - Mol - Morris - Morris - Morris - Morrison - Mostow - Naes - Nair - Nation - Navas - Nelson - Obregón - Olinghouse - Ouellette - Padakannaya - Pae - Paizi - Parault - Park - Park - Pasquarella - Perdijk - Perin - Peterson - Petscher - Phillips - Piasta - Pitchford - Pittman - Pollo - Powell - Priebe - Purpura - Ramirez - Ramirez - Raynolds - Reitsma - Rickard Liow - Ricketts - Rimrodt - Ring - Robbins - Roberts - Rosenthal - Roux - Rupley - Sabatini - Sadoski - Sainz - Santos - Schatschneider - Schaughency - Schwanenflugel - Scott - Segal-Seiden - Segers - Serrano - Sevcik - Shafman - Shankweiler - Shapiro - Silvestri - Skibbe - Smith - Soden - Sparks - Sparks - Spear-Swerling - Su - Sun-Alperin - Tao - Taylor - Thomson - Tilstra - Tong - Torpa - Trathen - Tsai - Uhry - Underwood - Vagh - Valdois - Vale - Valencia - van Bon - van Daal - Verhoeven - Wade-Woolley - Wagensveld - Wang - Wehby - Welcome - Whalley - White - Whitney - Williams - Wilson - Wise - Wiseheart - Wolf - Wu - Wu - Yaghoub Zadeh - Yeong - Zaretsky - Zhang
Stephanie Al Otaiba (Southern Methodist University)Connor, Carol; Meadows, Jane; Petscher, Yaacov; Greulich, Luana; Sidler, Jessica; and Lang, Laura - Responsiveness to Kindergarten Reading Instruction: Examining the Interactions among Student Characteristics, Reading Instruction, and Student Outcomes
Purpose This study describes kindergarten literacy instruction, exploring the relation between amount of instruction, how this instruction is implemented, and student outcomes, including child X instruction interactions. Our primary purpose is to learn more about the characteristics of children who do not respond to effective classroom reading instruction. Method Participants were 22 teachers and 250 children. Observations occurred three times per year during the 90 minute literacy block. Videos are being coded to determine the amount of time each target child received and the specific focus of instruction (phonological awareness, phonics, vocabulary, fluency, and comprehension), who managed instruction (teacher-mediated, or child-mediated), and what instructional materials were used. We also used an adaptation of Haager et al.'s (2004) likert rating scale toe valuate teaching behaviors. A large variety of standardized student measures (IQ, working memory, phonological awareness, language, literacy, and behavior) were administered to describe child X instruction interactions). Results and Conclusion The study has been completed; data analysis, using HLM to model growth and to explore interactions, is underway. Preliminary findings on student outcomes indicate that most children reached grade level reading skills; however, about 5% of children did not. We will examine child X instruction interactions that we anticipate will extend the literature on responsiveness to instruction.
Purpose: This study investigated the use of three assessment tools and two different methodologies to identify participants' responsiveness to early literacy intervention. Method: Thirty-two first grade children participated in the study: 21 were at-risk and 11 were developing normally. The WRMT-R, TOWRE, and the DIBELS ORF measures were given as pre-test and post-test measures. The children at-risk were provided an intensive early literacy intervention over an eight week period. Results: The three measures at post-testing did not identify the same children as at-risk for reading difficulties (WRMT-R =19%, TOWRE=62%, ORF=95%). The participants were divided into three groups: good growth (GG), poor growth (PG), and control based on ORF slope (3 points). The participants in the GG group made significantly larger grains in ORF score than both the PG and control groups. Based on slope, only 48% of participants remained at-risk, while 95% percent of the participants did meet the criterion level. The control group was significantly more accurate than the GG and PG groups on the initial ORF measure. Conclusion: The findings suggest that the specific assessment tool and methodology chosen to measure growth in early literacy skills has a specific impact on the number of students identified as at-risk.
Purpose: This study investigated the effects of dyslexia on four aspects of sentence production: fluency, grammaticality, response times and sentence structure choice. Method: Twenty-eight adults with normal reading (NR) and nineteen adults with dyslexia (aged 16-28) participated. Participants completed working memory (WM), vocabulary, and executive function tasks, plus a timed word reading task. In the experimental task, participants produced a sentence including 3 words: 2 nouns and a transitive verb from one of 4 types. Results: Participants with dyslexia produced significantly fewer fluent and grammatical sentences than NRs, even with word reading accuracy covaried. Participants with dyslexia responded slower than NRs, even with word reading time covaried. In both analyses, group interacted with stimulus conditions. Participants with dyslexia produced similar numbers of passive sentences as NRs, but these were distributed differently across stimulus conditions. Group differences in performance were related to word reading ability and WM. Conclusions: These findings suggest that dyslexia has pervasive effects on many aspects of sentence production when the language production system is stressed. Based on persistent group differences in fluency, grammaticality and response times even with word reading ability covaried, it appears that manipulating words into sentences can be difficult for adults with dyslexia
Purpose: This multiyear project evaluates the additive effects of a book exchange program and a parent education program for at-risk preschool children. Method: The Raising a Reader program involves weekly rotation of four new books in and out of children's homes as well as partnering preschool classrooms and families with neighborhood libraries. The parent education program involves monthly "Family Nights" in which parents learn about and practice shared reading techniques with their children. Content of the Family Nights program includes strategies often associated with Dialogic Reading interventions. This study reports on the first annual cohort of participants, which included children and families from 23 preschool classrooms. Classrooms were randomly assigned to one of three groups: Control, Raising a Reader, or Raising a Reader plus Family Nights. A variety of school readiness skills were assessed during pretesting and posttesting of 153 children. Most participating children came from ethnic minority and low SES backgrounds. Results: Multilevel ANCOVAs found that children in the Raising a Reader plus Family Nights group demonstrated significantly higher end-of-year language scores on standardized measures of complex language usage. Conclusions: These preliminary findings based on a limited sample, which will ultimately include 126 classrooms and 1000 children, suggest that putting high quality children's literature in the homes of at-risk children and providing parents with optimal shared-reading strategies can be an effective means of fostering children's oral language development, which is critical for reading acquisition and scholastic success.
Purpose: Little is known about vowel spelling development. Without solid developmental knowledge, the scope and sequence of instruction is limited. Method: In this study, we examined prevalence and accuracy of vowel spellings (short, long, diphthong, r-, and l-controlled) in simple morpheme words and the base of multi-morphemic words taken from written samples provided by 120 children in grades one through four. Prevalence was defined as at least one correct use of a vowel spelling; accuracy was the total number of correctly spelled productions divided by total opportunities for each vowel. We also examined differences in error types for short and long vowels (i.e., errors due to insufficient phonological knowledge, orthographic knowledge, or mental orthographic images). Results: Results revealed both within and across grade differences in prevalence, accuracy, and error type. For example, within vowel categories (e.g., long vowels), grade differences for prevalence existed for some vowels (i.e., a, o, u) but not others. Accuracy differences among grades differed by vowel category. Although grades did not differ on error types due to phonological knowledge, grade differences were noted for errors due to orthographic knowledge and mental orthographic images. Conclusion: Discussion will focus on the educational and clinical implications of the findings.
Purpose The purpose of the study was to explore differences and similarities in a mother's mediation styles when telling and reading a story to her kindergarten child. Method Participants were 50 mother-child pairs of middle SES. The children's means ages was 62 months (SD = 7.05). Two sessions, held in the families' homes, were videotaped. The mother read the child a story from an illustrated book with text in one session and told a story from an illustrated book without text in the other session. The order of the sessions was arbitrary. Results Results indicated differences in maternal mediation styles when telling a story compared with reading a story. Mothers mediated on a higher level when telling a story (asking higher-level questions, referring to the meaning of words, etc.). In addition, the children were more involved in interaction when their mothers told them a story compared with interactions when they were read a story. Furthermore, we found that mothers had a mediation style beyond the type of story. Conclusions While until now, the picture book genre has been associated with babies and toddlers, the results of this study indicate that this genre can also be suitable and beneficial for story telling interactions with kindergartners.
Dorit Aram (Tel Aviv University); Tova Most; Hanny Mayfit - Alphabetic and linguistics knowledge of Kindergartners with Hearing Loss: The Contributions of Mother-Child Joint Writing and Storybook Telling
PURPOSE. The study investigated mother-child joint writing and storybook telling as predictors of alphabetic and linguistic knowledge among kindergartners with hearing loss. METHOD. Participants were 30 Israeli kindergartners with hearing loss and their hearing mothers. Early literacy assessments tapped children's alphabetic skills (word writing, word recognition, and letter knowledge) and linguistic skills (phonological awareness, general knowledge, and receptive vocabulary). Each mother helped her child write words and told her child the story of a wordless book. Both interactions underwent videotaping and analysis. RESULTS. We found that maternal writing mediation correlated with the children's alphabetic skills. whereas maternal storybook telling correlated with their linguistic skills, A series of 3-step hierarchical regression analyses revealed that beyond children's age, children's degree of hearing loss, and storybook telling, joint writing predicted word writing (15%), word recognition (31%), and letter knowledge (36%). beyond children's age, children's degree of hearing loss, and joint writing, storybook telling predicted children's phonological awareness (22%), general knowledge (28%), and receptive vocabulary (18%). CONCLUSIONS. The study suggests that both joint writing and storybook reading interactions are productive contexts for promoting the major aspects of early literacy among children with hearing loss. Parents should not give up on any of these contexts.
Diane August (Center for Applied Linguistics); Carlo, Maria; Barr, Chris; Calderon, Margarita - Predictors of Growth in English Reading Comprehension for Young Spanish-Speaking English Language Learners
Purpose This longitudinal study investigated the development of English reading comprehension in eighty-two young Spanish-speaking English-language learners. Methods The study took place in schools implementing Success for All (SFA)/Éxito para Todos curriculum. Across sites, some students received English-only instruction, some bilingual instruction, and some Spanish-only instruction. Students' performance was measured at four time points between the beginning of their third grade year and the end of their fifth grade year. Growth modeling was used to examine between-group differences in English reading comprehension growth and to evaluate the relationship between English reading comprehension growth and initial levels of oral English and Spanish proficiency. Results Results indicated that students who received reading instruction first in Spanish and then transitioned into English (bilingual model) showed significantly higher rates of growth in English reading comprehension than students instructed only in English or only in Spanish. Initial levels of Spanish reading comprehension influenced the rate of growth in English reading comprehension for the bilingually-instructed students, but not for English-only or Spanish-only instructed students. Despite the differences across the groups in time devoted to English instruction, the English-only and bilingual groups did not differ in level of ultimate attainment in 5th grade English reading comprehension. Conclusion Findings highlight the important role of oral proficiency and instrutional context on the development of literacy.
This longitudinal study examined the effects of language status and reading disability on children's spelling development of 2 inflectional morphological markers: plural nouns and regular past tense verbs. The sample consisted of 46 English-as-a-second language (ESL) and 24 English-as-first language (EL1) students (35 non-RD, 35 RD) who were matched on a case-by-case basis on home language background and non-verbal ability. ANOVA'S and correlational statistics were applied to (a) compare the development of morphological skills in spelling plural and past tense markers in ESL and EL1 students, and (b) examine the extent to which correct spelling of stems relates to the spelling of the past tense and plural markers. Results indicated a similar pattern of development of the spelling of the target morphemes in ESLs and EL1s. Regardless of their home languages, RD students lagged behind their non-RD counterparts. Children who spelled the stems correctly were more likely to spell accurately plurals and past tense markers. The findings are discussed in terms of the ESL/EL1 and reading disabilities literature and implications for early assessment and intervention.
Purpose. We wished to assess the longitudinal predictors of reading comprehension skills in children learning to read in a highly regular orthography (Turkish). Method. Fifty -five children (mean age = 67.45 months, SD= 3.64) were assessed at the end of preschool (before reading instruction began), and at the end of grade 1 and grade 2, on a wide range of measures including listening comprehension, grammatical awareness, vocabulary, short-term memory, and working memory. Results. Longitudinal path analyses showed that Grade 1 reading comprehension was predicted by preschool measures of short-term memory, listening comprehension, and grammatical awareness. Grade 2 reading comprehension was predicted by grade 1 reading comprehension (autoregressor), vocabulary, and listening comprehension. In contrast, none of our measures of decoding skill (reading speed, reading accuracy, spelling) were unique predictors of reading comprehension. Conclusions. We relate our results to the Simple View of Reading. In the highly regular Turkish orthography efficiency of decoding skills relatively unimportant as determinants of variations in Reading Comprehension, while in contrast linguistic comprehension skills appear to assume great importance.
The impact of two training programs was evaluated through a repeated measure design with a control group. The first program combined phonological awareness activities with instruction in the alphabetic principle, teaching of Dolch words (May & Rizzardi, 2002) and of commonly used words in school (Scarborough, 2003). The second program combined teaching vocabulary through building larger semantic networks, phonological awareness activities coupled with instruction in the alphabetic principle (offered in the same amount of time of as in the first program), and instruction in commonly used words in school (Scarborough, 2003). The intervention targeted low socio-economic status preschool children (3 ½ to 4 ½), belonging to two language groups: native English speaking children and English-as-a-second language children (ESL). The results show that the children participating in the vocabulary intervention showed better performance on vocabulary than children in both the first training group and the control group (results of time and training interaction for PPVT approaching significance at p = .065). Additionally, children participating in both training groups showed significantly improved phonological awareness skills than children in the control group. The results show that it is possible to effectively train vocabulary with very young at risk children, while effectively training phonological awareness skills.
Purpose: Spina Bifida Myelomeningocele (SBM) is a neurodevelopmental disorder associated with good development of word decoding, but less skilled comprehension. Attention difficulties are also common in SBM. In children with AD/HD but no frank neurological impairment, behavioral ratings of inattention are highly predictive of some academic skills. Relations between reading skills and behavioral ratings of attention in SBM are tested. Method: Ratings of inattention and hyperactivity (parent SNAP Rating Scale) were used to predict reading comprehension (Paragraph Reading from the Test of Reading Comprehension and Passage Comprehension from the WJ-R), reading fluency (TOWRE), and listening comprehension (Making Inferences from the TLC-E) in 101 children with SBM and 37 typically developing controls. Results - After controlling for reading accuracy, neither inattention nor hyperactivity were related to reading comprehension or listening comprehension, but inattention and hyperactivity were related to reading fluency. The model accounted for 66% of the variance in TOWRE scores. Attention was not differentially predictive of any of the reading outcomes in the two groups. Conclusions - The findings are discussed with reference to current models of reading comprehension in SBM, and possible similarities and differences in neurocognitive mechanisms underlying attention and reading difficulties in SBM and neurologically intact individuals with AD/HD.
Roderick Barron (University of Guelph);Lovett, Maureen W; Frijters, Jan C; Lane, Sara; Ashrafhosseini, Diana; Morris, Robin D; Wolf, Maryanne; Sevcik, Rose A - Evaluating a delay versus deficit model of letter-sound learning in children with reading disabilities
Purpose Letter-sounds represent foundation knowledge for reading in English (Treiman et al., 1998) but little is known about disabled readers' letter-sound learning. For example, do their weak phonemic awareness skills impair ability to use the sounds (/b/) embedded in some letter name syllables (/bi/) in learning letter (B)-sound (/b/) associations? Method We identified beginning reader (N=150; CA=5yrs, 1mo.; word ID=5.1 words, word attack=1.6 nonwords) and disabled reader (N=220, CA=7yrs, 6mo; word ID=15.6 words, word attack=1.9 nonwords) groups. Letter-sound knowledge for letters whose names (B = /bi/, L = /εl/) contain sounds in the beginning (/b/) or end positions (/l/) of their letter names, or are not in their letter names, was assessed as well as for letters with one (T) versus two (E) associated sounds. Results Both groups showed mean performance patterns of 1) beginning> end > sound not in letter name and 2) one > two letter sounds. Phonemic awareness accounted for unique variance (7-15%) in each letter-sound category (age, RAN, PPVT-R controlled) for both groups. Conclusions A delay rather than a deficit model was supported; reading disabled and younger, normally developing children display similar patterns of performance and phonemic awareness skills in acquiring letter-sound knowledge.
Purpose: The purpose of this study was to evaluate the nature of reading problems in middle school students. Guiding questions include: What is the prevalence of difficulties in reading comprehension, fluency, and decoding skill? What proportion of struggling readers change over the course of the year? Method: Participants included 1867 students in grades 6-8. Participants were assessed in domains involving decoding, spelling, fluency, and comprehension. Participants were assigned to reading ability groups based on the presence-absence and types of reading deficits that were identified. Results: Results indicate that of students failing the Texas Assessment of Knowledge and Skills, 75% met criteria for difficulty in comprehension, decoding, and/or fluency. 19% showed comprehension difficulties. 81% decoding and/or fluency difficulties with 6% having only decoding difficulties, 12% having only fluency difficulties, 32% having problems in both domains, and 31% having problems in all domains. At year end, 73% exhibited difficulty in decoding and/or fluency with 4% experiencing having only decoding difficulties and 6% only fluency difficulties, 27% having decoding and fluency problems and 22% having problems in all domains. Conclusions: Accuracy and fluency difficulties continue to be problematic in adolescent struggling readers in addition to deficits in reading comprehension.
Purpose This study examined the impact of external disruptions in preschool classrooms on children's literacy growth, focusing particularly on children with low self-regulation. Method Participants included 151 3-4-year-old children from middle-class families. Self-regulation was assessed using Head-to-Toes, in which children are required to do the opposite of a prepotent response. Alphabet recognition was tested using a deck of uppercase-letter alphabet cards. Classrooms were observed for one full school day, and the duration of interruptions was coded in minutes. Interruptions included fire drills, loudspeaker announcements, and external teachers or students entering the room. Hierarchical linear modeling was used to examine the interaction between children's self-regulation and classroom interruptions on spring alphabet recognition. Results For children with low or moderate fall self-regulation scores, increasing classroom interruptions yielded poorer spring alphabet performance (approximately four fewer and two fewer letters, respectively). Children with high fall self-regulation actually showed an increase in alphabet recognition with increasing classroom interruptions. Conclusions These results document the importance of disruptions to the classroom environment for the academic growth of children with low to moderate self-regulation skills. Children with high self-regulation appear to be resilient to classroom interruptions. However, for children who lack this resilience, these interruptions significantly impacted an important emergent literacy skill.
Purpose-The purpose of this study was to determine if students who have been identified as requiring Tier 3 intervention (as defined in the Response to Intervention model) and who have comprehension deficits show progress after a specific intervention designed to stimulate imagery related to comprehension. Method-Students with comprehension deficits were identified from two districts using an RtI approach that included instruction using the Visualizing and Verbalizing for Language Comprehension and Thinking® Program in intensive intervention. The program is closely aligned to Paivio's Dual Coding Theory as a theoretical basis for comprehension instruction that integrates mental imagery and verbal processing. Once assessed with a battery of nationally normed language skill tests, students were placed in homogeneous groups and received daily instruction for at least one term before post-testing. Results-Assessment and any available progress monitoring results will be presented and aligned to a model of comprehension and its relation to decoding skills and imagery. Conclusion-Substantial gains were achieved on decoding and comprehension measures. The progress that these Tier 3 students made suggests that intensive intervention using instruction designed to stimulate the imagery-language connection can have a substantial effect over a relatively short elapsed time.
Many deaf students encounter great difficulty in learning to read (Gallaudet Research Institute, 2004). In general, their reading delay is explained by a deficit in phonological processing (Musselman, 2000). If phonological processing is acknowledged as an essential component of reading development (Demont & Gombert, 2007), recent studies have shown that morphological processing also play an important role in word recognition (Carlisle, 1995; Colé et Fayol, 2000). Considering deaf readers' difficulty in phonological processing, it is interesting to study morphological processing for at least two reasons: words' morphemes are transmissible through sign language, and morphological processing can be seen as a compensatory word recognition strategy. The objective of this study is to investigate deaf readers' knowledge of derivational morphology in written French. We used a word likeness choice task where subjects had to choose which of two pseudo-words resembled the most to a real word. Deaf subjects (n=26) aged 9 to 12 were matched to hearing subjects of the same chronological age (CA) and of the same reading age (RA). Results show that deaf subjects get lower scores than CA, but are comparable to RA. Results are discussed in terms of word recognition strategies in deaf readers and potential educational interventions.
Rebecca S. Betjemann (University of Colorado) ;Erik G.Wilcutt; Richard K.Olson; Sally, J. Wadsworth; Janice M.Keenan; John, C. DeFries; Bruce, F. Pennington - Accounting for the cognitive overlap between reading and attention: A genetic investigation of processing speed
Purpose There is high comorbidity between Reading Disability (RD) and Attention Deficit-Hyperactivity Disorder (ADHD), and research has shown they share common genetic influence. Recent phenotypic findings suggest that common deficits in processing speed may account for much of the cognitive overlap between reading and inattention. Here, we evaluate whether the genetic factors influencing processing speed also account for the genetic overlap between reading and inattention. Method Participants were twins from the Colorado Learning Disabilities Research Center, ages 8 - 18. We used Cholesky decomposition analyses to investigate the multivariate genetic influence on the following factors: Motor Processing Speed, Verbal Processing Speed, Word Reading, Attention Ratings, and Hyperactivity/Impulsivity Ratings. Results The genetic correlation between reading and inattention was .42, supporting previous findings of shared genetic influence between reading and inattention. The Cholesky decomposition showed common genetic influence shared between processing speed, reading, inattention, and hyperactivity/impulsivity. After that shared with processing speed, there was no additional genetic influence shared between reading and inattention. Conclusions Results indicate that genetic overlap between word reading and attention is accounted for by the genetic factors shared with processing speed. This provides evidence at the genetic level that processing speed may underlie the comorbidity between RD and ADHD.
Purpose Many tests presently available for assessing early reading skills have different test blueprints, e.g. varied item content and response format. This research sought to evaluate the evidence of construct validity of four assessments of early reading skills on phonics, vocabulary, and reading comprehension. Methods First (N = 198) and second (N = 201) grade students were assessed in a counterbalanced manner on STAR Early Literacy (SEL), Dynamic Indicators of Basic Early Literacy (DIBELS), Texas Primary Reading Inventory (TPRI), and Group Reading Assessment and Diagnostic Evaluation (GRADE). Confirmatory factor analysis with nested hypothesis testing was used to evaluate measurement equivalence and construct validity. Results Results indicated that the subtests intended to measure a common construct conformed to the hypothesized common factor model for all domains at both grades. Differences between subtest scaling and reliability within a common domain were found. Differences between maturational levels were also found in the underlying measurement model. Conclusion These results suggest that these tests of early reading skills might have different blueprints, but appear to measure a common construct. Differences in scaling and reliability of subtests should provide researchers and educators with important considerations when choosing assessments for an intended use.
Purpose. Passage vocabulary and topic inevitably affect reading speed and accuracy, but current fluency measurement either ignores the issue (single passage administrations) or evades it (median of multiple passage administrations). This study explores whether equating passages and the use of different metrics control for passage differences more effectively than traditional methods by comparing how scaling affects the fluency-comprehension relationship. Method. About 500 U.S. grade 5-8 students read 2-4 sixth-grade DIBELS oral reading fluency passages at three different times and took the GRADE reading achievement test twice. Four fluency metrics are examined: words-per-minute (wpm) and percent accuracy (accuracy) versus correct-words-per minute (cwpm), and errors-per-minute (epm). We use Rasch modeling to equate the passages and derive person measures, and with the latter predict comprehension using HLM. Results. Expository passages are more difficult than narrative passages and that passages also vary in difficulty within genre. Although cwpm is consistently more difficult than wpm, the difference varies by and within genre. Raw counts of errors (epm) appear to yield additional information beyond accuracy, especially among faster readers. Fluency-comprehension relationship investigations are ongoing. Conclusions. Differences between individual passages and metrics lead to questions about the use of cwpm for progress-monitoring and screening of intermediate students.
Jay Blanchard (Arizona State University); Atwill, Kim; Hisrich, Katy - The influence of L1 language proficiency in cross-language transfer: A four-year longitudinal study in L2 immersion-only classrooms (kindergarten-3rd grade)
Purpose. To study the influence of language proficiency on cross-language transfer of L1 (Spanish) language skills to L2 (English) language and literacy skills in L2 immersion-only classrooms. Method. Eighty kindergarteners were randonmly selected from a pool of L1 Spanish-speaking children with minimal L2 proficiency. Children were then tracked and assessed multiple times during their kindergarten thru 3rd grade years on L1 and L2 language and literacy skills. Descriptive, inferential and regressive techniques were used to analyze the data across the four years of the study. Results. Spanish-speaking children with limited L1 proficiency (receptive vocabulary) failed to keep pace with their L1 proficient peers in L2 language and literacy acquisition at all grade levels. Conclusions. The study results indicated that language proficiency in L1 will influence the development of L2 proficiency (Cummins-Developmental Interdependence Model). Thus bilingualism is not a homogeneous characteristic. If policy-makers assume that cross-language transfer will aid L1 to L2 language and literacy acquisition in young children, L2 immersion-only classrooms may not deliver on that assumption.
Bart Boets (Centre for Parenting, Child Welfare and Disabilities, K.U.Leuven, Belgium)Vandermosten, Maaike; Wouters, Jan; Ghesquière, Pol - Dyslexia as an auditory temporal processing deficit!? Results from a longitudinal study.
Purpose: The auditory temporal processing theory postulates that dyslexia results from a deficit in the processing of auditory temporal stimuli. This basic deficit is hypothesized to hamper accurate speech perception, hence disrupting subsequent development of adequate phonological representations and literacy skills. The present longitudinal study aimed to verify the presence of these postulated deficits and causal relations. Methods: 31 children with a family history of dyslexia and 31 matched low-risk controls were followed up from preschool to third grade. Low-level auditory processing, speech-in-noise perception, categorical perception, phonological ability and reading and spelling were assessed at various time moments. Results: A preliminary analysis where children were categorized in groups based on family-risk status and first-grade literacy achievement, indicated that those children showing both the family-risk and the literacy-impairment presented preschool deficits in FM-detection, speech-in-noise perception, phonological awareness and rapid automatic naming. Investigation of the relations further indicated that auditory processing was related to speech perception, which itself was related to phonological awareness and first-grade literacy development. In the current presentation, we will further elaborate these data by analyzing groups based on formal third-grade dyslexia diagnoses, and by modeling longitudinal relations using causal path analysis. Conclusions: This longitudinal study indicates that the auditory deficit generally precedes and predicts the literacy problem.
Purpose: We conducted an ERP experiment on the learning of word meanings by context and definition aimed at two related hypotheses: (1) A word is better integrated into the meaning of a sentence when it is learned through sentence contexts rather than definitions. (2) Integration is also affected by variation in sentence contexts. Method: Adults learned the meaning of unfamiliar words by reading sentences that contained the word or a definition. Sentence contexts were repeated or varied over 4 occurrences. Following learning, participants read sentences that contained a key word, either one they had learned or a matched control word, and judged whether the sentence made sense. Results: The N400, an index of word-to context integration, was reduced when a learned word made sense. However, this reduction was greatest for words that had been experienced in context, and reliably smaller for words experienced in definitions. The N400 was also reduced more following 4 varied sentence contexts than a single repeated context. Conclusions: The results provide clear evidence for both hypotheses. The advantage for learning word meaning from contexts is explained by the convergence of meaning features across various sentences, which allows stronger overlap of episodic memory traces. A similar explanation holds for the positive effect of context variability.
Recent research revealed that about three-quarters of students in special education in the Netherlands develop a substantial reading delay (van Bon, Bouwens & Broeders, 2006). These reading problems are primarily due to the quality of instruction (Vernooy, 2004, 2006). The aim of the methodology "How to teach children reading and spelling" (HTCRS; Schraven, 2004) is to improve teachers' quality of instruction. In the present research, the effectiveness of HTCRS was tested. During the school year 2006-2007, the development of reading and spelling skills of students in Grade 1 who attended schools for special education was investigated. Three schools participated and in one school HTCRS was used. At the start of the school year, reading and spelling skills of the three schools did not differ. At the end of Grade 1, reading and spelling skills of the students who were taught with HTCRS were substantially and significantly better than those of students in schools that did not use the methodology. More importantly, the reading skills of the HTCRS-students were similar to those of students in regular education, and the spelling skills were even better. The reading and spelling skills of the two other schools were at the national level for students in special education. The conclusion is warranted that HTCRS is also effective for students in special education. In my talk, I will explain what makes HTCRS exceptionally effective.
Regina Boulware-Gooden (University of St. Thomas)Carreker, Suzanne; Joshi, R.M. - Does online professional development lead to the same gains in spelling content knowledge for classroom teachers compared to teachers who take the professional development face to face?
Professional development is thought to be an important factor in increasing teacher knowledge in specific content areas (Moats, 1994; McCutchen, Harry, et al., 2002). While this is known to be beneficial for classroom teachers, the demands in the classroom as well as inadequate funding for substitute teachers have made it difficult for teachers to attend workshops during class time. With technology playing a larger roll in student learning outcomes, moving teacher development online seems the logical solution. Classroom teachers enrolled in an online class at Neuhaus Education Center in Scientific Spelling will be given a pre and post spelling test used to calculate their content knowledge of spelling rules and English orthography. Classroom teachers were given the same pre and post test, but their course was taken at the Neuhuas Education Center by trained staff. The study found no difference in teacher knowledge. This provides evidence that professional development given online is as effective as that in a face to face situation. Implications are important. With new initiatives, such as Response to Intervention, teachers' free time becomes increasingly scarce.
Abstract: Purpose This poster reports on a Grade 4/5 intervention (N=84) that investigated whether morphological instruction in the regular classroom affects morphological knowledge and/or vocabulary learning after controlling for initial vocabulary knowledge. Method Participants completed a standardized vocabulary pre-test measure and were randomly assigned by classroom to treatment and control conditions. Treatment consisted of twenty 50-minute classroom sessions addressing the building-block nature of morphemes (bases and affixes) that use consistent spelling despite pronunciation shifts (e.g. sign / signal). The three suffixing patterns, and the role of base words and bound bases (e.g. struct for structure) were taught through a problem-solving approach while the control group continued with regular instruction. Post-tests included measures of written morphological analysis and vocabulary. Results Hierarchical regression analyses controlling for initial vocabulary showed significant instructional effects on morphological analysis and vocabulary knowledge with words that were taught directly and novel words built on bases that were taught in the context of other derivations. No effect was found for words with an untaught base. Conclusions A short morphological intervention in the regular classroom setting was able to build word structure knowledge and vocabulary knowledge for words that were not directly taught.
Purpose - This study investigated the role of language level and non-verbal IQ in children's response to intervention. Method - 152 children participated in a randomised controlled trial evaluating both a Phonology with Reading programme (n=76) and an Oral Language programme (n=76). Participants with weak oral language skills were selected at school entry. Baseline language scores were then screened and 68 children identified with diffuse language impairments i.e. showing impairments on 2 out of 5 language measures. 29 of these children were further classified as SLI (non-verbal IQ in the normal range). Analyses were carried out looking at a) the outcomes of children with diffuse language difficulties compared to those with less severe difficulties, and b) the outcomes of children with SLI compared to those with a general delay. Outcomes were evaluated using a series of language and literacy measures given immediately after the intervention and after a 6 month delay. Results - Children with diffuse language difficulties showed poorer oral language outcomes than those with isolated difficulties, although fewer differences were found for literacy outcomes. Children with SLI showed a significant advantage over children with general delay on tests of literacy development but no difference was found for tests of oral language. Conclusions - These results suggest that response to intervention is mediated by the nature and severity of language impairment and indicate that children with SLI can use general ability to facilitate literacy development.
Purpose: We examined the influence of the home environment on the school readiness skills of young Latino children living in poverty. Method: A quasi-experimental design was used to assess the impacts of a school readiness intervention program comprised of home visiting (birth to 3) and preschool (ages 3-5) components. Results: A baseline model confirmed the direct effects of the intervention (the home language/literacy environment) and two environmental variables (mother's vocabulary and family income) at 36 months on children's later pre-k language, emergent literacy, and social skill outcomes. Two mediation models were tested to examine the indirect effect of the home language/literacy environment on pre-K outcomes: 1) at 36 months and 2) sustained through pre-k. The first model exhibited excellent fit as well as significant indirect paths to all pre-k child outcomes. The second model demonstrated a sustained effect of the intervention through pre-k on only one of four measured school-readiness variables and exhibited only adequate fit. Conclusion: The major impact of the home language/literacy environment on these Latino children's school readiness outcomes occurs before 36 months. Intervention focused on the home environment before age 3 is important to later success. Intervention resources need to be invested earlier rather than later.
Purpose - How does growth in reading skills such as fluency, word reading, and comprehension relate to end of year reading achievement in English and Spanish? To what extent do schools differ in these and what might that suggest about instructional and administrative effectiveness? Method - 771 Texas campuses provided data for 103,763 third grade students in growth tasks and reading performance in English and Spanish. The task data consisted of the Texas Primary Reading Inventory and the Tejas LEE: including reading fluency, word reading, and comprehension. For each task, a multilevel (students within schools) 3-time linear growth model was fit in order to predict end of year performance on the state reading proficiency exam. Results - The percentage of variability at the campus level was 7-10% English, and 10-24% in Spanish. In both languages, initial task status was more highly related to final achievement (r = .28 - .63) than was task growth (r = 0 - .35). Where significant, growth in component tasks was additionally helpful for achievement. Conclusions - While individual student ability and growth are important to achievement, there is substantial variability between campuses in both their growth and achievement rates. Instruction and other campus differences suggest further investigation.
Purpose: investigate the following question: Do struggling readers tutored in a one-on-one intervention format outperform their peers who receive similar intervention in a triad format (group of three)? Method: 125 struggling readers from 13 public schools, assessed as reading between mid-1st and end-2nd grades levels, were randomly assigned to an intervention format (1-on-1 or triad). Students in both formats averaged approximately 43 sessions over one school year. Intervention session time and content were held constant across formats and included assisted reading, word study, and fluency work. Each educator (15 certified teachers and 17 para-educators) had completed one full-year intervention practicum, and participated in another during the current study. Each educator worked with one singleton and one triad to control for teacher effects. The triad structure included rotating a "target student" for each session to provide a facsimile of 1-on-1. Pre-post data were collected for each student on norm-referenced and criterion-referenced measures: Woodcock Passage Reading, Woodcock Word Attack, and research-validated informal passage reading, word recognition automaticity, and spelling inventories. Results: HLM analysis using student data, tutor, and school as the level 1, 2, and 3 variables respectively, found significant differences in only one performance measure: Woodcock Passage Reading (comprehension). Conclusions: The extant literature provides few empirical investigations of the impact of group size on reading intervention effectiveness. In fact, a meta-analysis by Elbaum et al., (2000) identified only 2 studies (unpublished doctoral dissertations) that addressed this issue. The current study suggests that a triad format may approach a 1-on-1 tutorial in effectiveness. Contributions to this finding may have included the intensive professional development provided and a structure that prioritized regular, intensive work with a target student in the triad.
Purpose The preschool home literacy environment (HLE) has been linked to the development of a variety of literacy and oral language skills (e.g., Phillips & Lonigan, 2005). Most research in this area has examined older preschoolers. The origins and development of the HLE have not been extensively explored. We examined the HLE provided to very young preschoolers. Method HLE surveys (e.g., literacy activities with child, parental leisure reading habits) and adult and children's author recognition checklists were obtained from 262 mothers of children less than 19 months old. Results The children were exposed to a wide range of literacy activities and experiences, but many had relatively little literacy exposure. Shared reading was the most common literacy activity. Few children were exposed to activities specifically intended to teach literacy knowledge or skill, but approximately 15% of the older children were regularly taught using alphabet letter blocks or other techniques. Maternal reading habits and estimated reading ability were associated with the provision of more frequent and varied literacy experiences. Conclusions We demonstrated that differences in HLE activities and resources provided to preschoolers may begin at very young ages. Differences in HLE provision were related to parental ability, beliefs, and reading habits.
Purpose-To examine the effects of vocabulary-focused instruction and strategies-focused instruction on the vocabulary development and comprehension skills of students in primary grades who are adequate decoders, but non-proficient comprehenders of text. Methods- Using a pretest-posttest design, primary grade students (N=60), randomly assigned to two groups, received 32 sessions over eight weeks, of either vocabulary-focused instruction or strategies-focused instruction. Pretest and posttest data were collected using measures of expressive vocabulary, receptive vocabulary, and reading comprehension. Data were analyzed using a series of analyses of covariance with the pretests as covariates. Within and between group analyses were conducted. Results-A series of analyses of covariance revealed no statistically significant differences between groups on measures of vocabulary and comprehension. An analysis of covariance revealed a statistically significant difference between groups on a passage comprehension measure, favoring the vocabulary-focused group. A series of analyses of covariance revealed statistically significant differences within groups on measures of vocabulary and comprehension. Conclusion- Explicit vocabulary instruction and comprehension strategies instruction have a significant impact on both the reading comprehension and vocabulary of children in primary grades. With a more experienced adult or peer providing scaffolding, students' abilities can be expanded beyond what they are able to do alone.
Purpose The purpose of this study was to explore patterns of within-group variability in the emergent literacy skills of preschoolers at risk for academic difficulties. We used the person-oriented approach of cluster analysis to identify a normative taxonomy of the most common profiles of emergent literacy (i.e., oral language and code-related skills). We then examined the extent to which teacher report of emergent literacy differentiated profiles. Methods Participants were 492 preschoolers enrolled in publicly-funded programs (42-60 months). In the fall, children were administered eight measures of emergent literacy: four oral language measures (i.e., expressive/receptive grammar and vocabulary) and four code-related measures (i.e., alphabet knowledge, name writing, print concepts, rhyme). Controlling for age, hierarchical-agglomerative and K-means cluster analysis procedures were employed. Profiles were then compared on midyear teacher report, using a series of ANOVAs. Results Five internally-validated profiles emerged: highest overall (prevalence = 14%); three profiles with average oral language and differential code-related abilities (16%, 24%; 23%); and lowest oral language with broad code-related weaknesses (23%). Teacher report provided convergent evidence of external validity. Conclusions This study highlights the considerable heterogeneity of emergent literacy abilities within an "at-risk" group. Such information may be useful for characterizing intervention responsiveness in young children.
Purpose: Children with inattention and hyperactivity frequently have poor reading and language comprehension. We sought to determine whether inattention or hyperactivity make separate contributions to reading and listening comprehension. Method: In Study One, we explored the relations between teacher ratings of inattention and hyperactivity and word reading and reading comprehension in 8-11 year-olds. In Study Two, we compared reading and listening comprehension in three experimental groups selected for either high levels of inattention, or hyperactivity, or both. Their performance was compared with controls with age-appropriate levels of attention and (hyper)activity. Results: In Study One we found a strong relation between attention and both word reading and reading comprehension. Attention explained a small but unique proportion of variance in reading comprehension after word reading had been controlled. There was no evidence that (hyper)activity was related to reading. In Study Two, the hyperactive-only group did not perform more poorly than controls overall, but all three experimental groups obtained lower scores for listening than reading comprehension; the controls did not show this pattern of performance. Conclusions: Reading comprehension deficits in children with inattention are partly independent of word reading skill; children with inattention and hyperactivity are at risk of listening comprehension failure.
PURPOSE Adolescents (6th-8th grades) are in dire need of research based reading instruction designed specifically for their unique needs. While research has answered many question concerning specific instructional techniques necessary for the remediation of each sub-skill, it has not provided enough information concerning the most effective way to sequence and balance the instruction of each sub-skill to maximize responsiveness to instruction. METHOD The purpose of this study was to explore the best way to organize the instructional sequencing of the reading sub-skills to maximize adolescent's (N = 90) responsiveness to instruction. Nine classrooms were randomly assigned to three different sub-skill instructional schedules; a) Alternating (ALT), b) Integrated (ISS), and c) Sequential (SSS). All instruction occurred five days a week, 45 minutes a day for 25 weeks for a total of 97 hours of reading instruction. The reading sub-tests of the Woodcock-Johnson-Revised, the Gray Silent Reading Test, and ORF passages were used to assess gains. RESULTS Results demonstrated the SSS schedule significantly outperformed the ALT and ISS schedules in decoding, spelling and comprehension skill acquisition. The reading fluency assessments showed differing results between schedules. CONCLUSIONS Findings are discussed in regards to optimal instructional sequencing of reading sub-skills to maximize responsiveness to instruction for adolescents.
Purpose - To investigate whether student gains in reading and background knowledge related to specific types of science instruction and whether these effects depended upon students' initial language, literacy, and background knowledge skills (i.e., child-by-instruction interactions). Method - 88 second grade students in six schools, 33 classrooms of an urban Midwest school district participated in this study. Classrooms were formally observed three times per year (fall, winter, and spring). Children's language, literacy, and general academic knowledge were assessed in the fall and again in spring. Results - On average, teachers spent 92.4 minutes/day in language arts instruction (range 52.7 to 154.7 minutes/day) and 10.94 minutes/day in science instruction (range 0 to 75 minutes/day). Science instruction promoted rather than inhibited literacy skill growth for most students; however, the effect depended on students' fall vocabulary, reading, and background knowledge skills. Specific types of science instruction were associated with greater growth in children's reading, vocabulary, and background knowledge with child-by-instruction interaction effects. Conclusions - Results highlight the opportunity for even greater gains when instruction is individualized for students with respect to their current academic skills and knowledge. Data from this year's Individualizing Science Instruction pilot study will also be presented.
Purpose: The present study examined spoken language and emergent literacy skills in preschool children with cochlear implants. Method: Fourteen preschoolers, who received cochlear implants between 12 and 27 months, participated in the study. They were ages 48-59 months (5 female, 9 male). Children were administered six subtests of the Assessment of Language and Literacy (ALL): letter knowledge, rhyme knowledge, basic concepts, receptive vocabulary, parallel sentence production, and listening comprehension. Their scores were compared to those of 14 typically developing preschoolers from the ALL normative sample, matched for age, gender, SES, grade, ethnicity, and region of the country. Results: As a group, the children with cochlear implants performed more poorly than the children with typical language on the language composite but not on the emergent literacy composite. Letter knowledge was age-appropriate for most of the children with cochlear implants. In contrast, over 40% of the children with cochlear implants scored in the clinical range for rhyme knowledge (-1.5 SD). Conclusions: Rhyme knowledge, in addition to spoken language skills, should be explicitly taught to preschoolers with cochlear implants in order to secure a foundation for later reading development.
The purpose of this study was to determine the extent to which early elementary teachers' knowledge about reading contributed to their students' gains in word analysis and reading comprehension across a year. Method: The data were derived from the Reading First program in Michigan, through which high poverty, low achieving schools received funding to support improvement of early reading instruction. The 954 participating teachers instructed 17,703 students and were nested in 140 schools. Reading achievement for grade 1-3 students included fall DIBELS and spring subtest from the Iowa Test of Basic Skills. Teacher data were collected from a survey that included teacher information, an assessment of their knowledge about reading (Language and Reading Concepts, LRC) and a self-reported survey of practices. School-level data came from the Michigan Department of Education. Results: We stratified teachers on the basis of their propensity to have high/low teacher knowledge and modeled the effect of knowledge on student achievement through six (two per grade) three-level hierarchical linear models. Our analyses revealed significant evidence that teacher knowledge affected students' reading achievement. We found small but statistically significant effect sizes in both first grade subtests and in second grade for word analysis. A slightly smaller and statistically insignificant teacher knowledge effect on the grade two reading comprehension emerged. In grade three we found no evidence of an effect of teacher knowledge on the reading subtests. Fall measures of prior reading ability from DIBELS showed the strongest relationship to both outcomes in all grades. Students' race, economic status, and disability status played prominent roles in predicting both outcomes in all three grades. Conclusion: Teacher knowledge was linked to student growth on both ITBS subtests (word analysis and reading comprehension) in grades one and two. Third grade results are probably affected by the content of the teacher knowledge measure, which placed greater emphasis on decoding than comprehension.
Purpose There is substantial evidence suggesting an overlap between speech difficulties and literacy difficulties. The present study aims to elucidate the nature of this relationship by comparing four different groups of 4-6 year old children with: speech difficulties; a dyslexic relative; low nonword repetition; or no difficulties. Method Data collection is ongoing: currently there are 58 typically developing children and 47 children in total in the three risk groups. Ten of the twenty-six children with a family history of dyslexia (38.5%) had also received speech therapy, confirming a substantial overlap between groups. The children have completed a full battery of tasks measuring speech, language, literacy and phonological processing. They will be retested six months later to provide a measure of progress in literacy. Results The risk groups differ from controls in terms of their literacy, phoneme awareness, speech accuracy and language development. They also showed differences on tasks aimed at measuring phonological processing: nonsense word learning, mispronunciation detection and word classification. We will assess which of the measures at time 1 predicts poor literacy progress, and could perhaps be used as markers for possible future literacy difficulties. Conclusions This research will provide information as to the aspects of speech and language development which are most important for literacy development.
Kelly Cartwright (Christopher Newport University); Coppage, Elizabeth; Guiffre, Heather; Scarano, Kathleen - A Comparison of Metacognitive Skills and Cognitive Flexibility in Good and Poor Comprehenders
Purpose: Metacognitive skills and cognitive flexibility (an aspect of executive control that involves the ability to coordinate flexibly multiple features of tasks) contribute uniquely to comprehension (Author, 2002, 2007; Nagy, 2007). Good and poor comprehenders exhibit significantly different profiles of cognitive abilities (Yuill & Oakhill, 1991). This study was designed to extend this work and examine metacognitive abilities and cognitive flexibility in good and poor comprehenders. Method: Sixty-five university students completed the Woodcock Reading Mastery Test (WRMT) Passage Comprehension subtest to assess comprehension, WRMT Word Attack to assess decoding ability, WRMT Word Identification to assess word reading, Kaufman Brief Intelligence Test Verbal Subscale, measures of syntactic and phonemic awareness, and two measures of cognitive flexibility, graphophonological-semantic and semantic-syntactic, to assess participants' ability to coordinate flexibly these aspects of print. Twenty four low and 24 high comprehenders were matched on WRMT Word Attack for comparison. Results: Good and poor comprehenders differed significantly on all measures. Differences in syntactic awareness, phonemic awareness, and both measures of cognitive flexibility remained significant when verbal ability and word identification were controlled. Conclusion: These data complement existing work by demonstrating additional critical cognitive differences in metacognitive awareness and cognitive flexibility between good and poor comprehenders.
Purpose This study investigated the role of morphological awareness in vocabulary acquisition among Chinese-English bilinguals. Methods Participants included 48 kindergarteners, 34 first graders, 31 second graders, 37 fourth graders and 40 seven graders from a large Canadian city. All the children were of Chinese descent. They received parallel morphological awareness and vocabulary measures in Chinese and English, and a battery of linguistic and cognitive measures in English only. Result Regression analyses were carried out in each grade. To predict Chinese vocabulary, Raven and mother's education were entered in the first step, followed by phonological awareness in the second step, and Chinese morphological awareness in the final step. To predict English vocabulary, the same variables were entered in the first two steps, followed by English morphological awareness. Morphological awareness explained unique variance in Chinese vocabulary in every grade, whereas phonological awareness was not significant. In contrast, phonological awareness predicted English vocabulary in grade two, four, and seven, and English morphological awareness was significant only in grade four and seven. Conclusion Our results highlight the importance of morphological awareness in Chinese vocabulary acquisition. Moreover, we demonstrate that the importance of morphological awareness increases with grade level in English.
Paula Clarke (University of York, UK); Emma Truelove; Charles Hulme; Margaret Snowling - The York READing for MEaning Project: A randomised controlled trial of interventions to improve children's reading and language comprehension skills
Children with reading comprehension difficulties (poor comprehenders) typically show a wide range of oral language difficulties including weaknesses in semantic and grammatical skills and experience problems with the comprehension of both oral and written language. We report the results of a randomised controlled trial evaluating three theoretically motivated interventions designed to improve the reading and oral language comprehension skills of poor comprehenders. Children (n = 160) who were identified with poor reading comprehension skills in relation to their decoding ability were selected for inclusion in the study and randomly allocated to one of four groups, each of whom received instruction over two 10-week blocks: oral language training (OL), text comprehension training (TC), combined oral language and text comprehension training (COM) and a delayed treatment control (C). The OL training consisted of activities designed to improve vocabulary, listening comprehension, figurative language and spoken narrative and the TC training included activities to target metacognitive strategy use, reading comprehension, inferencing from text and written narrative. The COM training used a combination of all of the activities in the OL and TC programs. Our results show positive effects of the interventions on a range of measures of oral language and reading comprehension.
Leen Cleuren (Centre for Parenting, Child Welfare and Disabilities - K.U.Leuven); Kong, Yuk On; Latacz, Lukas; Verhelst, Werner; Ghesquière, Pol - Evaluation of a Speech Synthesis Based Reading Tutor to Speed up Reading in Poor Readers
Purpose. Within the SPACE project, a fully automated Dutch reading tutor, able to (1) track the child's reading progress and detect reading errors precisely by means of a speech recognizer; and (2) act as a fluent reading model and give adequate feedback by means of a speech synthesizer, is being developed. In order to explore the effectiveness of this tutor to improve a child's reading accuracy and fluency, an intervention study was carried out. Method. A pretest/posttest 4-week treatment with control group (no treatment) design was used. 10 pairs of reading disabled Flemish regular school children (grade 2-6) participated. Treatment consisted of non-repetitive reading sessions where the child could ask for phonic analysis help. When a child made an error but proceeded without asking for help, the speech synthesizer prompted phonic analysis feedback to the child in order to prevent the child from moving on before sounding out the word correctly. Results and Conclusions. In this paper we will discuss the effectiveness of the reading tutor mediated intervention sessions by comparing pre- and posttest reading skills for the experimental versus the control group. We will also have a closer look at how reading accuracy and fluency gradually improves with each reading session.
Purpose: Research has shown that some children with specific reading comprehension difficulties have concomitant impairments in listening comprehension and weaknesses in dealing with the non-phonological aspects of language processing. This longitudinal prospective study aims to investigate the hypothesis that weak oral language skills early in development may be a risk factor for children who go on to have specific reading comprehension difficulties in mid-childhood. Method: 250 children were recruited at age 4.5 years. Their cognitive, language and literacy skills were assessed at 4 time points over a 3-year period. 19 poor comprehenders and 19 control children, matched for age, nonverbal IQ and decoding, were selected at time 4 when the children were 7.5 years old. We examined and compared the early language abilities of these two groups. Results: The results support the hypothesis and indicate that the poor comprehenders who emerged at time 4 had a history of weaker non-phonological language abilities. There was also evidence to suggest that poor comprehenders had weaker pre-literacy skills than control children.. Conclusions: This study contributes to our understanding of what the early precursors or risk factors may be for children who go on to have specific difficulties with reading comprehension. The results also have implications for interventions for poor comprehenders.
The comprehension section of the Nelson-Denny Reading Test (NDRT; Brown, Fishco, & Hanna, 1993) is widely used to assess the reading comprehension skills of U.S. adolescents and adults. For example, the Law School Admission Council (LSAC) identifies the NDRT (Form G or H) as the "preferred measure" of reading comprehension for documentation of cognitive impairments (e.g., dyslexia) in students hoping to take the Law School Admission Test (LSAT) with accommodations (www.lsac.org). But is the NDRT a valid measure of reading comprehension? In the present study, we explored this question by asking normally-achieving university students (n=115) to answer the NDRT (Form G) multiple-choice comprehension questions without reading the passages. Overall accuracy rates were well above chance, and the students were particularly successful with inferential questions. These results raise serious validity concerns about the NDRT, which may be measuring verbal abilities (e.g., background knowledge; reasoning) rather than the intended construct.
Purpose This study explores the frequency and language of home reading practices of ELL preschoolers within a larger study on English vocabulary acquisition from storybook reading. It examines whether the frequency of home reading supports target word learning from storybook reading and how the language of home reading contributes to baseline L1 and L2 vocabulary. Method Seventy, 4-and 5-year-old speakers of Portuguese (L1) who were learning English (L2) were pretested in L1 and L2 receptive vocabulary, assigned to groups, and read stories with target vocabulary. Parents completed questionnaires about the content, frequency, and language(s) of home reading. Results Frequency per week of reading at home was a significant predictor of variance in target word learning. Children who were read to in Portuguese only had significantly higher baseline L1 scores than those read to in both languages or English only. Children who were read to in English only had significantly higher baseline L2 scores than those read to in both languages or Portuguese only. Conclusions The frequency of home reading helps L2 learning. Reading in a single language contributes significantly to children's vocabulary in that language. Findings are helpful for understanding how home practices support first and second language vocabulary learning.
Penny Collins (University of California, Irvine);Kemp, Susan - Predictors of Third Grade Reading Comprehension Among English Learners and Native English Speakers Predictors of Third Grade Reading Comprehension Among English Learners and Native English Speakers
This study examines the contributions of word reading, vocabulary, reading-related skills and reading related behaviors to third grade students' performance in reading comprehension. 186 third grade students from three suburban schools participated, which consisted of 126 native English speakers and 42 students who were English learners. At the beginning of the school year, students were administered standardized tests of reading fluency, reading comprehension, and vocabulary, in addition to measures of phonological awareness, morphological awareness, orthographic processing, and print exposure. The presentation will focus on an examination of (a) whether the relationships between basic reading processes and reading comprehension differ for native English speakers and English learners, and (b) whether the relationships between these skills and reading comprehension varies as a function of proficiency in English (as determined by the California English Language Development Test) among English learners. The generalizability of models of reading comprehension that have been developed within one's L1 to students who are learning to reading in their L2 will be discussed.
We assessed the value of Decoding Dynamic Assessment in predicting reading outcomes 12 weeks later in 105 1st graders. Children we pretested, progress monitored with Word Identification Fluency (WIF), and posttested. Pretest DA correlated significantly with pretest Word ID (r=.65) and Word Attack (r=.78). DA, WIF intercept, and WIF slope were significant predictors of posttest Word ID, Word Attack, and passage fluency, explaining 73%, 66%, and 76% of variance, respectively. After controlling for WIF growth, DA explained 12%, 20%, and 2% unique variance of Word ID, Word Attack, and passage fluency, indicating potential for DA as a predictor of learning to read in first grade.
Carol McDonald Connor (Arizona State University, Learning Sciences Institute, and the Florida Center for Reading Research)Morrison, Frederick J.; Fishman, Barry; Schatschneider, Christopher; Underwood, Phyllis; Crowe, Elizabeth - Individualizing Student Instruction in Two Studies: Technology, Instruction and First Graders' Literacy Outcomes
Purpose: Investigating the effect of child X instruction interactions on first graders' literacy outcomes, we focus on the results of the second of two randomized control field trials conducted during the 2005-2006 and 2006-2007 school years. In both studies, we examined teachers' use of Assessment-to-instruction (A2i) software, instruction in the classroom, and compared technology use for teachers who implemented individualized instruction with greater or lesser fidelity. Method: Ethnically/socioeconomically diverse schools were matched and randomly assigned to intervention or control conditions (n=10 & 7 schools, 46 & 26 teachers and 651 & 443 students in Studies 1 & 2 respectively). Intervention teachers used A2i software and received professional development (PD). In year 2, A2i was enhanced to include online PD resources and student-assessment graphs. Results: Replicating and extending Study 1, Study 2 HLM revealed that students in the intervention classrooms demonstrated significantly greater spring word reading skills compared to students in control classrooms (coefficient=2.99, p=.042, Effect size=.20) controlling for fall scores. There were differences in implementation and technology use within and across the two studies, which affected students' outcomes. Conclusions: Elucidating the practices of teachers who showed high levels of fidelity in these studies provide further evidence of the impact of child X instruction interactions.
Purpose: To design and evaluate a measure of orthographic knowledge. The measurement of orthographic skill has not advanced to the same degree as the measurement of other reading related skills, such as phonological skills. Method: To address some of the concerns with existing measures of orthographic skill (see Hagiliassis, Pratt, & Johnston, 2006), a new measure was constructed, using nonword pairs that sound similar but differ in terms of the frequency of the letter combinations (e.g., tays - tayz). Using both accuracy and response times, reliability analyses were conducted for both adult and child readers. To examine validity, performance was compared to performance on two "gold standard" measures of orthographic skill: the Orthographic Choice Task, which measures word specific orthographic knowledge, and the Word-likeness Task, which measures knowledge of graphotactic regularities. To further examine both convergent and divergent validity, a variety of measures of related domains, including reading, spelling, math, morphology, and phonological skill were also administered. Results: Evidence was found for both the reliability and validity of this measure. Conclusions: This measure of orthographic knowledge may measure orthographic knowledge with a greater degree of purity than previous measures, and contributes to our ability to measure and understand this core component of skilled reading
Purpose. This study follows up on Cormier and Desrochers (2007) who found that children take away more easily a phoneme from a stimulus word when the answer is a word than when it is a pseudoword. It had two objectives: to examine (a) whether this effect can be found in all types of phonemic elision tasks and whether (b) individual differences in working memory moderate it. Method. Children in grades 1, 3 and 5 (n = 112) did eight phoneme elision tasks (ten items each) in one of two lists: one in which the response was a word and another in which it was a pseudoword. The position of the target phoneme (e.g., beginning and end of CVC words, first and second position of CCVC words, etc.) varied across tasks. Children also performed a working memory task in which they repeated the last words in sets of sentences. Results. The word advantage was significant in all phoneme elision tasks, 3.61 < F's (1, 102) < 20.52, .06 > p's < .001, .03 < η2's < .17, except (C)VC and CV/C(C)V tasks, and after having controlled for the significant effects of working memory, 12.92 < F's (1, 102) < 81.14, all p's < .001, .10 < η2's < .45. Conclusion. This word advantage, a lexicality effect, in phoneme elision tasks raises questions about the nature of the relationship between phoneme elision and word decoding in school aged children.
Piers Cornelissen (Department of Psychology, University of York, UK)Hansen, Peter; Kringelback, Morten; Whitney, Carol; Holliday, Ian - Early activation in Broca's area during visual word recognition: evidence from MEG
Purpose: Recent neurophysiological evidence has suggested that the interaction between visual and speech areas starts very early (~130ms) during visual word recognition (e.g. Pammer et al., 2004). Here we used magnetoencephalography (MEG) to seek evidence for early activation of posterior superior IFG (Broca's area) in response to visually presented words. Method. Participants continuously monitored and responded to any colour change in a fixation cross while faces and 5-letter strings were presented centrally. We used synthetic aperture magnetometry (SAM) to calculate the timeseries in six regions of interest: left and right middle occipital gyrus (MOG), left IFG (pars opercularis) and its RH homologue, and the visual word form area in the left mid-fusiform gyrus (VWFA) and its RH homologue. We carried out frequency and amplitude domain analyses on the resultant data. Results: We found that strings, but not faces, activated the posterior superior inferior frontal / precentral gyrus at the same time as the VWFA. This frontal response was significantly greater for words than for consonant strings. Conclusions: These findings confirm very early interaction between the vision and language domains during visual word recognition. The implications for neurobiological models of orthographic processing will be discussed.
Purpose: Two studies were conducted to investigate the reading and spelling abilities of 72 Brazilian youth and adults with little or no formal education, varying from 16 to 81 years of age. Method: The first study investigated the relationship between phonological awareness (PA), rapid serial naming (RSN), and verbal working memory (VWM), on the one hand, and the ability to read and spell words, on the other. In a second study, we compared our participants who had started to read (N = 61) with typical 1st to 3rd grade children of the same reading level, with regard to performance on the PA, RSN and VWM measures. Results: Both PA and RSN contributed unique variation to literacy ability, even after differences in arithmetic skills (used as a proxy to educational level) and letter name knowledge were partialed out. In contrast to what has been found in studies with US adult literacy students (e.g., Greenberg et al., 1997), no differences were found between the adults and the children in the PA measures. Furthermore, the adults performed relatively better on most of the RSN measures. Conclusions: Possible reasons for these discrepancies (e.g., differences in participants' reading level and/or writing systems) are discussed.
Purpose: This research investigates the use of rhythm assessment as a tool for evaluating pre-reading ability in young children. The study also examines the ability of rhythm skills to predict reading ability in early readers. Method: Sixty children aged 3-6 have been followed longitudinally. The group was assessed using standardized measures of phonological awareness (PIPA, CTOPP), vocabulary (PPVT), spelling (PALS), intelligence (KBIT-2), and reading (Word-ID, Word-Attack). Both children's receptive rhythm ability (discriminating between two 'beats') and expressive rhythm ability (tapping in time to a beat) have also been examined. Results: We report results from the first 3 timepoints of the study (2 academic years). Both receptive rhythm and expressive rhythm show associations with children's rhyming ability, predicting approximately 50% of the variance. Letter-sound awareness is strongly associated with other phonemic awareness measures. As children begin to read, both letter-sound awareness and rhythm predict separate amounts of variance. Thus, both rhythm awareness and letter-sound awareness are important - but separable -- predictors of reading readiness. Conclusions: These results suggest rhythm can be used as an early predictor of reading ability. We anticipate that this could help to identify children at risk for reading difficulties before they even begin to fail.
Michael Coyne (University of Connecticut); Deborah C. Simmons; Shanna Hagan-Burke; Oiman Kwok; Athena Lentini - Early Intervention in the Real World: An Experimental Evaluation of the Early Reading Intervention Program
This presentation will describe a program of research dedicated to evaluating the efficacy of a kindergarten beginning reading intervention designed to help children at risk for reading disability establish critical phonemic and alphabetic skills. An 8-month randomized control trial with 57 school-based interventionists and 209 kindergarten students was conducted in Texas and Connecticut. Screening measures of letter knowledge and phonological awareness were used identify the four most at risk students from each kindergarten classroom at participating schools. Each group of four was randomly assigned to implement wither the experimental intervention or school designed intervention. Both treatment and comparison groups met for 30 minutes per day for approximately 130 days. Initial results suggest statistically and educationally significant benefits of the intervention on phonologic, word attack, and spelling measures. Findings will be discussed in the context of conducting large scale research studies in real world settings and scaling up early intervention efforts.
Purpose: This study examined relationships between dialect shifting toward Standard American English and reading achievement for African American English (AAE) speaking students. Dialect shifting was determined by comparing feature production rates during generation of an oral and a written narrative. It was hypothesized that rates in the oral narrative task would bear no direct relationship to standardized reading scores whereas rates in writing would be lower, and would be inversely related to reading outcomes. Method: Participants were 165 typically-developing African American first-fifth graders. Measures of feature production rates (dialect density; DDM) during oral and written narratives and of receptive and expressive vocabulary and sentence structure were collected and examined for relationships to standardized reading scores using Structural Equation Modeling. Results: DDMs decreased significantly between the oral and written narratives providing strong evidence of dialect shifting. The other oral language measures clustered into a Comprehension and a Production Factor. After controlling for socio-economic status and the two language factors, written DDMs exerted both direct and indirect effects on reading scores whereas oral DDMs exerted only an indirect effect through the Comprehension Factor. Conclusions: The results indicate that the ability to dialect shift contributes positively to the reading achievement of AAE-speaking students.
Purpose: We wished to test a popular model of motivation, cognition, and reading comprehension (Greene, Miller, Crowson, Duke, & Akey, 2004) and compare its fit in four countries. We used the large international PISA dataset to compare results: are the effects in the model the same in all four countries? Method: We used a hierarchical linear modeling to test an a priori model while accounting for the fact that sampled children are nested within schools. Participants were 15-year-old students: n = 2129 from the US, n = 2860 from Australia, n = 2308 from Norway, and n = 2757 from Korea. Materials were the PISA-designed and translated reading comprehension measure (see Chiu, Chow, & McBride-Chang, 2007) and various questions from the PISA motivation questionnaire (see Marsh, Hau, Artelt, Baumert, & Peschar, 2006). These were both theoretically-based, developed by international teams, extensively validated, and show Cronbach's alpha reliabilities above .85. In our analyses, we tested the effects shown in Greene et al's (2004) model using regression in HLM, and standardized the results for each country. Results: Results showed different patterns of effects across the four countries. Some results were surprising-for example, Control Strategies showed very different effects on comprehension across the four countries after accounting for self-concept (effect sizes from .63* to -.05ns). Other results were as predicted, e.g., very large effects for Performance-Approach Goals on Control Strategies in Korea (effect size = 1.38). Conclusions: Prior research that has been reported as culturally universal may in fact be reporting culturally-specific patterns.
Purpose: Previous research suggested a relationship between reading development and types of word association responses. Adults respond with more paradigmatic responses, same form responses, and young children give more syntagmatic responses, or words that follow other words in discourse. After reading acquisition children tend to give more paradigmatic responses. The present research examined this relationship with Hierarchical Linear Modeling and Structural Equation Modeling in a longitudinal study. Method: Some 130 children were tested in preschool and followed through third grade when 97 children remained in the study. They were given the Woodcock Reading Mastery Test, vocabulary tests, and word association tests. The word association responses were assessed as to the number of paradigmatic and syntagmatic responses. Results: HLM growth trajectories found that paradigmatic responding was significantly related to the Woodcock Word Identification scores, but that syntagmatic responding was not. The growth of paradigmatic responding was not related to vocabulary in kindergarten nor first grade, but there was a significant relationship in second grade. Further analyses related word associations to the various components of reading. Conclusions: These findings suggest complex relationships between reading development, vocabulary development, and comprehension.
Purpose The purpose of this study is to investigate the constellation of language and literacy skills that comprise proficient reading comprehension (+1SD) among fifth graders. Research on above average students has predominately focused on the affective domain, whereas the nature of achievement among these students is vague and in need of clarification. Method 989 fifth graders (NICHD Study of Early Child Care and Youth Development) comprise this academically, economically and culturally diverse sample. Picture vocabulary, letter/word identification, and passage comprehension (Woodcock Johnson) were assessed at the end of fifth grade. Latent class analysis (Mplus) was used to model class membership for all fifth graders in our sample, which was not limited to students who have already been classified as above average. Results Preliminary analyses of class memberships reveal that above average reading comprehension clusters differentially with letter/word identification and vocabulary. This suggests that differential skills are present in students who, overall, appear to be ahead in reading. Implications Differences in achievement among highly proficient readers may indicate areas of reading strengths and needs that are overlooked by broader assessments of ability or achievement and cut points that arbitrarily establish student classifications.
Purpose: The study investigated the affect of two Computer Aided Language Learning (CALL) programs in comparison to regular ELL instruction on language and literacy skill development. Method: Seventy-seven ELL students from grades 4 to 6 participated in a pretest-intervention-posttest design and were randomize assigned to one of thee conditions: 1) CALL-A listens to students reading and gives feedback on misread words, 2) CALL-B displays text while it reads to students, and 3) regular ELL class instruction. Students were administered language and literacy measures. Results: For all conditions, students made significant pretest to posttest gains across measures, with the exception of phonological awareness for which a trend was found. The interaction between stage of English language and literacy development and language and literacy skill gains was also examined. Regardless of treatment condition, early-ELLs made significantly greater gains in word level reading, and there was a trend towards intermediate-ELLs having greater gains in listening comprehension and reading comprehension. Conclusions: Though ELL students using CALL received less teacher instruction, they made the same gains as regular ELL class instruction. When teachers do not have time for one-on-one work, ELL students can us CALL as a supportive role in language and literacy skill development.
Neurobiological and behavioral mechanisms by which individuals learn the pronunciation and meaning of words is relevant to reading development. Twenty adult skilled readers learned the pronunciation and meaning to 40 pseudowords through two training methods: 20 words were learned via isolation training, while 20 were learned via context training. Faster RTs during the learning phases were observed for the isolation training condition; however, after fluency practice, both conditions produced similar RTs. Neurobiological findings revealed that trained words showed activation patterns similar to those seen for low frequency words; furthermore, context training resulted in greater consolidation in the left occipito-temporal region.
Deaf people's reading deficit is generally explained by their nonefficient word recognition processes (Colin et al., 2007), and specifically by poor phonological processing (Musselman, 2000). Another frequent explanation for deaf readers' reading deficit is linked to the fact that for those who primarily use a sign language to communicate, the majority language (French in our case) is considered a second language (Strong & Prinz, 2000). In order to test this last explanation, we studied phonological processing of written syllabic structures in deaf adults (n=35) matched on reading-level to adult Chinese learners of French as a L2 (n=21), and compared to expert readers of French (n=40). We used two computerized tasks of syllabic sensitivity: one epilinguistic task and one metalinguistic task (cf. Gombert, 1992). The results showed that all groups performed better on the epilinguistic task than on the metalinguistic task. Interestingly, on both tasks, deaf readers got lower scores than the Chinese and the control readers. The Chinese and the control readers did not significantly differ. These results are discussed in terms of language analysis and cognitive control (cf. Bialystok, 2001) involved in the tasks and in terms of the role of syllabic sensitivity in reading in deaf people.
Purpose: This pilot study examines whether accuracy, as a measure separate from correct-words-per-minute (WCPM), contributes uniquely to predicting students' performance on high-stakes reading measures. Method: This study included two cohorts of students in second through third grades (N = 112). Both cohorts were followed for two years. All students were administered DIBELS oral reading fluency (ORF) passages four times a year for each of the two years. Outcome measures were the Maryland State Assessment (MSA) administered at the end of third grade and two other norm-referenced measures. Results: A high accuracy threshold was a stronger predictor of students in second and third grades scoring at-risk on the MSA than WCPM, and similar results occurred for the other two measures. We arbitrarily set accuracy thresholds at 90%, 95%, and 98%. A 98% accuracy cutoff identified all third graders who were to be identified as at-risk on the MSA at the end of 3rd grade, while using established DIBELS benchmarks of CWPM only identified 75% of the at-risk students. Conclusions: It may be important to consider setting a high accuracy threshold when calculating cut-off scores on ORF measures in second and third grades when determining whether a student should be classified as at-risk.
Purpose A one-year longitudinal study was performed to test Chliounaki and Bryant's (2007) hypothesis that children's word-specific learning is a causal determinate in their general word learning. Method 62 Kindergarteners were asked to read and spell high frequency real words and matched pseudowords (e.g., with-bith) in October, January and April (Sessions 1, 2 and 3). Results Cross-lagged panel correlation analyses indicated that from the beginning to middle of kindergarten (from Session 1 to 2) children's early scores for high frequency words predicted their later scores for matched pseudowords better than their early pseudoword scores predicted their later high frequency word scores. This was the case both for reading and spelling. Between the middle and end of kindergarten (from Sessions 2 to 3) the effect of early word specific knowledge on the later acquisition of word general knowledge was still significant for spelling but not for reading. Conclusions The findings support Chliounaki and Bryant's hypothesis that children's knowledge of specific words plays a causal role in the early development of code skills. Further, the findings extend their model to include simple grapheme-phoneme correspondence rules and an educational context where correspondence rules are explicitly taught.
Bronwen Davis (University of Guelph);Evans, Mary Ann; Reynolds, Kailey - Parental Feedback During Shared Alphabet Book Reading: The Role of Child Miscues, Early Literacy Skills, and Alphabet Book Features
Purpose: We examined parent-child dyads reading an alphabet book together to identify the types of errors children make during alphabet book reading, the nature of parental feedback to these miscues, and whether miscues and feedback relate to literacy skills and to alphabet book features, such as the saliency of the featured object and familiarity of the object name. Shared storybook reading research indicates that children whose parents respond to errors with graphophonemic clues (e.g. pointing out letter sounds) demonstrate better subsequent word reading skills than children whose parents supply the miscued word. Alphabet books are absent in this research. Method: Early literacy skills were assessed in 55 non-literate, senior kindergarten children. Sessions of parent-child dyads reading an alphabet book were audiotaped, and later transcribed and coded for child miscues and parental feedback. Each page of the alphabet book was also coded for several characteristics. Results: Relations between child miscues and parental feedback will be examined, and related to current child literacy factors and specific alphabet book features. Conclusions: Findings will contribute to the understanding of factors that influence parental feedback early in their child's literacy development, and provide support for the development of empirically based criteria for alphabet book construction.
Purpose: We examined (1) the stability of latent classes associated with reading disability (RD) and typical development (TD) across time and (2) early indicators of late-emerging RD. Method: Analyses were based on a longitudinal sample of 177 students. At the end of 1st, 2nd, and 4th grades, students were assessed on word identification, sight word efficiency (SWE), and passage comprehension. Latent transition analysis models were developed to examine classification transitions from 1st to 4th grade and 2nd to 4th grade. Results: Results suggested that a small group of children with late-emerging RD could be identified in the fourth grade. In addition, SWE measured in 2nd grade was found to reduce the number of false negatives for RD and therefore was important for the classification of RD. In regards to early predictors of late-emerging RD, participants' performance on a listening comprehension task was a promising indicator. However, listening comprehension produced an unacceptably high rate of false positives. Conclusions: Results from this study indicate that RD/TD classification is relatively stable, however a set of students were identified having late-emerging RD. Results highlight the need for a set of early indicators that can identify children who are at risk for developing late-emerging RD.
According to Share's self-teaching hypothesis, phonological recoding is the mechanism by which beginning readers acquire (word specific) orthographic knowledge without the help of a tutor. The hypothesis is supported by a large number of studies in which children demonstrated orthographic learning of the spellings of pseudowords after reading aloud texts that contained these pseudowords. More recently, orthographic learning after silent text reading has also been shown. However, it is yet unclear whether orthographic learning during silent reading is due to phonological recoding, because independent evidence of the use of phonological recoding during silent reading is hard to get. Two studies will be presented in which we hypothesized that 1) there is phonological recoding of pseudowords during silent reading and 2) orthographic learning during silent reading is due to phonological recoding. In the first study second grade children read texts either silently or aloud. In the second study, a lexical decision task was administered to second grade children in two conditions. In the standard condition the task was done silently. In the second condition phonological recoding was suppressed. The results of both studies provide support for the first hypothesis but not for the second.
Purpose: A key question in spelling development lies in the extent of children's sensitivity to the role that morphemes, or the smallest units of meaning in language, play in spelling. We examine here whether this appreciation is similar for derived and inflected words, given that the former typically involve changes in word class from the base form, while the latter do not. Methods: To examine this question, we asked children in grades 2 to 4 to spell quadruplet sets of words that included the root, inflected and derived forms, as well as a control word (such as rock, rocked, rocker, and rocket). Results: Preliminary analyses indicate that children were more likely to spell the initial letter-sound sequence correctly when this represented the root of the word than when it did not (e.g., rock in rocked and rocker than in rocket). Most intriguingly, this effect was similar for inflected and derived words. Further analyses will examine whether the same pattern of results emerges in coding of spelling consistency. Conclusions: This study will inform us as to how much relatively young writers take advantage of the place that units of meaning play in spelling, permitting a refinement of our understanding of spelling development.
Carolyn A. Denton (University of Texas Health Science Center - Houston);Barth, A. E.; Cirino, P. T.; Wexler, J.; Vaughn, S.; Romain, M.; & Fletcher, J. M. - The Relationship between Oral and Silent Reading Fluency and Comprehension in Middle School: How Fluent is Fluent Enough?
The purpose of this study was to investigate the relationships between both oral and silent reading fluency and reading comprehension of middle school students. Guiding questions were (a) What is the relationship between oral or silent reading fluency and reading comprehension. (b) What proportion of variance in reading comprehension is accounted for by oral and silent reading fluency? (c) Is there a fluency rate that is "fluent enough" for adequate comprehension? Study participants were 1763 students in grades 6-8. Assessments included Woodcock-Johnson Tests of Achievement III Passage Comprehension, the Group Reading Assessment and Diagnostic Evaluation, the Middle School Progress Monitoring System, AIMSWEB Maze Measure and the Test of Sentence Reading Efficiency. Results indicate that the correlation between fluency (oral and silent) and reading comprehension is .57. The proportion of variance in comprehension accounted for by reading fluency (oral or silent) is .35. Results also indicate that there is not a reading fluency rate that is "fluent enough" for adequate comprehension. That is, if a high comprehension criteria is chosen (50th %ile), then a dysfluent reader has a 95% probability of not meeting this criteria, however a fluent reader has a 67% chance of being below this criteria.
Alain Desrochers (University of Ottawa) ;Glenn Thompson; Alain Marchand; Pierre Cormier - The Development of Alphabetic Knowledge among French-speaking Children over the Primary Grades The Development of Alphabetic Knowledge among French-speaking Children over the Primary Grades The Development of Alphabetic Knowledge among French-spe
The writing system of modern French is based on an alphabet of 26 letters and auxiliary marks, and a set of about 130 single-letter or multiple-letter graphemes. We investigated the development of these graphic units among French-speaking children from Kindergarten to Grade 6 in a cross-sectional study (N = 777). Younger children, up to Grade 2, were asked to name all 26 letters plus three letters that carry a diacritic mark (é, è, ê). All children were asked to sound out 58 graphemes, consisting of single-letter or multiple-letter vowels or consonants. Separate multi-level analyses were carried out on the letter-naming and grapheme-sounding data, with stimulus characteristics (e.g. print-to-sound consistency) as Level 1 and children's characteristics (e.g. grade level) as Level 2. The results indicate that single-letter graphemes are named and sounded out more accurately when they are vowels rather than consonants, when they have an inconsistent print-to-sound correspondence, when they do not carry a diacritic mark, and when the letter name begins with the phoneme represented by the letter itself. Similar results are observed with multiple-letter graphemes except that consonants are sounded out more accurately than vowels. Performance is found to improve from Kindergarten to about Grade 2 and then to level off.
With the goal of examining morphological relationships, the author compared the first 2000 words from both informational and narrative texts at the second-grade level. Content vocabulary words were grouped in morphological families and tagged according to linguistic form: inflection, compound, or derivation. Tokens per family were counted. Key vocabulary targeted for assessment in each text was compared against morphological relatedness and word repetition factors. Results revealed that morphological relatedness varied greatly across the two texts. Compared to the narrative, the informational text included twice as many average tokens per morphological family. In addition, greater text-assessment alignment was found in the informational text. These findings suggest that informational text might provide developing readers with greater opportunities to learn targeted vocabulary, and unique opportunities to develop automatic word recognition, fluency, and vocabulary. Informational texts may also better promote morphemic processing during reading. This may be especially helpful for English Language Learners.
Purpose: a) to investigate the role of working memory in comprehension outcomes after accounting vocabulary ability, b) to explore patterns of individual differences relative to vocabulary ability, working memory capacity, and reading comprehension among adults. Methods: Two hundred thirty twins and family members from the Minnesota study of twins reared apart were asked to read three passages of text aloud. After reading each passage participants were asked to recall everything that could be remembered. Coding skills, vocabulary and working memory were also assessed. Results: The regression analysis indicated that working memory explains a small but significant amount of variability in comprehension. Results from the Pattern and Level analysis indicated that individuals who have their highest predictor score on vocabulary and their lowest score on coding skills tend to score higher on comprehension. Level and Pattern effects significantly accounted for 41% of the variability in comprehension. Conclusion: Adults monitor and adjust their intake of information via coding skills in order to accommodate their working memory. The profile suggests that individuals with higher relative vocabulary ability read slower and make more coding skill errors yet have higher comprehension scores. Substantiation for a verbal preservation theory is discussed relative to aging and comprehension processes.
Linnea Ehri (City University of New York, Graduate Center); Eric Satlow; Irene Gaskins - Grapho-Phonemic enrichment strengthens keyword analogy instruction for struggling young readers Grapho- Phonemic Enrichment Strengthens Keyword Analogy Instruction for Struggling Young Readers Grapho- Phonemic enrichment st
First, second, and third graders (N = 102) who had completed at least one year of literacy instruction in other schools and had experienced failure entered this private school for struggling readers and received instruction in either of two types of systematic phonics programs over a four-year period. One group received a keyword analogy method (KEY) which taught them to decode words by analogy to 120 keywords. The other group received KEY enriched by instruction in grapho-phonemic analysis (KEY-PLUS). Results showed that KEY-PLUS students read and spelled words significantly better during the first two years of instruction than KEY students. The same differences remained evident, although not significant, during Years 3 and 4. The programs did not differentially improve reading comprehension. Results are consistent with developmental theories indicating the foundational importance of grapho-phonemic analysis for retaining written words in memory to facilitate word reading and spelling. Some effects of IQ were found. During Year 1, KEY-PLUS instruction was especially beneficial in helping average IQ students learn to decode pseudowords. High IQ students showed superior reading comprehension at the end of Years 3 and 4. High IQ students remembered word spellings better over the summer months than average IQ students.
In this presentation we present initial results examining the potential of an experimental dynamic assessment to aid in the identification of children at high risk of developing reading comprehension problems. The dynamic assessment measures children's ability to make causal inferences in text by making a series of increasingly difficult cause and effect judgments. Increasing scaffolding is provided to the children to aid them in making the judgments. The amount of scaffolding required will be correlated with various measures of reading skill (i.e., word reading, passage fluency, and reading comprehension) in a sample of 100 children assessed at the end of second grade.
Dion Eric (Université du Québec à Montréal); Monique Brodeur; Catherine Gosselin; Marie-Ève Campeau; Douglas Fuchs - Reading Problems among Students of Low Socioeconomic Status: An Experimental Test of the Prevention Model
Formal reading instruction has traditionally begun in first-grade, but the prevalence of reading disabilities among low socioeconomic status students has created an incentive to start reading instruction earlier. A relatively large-scale randomized study was conducted to determine the added-value of kindergarten reading instruction when first-grade reading instruction was improved. The study was conducted in a high-poverty area (Montreal, Canada). Kindergarteners (N = 270) and teachers were assigned to one of two conditions: evidence-based reading instruction in both kindergarten and first-grade (combined condition) or evidence-based reading instruction in first-grade only (first-grade condition). The program used in kindergarten, La forêt de l'alphabet, is adapted from the Project Optimize (Simmons & Kame'enui, 2003), and is designed to support letter sounds learning. The program used in first-grade is Apprendre à la lire à deux, is adaptated from the First-Grade Reading PALS (Fuchs et al., 2001), and is designed to support word-recognition. The pre-test kindergarten rapid letter naming was used to identify high-risk students. Compared with their first-grade condition peers, combined condition high-risk students had superior (e.s. = .54) end of first-grade decoding skills. A similar difference was not observed for their classmates. Early reading instruction thus appear especially important for high-risk students.
Mary Ann Evans (University of Guelph); Jean Saint-Aubin - Eye Movements of Senior Kindergarten Children Reading an Alphabet Book and Relationship to Their Letter Knowledge Eye Movements of Kindergarten Children Reading an Alphabet Book and Relationship to Letter Knowledge
Previous research tracking children's eye movements has shown that children rarely look at print when read storybooks. Alphabet books are referred to as "print salient books" with the presumption that they elicit greater attention to print. The purpose of this study was to describe children's eye movements when reading an alphabet book on their own, and to examine the relationship to their alphabetic knowledge. Each page displayed a single letter, a clear accompanying single object and corresponding printed word, and a bear which reappeared throughout the book. Twenty children ages 59 to 71 months were asked to read the book while their eye movements were monitored, and were tested on their letter name knowledge and receptive vocabulary. Results revealed that children spent significantly more time on the illustrations than on the letter or the word, which did not differ from each other. After controlling for vocabulary knowledge, the number of letters known by a child accounted for a significant amount of variance in the latency before fixating the letter, and the time spent fixating the word. Thus in order for alphabet books to elicit their attention to print, children must have acquired a critical mass of letter knowledge.
Fataneh Farnia (University of Toronto (OISE));Geva, Esther - A Longitudinal Examination of the Reciprocal Relations Between Reading Fluency and Reading Comprehension in Monolingual and ESL Students A Longitudinal Examination of the Reciprocal Relations Between Reading Fluency and Reading Comprehension in Mono
This study examined the longitudinal mutual facilitation between academic word reading, text reading fluency, and reading comprehension (RC) in a study involving 156 English native speakers (NS) and 417 English-as-a-second-language (ESL) students in Grades 4-6. Growth curve modeling was used to (a) describe and compare the patterns of growth in word and text reading fluency and RC, and to examine the contribution of(b) individual/group differences in Grade 4 acadenuc word and text reading fluency to Grade 6 RC, and (c) contribution of variation in Grade 4 academic word reading and RC to individual/group (i.e., ESL/EL1) differences in Grade 6 expository text fluency. Results indicated that (a) the ESL group performed significantly lower on RC, but similarly on text fluency in Grade 6, (b) Grade 4 expository word and text fluency each explained unique variance in Grade 6 RC, and (c) Grade 4 expository word and RC uniquely explained variance in Grade 6 text fluency in both ESL and EL1 students. The findings are discussed in terms of the differences in the development of reading in NS and ESL students beyond the primary years.
* Purpose: This study investigates universal and script-specific processes in reading eye movements, and specifically, how eye movements are planned in scripts that do not mark word boundaries with spaces. Most models of reading eye movements presume that words are visually delimitated by spaces, which then allow for effective parafoveal previewing and oculomotor targeting. Chinese and Japanese scripts, however, do not have any spaces between words. * Method: Adult readers from 4 countries read novels in native language while eye movements were recorded. Occasionally texts on the screen were shifted horizontally during saccades so that the subsequent fixation would miss the intended word. The frequency and amplitude of corrective saccades following manipulation were used as indices of eye movement planning strategies in reading. * Results: In English and Korean, where spaces are available, readers target specific words in the peripheral vision. Japanese readers target Kanji characters but apparently not Hiragana characters. Chinese readers have a complex saccade targeting strategy. * Conclusions: Skilled readers exploit statistical regularities in the orthography in programming eye movements. Implications to current theories of reading eye movements will be discussed.
Purpose This paper examines the effectiveness and efficiency of two types of oral vocabulary instruction that were delivered in addition to phonological awareness and decoding (PAD) instruction. Method Participants were first grade students (N = 79), primarily English Learners, in a low-performing school. Small instructional groups were formed based on risk status as indexed by nonsense word fluency, and these groups were randomly assigned to one of three conditions. The two vocabulary-added conditions differed in instructional emphasis: morphological awareness (MA) versus semantic relations (SR). In both, 30% of the instructional time was allotted to PAD instruction, versus 100% in the treatment-control (PAD) condition. Results Both treatment groups demonstrated greater growth in vocabulary than did the treatment-control group, and furthermore maintained equal or greater gains on other reading measures. Effects on PA and reading performance were largest for the SR condition, and effects on vocabulary and listening comprehension were largest for MA. In no case were effects for the PAD condition the largest. Conclusion The strategy based vocabulary instruction combined with direct teaching of phonological and decoding skills was more efficient at improving not only foundational skills, but also higher order skills crucial to comprehension of connected text.
This study utilized early reading assessment data in Spanish from a randomized trial of 210 schools in Texas to examine contextual effects on risk prediction in first and second grade classrooms. The primary objective was to examine the roles of (a) individual differences, (b) the grade 1 classroom, and (c) the pairing of first and second grade teachers in determining grade 2 outcomes in word reading and fluency. Participants were 1,743 Hispanic students participating in Spanish reading/language arts instruction in first and second grade. Analyses revealed that for fluency, a combination of student pretest centered around the class mean and mean of pretest classroom was the best predictor, in contrast to word reading where student pretest centered around the grand mean was the best predictor. Additionally, the effect varied by teacher-pair. These are important findings because (a) risk prediction differs for fluency versus word reading outcomes in Spanish reading, (b) risk prediction for fluency needs to be based on classroom context rather than student pretest scores alone, and (c) the particular pairing of teachers a student has when moving from first to second grade makes a difference to reading outcomes.
Purpose. Until recently the field has used primarily raw fluency data for both practical and research purposes. Although each paper takes a slightly different approach to investigating the measurement and nature of reading fluency, all demonstrate that raw fluency data has critical limitations, whatever metric is chosen and for whatever purpose. This paper serves to synthesize and discuss these findings in light of previous and ongoing research at University of Houston (UH) and in the field at large. Method. The presented papers will be discussed in relation to fluency investigations at UH, which have evolved from an equipercentile equating approach to using incremental models of invariance. UH studies experimentally manipulate passage order in order to isolate passage effects from order effects. Results. In general, across group invariance could be established for story factor loadings and intercepts. Further, across group-across story factor loading invariance was established, but across group-across story intercept invariance could not be established. Conclusions. The combined results from the UH fluency studies and the four papers presented strongly suggest that direct comparisons between observed scores should not be made between different DIBELS (or other fluency measure) forms. The relative merits of the different equating approaches will be discussed.
Olsen and Kenny (2006) have described a general strategy for analyzing data from indistinguishable pairs, which addresses the issue of non-independence within pairs. This general strategy was adapted to analyze the phenotypic relationship between parental education and growth in word recognition performance for 489 pairs of monozygotic and dizygotic twins in the Colorado sub-sample of the International Longitudinal Twin Study project while controlling for genetic relationship. Results from maximum likelihood analyses demonstrated that parental education predicts starting performance in kindergarten, but that it does not predict differences in the slope of growth in word recogntion from kindergarten to second grade.
Purpose: Vocabulary researchers have noted that semantic knowledge is typically acquired over many exposures, suggesting that people may have incomplete knowledge of some words. Although this idea has intuitive appeal, there has been little research on the neurocognitive processes associated with different stages, or "degrees," of semantic knowledge. Method: To examine neural correlates of meaning acquisition, we measured ERPs as learners encountered familiar and rare words in sentences that were either high or low constraint (providing either strong or weak cues to meaning). To probe immediate and delayed effects of learning, ERPs were time-locked to words presented just before each sentence, immediately after, and after a moderate delay (30 minutes or longer). Results: During the training, high vs. low constraint contexts lead to reduction in an ERP component associated with familarity-based memory (the MFN). In the delayed ERP task, we observed an N400 reduction for low-frequency, but not rare, words trained in high constraint sentences. Rare words continued to show an MFN that was sensitive to sentential constraint. Conclusions: These results suggest that the earlier MFN effect may reflect partial semantic knowledge, whereas the later N400 effect may reflect a more complete or robust representation of word meaning.
Stephen Frost (Haskins Laboratories); Mencl, W. Einar; Sandak, Rebecca; Landi, Nicole; Fulbright, Robert K.; Jacobsen, Leslie K.; Grigorenko, Elena; Constable, R. Todd; Pugh, Kenneth R. - Relating Phonological Awareness to Brain Activation Patterns for Reading
We examined the relationship between activation patterns at the level of the emergent functional neuroanatomy for reading in beginning readers and a behavioral marker of reading readiness, phonological awareness (PA). Twenty-seven early readers (ages 6-9) participated in an event-related fMRI task involving a match/mismatch judgment of speech and print targets to a picture cue. Brain-behavior analysis revealed that the magnitude of the modality effect is significantly correlated with PA in the LH superior temporal gyrus (STG), a primary area for spoken language processing. A median split on PA revealed that speech stimuli activate STG for both low and high PA children; however, printed stimuli activate STG only for children with high PA. The study provides a better understanding of the neurobiological signature associated with attaining early phonological awareness and provides support for a developmental trajectory towards integration of spoken and written language processing both cognitively and with respect to neural circuitry for those learners who are on the path toward successful literacy acquisition.
Proficient reading not only requires decoding accuracy but word identification automaticity that preserves cognitive resources for the demands of text comprehension (LaBerge & Samuels, 1974; National Reading Panel, 2000; Nathan & Stanovich, 1991; Perfetti, 1985, 2007; Rayner, Chace, Slattery, & Ashby, 2006). Researchers posit that successful readers process words in text between 200 and 400 msec (Gibson & Levin, 1975; Gough, 1972; Rayner, 1997; Rayner and Pollatsek, 1989). Torgesen, Rashotte, and Alexander (2001) maintain that rate and accuracy are important aspects of reading fluency because they can be reliably measured. Yet, scant research exists on the efficacy of assessing word-level automaticity and on how timed assessments of single word identification may predict other behaviors we associate with fluent reading. Eighty third and 80 fourth graders read graded lists of isolated words presented on a computer. Subjects were assigned to one of four presentation conditions (300 msec; 650 msec; 1000 msec; and 2000 msec duration time) and received an accuracy score that is considered a measure of word-level automaticity for each graded list. Subjects also orally read graded passages (IRI), yielding reading rates, accuracy, and comprehension. Scores from STAR Reading Test and Woodcock Reading Mastery Test were also collected. Preliminary results reveal that word identification scores for faster presentation rates (300 and 650 msec) predict oral reading rates significantly better than scores from slower presentation rates (1000 and 2000 msec) and also correlate more highly with other measures of reading proficiency. These data suggest that rapid single word identification scores predict other measures of reading fluency in connected text. In this poster session, we will present the full findings from our research as well as implications for theory and practice. References Gibson, E. J., & Levin, H. (1975). The Psychology of Reading, Cambridge: MA, MIT Press. Gough, P.B. (1972). One second of reading. In J. F. Kavanagh and I. G. Mattingly (Eds.), Language by ear and by eye. Cambridge: MA, MIT Press. LaBerge, D., & Samuels, S. J. (1974). Toward a theory of automatic information processing in reading. Cognitive Psychology, 6, 293-323. Nathan, R. G., & Stanovich, K. E. (1991). The causes and consequences of differences in reading fluency. Theory into Practice, 30(3), 176-184. National Reading Panel. (2000). Teaching children to read: An evidence-based assessment of the scientific research literature on reading and its implications for reading instruction. Washington, DC: National Institute of Child Health and Human Development. Perfetti, C. A. (2007). Reading ability: Lexical quality to comprehension. Scientific Studies of Reading, 11, 357-383. Perfetti, C. A. (1985). Reading ability. New York: Oxford Press. Rayner, K. (1997). Understanding eye movements in reading. Scientific Studies of Reading, 1, 317-339. Rayner, K., Chace, K. H., Slattery, T. J., & Ashby, J. (2006). Eye movements as reflections of comprehension processes in reading. Scientific Studies of Reading, 10, 241-255. Rayner, K., & Pollatsek, A. (1989). The psychology of reading. Englewood Cliffs, NJ: Prentice Hall. Torgesen, J.K., Rashotte, C. A., & Alexander, A. W. (2001). Principles of reading fluency instruction in reading: Relationships with established empirical outcomes. In M. Wolf (Ed.), Dyslexia, fluency, and the brain 333-355. York Press.
This study's purpose was to determine the type of teacher support necessary to scale up Kindergarten Peer-Assisted Learning Strategies (K-PALS) in Tennessee, Texas, and Minnesota. 145 teachers and their 1,674 kindergartners in 48 urban, rural, and suburban schools participated. Teachers were assigned randomly to 4 levels of teacher support: (1) a control group, members of which used the regular reading curriculum; (2) a workshop-only group (Workshop) that attended a 1-day K-PALS workshop; (3) a workshop plus booster group (Booster) that attended the workshop plus 3 booster sessions across the year in which a K-PALS "expert" provided teachers with problem-solving strategies; and (4) a workshop plus booster plus mentor group (Mentor) that received weekly technical assistance in addition to the workshop and booster sessions. All teachers (except controls) implemented K-PALS 4 times per week for 20 weeks. Data on students' performance showed that all three K-PALS groups outperformed controls across sites on phonological awareness measures. Results were more variable across sites for word-level and fluency measures. Contrarily to expectations, students in classes of Booster and Mentor teachers did not always outperformed students in classes of Workshop teachers. Implications for implementing evidence based practices at scale will be discussed.
We asked whether potential learn to algebra at the beginning of 3rd grade forecasted later development of word-problem skill. In the fall, 123 students were assessed on language, nonverbal reasoning, attentive behavior, calculations, and DA. Students were randomly assigned to 16 weeks of validated or conventional word-problems treatment and then were assessed on word-problem measures proximal or distal from treatment. DA measured a distinct dimension of pretreatment ability and proximal and distal word-problem measures were needed to account for outcome. Treatment accounted for outcome proximal to treatment. By contrast, DA proved sufficient to forecast learning on distal word-problem measures.
Correct use of the definite article is difficult to acquire for ESL students. Japanese ESL students experience particular difficulty because Japanese does not contain an article system. In order to develop an effective method in teaching the correct definite article usages, individual difference factors other than proficiency level should be taken into consideration. The purpose of the current research was to investigate whether one's cognitive style influences the acquisition of four different the usages in Japanese ESL students. Twenty-seven Japanese undergraduate students studying in New Brunswick were administered measures of cognitive style, definite article, and overall English proficiency. Data were analyzed using Pearson partial correlation coefficients and a repeated-measures ANOVA. With overall English proficiency partialed out, cognitive style measure positively correlated with scores on two of the four the usages that appear to require more grammatical judgment. Cognitive style measure positively correlated with overall English proficiency. Scores on the cultural use, which can only be learned through exposure, was significantly lower than the other three uses. Findings suggest that cognitive style is strongly related to the acquisition of the. Cognitive style may be an important factor to consider in developing an effective pedagogy for teaching definite article usages.
Purpose This study explored topic interest as a potential protective factor during challenging reading tasks in buffering the negative effects of challenge on children's motivation, attributions for failure, and persistence. Method Children (N = 56) ranging from 10 to 14 years read a passage that was two or four grade levels beyond their current reading ability under two matched experimental conditions: an Interesting and a Non-Interesting Passage condition. Motivation for reading and attributions for failure with reading were evaluated before and after the reading task. Persistence was measured with observations of task continuation when given the option to quit. Results Developmentally inappropriate challenge threatened children's interest/enjoyment for reading, perceptions of competence, and perceived effort. However, interest in the story topic buffered some of the negative impacts of challenge, sustaining children's interest/enjoyment, adaptive attributions, and persistence with the reading task. Conclusions Interest has the potential to protect certain children against less than optimal reading experiences in helping to sustain motivation, adaptive attributions, and persistence. In creating optimal and motivating challenging activities for reading, it is important that the task is suited to children's abilities and interests.
A majority of research on developmental dyslexia is performed in English-speaking countries. From this investigation there is a consensus that a phonological core deficit is the main cause underlying developmental dyslexia. English, however, is characterized a deep orthography with many inconsistent grapheme-phoneme and phoneme-grapheme correspondences. In more transparent orthographies, including most European orthographies, these correspondences are far more consistent. Consequently, it has been suggested that predictors of developmental dyslexia might differ across alphabetic writing systems. In this study the main purpose was to directly compare if and to what extent equivalent versions of preschool cognitive and language skills (i.e. phonological awareness, print knowledge, rapid naming, verbal memory, vocabulary, grammatical, and morphological skills) differentiate the group with and without reading and spelling disabilities across transparent (Norwegian/Swedish) orthographies and English (US/Australian) during the initial phase of literacy development.
Wei Gao (University of Maryland)Wang, Min Ph.D. (email@example.com / University of Maryland, College Park) Voting member - Subsyllabic Unit Preference of Young Chinese Children in Reading Pinyin
Purpose--Two questions were addressed: Do Chinese children show body unit preference in reading Pinyin like they do in processing spoken Chinese as shown in previous research? Do tone and initial consonant sonority have effects on children's subsyllabic unit preference? Method In Experiments 1 and 2, Pinyin analogy tasks were designed for monophthong and diphthong pseudo-Pinyin words respectively, with different shared subsyllabic units (i.e., body, rime, onset-coda) read in tone matched and unmatched conditions. Experiments 3 and 4 were short-term recall tasks with pseudo-pinyin words in tone matched and unmatched conditions. Fifty-two 5-year-old children participated in Experiments 1 and 2, and 23 children of same age in Experiments 3 and 4 respectively. Results In the analogy experiments, target words that share rime units with the clue words were read most accurately. Rime unit preference was also found in short-term recall tasks. Children performed better when tone is matched. They retained least subsyllabic units when initial consonants are liquid. Conclusion These results suggest that Chinese children have a different subsyllabic unit preference pattern in reading Pinyin compared to processing spoken language. We argue that a teaching approach that emphasizes rime units may play an important role in the rime unit preference.
Russell Gersten (Executive Director, Instructional Research Group);Dimino, Joe; MJayanthi, adhavi; Santoro, Lana ; Kim, James - Lessons Learned from an Observational study of Comprehension and Vocabulary Instruction for English Learners in a Reading First District
Purpose: We will discuss the findings of classroom observations that were conducted to assess the quality of reading comprehension and vocabulary instruction in Reading First classrooms with English Language Learners (ELLs). Method: The observational data was collected as part of our multiple-site, randomized controlled trials study of professional development. Our goal was to assess the extent to which teachers (N = 32) implemented research-based approaches for teaching comprehension and vocabulary. We used the Reading Comprehension and Vocabulary (RCV) Observational Measure - a moderate inference classroom observational measure - to assess the quality of teacher instruction. Observational items reflect two major pedagogical aspects of effective instruction: explicitness of instruction and nature of the interactive instruction (i.e., the amount of scaffolding practice and feedback provided). Results: We found significant differences between experimental and control groups in the number of times effective teaching behaviors relating to comprehension and vocabulary were observed. The descriptive data (especially from the comparison group) provide a vivid picture of current comprehension and vocabulary instruction for ELLs. We found some use of evidence-based practices, but consistent use was rare. Conclusions: We will discuss implications for teacher practice, specifically drawing attention to areas where effective instruction was minimal.
Esther Geva (Dept of Applied Psycholog & Human Development,OISE- University of Toronto); Gottardo, Alexandra - A longitudinal examination of language and literacy development in ESL children: Is "Transfer" a valid framework?
In a cross-linguistic longitudinal study we tracked the linguistic and word-level reading skills of ESL children whose home language is Chinese, Portuguese, or Spanish over a period of three years. Parallel measures were administered in the home language (L1) and in English, the L2. These measures included various language and reading and cognitive underlying processing tasks. In this presentation we will focus on the parallel development of L1 and L2 phonemic awareness, vocabulary skills, and word reading in the three language groups and discuss commonalities across the three groups that support a cognitive interpretation of transfer, and topological differences that may be interpreted as "positive" or "negative" transfer, depending on the nature of the task. From a developmental perspective, of particular interest is the extent to which these topological differences, based on phonological and orthographic properties of the L1, that are noted at the onset of schooling persist over time with regard to different linguistic units. We conclude that the different transfer frameworks are complementary.
Jennifer K. Gilbert ()Compton, Donald; Fuchs, Douglas; Fuchs, Lynn S.; Schatschneider, Christopher - The Additive Effects of Tier-1 and Tier-2 Interventions on Struggling Readers within an RTI Framework
Purpose - The purpose of this research was to examine whether Tier 1 validated classroom instruction enhances the effects of small-group Tier 2 intervention. Method - Participants (N = 83) were the poorest first-grade readers, based on word identification fluency (WIF) and rapid letter naming, from 42 classrooms in 16 schools in 2 school districts. In September, the 42 classrooms were randomly assigned to validated classroom instruction, Peer-Assisted Learning Strategies (PALS), or to conventional classroom instruction (i.e. no-PALS). Participants were assessed on WIF over 13 weeks in the fall and over 11 weeks in the spring. Students deemed unresponsive to fall classroom instruction were given supplementary Tier 2 Tutoring in the spring. Because these groups contained students from different classrooms, students were cross-classified by small groups and classrooms, and reading growth was estimated by a 3-tiered cross-classified piecewise linear growth model. Results - Results showed that students receiving Tier 2 Tutoring made significant reading growth, but that the validated Tier 1 PALS did not moderate this effect. Conclusions - Implications are discussed in terms of a Responsiveness-To-Intervention framework.
Stephanie Glasney (Florida State University and the Florida Center for Reading Research); Carol Conner; Claire Cameron Ponitz; Beth Phillips; Q. Monét Travis; Frederick Morrison - Individualized Instruction and Children's Literacy, Behavioral and Social Self-regulation in First Grade Classrooms
The present study examines the relations between students' growth in literacy skills and self-regulation and teachers' implementation of an individualized instruction intervention program. In this randomized control field trial, we examined first graders' growth (n = 452) in self-regulation and literacy outcomes and observed their classrooms throughout the year. Children's self-regulation was measured directly using the Head-Toes-Knees-Shoulders Task and teachers completed reports concerning children's social skills and behavior problems, using the Social Skills Rating Scale. Children's social and behavioral regulation skills significantly and uniquely predicted reading comprehension and vocabulary skill growth as measured by the Woodcock-Johnson Achievement Test-III. Moreover, students whose teachers individualized instruction with greater fidelity demonstrated stronger reading comprehension growth than did students whose teachers were in the control condition or who demonstrated less fidelity to the intervention. More time spent in instruction was also associated with more growth in self-regulation. This study suggests that using evidence-based instruction that is individualized may lead to higher levels of behavioral competence and literacy achievement for children.
Lucie Godard (Département de didactique des langues)Laplante,Line; Morris, Lori - The lexical knowledge, word recognition skills and spelling ability of L1 and L2 beginning readers in French: The importance of mother tongue and socioeconomic status
Purpose This study provides a detailed analysis and of the lexical knowledge, word recognition skills and spelling ability of 269 L1 and 154 L2 beginning readers. It was designed to tease apart the influence of linguistic and socioeconomic status. Method Lexical knowledge and word recognition were measured by means of multiple choice word-picture association tasks. The spelling assessment was an eleven-word dictation task. Data were analyzed by splitting the participants into a lower, middle and upper third and by examining the performance of each linguistic and socioeconomic group by third. Results The assessment of lexical knowledge showed an overall effect for mother tongue, with L2 participants, regardless of SES, performing worse than their L1 counterparts. The L2 lexical disadvantage was exacerbated by low SES in the lowest-achieving third. Word recognition results showed no L1-L2 difference but SES proved to be a factor in the bottom third. While a higher percentage of L2 than L1 children placed in the bottom third on the spelling task, there was no difference by mother tongue when SES was controlled for. Conclusion The full extent of the difficulties faced by L2 beginning readers in French is only revealed by separating linguistic and socioeconomic factors.
Jorge E. Gonzalez (Texas A&M University, VS); Durodola, Sharolyn-Pollard; Simmons, Deborah C. - Accelerating Preschool Children's Vocabulary and Background Knowledge: Effects of a Shared Book-Reading Intervention
Purpose The purpose of this study was to investigate an adult-child preschool shared-reading program premised on three interdependent constructs including word and background knowledge and knowledge of text structure. Research questions were: (a) What is the effect of the shared-book reading program on standardized and researcher developed measures of receptive and expressive vocabulary, and (b) is there differential impact based on student status at entry to preschool after controlling for the corresponding pretest scores and demographic variables including age, gender, bilingual status, ethnicity, school district, and teacher's teaching experience, and (c) are there teacher/student engagement/interaction features that predict language outcomes. Method A randomized field trial was using 20 classrooms assigned to the intervention or "practice-as-usual." Standardized measures: Peabody Picture Vocabulary Test III and Expressive One-Word Picture Vocabulary Test. Researcher-Developed Measures included a Receptive Vocabulary Test and an Expressive Picture Vocabulary Test. Multi-level modeling was used in the present study. We evaluated the intervention effect on the posttest measures by using the ANCOVA model with the pretest measures and other demographic variables as covariates. Results The intervention group scored substantially higher on both researcher developed measures at posttest. Students who had higher entry-level vocabulary had higher scores on the posttest measures. Additionally, Caucasian students scored higher on both EOWPVT at posttest than the African American students after controlling for their pretest vocabulary level and all other demographic variables.
Reading comprehension is the failed subject in Spanish reading learning (Pisa, 2006). Prosody is a factor related to reading comprehension, and rhythm is considered as a prosody component. Recent studies have shown the contribution of rhythm to the development of reading skills (David, Wade-Wolley, Kirby and Smithrim, 2006). Research aim was to determine if rhythm makes an independent contribution to the variance of reading ability. MCP and rhythm -without linguistic component-, decoding, reading fluency and reading comprehension were evaluated in 66 Spanish third graders. Hierarchical regression analysis showed that MCP explains a significant part of the variance, but independently rhythm explains another significant part of the variance of the dependent variables. Results suggest that rhythm plays a role in reading ability, in general, and reading comprehension, in particular. Rhythm role is discussed.
Debbie Gooch (Royal Holloway, University of London)Snowling, Maggie; Hulme, Charles - Time Perception and executive functions in children with Reading Disorder and Attention Deficit/Hyperactivity Disorder
Purpose To investigate whether deficits in time perception are specific to ADHD or RD, or whether they reflect comorbidity between these two disorders. Method The performance of four groups of children (11 ADHD, 13 RD, 18 ADHD+RD and 41 Controls) on measures of reading skills and executive functioning was compared. A duration discrimination task and a time reproduction task were used to assess time perception. 2x2 analyses were conducted to determine the influence of ADHD and RD on children's task performance. Results A trend for children with RD to obtain larger thresholds on the duration discrimination task compared to children without RD was found. However, only the ADHD+RD group performed significantly worse than controls on this task. Children with ADHD made larger errors on the time reproduction task compared to children without ADHD. The relationships between time perception and other executive skills were also examined. Conclusions Time perception deficits were found to be associated with ADHD or ADHD+RD rather than with pure RD. The results help us to more fully understand the cognitive deficits associated with comorbid ADHD+RD and highlight the importance of screening for comorbid ADHD when investigating skills such as time perception in children with RD.
The Simple View of Reading (SVR) states that reading comprehension is related to word reading (decoding) and listening comprehension, RC = D X LC (Gough & Tunmer, 1986; Hoover & Gough, 1990). Other views of reading comprehension suggest that when word reading skill is well established, vocabulary knowledge alone predicts reading comprehension (Carver, 1997). The SVR has been tested with monolingual speakers (Catts, Adlof & Weismer, 2006) and is beginning to be tested with second language (L2) learners, mostly Spanish-English speakers in the United States (Mueller & Gottardo, 2006; Proctor, Carlo, August & Snow, 2005). Method: Data were collected from third grade children who were learning English as L2 in Canada. The children had a wide range of first languages, Portuguese, Chinese and Spanish. A variety of decoding and oral language measures including vocabulary knowledge were administered. Results and conclusions: The results of preliminary regression analyses showed that vocabulary and word reading accuracy were related to reading comprehension. These results support the validity of the SVR in explaining variance in reading comprehension in L2 learners. In addition, Structural Equation modeling will be used to determine if both components of the SVR, D and LC, are necessary for predicting reading comprehension in these L2 learners.
Purpose - This study compared the efficacy of two parent-tutoring programs on reading achievement. Method - Thirty-two children in grades two to four were randomly assigned to one of three groups; Paired Reading (PR; Topping,2001), modified PR, or a control condition. The modified PR intervention included the word identification strategies of the Phonological and Strategy Training Program (PHAST; Lovett, Lacerenza, & Borden, 2000). Reading measures were administered before and after the 16 week intervention. Treatment integrity was measured via reading diaries and audio-taped reading sessions. Parental report of home literacy activities and satisfaction with the tutoring program were assessed with questionnaires. Results - Preliminary findings using one-way ANOVAs indicated the modified PR group achieved significant gains in standardized word, nonword and passage comprehension measures and on experimental measures used to measure the PHAST content, in comparison to the PR and wait-list control groups (Effect size M = 1.42,range from 0.37 to 2.36). No differences were found between the PR and control groups. Conclusions - Findings suggest significant reading gains can be achieved at home with a modification of the PR technique that incorporates teaching the word identification strategies of the PHAST Program.
Little is know about the writing practices of secondary teachers. In this national survey of secondary classroom writing practices, 361 teachers (a 51% response rate) completed a questionnaire asking them about their views of writing, their preparation to teach writing, the types of writing activities their students were asked to complete during the school year, their use of evidence based writing practices, types of instructional adaptations they made to meet individual students' needs, and their use of specific assessment procedures. In addition to describing the writing practices of secondary teachers in general, we examined if English, science, and social studies teachers differed in how they taught writing. Some of the findings (not all can be reported in an abstract) were: 1) secondary teachers reported that they received minimal preparation to teach writing during college, and only slightly higher than minimal preparation after college (English teachers reported being better prepared than social studies teachers, who reported being better prepared than science teachers); 2) the most common writing activities teachers' students engaged completed involved short answer responses and completing work sheets; and 3) the use of evidence-based writing practices were uncommon.
The current study investigated differences in performance on several measures of reading between two groups of English Language Learners (L2; Spanish and Portuguese) and a control group of students who spoke English as a first language (L1). Preliminary analysis show the uniqueness both within and between samples of Grade 3 L2 students. An L1 English group seemed to perform higher on assessments of receptive vocabulary, word identification and nonword reading. However, there are also significant differences between two cohorts of L2 Spanish speakers recruited from the same area and even the same schools in a large metropolitan area. More specifically, this study sought to determine what relationship print exposure (i.e., extracurricular reading) plays in the development of reading comprehension and word reading in these groups of students. Print exposure seems to play a mediating role in groups who are more at-risk for reading failure. In the Spanish L2 group, word reading was highly related to reading comprehension (r = .82). However, this relationship seems to be at least partially mediated by the title recognition test. Print exposure does not seem to predict unique variance in reading comprehension in students who are not struggling as much with attaining reading skills, such as the English and Portuguese groups in this sample. These findings have implications for the assessment of children learning to read in other languages, and also to models of reading comprehension.
Purpose: This study addressed the efficacy of different reading instructional approaches for adults: decoding and fluency (DF); decoding, reading comprehension, and fluency (DCF); decoding, reading comprehension, fluency, and extensive reading (DCEF); extensive reading (ER); a control comparison. Method: 206 adults reading at the third through fifth grade levels were randomly assigned to each of the five groups. They were administered a pre and post reading battery. Results: An analysis of covariance was conducted to determine whether post test scores differed for the groups after controlling for pretest scores and minutes of class attended. In sight word reading, the DCEF group scored significantly higher than the ER and control groups while the DF and DCF groups scored significantly higher than the control group. In fluency, the DCF and DCEF groups scored significantly higher than the control and ER groups while the DF group scored significantly higher than the ER group. In decoding, the DCEF, DF, DCF, and control groups all scored significantly higher than the ER group. Conclusions: Adults who read between the third and fifth grade reading levels benefit from explicit instruction in decoding, comprehension, and fluency. They also benefit from extensive reading when it is added to explicit instruction.
Noel Gregg (University of Georgia)Gregg,Noel.,Lindstrom,Jennifer.,Nelson,Jason.,Coleman,Chris., & Lindstrom,William. - Mediating Cognitive and Linguistic Processes Predictive of Reading Accommodation Selection: A Psychometric and Clinical Perspective
Purpose The purpose of this study was to (1) investigate the relationship of reading eligibility criteria and accommodation selection; and (2) identify the cognitive and linguistic processes predictive of reading accommodations using a psychometric and a clinical method. Method Eligibility Criteria and Cognitive Profiles We studied the characteristics of 500 young adults with LD (dyslexia) or AD/HD (underachieving in reading) based on a variety of cut-off metrics and found differences in learning profiles depending on the eligibility criteria used. We studied the interaction of disability category, eligibility criteria (performance < 16th %ile and > 16th %ile), and reading skill (decoding, comprehension, and fluency) with accommodation selection. Cognitive Predictors for Accommodation Selection - Psychometric and Clinical Psychometric Data: All participants underwent a comprehensive psychological evaluation. Based on measures used to diagnose dyslexia identified in the literature, specific standardized tests were used to investigate their predictability in accommodation selection. The domains included: knowledge, reasoning, working memory (attention, executive functions, processing speed), long term memory, phonemic and orthographic awareness, oral language, and grapho-motor abilities. Clinical Decision Making: Accommodation selection is based on scores across measures, background information, and behavioral observations. Methods included systematically coding the cognitive deficits believed to underlie functional limitations in reading. Both the single standardized score criterion and the clinical criterion were analyzed as to their predictive power for accommodation selection.
Qun Guan (University of Science and Technology Beijing & Florida State University)Roehrig, Alysia - Faster is not necessarily better: the role of individual differences in processing elaborative inferences for a coherent text memory
What is contributive to a coherent text memory? The study examined the individual differences among college readers (N=73) in terms of their online inference-making processes and offline text memory. Measures of individual differences included reading efficiency, working memory capacity, and naming times on the probe words of the salient elaborative inferences in text. Independent text variables included saliency of elaborative inferences and text coherence. A 2 (high and low salient elaborative inference) X 2 (strong and weak coherence of text) X 2 (high vs. low working memory span) X 2 (good vs. poor reading efficiency level) mixed factorial design was applied. Regression approach on the repeated measure design was conducted. Results suggested that text coherence was not predicted by the reading efficiency (i.e., the scores of reading efficiency test, and naming time on the inferential probe words). A significant interaction between text coherence and working memory capacity suggested that the coherent offline text memory was contributed by a robust online processing mechanism. In order to achieve a coherent memory after reading, low-working memory and high-working memory readers were employing different online processing mechanisms. Their offline coherent text memory was facilitated when they were engaged in slow but elaborative online processing.
Ying Guo, Florida State University, firstname.lastname@example.org Membership Status: Non-member Alysia D. Roehrig, Florida State University and Florida Center for Reading Research email@example.com Membership Status: Active member Title: The role of vocabulary knowledge, syntactic awareness and metacognitive awareness in reading comprehension of adult English language learners Interactive Paper Abstract Purpose: We examined whether vocabulary knowledge, syntactic awareness and metacognitive awareness were distinguishable constructs, and their strength of relations with reading comprehension. We also examined whether 1) poor readers were inferior to good readers on the constructs and 2) the relations between constructs and reading comprehension differed across poor and good readers. Method: 278 undergraduates from 3 Chinese universities participated. Those with TOEFL reading scores in the sample's top and bottom 25% were identified as good and poor readers. 8 assessments were administered concurrently, with 2 measures representing each construct. Analyses included confirmatory factor analysis, structural equation modeling and MANOVA. Results: CFA and SEM suggested the two-factor model of Vocabulary knowledge/Syntactic awareness and Metacogntive awareness offered the best fit and Vocabulary knowledge/Syntactic awareness has a stronger relationship to reading. MANOVA indicated significant differences between poor and good readers in both constructs. Multigroup analyses using SEM suggested the correlation between the 2 constructs in poor readers was stronger than in good readers, but the pattern of relations of each construct to reading remained constant across groups. Conclusions: Linguistic knowledge is important for successful L2 reading. Interdependency between vocabulary and syntactic awareness indicates performance on syntactic awareness partially depends on vocabulary.
Reading storybooks to preschoolers are believed to promote children's reading development. This study investigates children's real-time print exposure and story comprehension during shared book reading. Eighteen 6-year-old Chinese-speaking preschoolers and twenty-six 4-year-old English-speaking preschoolers listened to electronic picture books read by an experimenter in two different styles, verbatim and interactive. Children's eye movements were monitored with a Tobii eye-tracking system. Adults' storytelling style did not significantly affect the overall proportion of time children spent looking at texts in either countries. However, children spent more time looking at texts during the beginning time of each page when the story was read in interactive style versus verbatim style, suggesting children sought information from texts when they were expected to make sense of the story on their own. On the other hand, data also suggested children were more engaged in story comprehension during verbatim reading. We measured the time between when the storyteller named of a main character or object on the page and when the child first looked at the person or object. The verbatim condition is 1.8 times as fast as the interactive condition (a significant difference). Effects of storytelling strategies, cross-linguistic differences, and implications to reading development will be discussed.
Some recent studies suggest that prosody may affect early stages of reading acquisition (e.g., see Journal of Research in Reading 2006 monograph). Furthermore, prosodic skills may help to automate some reading tasks (e.g., stress assignment or syntactic processing). If so, prosody should also be related to adult reading performance. This study aimed to test that hypothesis. It used phonological awareness (oddity and deletion) and stress awareness tasks (i.e., to indicate the stressed syllable in a three-syllable word). Dependent variables were performance at reading comprehension and reading aloud pseudowords. Results showed that stress awareness explained a significant amount of variance for pseudoword reading scores (reaction times). This datum is discussed in the context of the role of prosodic skills in the acquisition and automation of reading prosody.
It has been proposed that language-based learning disorders such as dyslexia may be attributable to low-level deficits in auditory processing. However, studies seeking evidence for auditory processing deficits in these groups have reported inconsistent results. A recent theory suggests that children with dyslexia may be delayed in their development of auditory processing abilities (Wright & Zecker, 2004). To assess this theory, we measured frequency discrimination and frequency modulation (FM) detection thresholds in 78 children with dyslexia and 84 controls aged between 7 and 14 years. As a group, children with dyslexia performed significantly more poorly than controls on both tasks. However, the profile of impairment differed both as a function of age, and between the two tasks. In general, thresholds for the dyslexic group did not differ from those of controls who were between 2- and 4- years younger. Together, our results support the theory that children with dyslexia may be delayed in their development of auditory processing skills. Further, our findings suggest that discrepancies in the literature may be explained, at least in part, by differences both in the ages of the participants and in the normal developmental trajectories of the auditory processing tasks assessed.
Purpose: Studies of letter recognition have mainly been concerned with recognition of letters of the Roman alphabet. The purpose of this study is to investigate letter recognition in Arabic by both native speakers and FL learners to whom L1 is written in the Roman alphabet. Method: Based on a comparison of letter architecture in the two alphabets, Arabic medial letters are grouped in a system similar to Bouma's (Bouma, 1971) grouping of Roman minuscules. This analytical grouping is then verified empirically, employing pseudo-word reading by both native speakers and adult FL learners of Arabic at different levels. Results: The empirical investigation generally supports the letter architecture analysis but it also points at considerable phonological interferences in FL learners and, more surprisingly, native speakers as well. A number of mistakes seem to be caused by rather complex interferences between letters with both graphemic and phonemic similarities as well as ambiguous phonology due to dialectal influence (for native speakers) and foreign consonantal phonology (FL learners). Conclusions: As for native speakers, the results seem to support incipient criticism pointing at general educational problems in the Arab world due to the major differences between spoken and written Arabic (e.g. Maamouri, 1998). Regarding FL learners, the study indicates a need for purposive training of grapheme-to-phoneme correspondences within specific groups of consonants. ----- Bouma, H. (1971). Visual recognition of isolated lower-case letters. Vision Research 11:459-474. Maamouri, M. (1998). Language Education and Human Development. Arabic Diglossia and its impact on the quality of Education in the Arab Region. Mediterranean Development Forum, Marrakech, Morocco, International Literacy Institute, University of Pennsylvania, USA.
Perhaps the most frequently scientifically tested writing intervention is the Self-Regulated Strategy Development (SRSD) model. This approach involves teaching students writing strategies, self-regulation procedures for managing those strategies, and needed skills for using the strategies. This meta-analysis examined the effectiveness of SRSD in writing when tested with experimental group studies as well as single-subject design studies. In a comprehensive search of the literature (including journal articles, dissertations, presentations, and so forth), we identified 15 true- and quasi-experimental studies and 27 single subject design investigations examining the effectiveness of SRSD writing interventions. Effect sizes for the group experimental studies were calculated using Cohen's d, yielding an average weighted effect size for writing quality of 1.20 (effect sizes ranged from .39 to 3.32). The confidence interval surrounding this average weighted effect size was 1.05 to 1.34. For the single subject design studies, the average percent of non-overlapping data (the most common effect size metric for this type of study was 97% for schematic structure of compositions (considered a strong effect). Thus, these findings provide strong support for the use of SRSD as an effective writing intervention.
Emily Hayden (University of Nebraska Lincoln)Trainin, Guy; Wilson, Kathleen; Rankin-Erickson, Joan - Developing fluent readers: Using a research based fluency curriculum in print and technology formats
Purpose Examines the impact of a fluency curriculum in two different formats (print, print-plus-technology) on reading fluency, vocabulary, comprehension skills for students grades 2-5. Method 1484 students, representing demographic achievement levels, environments, ethnicity, mobility, SES, special education. Grades 2-5; 76 classrooms across 9 schools, in Midwestern city. Experimental design, stratified random sampling: 26 classrooms used print fluency curriculum, 27 print-plus-technology, 23 controls used standard district fluency curriculum. Experimental groups used the fluency instructional materials and format: fifteen minute sessions/three times/week; nineteen weeks. Outcomes analyzed across 2-3 timepoints using DIBELS ORF, IRAS-R vocabulary, Gates MacGinitie comprehension in three level HLM. Results Growth slope for students using the fluency curriculum was significantly steeper than controls for fluency, vocabulary, and comprehension across all grades. Print-plus-technology format results were not significant over print format of the curriculum. Conclusions Both formats resulted in significant student gains in fluency, vocabulary, comprehension over controls. Results generalized across achievement groups, ethnicities, ELL level. Study reinforces the value of implementing specific, frequent fluency instruction with research supported texts and procedures. Demands on time, variations in teacher fidelity and commitment are barriers. Organization of the curriculum and support during study resulted in commitment to the instruction that impacted student growth.
Purpose: We examined predictors of spelling skill in deaf children with cochlear implants and compared their performance to that of hearing children matched on age. Method: 44 orally educated children with cochlear implants (mean age 8.8 years) and 44 hearing children (8.6 years) spelled 80 words from pictures. Multilevel logistic regressions predicted spelling accuracy on a by-trial basis from characteristics of the child and of the attempted word. Results: Age at implantation did not independently contribute to spelling accuracy. Deaf children with cochlear implants were poorer spellers (51% correct) than hearing age-mates (64%), but similar characteristics were associated with spelling success for both hearing and deaf children: age, reading skill, whether the child pronounced the target word correctly, and lexical properties such as length and frequency. Conclusions: Many studies of children with cochlear implants have found that age at implantation is an important predictor of language and reading success. For spelling, it appears to play no role beyond its contribution to reading skill. Instead, spelling success is best modeled with the same factors that are important for spelling performance in hearing children.
Purpose The purpose of this study was to examine how skills in morphological awareness (MA), specifically regarding stress-shifting and stress-neutral English suffixes, and linguistic prosody contribute to the prediction of successful reading in children in grades 3, 5, and 7. Method In addition to control, reading, and linguistic prosody measures, participants' MA was tapped using a task adapted from Carlisle (1988) wherein items were divided into 4 categories: No Change (e.g. appear-appearance); Phonological Change (e.g. elect-election); Lexical Stress Change (e.g. artist-artistic); and Both Phonological and Lexical Stress Change (e.g. electric-electricity). Participants were native English speakers in Grades 3, 5, and 7 (n=90) from elementary schools in Eastern Ontario. Results Results showed significant correlations between measures of linguistic prosody, MA, and reading skills (fluency, comprehension, and decoding). Prosody was a significant predictor of MA after controlling for intelligence, receptive grammar, working memory, and PA and better MA skills predicted greater ability in advanced reading. Further, sensitivity to linguistic prosody better predicted ability in the MA of stress-shifting morphemes than the MA of stress-neutral morphemes, and the MA of stress-shifting morphemes better predicted advanced reading ability. Conclusions Sensitivity to stress placement, driven by the morphological rules of English, is related to decoding and comprehension in developing readers. This relationship is moderated by MA; developmental differences are observed in stress-shifting and non-stress-shifting morphemes.
Purpose The time course of homonym meaning activation was investigated in single-word and sentence context in children with poor reading comprehension (PC; n=21) and verbal age controls (n=21) to explore whether individuals with PC are less able to suppress redundant or inappropriate information when forming mental representations. Method In Experiment 1 children heard homonym word primes (BANK) or unrelated word primes (CAKE) and named dominant (MONEY) or subordinate (RIVER) picture targets. In Experiment 2 the homonyms were in sentence context to investigate appropriate and inappropriate facilitation. Results Overall, Experiment 1 revealed priming effects for dominant associates at 250ms and 1000ms ISI; subordinate priming was only evident at 250ms. However, as a group the PCs did not show subordinate priming at either ISI. In Experiment 2 children with PC demonstrated appropriate facilitation at 250ms and marginally significant facilitation at 1000ms ISI, similar to controls. Both groups showed inappropriate facilitation at 250ms; only the PC group showed inappropriate facilitation at 1000ms. Conclusions These results suggest that children with PC may have difficulties with suppressing contextually inappropriate homonym meanings, but they are also less able to activate subordinate meanings in single-word context. The extent to which a suppression deficit may account for the results is discussed.
School-Family Partnership and Code and Vocabulary Skills of American Head Start Preschoolers Purpose This study examined (1) the nature and extent of school-family partnership practices in Head Start, an American preschool program for children in poverty, as well as (2) their associations with code- and vocabulary-related learning (a) from fall to spring of preschool and (b) through kindergarten, and (c) the degree to which these effects depended in part on children's initial levels of skill. Method Participants (n=816 children and families, n=266 teachers, n=209 center directors) were involved in the national Head Start Family and Child Experiences Study. In fall and spring of Head Start and spring of kindergarten, children were administered the Woodcock-Johnson Letter-Word and Dictation subtests of code-related skills and the Peabody Picture Vocabulary Test. Parents and center directors were surveyed on their partnership practices in the spring of Head Start. Results Question-1: Results revealed great variability in family and center partnership practices. Question-2a: Controlling for family background variables and classroom quality, HLMs revealed that book reading, classroom volunteering and parent-teacher meetings were associated with code and vocabulary skills. Question-2b: Growth models showed these effects to be stable through the end of kindergarten. Question-2c: Book reading was most associated with growth for children with initially low vocabulary. Conclusions Over and above other factors, specific school-family partnership strategies relate to the early literacy of high-poverty American preschoolers.
Michael Hock (University of Kansas);Brasseur, Irma; Deshler, Donald; Kingston, Neal - What Is the Nature of Struggling Adolescent Readers in Urban Schools and What Reading Component Skills Best Predict Reading Comprehension?
Purpose: The purpose of this study was to examine the component reading skills of adolescent struggling readers attending urban high schools. Method: Eleven measures of reading skills were administered to 345 adolescent readers in order to gain a research-based perspective on the reading skill profile of this population. Adolescent readers were assessed in the domains of Alphabetics, Fluency, Vocabulary, and Comprehension. Results: Analysis of the results found that 63% of the struggling adolescent reader group had significant deficits in all of the reading components listed above. Subgroups of struggling readers showed similar but more severe patterns. For example, students with learning disabilities scored significantly below the levels of the struggling reader group at large. In contrast, most proficient readers scored high on all measures of reading with above-average component reading skills in word level skills, vocabulary, and comprehension. The lowest skill area for the proficient reader group was fluency. In a predictive analysis of the same data, the component skill of vocabulary was found to be most predictive of reading comprehension. Conclusions: Given the varying profiles of reading problems in urban settings, and the predictive analysis of these data, the findings have implications for policy and instructional programming.
Purpose: To determine the influence of phonological and lexical test item characteristics on phonological awareness performance. The Phonological Deficit Hypothesis and the Lexical Restructuring Model provided the theoretical basis for the study. Method: Typically-developing kindergarteners (ages 5;0-6;0) completed an initial phoneme deletion task containing 20 CVC words varying in phonological (sonority of sound-to-be-deleted: low sonority vs. high sonority) and lexical (neighborhood density: dense vs. sparse) characteristics. A repeated measures ANOVA was used to analyze accuracy data. Results: Results revealed a significant interaction between sonority of sound-to-be-deleted and neighborhood density. Children were able to delete sounds from dense words more accurately regardless of the sonority of the sound-to-be-deleted. However, in sparse words, low sonority sounds were easier to delete compared to high sonority sounds. Conclusions: Sonority of the sound-to-be-deleted and neighborhood density influenced phonological awareness performance in typically developing children. Thus, both theoretical models were supported. The manipulation of these characteristics may increase the sensitivity/specificity of phonological awareness tasks for the early detection of dyslexia. Moreover, considering these factors may increase treatment efficacy.
Purpose A growing amount of research has demonstrated that speech rhythm sensitivity is related to reading independently of phonological awareness. However, we do not yet know whether these observed speech rhythm deficits represent a specific deficit in children at risk of dyslexia, or whether the relationship is due to a developmental delay. Although studies have already compared dyslexic children with age-matched controls in terms of their speech rhythm sensitivity, no study to date has also included a reading-age matched control group, which is necessary to inform this debate. Method Fourteen children aged 9 to 11-years-old classified as 'at risk of dyslexia' were gathered, along with fourteen age-matched controls and fourteen reading-age matched controls. Phonological awareness assessments were administered along with a battery of speech rhythm assessments from the recent literature. Results The results showed that after controlling for receptive vocabulary and phonological awareness, there were significant group differences only on the stress sensitivity measure from the speech rhythm battery. Post hoc analyses revealed no significant group differences between children 'at risk of dyslexia' and the reading-age matched controls, which is suggestive of a maturational lag as opposed to a specific deficit. Speech rhythm was also found to be significantly correlated with reading and phonological awareness, but only in children with typical reading development. Conclusions The findings from this study emphasise the importance of speech rhythm in children's reading development and contribute to our current understandings of segmental phonology in reading.
Purpose: The purpose of the study is to investigate whether growth in a working memory system predicts early writing development of young children aged 3 to 8 years. The questions addressed concern the nature of the working memory system that underlies growth in writing skills. Method: 225 children were enrolled in the longitudinal study of working memory, language, and writing skill development from the ages of 3 - 4 years to 7-8 years. Data was collected in 4 waves; multilevel modeling procedures were used to estimate how growth in working memory capacity influenced growth in writing skills. Findings show that children's working memory capacity increases with age; however, this rate of growth is slow during the preschool years and increases once the children enter school. Change in writing performance (spelling, printing, composing) is mediated by working memory development, but other factors also play a role. Implications for a model of early writing development among young children prior to school entry are discussed.
Purpose The Instructional Content Emphasis-Revised (ICE-R) is used to measure teachers' use of instructional time. However, research into how well the emphasis areas predict student reading achievement have not indicated strong relations. One hypothesis for this is that the level of specificity (having 10 broad areas each with subcategories) may diffuse the measure and create floor effects that are unable to be addressed through transformation of the variables. Method 100 Grade 1-3 teachers were observed during their reading block as part of their participation in one state's Reading First project. Data were analyzed using a) the ICE-R coding criteria and aggregation scheme, and b) aggregation of data based on the big ideas of beginning reading. Due to the clustered nature of the data, HLM was used. The dependent variable was performance on the state criterion-referenced test. Results Differences in overall prediction as well as individual component prediction were found between the two aggregation approaches. Prediction also varied by grade. Conclusions Level of aggregation of the target behavior for observation measures can affect the relation between the target behavior (reading instruction) and outcomes to be predicted (student reading achievement). Implications for research using observation measures will be discussed.
Purpose This paper contributes to the research base on cross-linguistic transfer in the domain of reading comprehension by addressing the following research question: among native Spanish speakers in dual language programs, is first language (L1) reading ability significantly associated with second language (L2) reading ability? If so, does that relationship change over time? Method The sample includes 82 native Spanish speakers enrolled in four dual language programs across the United States. The sample was followed from second through fifth grade, and was repeatedly administered a battery of subtests from the Woodcock Language Proficiency-Revised in both English and Spanish, including picture vocabulary, a decoding cluster, and passage comprehension. Data were analyzed using multiple regression in order to compare the potentially varying effects of L1 literacy on L2 literacy over time. Results Controlling for English oral vocabulary and English decoding ability, Spanish reading comprehension was significantly and positively associated with English reading comprehension at all grade levels, and these effects increased over time, from .24 points in 2nd grade (p<.01) to .34 points in 5th grade (p<.001). Conclusions These findings extend other cross-linguistic research by providing evidence of the ongoing and increasing relationship between first and second language reading comprehension over time.
In alphabetic languages, it is often argued that during the preschool and early school years, children's phonological awareness progress through three levels: from awareness of syllables to awareness of onset-rime and finally to phoneme awareness. The literature to date suggests that syllable and onset-rime awareness develop on the basis of cognitive maturation; however, phoneme awareness develops on the basis of some specific instruction. In this study, we investigated the development of phonological awareness of Taiwanese children from 5 to 7 years old by comparing phonological awareness at the rime and phoneme levels. We further explored the existence of cross-language transfer of phonological awareness from Mandarin Chinese to English. The subjects were 35 kindergarten students, 37 first graders, and 37 second graders. Three phonological awareness tasks were used, including rhyme oddity, initial phoneme isolation, and initial phoneme deletion. In addition, subjects' intelligence, vocabulary ability, working memory were measured as well. The results showed that the developmental patterns of phonological awareness of Taiwanese children were similar to those observed in alphabetic languages, that is, the rime awareness developed earlier than phoneme awareness. In addition, the instruction of Chinese phonetic symbol (Zhu-Yin-Fu-Hao) did facilitate the development of phoneme awareness. Moreover, the cross-language transfer of rime awareness was found in 5 to 7-year-old children, but the transfer of phoneme awareness was influenced by the characteristics of Chinese syllabic structure.
Purpose While early literacy development by bilingual and multilingual children is receiving attention from researchers, multilingual low-SES adolescent English literacy acquisition remains mainly uncharted. This study challenges the assumption that by secondary grades minority language students have acquired basic English literacy skills. Method A two-year longitudinal evaluation of reading, fluency, and spelling of inner city grade 7 and 8 low SES multilingual learning disabled (LD) and non-LD students was undertaken as part of a literacy enhancement project. Standardized tests measured Reading attainment, while Spelling was assessed via a developmental word features test based on error analyses studies. Repeated Measures were performed on all data for both groups at the end of grade 8. Results All measures showed large ranges for both groups. Significant differences occurred between and within groups on all measures, with the smallest change occurring on the spelling factor implicating decoding issues. The reading comprehension age means showed only grade 4-5 level for the LD group, and grade 6-7 for the non-LD group. Conclusions These adolescents literacy development is far from over, and the erroneous assumption that they mastered basic English literacy may contribute to the growth in 'environmental dyslexics'. This high risk population requires explicit instruction in comprehension, decoding, and encoding.
Purpose: To determine whether phonological-orthographic processing deficits were present in a group of adult Spanish-speakers who reported severe difficulty learning English. Method: 30 adult participants (12th+ years of education) were divided into three groups: Poor (PELL), Good (GELL), and Average (PEER) English-Language-Learners, according to self- and teacher- rating scales. They were tested on nonword-decoding, word-recognition and sentence-reading in Spanish. Performance was compared among the groups in terms of time and accuracy. Results: The PELL (N=7) took longer time than the GELL (N=6) in completing all three tasks: nonword-decoding ( p<.014), word-recognition (p<.001), and sentence-reading (p<.002). The PELL also took longer than the PEER (N=16) in nonword-decoding (p<.040), word-recognition (p<.001) and sentence-reading (p<.002). Accuracy levels varied. The PELL were less accurate than the GELL, but not the PEER, in nonword-decoding (p <.001). There were no differences among any of the groups on word-recognition and sentence-reading. Conclusions: Phonological-orthographic processing deficits in the native language may adversely influence the learning of English as a Second Language. Processing time, but not always accuracy, may identify reading-language learning problems in a transparent orthography such as Spanish. Implications are considered for the learning of English, which has a more opaque orthography than Spanish.
Purpose: Investigate whether naming speed correlates with reading ability for adult second language (L2) readers when they read an orthography that is distant from their first language (L1). Method: Measure the English letter naming speed of sixty-two Japanese adult readers with at least 6 years of English education, using discrete and serial formats, as well as their cognitive and linguistic abilities. Results: Serial naming speed correlates with word-level reading accuracy but does not explain additional variance when entered into regression analyses after working memory and phonological processing skills. Discrete letter naming, in contrast, correlates with word reading speed and explains unique variance after controlling for working memory and phonological processing. Conclusions: This study shows that naming speed correlates with reading ability for L2 adult readers, independent of phonological processing skills. This is consistent with the literature for children (e.g. Geva and Yaghoub-Zadeh, 2000; Mannis et al. 1999). It also demonstrates that discrete letter naming is a stronger predictor of word-level reading ability than is serial letter naming for adult readers from non-alphabetical orthography backgrounds, in contrast with studies of L2 adult/adolescent readers that found serial naming speed plays a significant role (Morfidi et al, 2007; Nassaji and Geva, 1999).
Purpose: This study was a replication and extension to a much larger sample of previous findings that children's routine writing samples could be scored reliably and showed more improvement in features such as quality of spontaneous spelling in classrooms in which the teaching was rated as more consistent with program objectives. Method: Participants were kindergartners enrolled in classrooms assigned to the Breakthrough to Literacy (BTL) curriculum as part of a large-scale randomized intervention study. In this very large Midwestern city, most schools serve high proportions of students who come from low-income families. A total of 1844 monthly writing samples were collected from 208 children in 21 classrooms. These writing assignments were a routine part of the BTL curriculum. Intensive observations of these classrooms were done using the OMLIT scale (Goodson, Layzer, Smith, & Rimdzius, 2004), which includes a subscale (QUILL) that focuses on writing instruction. Results: Writing samples have been prepared for coding and the OMLIT data are available, but results have not yet been determined. Conclusions: We expect to find that features of kindergartners' writing such as spontaneous spelling develop more rapidly in classrooms that observers rate as high in quality of instruction.
Purpose Social dialects vary on oral language parameters, such as phonology and syntax. We examined whether the degree of non-mainstream dialect played a role in accurate derived word stress production and in decoding proficiency. Method Sixty-four third graders in Memphis, TN participated in the study. Our population included speakers of African-American English (AAE) and Southern American English (SAE). Degree of dialect was determined by identifying AAE and SAE morphosyntactic features on a language test, then calculating the total number of dialect features divided by the number of test items. Socio-economic status (SES) was estimated from parent occupations. Stress accuracy was measured by having children produce derived words with stress-changing suffixes (e.g., put -ity on the end of festive to produce festivity). The WRMT provided the decoding measure. Results Degree of dialect (DD) was significantly and negatively correlated with all variables: More dialect features were related to poorer performance. DD contributed uniquely to stress accuracy performance, after controlling for age, sex, and SES. When decoding is regressed on DD and SES, DD accounts for 12.6% of the variance in decoding scores. Conclusions Degree of dialect influences performance on our stress accuracy task. Considering the strong relationship between stress accuracy and decoding, we may want to further explore the influence of dialect on reading development.
Purpose: This study seeks to improve the identification accuracy of second and third graders at risk for reading problems, using end of grade 3 levels on a standardized assessment (SAT 10) and a state assessment (FCAT) as the criterion measure. Methods: Using a large extant database from the Florida Center for Reading Research, we: 1. Establish levels of performance on the outcome measures that separate 'successful' from 'unsuccessful' reading. 2. Separate the data files into three subsamples to run classification analyses with an initial group, and cross validate scores with subsequent groups. 3. Compute discriminant analyses using a set of predictors (PPVT, Reading Vocabulary, previous SAT scores) available at each grade level. 4. Evaluate the classification statistics and cross-validate with another sample 5. Determine cutpoints that yield sensitivity of 90% or greater and specificity rates 6. Generate ROC curves to examine AUC using a combination of predictors. Results: Data sets were received December 6, analyses to be completed in January 2008. Results will report the best set of predictors using sensitivity/specificity and ROC analyses. Conclusions: This study is unique in that it sets a minimum sensitivity level (> 90%) and tests improvements in specificity as a function of additional predictors.
Bonnie W Johnson (University of Florida) ;Lombardino, Linda; Farrar, M Jefferey; Dugger, Cynthia; Beresheski, Marcelline; Hogue, Rachel - The relation between emergent literacy skills and theory of mind in preschool children
Purpose: This study examined whether cognitive flexibility, as measured by theory of mind, is related to emergent literacy skills. Method: Thirty-four children with typical language development, ages 48 - 65 months (X=55.44, SD=4.57), participated in this study. Theory of mind tasks included appearance/reality, representational change, and false belief items. Language and emergent literacy tasks measured children's receptive vocabulary, basic knowledge, letter knowledge, rhyme, and phoneme deletion abilities. Results: Correlation and regression analyses were used. Maternal education and/or receptive vocabulary, but not theory of mind, accounted for a significant amount of variation in all emergent literacy tasks. Conclusions: Theory of mind has been found to be related to many aspects of language development. In this sample of preschool children it did not predict children's emergent literacy skills. Cognitive flexibility, as measured by theory of mind, may play a role in some aspects of children's development of reading, but it does not appear to play a role in phonological awareness skills in preschool children.
Purpose: The relationship between Rapid Automatized Naming (RAN) and reading ability is well documented (see Wolf & Bowers, 1999), but little is known about which component processes are important in RAN, and why developmental dyslexics display rapid-naming deficits. We investigated whether RAN and RAN-deficits are influenced by extra-phonological influences (Wolf & Bowers, 1999) or by phonological processes only (Clark, Hulme, & Snowling, 2005). Method: We conducted an eye-tracking study that manipulated phonological and visual information (as representative of extra-phonological processes) in RAN. We obtained articulation measures (the Eye-Voice span) in addition to processing times (gaze duration). Results: Results showed that both phonological and visual processes influence naming speed for both dyslexic and non-dyslexic groups, but the influence on dyslexic readers is disproportionately greater. Conclusions: Our findings suggest that whilst phonological processes are critical in determining RAN performance, and contribute to the naming-speed deficit in dyslexia, extra-phonological processes are also influential. We discuss the implications for causal theories of dyslexia.
R. Malatesha Joshi (Texas A & M University)Aaron, P.G.; Binks, Emily; Boulware-Gooden, Regina; Carreker, Suzanne; Graham, Lori; Washburn, Erin - Where does the disciplinary knowledge for reading instruction come from?
Purpose: McCardle and Chhabra asked a fundamentally important question, "Why is it that the colleges of education do not arm teachers with this vital information before they enter the classroom to teach our children?" In order to answer this question, we hypothesized that perhaps (a) teacher educators themselves may not be knowledgeable about the linguistic constructs and/or (b) the textbooks used in the reading education classes may not be providing the information about evidence based reading research. Method: In order to test the above two hypotheses, we measured the disciplinary knowledge of teacher educators who prepare preservice teachers through survey as well as interview and also examined the textbooks used in reading education courses for the content knowledge. The results showed that many teacher educators are not knowledgeable in many of the basic linguistic constructs needed for literacy development and many of the textbooks used in reading education courses in colleges are not providing the evidence-based reading instruction programs. It is concluded that (1) teacher training programs should explicitly teach the components in effective reading instruction and (2) textbooks used in pre-service reading education courses be carefully selected to include all components of scientifically-based reading research
Laura Justice (Ohio State University); Anita McGinity; A. Breit; Sonia Cabell; Joan Kadaravek - Enhancing Emergent Literacy Skills in Children with Language Difficulties: Findings from Two Intervention Studies
Findings from two experimental intervention studies demonstrate the effect of meaning-focused and print-focused book reading on the print knowledge of preschoolers with language difficulties (-1 SD on a standardized measure of language). Intervention 1 (12 weeks) involved 48 parent-child book reading sessions where parents focused questions/comments on the book's meaning (comparison, n = 13) or print (treatment, n = 15) throughout the session. Intervention 2 (30 weeks) involved teachers conducting daily interactive reading (comparison, n = 21), daily interactive reading with a print-focus (high-intensity treatment, n = 34), or twice weekly interactive reading with a print focus (low-intensity treatment, n = 27). Preliminary findings comparing children's gains in print concept and alphabet knowledge across conditions using ANOVA and calculations of bias-corrected Cohen's d suggest a significant advantage of print-focused reading for enhancing children's print concepts in studies 1 and 2 (p < .05; d = .92, 4.63, respectively). However, Study 2 found the print-focused advantaged in the low-intensity treatment condition, only. Planned analyses will consider effects of condition after controlling for the socio-emotional quality of the intervention context and child attentional characteristics. Implications for the design of early literacy prevention programs for children with language difficulties will be discussed.
Purpose: The current study explores pre-reading children's understanding of the elements of written words. Previous studies have shown that children expect written language to be linear and composed of symbol-like elements (Bialystok, 1995). Additionally, first graders judge letter strings to be better words if they contain common rather than uncommon letter combinations (Treiman, 1993). The current study uses eye-tracking technology to explore children's implicit orthographic knowledge. Method: We presented 3-5-year-old children with a series of non-words and asked them to judge whether symbol strings were 'words'. We tracked their fixation patterns to determine where their attention was directed. Results: Preliminary results show that pre-readers fixate longer on symbols or numbers than letters when determining word status; however their fixation rate is approximately equal for both legal and illegal consonant clusters. This relationship between this and other indicators of emergent literacy, like letter knowledge, phonemic awareness and symbolic understanding are also addressed. Conclusion: This suggests that prior to instruction in reading children have a sense of which individual symbols may or may not be used to assemble words; however, at this stage they do not use orthographic legality when judging words from non-words.
Purpose: We investigated variation in family literacy practices in a diverse rural sample, and its associations with children's language development, using a unique family literacy survey designed for use with rural families. Method. We estimated the relations among social address characteristics of children, family literacy practices, and children's developing language skills (i.e., standard scores from the Preschool Language Scale) within a sample of 1005 children participating in the Family Life Study (43% African-American; 49% Male), a longitudinal study of children's development and family practices in rural America. Results from initial model. In the context of simultaneous linear regression, access to technology and child emergent literacy behaviors account for children's language skills above and beyond social address variables. However, these findings represent main effects, which most likely mask the diversity of relations between family processes and child outcomes in this rural sample. For Presentation at SSSR. We will present rich description of four categories of family literacy within a rural sample that varies by state location, race, and family income. Additionally, we will further probe the present findings to illustrate more clearly the interactive effects of race, income, and family literacy on children's developing language skills.
Researchers of literacy development have found that children's oral language development contributes to their early literacy and later school achievement. The role of literacy environment in bilingual children's oral language competency in their language of schooling is, unfortunately, less researched. Purpose: This study is an exploratory examination of the role of home literacy environment in middle-income Indian bilingual children's English receptive vocabulary. Method: Data were collected from 90 kindergarten children (mean age = 51.26 months) from two schools in Bangalore, India. Home literacy environment was assessed using a parent questionnaire. Children were administered the Peabody Picture Vocabulary Test (PPVT) in school. Using structural equation modeling the relationships between book reading practices, degree of English parents spoke to child, children's age, demographic variables (income and parental education levels) and children's PPVT scores were analyzed. Results: The home environment variables along with children's age explained 83% of the variance in children's PPVT scores. Home book reading practices explained the most variance in children's vocabulary. Conclusion: Although the sample size is relatively small, and the study exploratory, these findings illustrate the importance of parental involvement in Indian children's oral language development. These findings are discussed in relation to implications for children's literacy development.
Purpose Emergent literacy skills detected during the preschool years have been shown to be predictive of later reading success. However, gathering reliable data during the preschool period is sometimes a difficult task and there is often disagreement about which child is truly at risk for reading problems. Two ways to improve classification accuracy in testing were examined: alternate assessment techniques and alternate scoring methods. Method In a longitudinal study, preschool students were randomly assigned to three assessment groups: regular, scaffolded, and dynamic assessment administration and given three phonological awareness measures at three times and a print awareness task. In a second study of preschoolers, a traditional unit weighted scoring approach was compared to a regression weighted scoring approach for three phonological awareness measures. Results Study 1 results concluded that student reading performance was better predicted in the scaffolded and dynamic assessment groups when compared to the regular group. Study 2 results indicated that the regression weighted scoring approach was a better predictor of reading ability as measured by a print awareness task. Conclusions The results of both studies confirmed the hypotheses that alternative testing and scoring methods may be useful for better identifying students at risk for reading difficulties.
Purpose - Literacy instruction in the UK is widely supported by computer software programs aimed to improve literacy skills. This study investigated the effectiveness of a whole-word multimedia software program, 'Oxford Reading Tree for Clicker', compared to traditional printed texts using 'Big Book'. Method - Sixty-one typically developing readers, aged 5-6 years, were drawn from four Year-1 classes across three primary schools. Each child was given two interventions ('Clicker' and 'Big Book'). A different story from the Oxford Reading Tree scheme was used in each intervention, which was delivered over five one-hour sessions over the course of one week. The order of intervention was counterbalanced across participants. Performance on tasks of written word recognition, written word naming, and phonological awareness, was measured pre and post each intervention week, so that performance gains could be determined for each intervention. Results - Gains in performance were found following both interventions. However, significant gains were found only after the 'Clicker' intervention, and only for written word recognition. Conclusions - 'Clicker' is more effective at promoting visual whole-word recognition skills than are traditional printed texts, possibly because 'Clicker' highlights words as they are spoken aloud by the narrator, thus drawing attention to the written form.
Purpose: Children with ADHD typically show deficits in reading comprehension. Is this because of a comorbid word reading disability that produces their comprehension difficulties? Or is it because they have specific deficits in comprehension? We address this question by examining the comprehension skills of children with ADHD who do not have word decoding deficits, and in fact, are matched to controls on word decoding. We examine not only reading comprehension, but also listening comprehension. Method: 76 children with ADHD and 76 controls matched for word reading skill were given multiple reading and listening comprehension tests. Results: We found that children with ADHD show comprehension problems relative to controls even when the two groups have equivalent decoding skill. This shows that children with ADHD have specific comprehension difficulties that cannot be attributed to comorbid decoding skills. Further support for these comprehension deficits comes from our finding that they are poorer not only in reading comprehension but also in listening comprehension. Interestingly, while controls show no difference between listening comprehension and reading comprehension, children with ADHD tend to perform worse while listening than while reading. Conclusion: We conclude that there are comprehension deficits in ADHD that are independent of word decoding deficits.
Purpose The present study examines the relationship between passage cohesion and reading fluency on students in grades 3, 7 and 10 (n = 213, 186, 176 respectively). Method Oral reading fluency scores were collected using 3 passages from an oral reading fluency assessment, 3 passages from a standardized assessment, and 3 grade level passages taken from texts. The texts were then coded using Coh-Metrix, a computer program that analyzes texts on multiple measures of cohesion, language, and readability (Graesser et al., 2004). Correlations between Coh-Metrix output and fluency scores were computed for each grade. Results Results revealed strong correlations between level of cohesiveness and reading fluency for 3rd grade, such that passages with similar syntactic structure among sentences and paragraphs were read more fluently than those with less syntactic structure. However, in 10th grade significant correlations were found between vocabulary and level of cohesiveness, such that passages with a higher incidence of concrete content words were read more fluently than passages with a higher incidence of abstract content words. Conclusions These results have implications for the creation of future curriculum based measurement assessments.
Heidi Kiefer (Ontario Institute for Studies in Education/University of Toronto)Kiefer, Heidi; Chen, Xi; Girard, Lisa - Different Strokes for Different Folks: Differences in Cross-Language Orthographic Transfer for Spanish-English and Chinese-English Bilinguals
Purpose To investigate the relationship between orthographic knowledge and recognition of basic writing units within both alphabetic (Spanish) and nonalphabetic (Chinese) systems, as well as cross-language transfer of orthographic skills to English. Spanish and Chinese orthographic skills were hypothesized to significantly contribute to within-language reading. Cross-language orthographic transfer was predicted only with Spanish-English bilinguals due to similarity of writing systems. Method Spanish-English (n=98) and Chinese-English (n=78) bilinguals in grades 4 and 7 completed tasks assessing orthographic knowledge and character and/or letter-word reading in their respective languages. Separate Spanish and Chinese initial analyses consisted of step-wise regression with orthographic knowledge as predictor of same-language character/word reading skills, controlling for age, mother's education, intelligence, memory and phonological skills. Secondly, parallel analyses were conducted with English letter-word identification as dependent variable to explore cross-language prediction. Results Both Spanish and Chinese orthographic processing significantly predicted same-language letter-word (Spanish) or character (Chinese) reading. In terms of cross-language transfer, Spanish orthographic skills contributed a significant amount of unique variance to English letter-word identification. This was not evidenced with Chinese orthographic skills. Conclusions Findings suggest orthographic awareness impacts basic reading in Spanish and Chinese but cross-language orthographic transfer occurs only between languages with similar phonetic-based writing systems.
Michael Kieffer (New York University)Lesaux, Nonie - Dimensions of Vocabulary Knowledge among Adolescent Language Minority Learners and Monolingual English Speakers: Morphology, Polysemy, Contextual Sensitivity, and Academic Language
Purpose: This study examines the multiple dimensions that comprise the vocabulary knowledge construct among adolescent language minority (LM) learners and monolingual English (EO) speakers. Method: Five-hundred, sixty-one students in sixth grade (392 LM, 169 EO) were assessed with standardized and researcher-created reading vocabulary measures, including general vocabulary, polysemy, contextual sensitivity, word association, academic vocabulary, and real-word and pseudo-word morphological awareness tasks. Exploratory Principle Components Analysis (PCA) and Confirmatory Factor Analysis (CFA) will be used to examine the factor structure of vocabulary knowledge, and to investigate whether the same structure describes the vocabulary knowledge of LM learners and native English speakers. We hypothesize that these measures tap two related latent dimensions, one representing breadth of vocabulary (i.e., the number of words children know) and another representing depth of vocabulary (i.e., how well children know those words). Results: The first set of analyses using PCA suggested two factors with loadings that support the hypothesized two dimensions. The next set will confirm this factor structure using multiple-group CFA and test whether LM learners and native English speakers demonstrate the same factor structure. Conclusions: If confirmed, these findings suggest that multiple dimensions of vocabulary should be considered when building theoretical models of reading comprehension.
Purpose To investigate the relationship between auditory rhythmic timing cues, phonological awareness and reading in Garden Variety (GV, children with depressed IQ and reading level) poor readers. Method A battery of phonological and auditory processing tasks was administered to four groups; GV poor readers, a group of dyslexics matched on reading and chronological age, typically developing chronological age controls and typically developing reading age controls. Results Contrary to expectations, GV and dyslexic children were not significantly different on the auditory processing tasks. All children with poor reading showed poor auditory processing and poor phonological development. The GV poor readers showed more severe auditory processing deficits, however these were found across all tasks, whereas for dyslexics the auditory deficits were specific to certain auditory tasks. Conclusions These findings suggest that children with poor IQ and poor reading development share with dyslexics a core phonological deficit. The potential role of auditory processing in causing this core deficit will be discussed.
Purpose This study examined the relationship of growth trajectories of oral reading fluency, vocabulary, letter naming fluency, and nonword reading fluency from grade 1 to grade 3 with reading comprehension in third grade. Method Data from 12,000 children who were followed from kindergarten to third grade longitudinally were used. These children were administered DIBELS, PPVT, and reading comprehension (SAT 10) tasks multiple times in each year. Using Hierarchical Linear Modeling, we estimated children's initial status and rate of growth in each predictor within each grade. In addition, dominance regression was conducted to examine relative contributions that the predictors made to the outcome, reading comprehension in third grade. Results Among the first grade predictors, individual differences in growth rate in oral reading fluency and vocabulary skills made the most contribution to reading comprehension in third grade. Among the second and third grade predictors, children's initial status in oral reading fluency and their vocabulary sizes had the strongest relationships with their reading comprehension skills in third grade. Conclusions These results indicate the importance of both oral reading fluency and vocabulary for reading comprehension achievement in third grade. Furthermore, growth of oral reading fluency in first grade appears to have important implication for later reading comprehension achievement.
James Kim (Harvard University)Guryan, Jonathan University of Chicago, Graduate School of Business This study was funded by The William T. Grant Foundation. - Can Parents and Family Members Enhance the Effectiveness of Voluntary Book Reading During Summer Vacation? Results from a Randomized Experiment
Purpose Numerous empirical studies have shown that children from low-income and minority families fall behind in reading during summer vacation. To prevent reading loss, we designed an intervention in which children received books matched to their reading level and were instructed to use fluency and comprehension strategies while reading books at home. We undertook an experiment to examine whether parent training could improve student outcomes. Method A total of 400 low-income Latino children in fourth-grade were randomly assigned to 1 of 3 conditions: (1) control group, (2) treatment group (leveled books only) and (3) a family literacy group, in which children received the core treatment and parents were trained in biweekly family literacy events to use comprehension strategies to support children's book reading activities at home. Children were pre and posttested on measures of home literacy activities and reading comprehension (Gates-MacGinitie Reading Test; GMRT). Results Children in the family literacy group reported reading significantly more books and engaging in more literacy activities with family members than the control group. We also found significant effects (ES = .24) on the GMRT for children who received well-matched books (i.e., independent reading level). Conclusion The National Literacy Panel noted that "evidence for the impact of parent involvement efforts on children's achievement is surprisingly thing." Results from this experiment suggest that family literacy programs can enhance student outcomes and also underscore the importance of providing books that are at children's independent reading level.
Purpose The effects of e-book reading on children's emergent reading with and without adult mediation were investigated. Method Educational electronic storybooks developed to support young children's early literacy were used. Ninety-six 5 to 6 year-old kindergarten children from low SES families were randomly assigned to one of three groups (32 children each); (1) independently reading the e-book; (2) reading the e-book with adult mediation, (3) receiving the regular kindergarten program (control). The two intervention groups included four book-reading sessions each. Pre-and post-intervention emergent reading measures included CAP, word reading, and phonological awareness. Results Regression analysis showed that only the activity of reading the e-book with adult mediation contributed significantly to the children's emergent word reading and their CAP knowledge compared to the control group. Reading the e-book with and without adult mediation also contributed to the children's phonological awareness compared to the control group. Conclusion Our discussion will focus on the type of support that young children need in order to show progress in different emergent reading skills when using e-books.
Purpose This study investigated the effect of early bilingualism on language development beyond cross-language transfer. Drawing on the cognitive flexibility theory (Hakuta & Diaz, 1985; Kuo, 2006), which posits that bilinguals may universally develop a greater readiness to reorganize linguistic input, the study focused on the processing of phonotactics (i.e., how phonemes are combined to form acceptable syllables in a language). Method Participants were 193 Mandarin-speaking kindergarteners, first-graders and second-graders in Taiwan with different degrees of exposure to a Taiwanese dialect. Experiment 1 compared bilinguals to monolinguals in their ability to judge the well-formedness of Mandarin syllables. Experiment 2 examined how they acquired a novel phonotactic system with phonemes in Mandarin but syllables that do not exist in their linguistic repertoire. Results Bilinguals showed a disadvantage in judging the well-formedness of Mandarin syllables at the kindergarten level, which may be attributed to the overlap of their two developing phonotactic schemata. However, by first grade, bilinguals caught up to their monolingual peers and even demonstrated an advantage in the ability to acquire a novel phonotactic system. Conclusion The study provides support for the cognitive flexibility theory and calls for a need to reconceptualize bilingual effect beyond the prevalent cross-language transfer view.
Purpose: This study examined relationships between children's development of early syllable structure, familial risk for reading disability (RD) and later reading status. The implication for differences in early syllable structure is discussed in terms of its utility as an indicator of phonological complexity. Method: Babbling between the ages of 6 and 18 months was examined in 22 children. (7 were eventually identified with RD, 8 were normally reading counterparts at low familial risk for RD, and 7 were at high familial risk for RD but were normal readers). Levels of syllable complexity were coded using a hierarchy adapted from Paul and Jennings (1992). Development of syllable complexity was examined both across individual trajectories and cross-sectionally at 'early' and 'late' babbling stages. Results: Preliminary findings indicate that children at familial risk for RD showed trends toward fewer syllables per utterance in early babbling (prior to 14 months) and used significantly fewer consonant clusters per syllable after 14 months. Conclusion: As syllable complexity is an early indicator of phonological sophistication, differences at this level may offer a window into how the phonological system of children with RD is structured during a period when single words are emerging.
Purpose: Deficits in speed of processing may contribute to higher level language and literacy impairments. The present study aimed to assess the relationship between speed of information processing and reading ability in typically-developing children, whilst controlling for the influence of well-established predictors of reading performance. Method: A sample of 74 children aged between 6 years 10 months and 10 years 9 months were recruited from main stream schools in the UK. The test battery included standardised tests of expressive and receptive vocabulary knowledge, non-verbal ability and processing speed, together with measures of phonological processing and verbal memory skills. Children also completed a series of experimental processing speed tasks on a laptop computer, and measures of single word reading and speeded reading. All children were seen individually in their schools on two separate occasions to complete the test battery. Results: Findings suggest that after controlling for vocabulary knowledge speed of processing is a significant unique predictor of reading. However, after reading-specific predictors (e.g. phoneme deletion) are entered into the analyses, the influence of speed of processing becomes negligible. Conclusions: The implications of the findings for typical and atypical reading development will be discussed.
Amy Lederberg (Georgia State University);Easterbrooks, Susan R.; Miller, Elizabeth Malone; Page, Jessica Robin; Connor, Carol McDonald - The Development of Phonological Sensitivity, Vocabulary, and Early Literacy Skills in Deaf Preschoolers with Cochlear Implants
Purpose: To describe emergent literacy skills for young children with hearing loss. Method: 26 children (average age = 60 months) enrolled in self-contained early childhood classes for children with hearing loss were assessed in the fall and spring of the school year. All children had at least one cochlear implant. Average age of implantation was 20 months. Results. Standard Scores showed that knowledge of letter names and common written words was age appropriate. Vocabulary and phonological sensitivity skills lagged far behind that of their hearing peers. Across the school year, raw scores significantly increased on tests of Letter-Word Identification (WJ), Passage Comprehension (WJ), Letter-Sound Association, Alliteration, Elision, and Blending (TOPEL). Scores did not change for Rhyming, Syllable Segmentation, and Vocabulary (WJ). Vocabulary, phonological skills, and memory (verbal word span) significantly correlated with literacy skills both across time and at each testing. Conclusion. Future research needs to explore whether rhyming and word segmentation are unusually difficult for deaf children or if these results are due to the nature of their classroom instruction.
The orthography to phonology correspondence of Chinese can be either described by a dichotomous index of regularity or by a quantitative index of consistency. This study aims to explore when and how the knowledge of phonetic consistency is acquired by children in learning to read different types of Chinese characters. Children of 4th and 5th grades were asked to name a set of phonograms, which can be divided into two consistency levels (high and low) and three word types (regular, irregular, and bound phonetics). The 4th grade children showed higher accuracy for reading high consistent characters than for reading low consistency characters and this effect was found in reading regular and irregular characters but not for reading bound phonetic. For 5th graders, the accuracy data showed the same pattern consistency effect for reading all types of characters. As for reading time, the consistency effect was only found in reading regular characters. These findings suggest that the awareness of regularity helps to aware the phonetic consistency.
Nonie Lesaux (Harvard University Graduate School of Education);Kieffer, Michael - Qualitative Differences or a Question of Degree? The Prevalence and Sources of Reading Comprehension Difficulties among Language Minority Learners and their Classmates in Urban Middle Schools
Purpose: This study was designed to investigate the prevalence and nature of reading comprehension difficulties among a sample of language minority (LM) learners and monolingual English (EO) speakers drawn from the same urban middle school classrooms. Method: Five hundred and eighty-one sixth grade students (399 LM; 182 EO) were screened for reading comprehension difficulties using a standardized measure. Students who scored at or below the 35th percentile were administered measures of general vocabulary, academic vocabulary, nonword reading, word reading efficiency, passage reading fluency, and working memory. Results: Sixty-percent of LM learners were identified as struggling readers, compared to 40% of EO students. However, among these struggling readers, there were no mean differences on measures of language and literacy skills, with the exception of a small to moderate difference in general vocabulary between LM learners and EO students. Using latent class analysis, the results demonstrate that LM and EO students were evenly distributed among four distinct skill profiles of struggling readers. Conclusions: Although the population of LM learners may be at slightly greater risk than EO students for reading comprehension difficulties, in urban middle schools, the two groups show comparable and similarly diverse patterns of reading difficulties.
Purpose: The present study examined the quality of lexical representation in Chinese children and proposed a model as a general framework for understanding Chinese word reading. According to the Lexical Quality Hypothesis (Perfetti, 1992), a high quality representation would have a fully specified orthographic representation and redundant phonological and semantic representations. Method: Four literacy tasks and six measures representing mappings of the orthography-to-phonology route and orthography-to-semantics route were administered to 94 Grade 2 to Grade 4 Chinese children in Hong Kong. Results: Regression analyses showed that both routes were significant in predicting Chinese word reading and meaning. Moreover, the result that high frequency words were read aloud correctly but less well understood in meaning implied the earlier development of orthography-phonology connection. Higher graders showed more effective use of radicals and rapid retrieval of pronunciations and meanings of characters as compared with their younger counterparts, which indicated that they possessed higher quality and reliable representations. These results show that the lexical and sublexical links become more interconnected and redundant with increasing grade levels. Conclusions: It is concluded that both orthography-to-phonology and orthography-to-semantics connections are important in reading and understanding Chinese characters. Automatic lexicon for efficient reading in Chinese requires precise, redundant and integrated lexical representations in both routes of all three constituents.
Young children's ability to read and spell their own and classmates' personal names was studied. Hebrew-speaking pre-literate kindergartners (N = 60) were given several tasks assessing personal name knowledge in and out of context as well as emergent literacy and cognitive skills. Most children knew how to read and spell their own names. Children varied greatly in their knowledge of classmates' names. Cues inherent in the written names rather than environmental cues supported identification. As indicated by regression analyses, personal name knowledge was explained not by general cognitive skills (i.e., vocabulary or math) but primarily by letter name knowledge and secondarily by phonological awareness. Knowing over half of the letters provided a major boost to children's success on the personal names tasks. Beyond two-letter names which were the easiest to read and spell, differences were not observed indicating that longer names were remembered using partial cues. We theorize that informal learning about personal names is fueled by children's needs or interests and emerges from a foundation of letter knowledge and phonological awareness combined with exposure to names. Findings suggest that teachers might capitalize on this course of development by incorporating personal names of classmates into emergent literacy instruction.
Predictors of Reading Comprehension in Chinese English-immersion Students Purpose. We investigated the prediction of English reading achievement from English and Chinese phonological awareness (PA) and naming speed (NS) in Chinese English-immersion students, to test the Double Deficit hypothesis and look for cross-linguistic transfer. Method. Participants were 49 Grade 2, 48 Grade 4, and 40 Grade 6 students in immersion programs in China. Measures were, for PA, Chinese and English sound detection, and Chinese tone detection; for NS, Chinese and English digit naming; for reading achievement, Cambridge Young Learners English Tests. Results. Hierarchical regression analysis indicated that at Grades 2 and 4, English PA and NS were significant predictors of reading; Chinese PA and NS did not add any further variance. At Grade 6 the English PA measure did not have adequate variance, but English NS predicted reading; Chinese PA added variance at p = .08 level. When the Chinese predictors were entered first, Chinese PA predicted at each grade level, but Chinese NS only at Grade 4. English NS added significant variance at Grades 2 and 6. Conclusion. Both English predictors were stronger than both Chinese predictors. There was no evidence that Chinese PA or NS had any additional effect after controlling English PA and NS.
Purpose. The Canadian educational system is challenged by an increasingly diverse population. The purpose of this study was to examine the profiles of at risk fourth grade readers on the Progress in International Reading Literacy Study (PIRLS) in terms of contextual and student ability factors. Method. The participants included 4295 students (1531 students from French-language schools) and were divided into 2 groups: those who scored above and below the 25th percentile on the reading achievement benchmarks (i.e., Overall Reading, Reading for Literacy Purposes and Reading to Acquire and Use Information). Items were grouped into the following factors: SES, home literacy, general home, school literacy, general school, and student ability upon entrance at Grade 1. Results. The means for the at-risk group were significantly lower than their typically achieving peers on all factors examined. When hierarchical regression analyses were completed, first controlling for SES and language background, only individual student ability factors were consistently significant predictors of at risk reading status. Conclusion. The results of the profile analysis revealed that Francophone English language learners differed from their peers on several contextual home and school factors, but that when SES and language background were controlled, skills upon entrance to school most consistently predicted reading status.
*Purpose - This study had two purposes: 1) to compare the development of phonological awareness between children in China and Chinese-speaking children in Canada, and 2) to examine the role of phonological awareness in character reading among both groups of children. *Method - Participants were 93 Chinese children and 102 Canadian children from kindergarten and grade 1. They received a battery of phonological awareness and reading measures in Chinese. *Results - Chinese children outperformed Canadian children on syllable awareness, whereas Canadian children scored higher on the phonemic deletion task. There was a significant interaction between country and grade on phonemic awareness. Interestingly, first graders in China, who had received pinyin instruction, performed similarly on phonemic awareness as Canadian first graders. We also found that both syllable and phoneme awareness predicted unique variance in Chinese character reading in China and Canada after controlling for age, mother's education, rapid digit naming, and PPVT. *Conclusions - Because syllable is more salient in Chinese and phoneme is more salient in English, these results support that children's development of phonological awareness is shaped by their early spoken language experience. Moreover, learning an alphabetic system significantly improves children's phonemic awareness. Finally, phonological awareness plays an essential role in Chinese reading.
Purpose This study addresses the following questions: 1) Does socioeconomic status (SES) relate to reading comprehension in the elementary grades? 2) Do SES components vary in predictiveness? 3) Does race, gender, or time of assessment affect these relations? Method A meta-analytic review was conducted on studies implemented in the United States which examined individual-level SES measures relation to reading comprehension outcomes. Electronic searches of ERIC, PsycINFO, and Sociological Abstracts, as well as, hand searches of prior literature reviews and references lists of studies identified were conducted. Effect size and descriptive data was coded for the 12 eligible studies with interrater agreement at 90%. Results A random effects model using the method of moments technique was employed identifying an overall mean of 0.25 (p ≤ .01). The findings suggest that measures of SES may serve as better predictors of reading comprehension performance when-multiple indicators are included, their assessed prior to grade 1, involve a racially homogeneous sample; and/or relate to comprehension specific measures. Conclusions These findings emphasis the importance of researchers systematically considering how performance estimates vary based on the study's features. Understanding the relation between SES and reading may help unveil the social factors involved and guide instructional development.
Purpose: Serial naming is often used to predict reading ability, however, the cognitive components of this task that are related to reading are still being debated. As naming letters in isolation requires many of the same cognitive demands as serial naming (both tasks require participants to see the stimuli, retrieve its phonological representation, and respond verbally), we hypothesize that any differences in performance between these two tasks ¬should be unrelated to the phonological process of converting visual stimuli to verbal outputs. Thus, the purpose of this study was to explore and manipulate the differences between serial naming and isolated naming; allowing us to isolate the differences between the two tasks and determine which difference is responsible for the unique relation of serial naming with reading. Method: In this study, differences between the tasks are systematically varied, resulting in seven locally-developed conditions of the naming task. Fifty college-age students were given all seven conditions in addition to measures of reading ability. Results: Individual differences on the seven conditions were unexpectedly unrelated to reading outcomes, possibly due to range restriction. Mean differences in letters read per second suggest that the variance in serial naming is a result of multiple letters being visible at once. Conclusion: Results indicate that parafoveal information is an important component in the serial task, and should be considered and examined more explicitly in reading.
Purpose: This study examined comprehension of passive and complex sentences in adults with dyslexia. Method: Thirty-three normal readers (NR) and seventeen adults with dyslexia (aged 16-28) participated in the study. Participants completed a battery including working memory (WM), executive function, and vocabulary tests, plus a timed word reading task. Participants saw a sentence on a computer screen over two pictures arrayed horizontally. Distractor pictures showed the same participants with roles reversed. Experiments contrasted comprehension of active and passive sentences and four types of relative clause sentences: center- and right-branching, subject and object relatives. Results: Participants with dyslexia were significantly less accurate than NRs at comprehending passive sentences but not active sentences, even with word reading covaried. With word reading covaried, participants with dyslexia were significantly less accurate than NR comprehending sentences with center-embedded, objective relative clauses, but accuracy did not differ between groups otherwise. Groups did not differ in response times in either experiment. Covarying WM eliminated differences in performance in both tasks Conclusions: These findings suggest that adults with dyslexia may experience difficulty comprehending structures that require WM. We hypothesize that impaired, underspecified phonological representations place an increased load on WM resources, affecting complex language processing at many levels.
Christopher Lonigan (Department of Psychology)Lonigan, Christopher; Farver, JoAnn - Development of reading and reading-related skills in preschoolers who are Spanish-speaking English-language learners
Background & Purpose. As a group, children who are English-language learners (ELL) are at considerable risk of educational difficulties, particularly in reading. In the United States, Spanish-speaking children constitute both the largest ELL subgroup and the fastest growing school-age population. Significant achievement gaps exist between Spanish-speaking ELL students and non-ELL students. This study was designed to examine how ELL preschool children's development in reading-related skills in both English and Spanish were related to their reading skills when they were in kindergarten and first grade. Method. Assessments of language, phonological awareness, and print knowledge were collected for 185 4-year-old ELL children, and measures of reading and phonological awareness were collected when these children were in kindergarten and first grade. All measures were completed in both English and Spanish. Results. Cross-sectional and longitudinal analyses revealed substantial predictive relations between preschool measures and later reading skills. In general, the degree of variance accounted for by the constructs was greater than the variance accounted by the language of assessment. Conclusions. These results provide evidence of the importance of key domains of Spanish-speaking ELL preschoolers' early literacy skills for later reading, as well as the relative importance of the language used to assess those skills.
In content-area learning such as social studies, students' comprehension is impaired if they do not have an understanding of the overall theme or "big idea" of the text. Various tasks have been used to assess comprehension of theme: How well do they do the job? Are some history themes easier to comprehend than others? Students were randomly assigned to one of three comprehension tasks (multiple-choice, sorting, and generating stories). All read the same twenty-four short theme-based stories, which represented eight themes typically found in elementary school social studies curriculum (rebellion, invention, migration, discovery of the unknown, taxation, trade, representative government, compromise). These themes were selected on the basis of teacher interviews and textbook analysis. Following the experimental task, all students were asked to define each theme. ANOVA indicated that students' performance on the multiple-choice task was higher than their performance on the sorting task and the generating stories task. There were substantial differences in the comprehension of stories containing the various themes, as well as the ability to define the themes. Students understand some themes, which will allow teachers to make connections between those themes and the history content they are encountering. If teachers are aware of which themes students struggle with, they can pre-teach these themes using familiar content.
Maureen Lovett (The Hospital for Sick Children, Learning Disabilities Research Program); Robin D. Morris; Maryanne Wolf; Rose A. Sevcik; Karen A. Steinbach; Jan C. Frijters - Multiple component remediation for young children with reading eisabilities: Can early intervention facilitate 'closing the gap'?
Purpose: The efficacy of multiple component reading remediation was compared for children with reading disability identified at the end of kindergarten, first, or second grade. Program efficacy was evaluated relative to a curricular control sample with RD. Children received intervention in first, second, or third grades. Method: Remedial intervention was delivered in small groups of four children. In this multi-site study, the 125 hour intervention was the Triple-Focus Program, combining phonologically- and strategy-based instruction (i.e., PHAST Reading, Lovett, Lacerenza, & Borden, 2000) and fluency/comprehension-based programming (i.e., RAVE-O; Wolf, Donnelly, & Miller, 2000). Experimental children were tested at pretest, 70 hours, 125 hours, and one-year follow-up, and control children at pretest and 125 hours. Results: Intervention children achieved superior outcomes to control children on measures of decoding, word identification, and reading comprehension at 125 hours and continued skill growth one year later. Most children receiving intervention in first grade demonstrated age-appropriate achievement in decoding (88%), word identification (92%), and passage comprehension (80%). More first grade children with RD were normalized by the end of intervention than children in second or third grade. Conclusion: Multiple component remediation is associated with positive reading outcomes for children with reading disability and is particularly effective when given early.
Purpose Several studies found that students with reading fluency deficit perform weakly in naming speed; these studies, however, did not examine students' working memory. We hypothesized that working memory may play as an important role as naming speed in reading fluency. Method The participants were students in the first to fifteenth grades, males and females who demonstrated a reading fluency difficulty in clinical assessments. We examined these students' abilities in three components of phonological processing (i.e., phonological Awareness, working memory, and naming speed) from the Comprehensive Test of Phonological Processing (CTOPP). Qualitative data were also used to support the quantitative data. SPSS was used to analyze the descriptive data. Results Our data showed that students with reading fluency were deficient in working memory, which was assessed with the memory for digits and nonword repetition subtests. However, their scores in rapid letter naming and rapid digit naming tasks fell in the average range. Conclusions Our results suggest that working memory may play a more important role in reading fluency than naming speed. Instruction designed to improve students' reading fluency may need to address working memory issues in addition to word recognition and retrieval speed. .
Purpose The present study investigated the development of two types of Chinese orthographic knowledge, i.e., component position and component structure, and their roles in concurrent and subsequent Chinese character reading. Method Forty-eight kindergarteners and forty-five first graders were recruited from Beijing, China. The participants were tested twice, in September 2006 and June 2007 respectively, on component position, component structure, character reading, nonverbal IQ, phonological awareness, and visual discrimination. Results Both kindergarten and first grade children performed better on component position than component structure. For kindergarteners, no difference was found between pre- and post-tests in either types of orthographic knowledge, whereas first graders significantly improved on both measures. Moreover, first graders' pretest component position uniquely predicted concurrent and subsequent character reading above IQ, phonological awareness and visual skills. For kindergarteners, only posttest component position predicted unique variance in concurrent reading. Conclusion Our results indicate that the two types of orthographic knowledge have different developmental trajectories and play different roles in character acquisition. Compared with component structure awareness, component position awareness develops earlier and is more important for character acquisition. Therefore, explicit instruction on component position may benefit Chinese literacy development in early years.
Stuart Luppescu (University of Chicago); Gina Biancarosa; Anthony Bryk; David Kerbow - Testing the fluency-comprehension relationship through equating: A Rasch equating of DIBELS and Terra Nova in grades K-3
Purpose. Reading fluency measures are used as indices of reading achievement under the assumption reading skill represents a unidimensional construct. This paper directly tests that assumption by examining how DIBELS fluency measures and Terra Nova comprehension measures scale together for Kindergarten through third grade students. Method. In the fall and spring of two academic years, almost 10,000 K-3 students took grade-appropriate DIBELS fluency subtests: initial sound, letter name, phoneme segmentation, nonsense word, and oral reading. Consistent with a common-persons approach to equating, in the fall, subsamples also took subtests for the prior or next spring form of the subtests. All grade 1-2 students took the Terra Nova around the spring DIBELS administration. Rasch modeling was used to equate item level data for both tests. Results. DIBELS fluency subtests scale together successfully into a single vertical scale of early literacy development. Although Terra Nova equating is ongoing, previous research demonstrates equating reading fluency and comprehension data is both possible and meaningful. Conclusions. Should the two tests scale together successfully, results would constitute a new line of evidence of the fluency-comprehension relationship and the unidimensionality of reading proficiency. Because curriculum-based fluency measures are used frequently in schools, results also have practical significance.
Purpose & Methods: Children with and without familial risk for dyslexia have been followed from birth to school age in the Jyväskylä Longitudinal study of Dyslexia (JLD). We have attempted to identify precursors of dyslexia as well as examine the predictive developmental cues of later reading acquisition among readers of the highly orthographically regular Finnish language. Results: Our findings suggest that atypical speech processing and categorical perception of speech sounds (especially phonemic quantity, a highly relevant sign of speech-related processing in Finnish) are the most likely precursors of dyslexia. Atypically slow retrieval of spoken responses corresponding to a visual stimulus may be a second, partly independent precursor. Conclusions: Some of our predictive observations may be applicable to practice, to identify children at risk years before school age. The predictor easiest to assess is delayed learning of letter names, which successfully identifies all those who may end up with problems 1-3 years before the start of formal reading instruction. Adding rapid naming and phonological measures increases the accuracy of avoiding false positives. Expressive speech is predictive from a very early (<2 y) age in differentiating children who later face reading problems from the larger pool of children defined to be at familial risk.
Purpose: In this talk, we will present the results of a study of an enhanced decoding curriculum for adults reading at the low-intermediate level. The curriculum teaches metalinguistic concepts about phonology and orthography, includes both spelling and decoding, applies decoding skills to multisyllabic words, and teaches cognitive and metacognitive strategies to support application of new skills in reading. Method: This curriculum was evaluated in an experimental study involving 68 adult reading classes in 23 adult literacy programs in 12 states across the country. Sixteen programs were randomly assigned to receive either the enriched curriculum or to continue their existing reading instruction. In addition, seven of the 23 programs were selected because they were already using published decoding curricula; these programs serve as an alternate treatment group. Results: We will present findings on multiple reading outcomes and information about the effects of demographic and background factors on outcomes. Conclusions: Adult literacy learners at the low-intermediate level can make progress in their decoding skills when provided with instruction. The relationships among outcomes and background factors are complex.
The aim of this study was to test whether the transposition effects found in some Indo-European languages that use Latin script, where transposition of consonants did not slow down word recognition (e.g. Andrews, 1996, Perea and Lupker, 2004), could be obtained in Arabic. Our hypothesis was that due to the central role of the consonantal root in Arabic such transposition would disrupt word recognition. Method: In two masked priming lexical decision experiments, we tested the effects of transposition of root consonants in triliteral Arabic words on word recognition compared to two controls where either one or two of the transposed letters in the experimental condition were substituted. We examined the transposition of the last two root consonants in Experiment 1 and the first two roots consonants in Experiment 2. Forty eight different Arabic-speaking university students took part in each experiment. The results showed that while transposing the last two root consonants primed targets, the transposition of the first two consonants was not different from controls, despite the fact that all words started with a prefix that is at least one letter long. Conclusions: These findings are different from what has been found in Indo-European languages and cannot be easily accommodated within existing models of word processing. References Andrews, S. (1996). Lexical retrieval and selection processes: Effects of transposed-letter confusability. Journal of Memory and Language, 35, 775-800. Perea, M., & Lupker, S. J. (2004). Can CANISO activate CASINO? Transposed letter-similarity effects in nonadjacent letter positions. Journal of Memory and Language, 51, 231-246.
Purpose This paper describes the development and validation of a Cognate Awareness Test(CAT)which measures cognate awareness in Spanish-speaking English Language Learners (ELLs) in fourth and fifth grade. The CAT was developed as part of a larger study (August, Carlo, Barr & Calderón) whose purpose was to investigate the development of second language literacy in young English-language learners. Methods Participants in the study were 173 Spanish-speaking English-language learners who were participants in the larger transfer study. Rasch analyses were used to determine the internal validity of the assessment and correlational analyses were used to determine external validity. Results An investigation of differential performance on the two subtests of the CAT (cognates and noncognates) provides evidence that the instrument is sensitive to English-Spanish cognate awareness among elementary school- age Spanish-speaking English-language learners. The correlation between CAT cognate vocabulary and the Spanish WLPB-R/PV was moderate (r=.38, N=132, p<.01). Again, there was no correlation between CAT cognate vocabulary knowledge and English WLPB-R/PV (r=-.01, N=132, p>.01). However, English vocabulary knowledge as measured by the English WLPB-R/PV was moderately related to CAT non-cognate vocabulary (r=.37, N=132, p<.01), while Spanish vocabulary as measured by the Spanish WLPB-R/PV was not (r=-.05, N=132, p>.01). Conclusion Findings help confirm that the Cognate Awareness Test as a valid measure of cognate awareness for Spanish-speaking students in the sample.
Jeannette Mancilla-Martinez (University of California, Irvine) - From Receptive to Productive Word Knowledge: Exploring the Relationship between Vocabulary Instruction and Writing amongst Fifth-Grade Language Minority Learners
Purpose: The effect of a 20-week pilot vocabulary intervention on Language Minority (LM) students' writing was examined. Method: Twenty-four fifth-grade LM learners from one predominantly Latino, low-income school in the Northeastern U.S. participated. Students wrote persuasive essays at the end of each vocabulary intervention week and all essays were transcribed. Frequency of occurrence of the taught vocabulary was compiled using the Computerized Language Analysis (MacWhinney, 2000) program. To determine whether students' overall writing quality improved across the 20-week intervention, scores from a quality coding scheme were analyzed with individual growth modeling using the multi-level model for change (Singer & Willett, 2003). Results: Students incorporated the current week's, and previously taught, target vocabulary in their essays each week. Further, improvements in students' overall writing quality were found, with the bulk of writing growth occurring during the last 10 weeks of the intervention. Conclusions: This study extends our understanding of the relationship between vocabulary instruction and writing amongst LM learners. Results suggest that students need many opportunities to productively use newly taught words in both oral and written language and that increased word knowledge may help students exert greater control over the precision of their writing, thus improving their overall writing quality.
Purpose: Can researchers utilize monolingual research-based criteria to identify reading disabled children among Spanish-speaking English language learners (ELLs)? Method: We identified 45 fifth graders at risk for reading disability (RD) from a larger, longitudinal sample, based on low standardized word/nonword reading skills, and equated them to 90 non-RD children, matched on spatial ability (average range), receptive oral language in English (below average, but equated across groups), gender and age. A battery of English and Spanish reading and cognitive tests were given in grades K-8. Results: The RD group was lower in grade 5 on all predicted measures in Spanish and English, particularly in phonological awareness, rapid digit naming, and reading comprehension. Predictive analyses indicated print knowledge, letter name knowledge and rapid object and digit naming predicted group status (RD, non-RD) from kindergarten (Spanish) and 1st grade (English). Follow-up analyses revealed the RD and non-RD groups were moderately stable into 7th grade, and reading fluency was also affected. Conclusions: Once allowance is made for lower overall language achievement, RD children can be identified in an ELL sample. RD children had a fundamental problem with components of reading acquisition in both languages that resembled deficits in the monolingual literature. Early identification of children at risk for reading disability seems feasible in an ELL population.
Sandra Martin-Chang (Concordia University);Levesque, Kyle; Kim, Michelle - Remembering the story, forgetting the words: evidence for different types of cognitive processing during contextual and isolated-word reading
Purpose: These investigations examined the cognitive processes involved when words are presented in increasingly meaningful contexts. Method: Sixty-two undergraduates (n=31) read three sets of words; one set was read in isolation, a second set was read in context and a third set was generated from definitions. A surprise recall task concluded each experiment. Implicit memory (memory without awareness) was measured in Experiment 1 ("Complete the word stems with the first word that comes to mind"). Explicit memory was measured in Experiment 2 ("Recall as many of the words as possible"). Results: Implicit recall was greatest in the isolated-word condition but lowest in the generated and context conditions. Explicit recall was highest in the generated and context conditions and lowest for the isolated-word condition. Conclusion: The seminal work of Jacoby (1983) has shown that a) implicit memory is greatest for words encoded perceptually and b) explicit memory is greatest when words are encoded conceptually. Together, the results from Experiments 1 and 2 suggest that reading words in a meaningful story elicits more conceptual processing than reading words in a list. Moreover, our findings suggest that reading words in context might be eliciting as much conceptual processing as generating words from definitions.
Rhonda Martinussen (Ontario Institute for Studies in Education) ;Rosemary Tannock; Peter Chaban - The relationships among inattention symptoms, oral reading fluency, and teachers' perceptions of children's comprehension skills
Purpose: The objective of this study was to examine the relationships among inattention symptoms, oral reading fluency, and teachers' perceptions of children's comprehension skills. Method: The participants were a community-based sample of 75 students (42 boys, 33 girls) who completed baseline testing for an intervention study. A stratified sample of participants (good attention or moderate to high levels of attention problems and academic difficulties) were recruited from participating classrooms. Results: Compared to children without attention problems, children with attention problems exhibited weaknesses in word and text reading fluency and they were also more likely to be rated by teachers as having decoding and comprehension problems. A subgroup analysis indicated that students with attention problems without single word decoding difficulties received lower teacher ratings of comprehension skills (e.g., drawing conclusions) and were less fluent readers than students who did not exhibit attention or decoding problems. Hyperactivity symptoms were not significantly associated with word or text reading fluency independent of inattention symptoms. Conclusion: The findings suggest that children with attention problems may be at particular risk for poor reading fluency and comprehension weaknesses and research is needed that examines the relations between reading fluency and text comprehension skills in this population.
The effects of self-regulated strategy development (SRSD) instruction for expository reading comprehension plus writing instruction on language and vocabulary development was tested in a secondary data analysis. Eighty-seven students with and without disabilities participated in the original experiment (Mason, 2007). Students were randomly assigned to SRSD for reading comprehension, SRSD for reading comprehension + writing, or control. Student essays (collected prior to instruction, immediately following instruction, 2 months following instruction, and in the following year) were coded into the Systematic Analysis of Language Transcripts program (Miller & Chapman, 2002). Syntactic variables included the mean length of utterance in words and morphemes, and a subordination index. The semantic variables included number of different words, number of total words, and type-token ratio. Results indicated statistically significant effects for the semantic measures at post-test (ES = .44 to .56) and for 2-month maintenance (ES = .58 to .66). No change was noted in the year following instruction. Post hoc testing indicated that for semantic variables both treatment groups outperformed the control group. The syntactic measures were not observed to be statistically significant by treatment group or testing time. Implications of these findings for reading comprehension and writing intervention research will be discussed.
What are the early markers of reading difficulties in Chinese children? From our longitudinal study of Hong Kong Chinese children's reading development in approximately 200 children, we selected 50 children who were in the bottom 25% of readers in our sample at age 6 (X of Chinese character recognition=6.99) and 50 controls (X of Chinese character recognition=42.14), matched for nonverbal intelligence (Raven's), parental education and income, and age. These groups were compared on a variety of reading-related skills at ages 4, 5, and 6. The groups differed significantly across all three years on Chinese character recognition, phonological awareness, and morphological construction (p<.05). In addition, they differed consistently in homophone awareness (tested at ages 4 and 6 only), tone detection (ages 5 and 6 only), and visual spatial knowledge (tested at ages 4 and 5 only) across times these tasks were tested (all ps<.05). Findings underscore the importance of phonological awareness, morphological awareness, and visual skill in distinguishing early reading skills in young, developing Hong Kong Chinese children. (For the conference, one additional follow-up year of data (i.e., skills at age 7) will be included in the poster as well.)
Purpose: Although reading to children is often touted as a key step toward developing print awareness, available evidence suggests that preschoolers do not look at text in that context (Evans & St. Aubin,2005; Justice et al., 2005). We were interested in whether the introduction of a novel word might affect children's scanning of picture books. Methods: Ten preliterate children aged 4-6 years were read a picture book modified to include a novel word. Eye movements were monitored using an ASL Mobil Eye setup. Number and duration of fixations, as well as latency to first fixation on the target referent were all analyzed. Results: Preliminary results indicate that 1) the previous finding of low levels of fixation on words in picture books is replicated, but 2) hearing the novel word read aloud does trigger a fixation on the target referent in children who learn the word. Conclusions: These results suggest a possible resolution to the paradox that reading to children promotes literacy but not via increased attention to the written text. The fact that pre-reading children look longer at text in the context of an unfamiliar word suggests a nascent understanding that print does contain useful information, something that literacy can capitalize on.
In this presentation, we will report on a study designed to determine the effectiveness of Kindergarten Peer-Assisted Learning Strategies (K-PALS) for English Learners (ELs). K-PALS is an evidence-based classwide peer tutoring program designed to increase the beginning reading skills of young students. K-PALS is supported by a strong research base, but little is known about its effectiveness for students from diverse cultural/linguistic backgrounds. Data for this study were disaggregated from a large-scale study that was designed to investigate implementing K-PALS at scale. Twenty K-PALS ELs were compared to 20 Control ELs and to 20 K-PALS non-ELs on early reading skill acquisition, using a pretest/posttest control group design with matched samples. We also compared proportions of ELs unresponsive to K-PALS to ELs unresponsive to traditional instruction. Teachers implemented K-PALS 4 times per week for 18 weeks. Following intervention, analyses of covariance on posttest measures indicated that K-PALS ELs performed reliably higher than Control ELs on measures of phonemic awareness and letter sound recognition, and performed similarly to K-PALS non-ELs. Further, higher proportions of K-PALS ELs were responsive to instruction than control ELs. Implications for implementing evidence-based practices with diverse groups of at-risk readers are discussed.
Purpose The aim of this study was to examine the theoretical and empirical relationships among word reading, text fluency, and reading comprehension. Some theories support the notion that being a fluent reader provides benefits for reading comprehension beyond those accounted for by efficient word reading skills (Fuchs et al., 2001; Schreiber, 1980; Wolf & Kadzir-Cohen, 2001), while others do not (LaBerge & Samuels, 1974; Perfetti, 1985; Schwanenflugel et al., 2006; Torgesen, 2001). Three theoretical models depicting how word reading accuracy, word fluency, and text fluency operate together to produce good comprehension were evaluated. Method Third- and fifth-grade students (N=190) completed a series of standardized measures of word reading accuracy, word fluency, text fluency, and reading comprehension. Results Results from the path analysis supported a text fluency model which states that word reading accuracy, word fluency, and text fluency each make unique contributions to comprehension in 3rdgrade children. However, the influence of these basic reading skills declined in the 5th grade children, suggesting that other factors may be needed to explain comprehension in older students. Conclusions Together these results suggest that text fluency is an important reading skill for elementary school children and relationships among reading skills may vary across development.
Books as Lexical Reservoirs for Preschoolers Purpose: Although book reading has proven effects on preschoolers' vocabulary, researchers rarely scrutinize the lexical inventories of books. We examined the size of books' lexical reservoirs (potentially rich book words), the relationship between book quality and reservoirs, and comparisons between overall book language, lexical reservoirs, child speech, and adult speech. Methods: We analyzed the language of 108 books (n = 103,950 words) using the Parental Frequency Corpus (CHILDES); a sample of child expressive language; Dale/Chall list; Biemiller's LWV ratings; and word tiers. We examined the vocabulary diversity (CLAN) and quality of books, relating estimates of quality to lexical inventories. Findings: A small group of materials offered larger than average lexical reservoirs (n=5) and higher estimates of quality. Books listed in our quality source had larger lexical reservoirs than other books. Predictable and concept books had less diverse and smaller reservoirs. One-half of the overall language of books provided potentially enriching language before intersecting with the expressive language of preschoolers. Conclusions: Lexical reservoirs differ by book genre and book quality. Researchers and teachers should pay attention to lexical reservoirs when choosing books. I am not a voting member and can provide additional information about the study. This abstract is an abbreviated description of a 75-page paper.
Purpose: We examined how ADHD impacts the text representation constructed during reading comprehension by evaluating recall of central versus peripheral information relative to controls matched for age and for decoding skill. Method: 20 children with ADHD and 20 controls read aloud a passage from the Qualitative Reading Inventory and then retold it. Results: Compared to controls, children with ADHD showed a specific deficit in the recall of central information, but no deficit in recall of peripheral ideas - a Centrality Deficit. Since ADHD is often comorbid with decoding problems, we matched a subset of each group on decoding and found the same interaction, indicating that the centrality deficit is due not to decoding problems, but to attentional problems that impact their ability to coherently connect the ideas within a text. Conclusion: The attentional deficits of children with ADHD appear to limit the resources required to construct a coherent text representation with the result being a representation in which central ideas are not as connected, and thus not as well recalled, as in controls. Furthermore, this centrality deficit cannot be explained by comorbid decoding deficits. ADHD thus appears to involve comorbid comprehension deficits.
Purpose Exposure to books is assumed to be a major source for the two pillars of learning to read: vocabulary and alphabetic knowledge. In the current meta-analysis, it is questioned whether storybook reading in classroom settings is effective as a result of tutorial behavior from teachers or simply because of children's encounters with books and whether interactivity affects both precursors of literacy. Method It is meta-analysed whether experimental studies implementing interactive book reading in classrooms reveal additional effects on the vocabulary and alphabetic skills of two- to six-year-old children. All included studies compare interactive book reading with reading books as usual. Results We quantitatively analyzed 31 (quasi-)experiments or 2049 children (NInteractive Reading= 1032; NControl= 1017), of whom 1030 attended day-care or preschool programs and 1019 were in kindergarten. The influence of moderators like group size, repeated reading, and program type (interactive reading with versus without book-related activities) is discussed. Conclusion The interactive character of book reading sessions promotes vocabulary growth in a younger as well as an older age group; it also affects alphabetic knowledge but only when children grow older.
Purpose To gain a better understanding of the lexical knowledge and word-finding strategies of language minority children, we examined the performance of 1086 grade 2 to 6 pupils (723 L1 and 363 L2) on a synonyms task. Method The participants were given a lexical prompt, the first two letters of the target synonym and the appropriate number of blanks to complete it. Results were analyzed by dividing each grade level into a top, middle and bottom third and by comparing the performance of L1 and their L2 peers by third, grade, socioeconomic status and item. Results As expected, the L2 children did not fare as well overall as native speakers. However, on items in which the two-letter cue split the graphemic representation of a phoneme, the L2 children proved superior. The impact of socioeconomic differences was minimal. In both linguistic groups, children in the top third in grades 2 and 3 were often more successful in finding synonyms than those in the bottom third in grades 5 and 6. Conclusions Results suggest that L1 and L2 children used quite different strategies to complete the synonyms task and that each group might benefit from instruction adapted to these differences.
Darrell Morris (Appalachian State University)Bloodgood, Janet; Frye, Beth; Kucan, Linda; Perney, Jan; Trathen, Woodrow; Mock, Devery; Schlagal, Robert - Reading Rate and Concomitant Abilities in Second- to Sixth-Grade Students
Reading Rate and Concomitant Abilities in Second- to Sixth-Grade Readers Purpose: (a) to investigate normative reading rates (oral and silent) at grade levels 2-6, and (b) to determine which abilities (e.g., automatic word recognition, oral reading accuracy, comprehension, spelling) best predict reading rate. Method: 220 students, randomly sampled from 8 schools in a rural county, were followed longitudinally from grade 2 to 6. Each year the children were individually assessed in the following 8 areas: automatic and untimed word recognition; oral reading accuracy, rate, and comprehension; silent reading rate and comprehension; and spelling. Results: Results showed that for students reading grade-level material (i.e., a third grader reading a third-grade passage), oral reading rate advanced approximately 10 wpm- silent reading rate, 20 wpm-between second and third grade and third and fourth grade. Surprisingly, there was a leveling-off of reading rate between fourth and sixth grade. Conclusions: Besides the normative data, several findings are of interest: (1) There was a strong relationship between and oral and silent reading rate at each grade level, suggesting a common underlying cognitive process; (2) An isolated measure of word recognition (flash word recognition) was a much better predictor of oral and silent reading rate than was a contextual measure of word recognition (oral reading accuracy); and (3) The leveling-off of reading rate (oral and silent) as material became more difficult in fifth and sixth grade is a finding that has not been previously reported in the literature.
Robin Morris (Georgia State University, Department of Psychology); Maureen W. Lovett; Maryanne Wolf; Rose A. Sevick; Karen A. Steinbach; Jan C. Frijters - Multiple component remediation of developmental reading disabilities: One year follow-up from a controlled factorial evaluation
Purpose: The effectiveness of two different multiple-component, deficit-directed intervention programs for RD children (PHAB/DI + RAVE-O and PHAB/DI + WIST/PHAST) were evaluated in a short-term study against alternative treatment control programs. This study evaluated their continuing impact one year following intervention. Method: 279 young disabled readers were randomly assigned to a program. Interventions were taught by trained research teachers for 70 hours on a 1:4 ratio at 3 different sites. Children were evaluated at 0, 35, and 70 hours of instruction and one year later. Individual growth curve methodology was used to analyze changes in each outcome measure. Results: Initial intervention results showed significant improvement for the experimental programs compared to the control programs on basic reading skills. These improvements were maintained at one-year follow-up. Children in the alternative treatment math control program year were randomly assigned to one of the multiple-component interventions during the follow-up year and showed significant gains in basic reading skills providing a secondary validation of the original findings. Conclusions: Reading gains from short-term multiple component interventions are maintained one year following the end of intervention with some evidence that continued reading growth momentum occurs in some areas.
Frederick Morrison (University of Michigan); Michelle Housey; Megan Hoffmann; Annemarie Hindman; Lori Skibbe - The Impact of Schooling on Children's Early Academic Skills from Pre-Kindergarten through First Grade
The school cutoff technique was used to examine how differences associated with pre-kindergarten and kindergarten instruction influenced children's development of literacy skills. Children (n = 60) whose birthdates fell within two months of the arbitrary cut-off date for kindergarten were administered four subtests of the Woodcock-Johnson III Tests of Cognitive Abilities and were asked to name all letters of the alphabet. Using a hierarchical linear model, schooling effects were found for children's development of phonological awareness, vocabulary, decoding skills, and alphabet knowledge, such that children in kindergarten scored higher on all measures than those in pre-kindergarten. In addition, the school cutoff technique was used to compare literacy growth for kindergartners and first graders (n = 79). However, in addition to investigating how naturally-occurring instruction relates to differences in literacy growth, the amount of instruction (observed via classroom observations) was also considered. Findings suggest that exposure to more advanced instruction benefits young children's literacy development, although this advantage may be explained by the differing amounts of language arts and higher order (e.g., writing instruction and listening comprehension) instruction that children receive across grades. Results highlight the impact of early schooling processes and discuss mechanisms that may explain patterns of differences.
Jack Mostow (Project LISTEN, Carnegie Mellon University); Beck, Joseph; Zhang, Xiaonan; Leszczenski, James - Does fluency growth transfer among related words? Longitudinal evidence from Project LISTEN's Reading Tutor
What is the atomic unit of skill learning in oral reading fluency growth? Prior research [1, 2] suggested that fluency is word-specific, i.e., that oral reading practice of one text improves fluency on another text only if it shares many of the same words. Does oral reading practice on connected text speed up the time to read not just the same words in new contexts, but orthographically similar words as well? To find out, two recent studies analyzed 6,213,289 words read by 650 children in grades 1-5 who used Project LISTEN's Reading Tutor during the 2003-2004 school year. Learning factors analysis  fit the data better by modeling words with the same root as the same fluency skill than as independent skills . Learning decomposition  of exponential growth curves estimated the impact of practice on later word reading time. Reading the word itself helped most. A word with the same root but different ending helped significantly, about half as much . So did a word with the same rime, but only about an eighth as much. Both analyses indicate transfer of fluency growth between similar words, especially for better readers. 1. Rashotte, C.A. and J.K. Torgesen. Repeated reading and reading fluency in learning disabled children. Reading Research Quarterly, 1985. 20(2): p. 180-188. 2. Martin-Chang, S.L. and B.A. Levy. Fluency transfer: differential gains in reading speed and accuracy following isolated word and context training. Reading & Writing: An Interdisciplinary Journal, 2005. 18: p. 343-376. 3. Cen, H., K. Koedinger, and B. Junker. Learning Factors Analysis - A General Method for Cognitive Model Evaluation and Improvement. Proceedings of the 8th International Conference on Intelligent Tutoring Systems, 164-175. 2006. Jhongli, Taiwan. 4. Leszczenski, J.M. and J.E. Beck. What's in a word? Extending learning factors analysis to modeling reading transfer. Proceedings of the AIED2007 Workshop on Educational Data Mining, 31-39. 2007. Marina del Rey, CA. 5. Beck, J.E. Using learning decomposition to analyze student fluency development. ITS2006 Educational Data Mining Workshop 2006. Jhongli, Taiwan. 6. Zhang, X., J. Mostow, and J.E. Beck. All in the (word) family: Using learning decomposition to estimate transfer between skills in a Reading Tutor that listens. AIED2007 Educational Data Mining Workshop 2007. Marina del Rey, CA.
Although Faroese orthography is relatively young (from 1846), the orthography is etymologically founded. The orthography is rather inconsistent in the writing direction. However in the reading direction it is more consistent when a number of rules are observed. Purpose: This study aims at showing, when Faroese pupils master certain orthographic patterns. Both etymological and morphological principles have impact on the orthography. Method: 191 pupils from different class levels participated. The wrote word dictation and free text. On the basis of quantity and quality analysis of correctly written words as well as spelling errors a description of how Faroese school children acquire writting/spelling competence is suggested. Results: The results of the study is presented i tables showing that spelling ability is developing throughout all 1.-9. class. Conclusions: The results are useful for developing didactics for teching Faroese spelling. One conclusion is also that the need for educational research in Faroese - both regarding reading and writing - is great and uncovered.
This study examines the vocabulary demands of middle school textbooks in both math and social studies. Textbooks widely used in regular education middle school classrooms were the subject of the analysis. Electronic versions of textbooks and vocabulary analysis software were used in order to develop comprehensive word lists for three textbooks in each of the two disciplines, which allowed for subsequent analysis as to how many of the words that recur in each text are potentially new and unknown words for many students. This study revealed that an average of 25% of the text consisted of words likely to challenge struggling readers. Given the conventional belief that 90 - 95% of text must be known in order for comprehension to occur, these finding suggests that limited vocabulary knowledge is likely to negatively affect academic performance in content area classrooms where independent reading is expected.
Purpose. This experiment investigated the item-level specificity of the relationship between semantic factors and word reading. It also asked whether the relationship between meaning and reading is moderated by orthographic consistency. Method. 7-year-old children read aloud words varying in orthographic consistency. We also measured their semantic knowledge of the same items. Analyses focused on item-level relationships between knowing a word and reading success. Results. There was a strong relationship between meaning and reading, moderated by orthographic consistency. Children's ability to read words containing typical sound-spelling correspondences was not associated with semantic knowledge for the same items. By contrast, semantic knowledge showed a close association with exception word reading. Consistent with this, known exception words were read more accurately than unknown exception words and the item-level relationship between semantic knowledge and reading success was closer for exception words than for consistent words. Conclusions. These findings demonstrate a semantic contribution to reading aloud in children in the relatively early stages of learning to read. They are consistent with an item-level relationship between semantics and reading aloud, as predicted by the triangle model.
PURPOSE: The purpose of the study was to investigate the relation of, decoding, phonological memory (capacity and processing complexity) and reading comprehension, in a transparent orthography. METHOD: Twenty-five children with no visual, auditory, neurological deficits, or reported learning difficulties participated in the study. All subjects performed text comprehension, naming and memory tasks. Naming tasks involved visually presented stimuli that differed in type (pictures, words and pseudowords) and extension (two or three syllables). Working memory was evaluated in three different conditions: a picture, word and pseudoword span task. RESULTS: Children that showed better decoding skills performed better on text comprehension (r= 0,697; p=0,00). Comprehension performance showed a moderate correlation with memory storage capacity for words (r=0,438; p=0,029) and pseudowords (r=0,426; p=0,034), but not for pictures. There was an effect of stimuli type (ANOVA, F=28,01; p=0,001) on the number of correct responses on the memory span task. Regarding the span level, there was also an effect of stimuli type; pictures > words (t=2,83; p=0,009), words > pseudowords (t=9,14; p=0,00). CONCLUSIONS: Stimuli type and extension influence performance on a working memory task. Decoding skills showed a greater influence on text comprehension than working memory skills for second graders, beginning readers of Brazilian Portuguese.
Purpose: Frequent words are presumably easy to read because we have high quality representations of words we encounter often. However, while word frequency may provide an overall estimate of word knowledge, word knowledge can vary along a number of dimensions, including orthographic, phonological, and semantic components (Perfetti & Hart, 2001). Two studies specifically examine how subcomponents of word knowledge affect online reading behavior, and more directly establish the relationship between knowledge and normal reading behavior. Method: Forty adults were trained on a set of rare words in 6 training conditions (combinations of orthography, phonology, and meaning training) with 1, 3, or 5 exposures. Eye movements were recorded while they read sentences containing those words. An additional 40 adults read passages containing undisclosed target words, and were later tested on their target word knowledge. Results: Results extend previous findings that orthographic knowledge, but not meaning knowledge, affects early eye movement measures, but that meaning knowledge becomes important in later measures. Conclusions: The findings suggest that sub-components of word knowledge are functionally distinct in terms of their effects on reading behavior, and that this should be taken into account in models using a single estimate of word knowledge, such as frequency, to make predictions.
Purpose Mostly, it's assumed that both foveas fixate conjointly during reading. fMRI evidence from Toosy et al. (2001) shows that monocular stimulation produces greater activation in contralateral than in ipsilateral cortex. Given a vertically divided split fovea (Shillcock et al., 2000), outer visual field (corresponding to inner hemifoveas) half-word presentations better correlate with correct word identification than inner visual field (corresponding to outer hemifoveas) presentations (Obregón & Shillcock, 2007, submitted). Here we explore the correct identification of overlapping and non-conjoint five letter words presented stereoscopically, predicting that text mostly presented to inner hemifoveas is preferred. Method We presented four letters from five-letter words to each eye, in one of three overlapping conditions (conjoint, 2-2 letter overlap, and 3-1 letter overlap), counterbalancing strings across inner and outer hemifoveas. Thirty bi-ocularly balanced native English speaking university students took part. Results Linear mixed effects analyses (with participant random effects) showed better word naming (all p's<0.0001) for inner hemifoveal presentations, for both conjoint and 2-2 letter overlapped cases, but no effects in asymmetric 3-1 overlapped text. Conclusions Given balanced overlaps, inner hemifoveas are more important for reading, while asymmetric overlaps underscore visual distinctiveness.
The purpose of this study was to describe the development of 92 second- and 101 fourth-grade students in narrative writing quality, writing conventions (correct word sequences), and vocabulary diversity throughout a school year. In addition, student characteristics (e.g., gender, reading, spelling, handwriting, working memory) were entered into the model to determine unique predictors of spring writing achievement or rate of change. Three waves of writing assessments were administered throughout the school. Hierarchical linear modeling was used to model growth. Results indicated that second- and fourth-grade students demonstrated growth in writing quality and vocabulary diversity from the fall to the spring. Second-grade students demonstrated growth in writing conventions, while fourth-grade students remained flat. Significant developmental differences were found for all writing outcome measures, in which fourth-grade students received higher scores on all writing outcome measures. The results indicated the importance of reading comprehension in writing quality, writing conventions, and vocabulary diversity. Additionally, low-level writing skills (e.g., handwriting, spelling) were unique predictors of writing outcome measures. The results add to a limited body of research describing predictors of elementary-level writing development using a multivariate approach.
Purpose: The purpose of this research was to evaluate the accuracy of the Simple View of Reading in accounting for reading comprehension in different age groups, and in the process consider if this influential theory fully acknowledges the role of oral vocabulary. Method: Children completing grade 1 and children completing grade 6 were assessed on a battery of measures encompassing phonological processing, listening comprehension, breadth of oral vocabulary, depth of semantic knowledge, decoding, visual-word recognition, and reading comprehension. Results: The primary analysis of interest here explored the role of oral vocabulary and semantics in predicting reading comprehension, beyond the components of the Simple View of Reading (i.e., decoding and listening comprehension). This was accomplished through hierarchical multiple regressions for each age group under study. Conclusions: The results of this study revealed that when oral vocabulary is thoroughly assessed (i.e., both in breadth and depth), its role in reading comprehension beyond the components of the Simple View of Reading, becomes evident. Results are discussed in terms of the changing roles of oral and written language skills in explaining reading comprehension across development, and interpreted with reference to Perfetti and Hart's not so simple Lexical Quality Hypothesis.
Purpose: To examine whether alphasyllabary orthographies are stored and processed alphabetically or syllabically. Method: We will present two studies. The first study probes segmental awareness by asking participants to delete the first sound of a word in Tamil that appears in the second positin in writing. In the second study we recorded RTs for processing nonwords with a nonlinear vowel marker appearing in initial, medial and final position and compared the same with processing of stimuli with linear elements. In both the studies the participants were fluent readers of Tamil. Results: The results of the first study were analyzed in terms of the frequency of types of errors including the deletion of the nonlinear vowel marker and the edit hypothesis, which would explain the errors in terms of the analogous editing operations involved in phoneme deletion. The results were largey in support of the hypothesis. There were significant number of errors that suggested that the Tamil alphasyllabry may have a segmental representation in the minds of native speakers. The Reaction time study more clearly suggested that the readers processed Tamil nonword elements alphabetically with RT being signifiantly higher for nonwords with nonlinear elements. Conclusion: Indian alphasyllabic scripts seem to have a segmental representation in speaker's mind and seem to be processed segmentally at least for words with nonlinear vowel diacritics.
Purpose This study investigated the role of naming speed and errors in reading fluency and comprehension skills of young children who had reading difficulties. Method Three groups of 191 young children participated in the study: (1) 66 English-speaking native students, (2) 52 Latinos who had English as an initial reading instruction (English R1), and (3) 72 Latinos who received Spanish as R1 (Spanish R1). The measures were the digit and letter naming subtests of the CTOPP and the fluency and comprehension of the GORT. Results The three groups performed differently in digit naming speed and letter errors (p < .05). When the naming latencies and errors were entered into a regression equation predicting fluency and comprehension, letter naming errors accounted for the most variance in all three groups. The magnitude of variance accounted for was the greatest for the Spanish R1 students, followed by the English R1 Latinos, and native students (26%, 19%, and 8%, respectively). Conclusions This study indicates that errors in the naming tasks predict more variances in reading fluency in second-language learners than native counterparts. The results suggest that accuracy of naming tasks and language of instruction in ESL students should be taken into account in reading.
Despina Paizi (Institute for Cognitive Sciences and Technologies (ISTC-CNR)&Dept. of Psychology, University of Rome);Zoccolotti, Pierluigi; De Luca, Maria; Burani, Cristina - List Context Effects in Italian Dyslexic and Typically Developing Readers
Purpose Italian is a transparent orthography with consistent grapheme-to-phoneme mappings. Contrary to proficient readers, Italian developmental dyslexics seemed to prevalently employ nonlexical reading, because of their slowness and sensibility to stimulus length. We aimed to challenge this view by manipulating list context. When words are presented mixed with nonwords, the lexical route could be de-emphasised in favour of the nonlexical that can read both types of stimuli, but is not sensitive to lexical effects; possibly resulting in diminished frequency and lexicality effects, especially for the dyslexics. Method Two experiments on eighteen dyslexic and thirty-six typically developing readers tested lexicality, word frequency and length in reading aloud words and nonwords in two different list contexts (pure/mixed blocks). Results Lexicality, word frequency and length effects were present for both groups, irrespective of list context. The dyslexics were generally slower, less accurate and more sensitive than the controls to stimulus length, confirming previous evidence. They performed better in pure - than mixed - blocks, similarly to skilled readers, indicating lexical reading and flexibility in changing reading routes according to list context. Conclusion The presence of lexicality and frequency effects in the context of nonwords, provides strong evidence against nonlexical reading for Italian developmental dyslexics.
Susan Parault (St. Cloud State University; Counselor Education and Educational Psychology)Rihana S. Williams - A closer look at word learning in context: An examination of number and types of context clues and word learning
Research suggests that a great deal of word learning results from incidental exposure to unknown words in written context. Many believe that context clues lead to this word learning. However, we currently have little knowledge about how context clues function in text to increase the probability of word learning. We hypothesize that the number of clues in text as well as the different types of clues will influence word learning. College students read 12 text passages (6 narrative and 6 expository) containing several pseudowords whose meanings were inferable from as few as one to as many as three different context clues. Participants read all of the passages and then chose the correct meaning for each pseudoword on a multiple-choice test. Initial regression analysis shows number of clues present in text to be a significant predictor of definitional accuracy in both narrative and expository text. In addition, with number of clues removed synonym, cause and effect, prior knowledge and compare/contrast type clues were significant predictors of definitional accuracy in expository text. Prior knowledge, words in a series, and definition clues were predictors of word learning in narrative text. Future analysis will examine participants written definitions for the pseudowords.
Jungjun Park (Baylor University); Lombardino, Linda; Kelley, Ronald; Riess, Melissa - Phonological Processing, Spoken Language Skills, and Reading Development of Children with Sensorineural Hearing Loss
*Purpose: Previous studies reported that hearing impaired (HI) children perform comparably to their age matched normal hearing peers (NH) on various measures of reading skills even with impaired phonological ability. This finding raises a critical question regarding the precise role of phonological skills in literacy. In this study, we asked, How are the phonological processing and reading skills of HI subjects different from those of normally hearing and dyslexic children? We hypothesized that HI children's phonological ability would show phonological skill and reading skill deficits when compared to their normal hearing peers. Method: Participants ranging in age from 7-12 with normal hearing/reading (NH: n=29), dyslexia (D: n=30), and sensorineural hearing loss (HL: n=19, better PTA=48 dBHL) were tested. Numerous tests of reading and reading-related skills, including the CTOPP, GORT-4, TOWRE, and WJ-ACH-III word identification and word attack tests, were given individually to each participant. MACOVAs were used (covariates=non-verbal IQ, grade, age) to analyze the data. Results: HI children's phonological scores were significantly lower than their NH peers in the half of tests given. Interestingly, HI children's reading scores were significantly lower than their NH peers on all reading tests with the exception of one timed test of decoding. In pairwise post-hoc analyses for the HI, NH, and D groups: (1) between the HI and the NH groups, elision, blending, and non-word repetition were significantly different. No significant differences were found for 'rapid naming (digit/letter)' and short term memory (digit span) (2) The HI subjects scored significantly higher than the dyslexic subjects on the RAN. The RAN was noted as a good clinical marker discriminating dyslexic readers from children with hearing impairment. Conclusion: Evidence supports that phonological processing skill is a necessary condition for reading development in children with and without hearing impairment.
The purpose of this study was to explore the Visual Attention Span Deficit Hypothesis in a large sample of persons who have developmental dyslexia. We predicted that serial visual attention (SVA) would be related to both rapid naming and word reading tasks that require serial attention and scanning. Sixty-five persons with dyslexia, ranging in age from 6-21 years, were given a standardized battery of cognitive tests that included: (1) visual matching (test of attention) (WJ-COG-III), (2) word reading, word decoding (WJ-ACH-III), and (3) phonological awareness, phonological memory, rapid naming (CTOPP). Partial correlations between the SVA and CTOPP tests revealed a moderately significant correlation for the rapid naming tests only (r=.613, p<.0001). Partial correlations between SVA and word reading tests were significant (word identification, r=.53, p<.0000; word attack, r=.412, p=002). Stepwise regressions showed that SVA explained a substantial and significant variance of untimed (29.8%) and timed word reading (21.8%), respectively. The blending and digit memory subtests of the CTOPP were also significant predictors, but they only accounted for 10% of unique variance of both untimed and timed word reading. These findings support the visual attention span deficit as an explanation for underlying processing deficits that affect reading performance in dyslexia.
Purpose - Adolescents learning English as their second language (L2) have significant difficultly learning the skills involved in oral language and reading comprehension. Virtually no research has investigated the acquisition of language and literacy in L2 adolescents learning English. Therefore, we know little about which specific skills influence the development of reading and oral production (August & Shanahan, 2006). Methods - The current study examined and compared factors related to reading comprehension in adolescents with English as a first language (L1, n=28) and L2 with less than two years in Canada (n = 49). Specifically we compared the factor loadings of measures of decoding and phonological awareness, vocabulary knowledge, and print exposure separately for each group. Results - Both groups produced a three-factor model. However, the factor loadings differed between groups. English L1 students had separate factors for (1) oral vocabulary, (2) phonological decoding, and (3) print exposure. Whereas, the model for the English L2 students had three different factors: (1) vocabulary and phonological decoding, (2) exposure to magazines, and (3) exposure to authors. Conclusions - The results demonstrate that in early English L2 learners reading skills are represented as one factor. These findings suggest that extensive research is required to examine models of L2 reading in adolescents who are beginning to acquire English. Further research should test and compare models of adolescent reading development of English L1 and L2 adolescents.
Purpose: The main purpose of this project is to study effects on word recognition caused by overlap in form and meaning between the word that has to be recognized and other word representations stored in the mental lexicon. Does the presence in the mental lexicon of an entry for homework speed up the recognition of the word work? Method: Children were chosen as population of interest, because they are learning many new words each day, and little is known about the role of related entries in the mental lexicon on word recognition for this particular group. Following a cross sectional longitudinal design the development of effects of word families on word recognition was investigated in second and fourth grade children who were followed over a two-year period using visual lexical decision. Results: The results show that even the youngest children in the study benefited from knowledge of words which are related to the word they are reading, both on word recognition speed and accuracy. Interestingly, this benefit does not seem to increase linearly by age, but depends on the need to use information about related words in decoding the word to be read. Conclusions: The data provide evidence for the family size effect in young children. The absence of a linear increase of this effect over age levels suggest a reorganization of the mental lexicon is taking place throughout the grades.
Purpose This presentation reports data on an intervention with community college students with low reading and writing ability. The study asks: (1) does guided practice in multiple reading and writing strategies improve reading and writing skills? (2) Does contextualization increase the effect of guided practice on skills improvement? (3) Do prior knowledge, topic interest, and student characteristics affect outcomes? Method Participants were n=600 low-SES students enrolled in remedial (developmental education) classes in three urban community colleges in the U.S. There were three conditions: contextualized and generic text treatment (randomly assigned), and no-treatment comparison. The intervention consisted of 10 self-paced, self-directed units providing guided practice in written summarization, clarification, and question-formulation skills. Repeated measures analyses of covariance and multiple regressions were used to analyze the data. Results and Conclusions Results will be reported for the following outcome measures: pre-post Nelson-Denny Reading Comprehension Test and locally-developed Science Summarization Test (representation of main ideas from source text, productivity, reproductions, and writing quality), and performance on written summarization, clarification and question-formulation items during the intervention, controlled for prior knowledge, interest, language proficiency, age, gender and secondary education credential. Implications for community college instruction will be drawn from the findings.
Purpose: Diffusion Tensor Imaging (DTI) was used to investigate the neural substrates of Reading Disability (RD) in children. Methods: DTI data was collected on children ages 7-14 years: 8 with RD and 9 controls. Voxel-wise statistical tests for between-group differences were performed on images of Apparent Diffusion Coefficient (ADC), Fractional Anisotropy (FA) and Principal Diffusion Direction (PDD) -- indicators of water mobility, white matter integrity and fiber orientation, respectively. Results: Consistent with previous studies, we found decreased FA for RD compared to controls in left temporo-parietal white matter. Additionally, ADC was decreased for RD in left superior temporal and middle occipital gyri, both of which are associated with reading in fMRI studies. In the right frontal lobe, PDD was found to be more medial-lateral in RD than controls and FA was correlated with an out-of-scanner word reading efficiency measure. Conclusions: Our findings are consistent with previous evidence suggesting structural differences in left temporo-parietal white matter in children with developmental language disorders, possibly suggesting a basis for fMRI activation differences in left occipito-temporal cortex in RD. Additionally, right frontal white matter was associated with reading-skill differences using two separate approaches. Overall, findings suggested that anomalous networks may be implicated in the neural substrates of RD.
Yaacov Petscher (Florida Center for Reading Research); Christopher Schatschneider; Young Suk Kim - A comparison of oral reading fluency trajectories using equated and non-equated data from DIBELS oral reading fluency
Purpose: This study seeks to examine the difficulty among selected DIBELS Oral Reading Fluency passages in 3rd grade and use scores from equated passages to compare individual growth curves from equated scores with non-equated scores. Methods: Using a large extant database from the Florida Center for Reading Research, we: 1. Examine the difficulty of 12 passages given during the 2005-06 academic year. 2. Estimate how much of the variability in DORF scores is between students and between passages 3. Conduct an equipercentile equating on the passages. 4. Using the equated scores, apply growth curve and/or piecewise growth curve models to evaluate the individual trajectories of the equated passage scores compared with curves when equating had not been applied. The sample (n = 13,335) comes from Florida's Reading First population. Participants were fairly equal in gender distribution (Male = 51%), and were more representative of White (37%) and Black (37%) students. Results: Data received December 6, analyses to be completed in February 2008. Conclusions: This study is unique in that it applies equipercentile equating to a large data set of individuals and demonstrates the differences in curves that will be observed based on the use of equated or non-equated passage scores.
PURPOSE Research on alphabet knowledge suggests that, contrary to the assumptions of typical alphabet assessments (scores 0-26), letters vary in how difficult they are to learn and may also follow a typical developmental sequence (e.g., children may learn letters A-B-C before X-Y-Z). The present study sought to empirically test these hypotheses using item response theory (IRT). METHOD Archival data (N = 1074) was compiled concerning the letter name knowledge of preschool children, all native English speakers, ages 26 to 81 months. Children were presented with randomly ordered letter cards and asked to produce the name of each letter. Using MPlus software, an IRT analysis was conducted on the dichotomously-scored responses to individual letters. Three IRT models were compared: a base model constraining all letter difficulties to be equivalent, and both 1- and 2- parameter models. RESULTS The 2-parameter model provided the best fit to the data. Difficulty and discrimination parameters varied across individual letters. Results suggested a developmental order for letters at the extreme ends of the difficulty continuum. Calibration and validation using two random subsamples supported model invariance. CONCLUSIONS Findings add to the growing literature exploring the nature of alphabet learning in young children, and have implications for both assessment and instructional practice.
PURPOSE Effective teaching of reading may require specialized knowledge to implement evidence-based practices. We hypothesized that student reading gains would depend on both teacher knowledge and the use of explicit decoding instruction in the classroom. Post hoc, we explored the implementation of decoding instruction by effective teachers, differentiating between those with above- and below-average knowledge levels. METHOD First grade teachers (n = 42) and students (n = 439) were assessed on the Teacher Knowledge (TK) survey and Letter-Word Identification subtest (WJ-Tests of Achievement), respectively. Videotaped classroom observations were used to determine the number of minutes each student spent in specific decoding activities. HLM analyses tested the direct impacts of TK and decoding instruction, as well as the TK x Decoding interaction, on student gains. RESULTS A TK x Decoding interaction predicted student word recognition gains. Students of high knowledge teachers achieved larger gains when more decoding instruction was provided while students of low knowledge teachers showed weaker gains. Post hoc analyses demonstrated differences among such teachers in the specific types of decoding instruction provided. CONCLUSIONS Results highlight teacher knowledge as a necessary but insufficient factor in promoting student reading skill; it must be enacted in the classroom to impact student learning.
Purpose: To investigate whether letter identification is influenced by the distributional regularities inherent to orthography. Specifically, to determine if letter position encoding is sensitive to positional letter frequency (i.e. the frequency with which individual letters appear in particular positions within written words of a specified length). Method: Skilled (N=28) and dyslexic (N=29) adult readers of English and skilled adult readers of Greek (N=24) were given a letter search task that required identification of a prespecified letter target embedded within a random five-letter string. Stimuli were drawn from the reader's native orthography and consisted of 25 English letters and 20 Greek letters. Response latencies to identify individual letter targets in each of the five string positions were correlated with positional letter frequency counts for the reader's native orthography. Results: Significant negative correlations were found for each of the three participant groups for letters appearing in the initial string position, reflecting shorter response latencies for more frequently occurring letters. A significant negative correlation between response latency and positional letter frequency was also found for skilled readers of English for letters presented in the final string position. Conclusions: Extraction of statistical regularities from orthographic input is a mechanism by which orthographic learning occurs.
Purpose: This study addressed ways in which spelling can be an underlying cause for students who struggle with writing. This study examined the type of spelling errors students commit in their writings. The errors were examined to determine if the amount of spelling errors cause the productivity of the narrative to be constrained. The following research questions were investigated: (1) What type of spelling errors do students in sixth grade commit when writing; (2) Do the amount of spelling errors committed by students in sixth grade constrain the productivity of their writing? Method: The participants included 142 sixth-grade students from a low SES school in a large, urban district in Southeast Texas. On two different occasions, the students wrote about a time when they were sad and their best day ever. Students' spelling errors were categorized as phonological, morphological, or orthographic errors. Using t-units, students' writings were analyzed to determine the amount of productivity for each narrative. Results: Students' spelling showed more morphological errors, especially with inflected endings. Students who were overall poor spellings wrote less, which means that these students' writings were constrained by their spellings. Conclusions: Implications for classroom instruction and interventions will be discussed.
Before children learn how letters correspond to sounds, their writing is often characterized as random (Gentry, 1982). We evaluated this characterization by analyzing the productions of prephonological spellers (mean age 4.8; 35 in Brazil, 23 in the US), who were identified by using string-edit techniques and randomization tests to ascertain that their productions were no closer to phonologically plausible spellings of the target words than to those of other words. Instead of being random, the spellings reflected the texts children experience: children's books, their own names, and alphabet listings. The frequency of the letters they produced correlated with their frequency in books from the child's country, over and above an effect whereby children favored letters from their own name. In addition, children's use of letter bigrams correlated with their frequency in text and with whether the bigram appears in alphabetical order. Doubled letters were avoided more by Brazilian than US children, reflecting their rarity in Portuguese but not English text. No evidence was found for reportedly universal prephonological patterns such as minimum word length. Our findings suggest children's prephonological writing is neither random nor universal, but reflects statistical learning of the visual patterns children encounter
Daisy Powell (Department of Psychology, Roehampton University, UK)Stainthorp, Rhonda; Stuart, Morag - A deficit in orthographic knowledge, but not orthographic learning, in children poor at Rapid Automatized Naming tasks
Purpose: This research forms part of a longitudinal investigation of the cognitive factors underlying the well-established relationship between rapid automatized naming (RAN) and reading development. We have previously reported data suggesting that a group of children with a single RAN deficit showed a visual processing deficit relative to controls matched on phonological awareness and other key variables. The aim of the current research was to investigate whether the visual deficit in the low RAN group may in turn lead to deficits in orthographic skills. Method: The performance of children with a single RAN deficit was compared with controls on a series of tasks assessing different aspects of orthographic knowledge and learning. The children's performance was further considered in the light of prior, longitudinal data on reading and reading related skills. Results: Consistent with our predictions, the low RAN group performed less well than controls on several measures of orthographic knowledge, even when controlling for reading skills. However, they unexpectedly showed an advantage over controls in an orthographic learning task. Conclusions: The results have implications for our understanding of the RAN-reading link. More generally, findings raise questions about the construct of orthographic knowledge, and how it is operationalized in different paradigms.
Purpose: The purpose of this study was to examine prior knowledge effects on oral reading accuracy. Does prior knowledge about a passage's topic lead to fewer miscues that change the meaning of the passage? Does prior knowledge have similar effects for reading disabled (RD) and control children? Method: 52 fourth-graders (26 RD, 26 controls) read aloud one passage of the Qualitative Reading Inventory, after being assessed for prior knowledge of the passage topic. The child's reading was taped and transcribed, and miscues were coded for semantic similarity. Results: A three way interaction was found between reading ability, prior knowledge, and type of miscue. Children with RD who had prior knowledge made more semantically similar miscues and fewer semantically dissimilar miscues than those without prior knowledge. Controls with prior knowledge did not differ from controls without prior knowledge in the number of either semantically similar or semantically dissimilar miscues. Conclusions: The results suggest that prior knowledge serves a compensatory function for the weakened lexical representations of children with RD. Just as context has been found to facilitate word identification more in children with RD than controls, so too does prior knowledge.
Print Concepts (PC), the knowledge of how print is used in books and how it represents language, is an understudied aspect of emergent literacy skills. Research has just recently begun to conceptualize and examine the structure of this domain. For example, Justice and colleagues (2006) indicate that PC, as assessed by the Preschool Word and Print Awareness measure, is a single unitary trait in preschool children three to five years old. However, they found considerable variability in scores, which may be attributable to age. It was the intent of the current study to determine if the unitary structure of PC is stable in both younger and older preschool children. In this study, 1,753 preschool children (25-71 months, 50% boys, primarily Caucasian and African American), as a part of four other studies, were assessed on Clay's (1979) Concepts About Print 'Sand Task.' Confirmatory factor analysis results support a unitary factor structure for PC in both age groups. However, differential item functioning analyses indicated that the items tapping punctuation knowledge function differently between the two age groups. Implications for assessment and future research will be discussed.
It has been consistently shown that dyslexics are relatively poor at categorizing phonetically similar but phonologically contrastive speech sounds, especially in noise. To examine this deficit further, a categorical perception experiment was used to assess dyslexics' and controls' categorization of synthetic speech sounds along /ada/-/aga/ and /aba/-/awa/ continua that were factorially combined with noise and corresponding visual speech cues. Results show that the dyslexics exhibited inordinate difficulties in categorizing /ada/ vs. /aga/ in noise. In noise, the dyslexics' /ada/-/aga/ categorical function was flatter in slope relative to controls, and generally less influenced by the introduction of visual cues. However, the dyslexics showed better categorization of noise masked /aba/-/awa/ audio cues, and greater reliance on the corresponding visual cues than controls. This suggests that dyslexics may find place of articulation contrasts more problematic than manner contrasts, regardless of the modality of presentation. Although auditory temporal processing theories assert that difficulties in processing brief formant transitions underlie the dyslexics' poor categorization of /ada/ vs. /aga/, they cannot easily account for difficulties in processing the corresponding visual cues for these sounds. Thus, a more comprehensive account of dyslexic speech perception impairment may make reference to the aural and visual processing of phonological features.
Purpose: The purpose of this study was to examine the within- and cross-linguistic contributions of morphological sensitivity to word reading and reading comprehension in Spanish-English bilingual children. Specifically, we aimed to answer the following questions: (1) Does morphological sensitivity explain unique variance in reading in Spanish and in English? (2) Does morphological sensitivity in Spanish affect reading in English, and vice versa? Method: Ninety-seven Hispanic-Canadian children from grades four and seven received a battery of measures on morphological sensitivity, non-verbal IQ, working memory, vocabulary, phonological awareness, word reading, and reading comprehension. Morphological awareness, vocabulary, and reading were measured both in English and in Spanish. Other measures were given in English only. Results: Regressions analyses indicated that morphological sensitivity made unique contributions to both word reading and reading comprehension within Spanish and within English after controlling for intelligence, memory, phonological awareness and vocabulary. Cross-linguistic effects of morphological sensitivity were found for word reading but not for reading comprehension. Conclusion: The results suggest the possibility of transferring some reading related skills between languages with overlapping characteristics such as English and Spanish. These findings have theoretical implications for second-language reading acquisition and practical implications for teaching and assessment of Spanish-English bilingual children.
This study examined the auditory perception and invented spelling of English stop consonants, which exist as different phones in Spanish and English. Data were collected in June of the kindergarten year and in June of first grade year from 13 pairs of monolingual and bilingual students matched for phonemic awareness and letter-sound knowledge. The children completed auditory perception and invented spelling assessments. A repeated measure analysis of variance on the kindergarten and first grade data showed no significant language-time interaction for auditory perception (p < .249). A near significant language-time interaction was found using a repeated measure ANOVA (p < .071) on the kindergarten and first grade invented spelling data. Significant differences in spelling were found using t tests at the end of kindergarten (p < .045) with monolingual outperforming bilingual students; but none were found at the end of first grade (p < .674). These results demonstrate that Spanish-English bilingual children, who are able to recite the names and sounds of English stop consonants as well as discriminate between them, may experience Spanish interference in the more complex task of invented spelling. For this sample, the interference appeared to disappear with time and/or learning to read.
There is a surprising lack of systematic research evaluating the effects of reading exercises for young deaf children. Therefore, for the present study two computer-based exercises were developed and learning effects were determined by post tests. One (spelling oriented) exercise was to select the correct word among three orthographically similar alternatives that corresponds to a drawing or a sign (digital video). The other (meaning oriented) exercise was to select the correct sign or picture among three alternatives that corresponds to a written word. Eleven deaf Dutch children with a mean age of 7;10 participated in the study. A first question was whether in single word exercises the meaning or the spelling of a word should be emphasized. A second question was whether there was any effect of using drawings or signs to refer to the meaning of the word. The results reveal that emphasizing the word-spelling is most effective for learning to read for deaf children and the findings also suggest that drawings are more efficient in the current exercises.
Purpose: The early linguistic environments of bilingual kindergarten children influence their invented spellings of English words. We examined whether predictable kinds of metalinguistic awareness can be found in bilingual children's spelling errors after they have been exposed to written and spoken English routinely. Method: Bilingual children (N=83, aged 9-10 years) from the same English-medium primary school but with three different language backgrounds (Mandarin L1, Malay-L1 or English-L1) completed a 120-item experimental test specifically designed to assess the effects of frequency, regularity and morphological status (verb type) on spelling performance. Results: Using a detailed classification system, three independent raters made judgements about the nature of the children's spelling errors: visual-orthographic (whip <winpe>), phonological (sewed <soad>), morphological (flooded <flood>). ANOVAs showed that Mandarin L1 children relied more on visual-orthographic processing (frequency effect), whilst Malay L1 children relied more on phonological processing (regularity effect). The English L1 group used visual-orthographic and phonological processing, and show greater sensitivity to the target word's morphological status. Conclusion: Theories of spelling development for unilingual English-speaking children invariably stress the role of phonological awareness but the metalinguistic footprints evident in bilingual children's spelling errors are more consistent with a modified version of Seymour's (1997) dual-foundation model.
Purpose: A word learning paradigm was used to teach children new word meanings with and without orthography. We investigated whether the presence of orthography facilitates word learning. Method: Fifty-eight children aged 8-9 years took part in the experiment. To approximate word learning, children learned 12 pairings between novel phonological forms and pictures of novel 3D objects. For half of these pairings, nonword orthography appeared above the picture. Children were not alerted to the presence or absence of orthography. After training, children were presented with a nonword-picture matching task to assess learning of nonword 'meanings'. Following this, children were asked to spell nonwords to dictation. Results: Performance was better for nonwords trained with orthography in three measures. During training children were more likely to learn nonword-picture pairings that appeared with orthography. Also, performance on the nonword-picture matching and spelling post-tests was improved where children had been exposed to orthography. Conclusions: Unsurprisingly, children's spelling of novel phonological forms was improved if they had experienced orthography during training. Of particular interest was the finding that orthography seemed to boost the ability to form associations between phonological and semantic information. The theoretical and practical implications of these findings will be discussed.
Purpose: In this talk we will present the results of study of reading in young adult college students. The focus of the study is to better understand the underlying behavioral and neurobiological correlates of reading and reading comprehension in both skilled and impaired readers in this age group. Method: Sixty-five young adult students with at least some college education, ages 18-24 years, were evaluated on measures of word recognition and reading comprehension as well as phonological processing, oral language and executive function. Participants also took part in an fMRI paradigm to examine neural correlates of word recognition and sentence comprehension in both auditory and visual modalities. Results: We will review contributions of various skills to reading comprehension as well as functional neuroimaging results. Conclusions: Various cognitive and neurobiological processes influence reading in this population.
Jeremiah Ring (Texas Scottish Rite Hospital for Children); Brown, Sasha; Black, Jeffrey L. - Reading Fluency Instruction for Students at Risk for Reading Failure: Effects of Two Approaches to Improving Reading Rate
Background. The importance of reading fluency in comprehension has been well documented. The ultimate goal of fluency intervention is for students to improve their rate of reading for all texts, not just passages that are practiced. The purpose of this project was to compare how two different models of reading rate instruction achieved this goal. Method. The data reported are from a semi-randomly assigned clinical trial of children who were considered at-risk for reading failure (i.e., Tier II) from their performance on state-mandated tests. Both treatments were published programs based on the repeated readings model of fluency instruction. The primary difference was that one program focused instruction at the text-level (i.e., short stories); the second focused instruction at the lexical and sub-lexical level of phonics concepts (e.g., CVC units). Results. Preliminary analyses suggest statistically significant gains on standardized measures of reading, comprehension and oral reading fluency for both groups. Treatment differences were not reliable. Analyses of measures of rapid decoding skills are planned. A second cohort of younger children will be added to explore developmental differences. Conclusions. The results of this study suggest that focused instruction on reading rate at the word level shows similar transfer to text-level reading fluency as text-level reading rate practice.
Purpose Little research has examined the link between spelling and decoding errors patterns. The purpose of this study is to analyze the relation between spelling and decoding error patterns across reading skill levels. This information will enable teachers to quickly identify deficits in student spelling and decoding skills to allow for strategic intervention implementation. Methods One hundred, second grade students were administered a decoding inventory and two spelling inventories. Letter sound patterns on the decoding inventory were matched with the letter sound patterns on the spelling inventories to examine the relation between spelling and decoding error patterns. The data analysis will include the Pearson product-moment correlation to compare student performance on the decoding and spelling inventories. In addition, 2 (high and low readers) x 3 (error types) analysis of variance with repeated measure on second factor (error types) will be conducted to determine if there is a statistical difference between the error types made by high/low readers on the decoding and spelling inventories. Results and Conclusions This study will add to the theoretical literature base on the relation of spelling and decoding. In addition, this study will inform teaching practices by examining assessment procedures teachers use and strengthening decoding/spelling interventions.
Purpose: This study examined the relationship between reading fluency and reading comprehension in first grade English language learners (ELLs). This study was motivated by the theoretical view that fluency may be a measure of overall reading comprehension. Method: Fluency was measured with a composite of curriculum-based words correct per minute measures. Comprehension was measured with the Stanford Achievement Test 9 (SAT9). Additional decoding-related and language-related competencies of 81 first grade children from two language groups were measured and then reading comprehension and fluency were regressed on to these measures. Phonemic awareness and Woodcock Johnson word decoding, Peabody-III, and the Khan- Lewis articulation test were measured. The relationship between fluency rate and achieving grade level comprehension was also examined. Results: Fluency and SAT 9 reading were highly correlated (r= .77). Articulation and Woodcock scores were significantly associated with both comprehension and fluency. Grade level comprehension was associated with lower fluency rates for ELLs than is typically found with English only children. Conclusions: Fluency is a reasonably good measure of comprehension in English language learners, although the fluency rate necessary for good comprehension may be lower than for English only children. Articulation is an important factor in ELLs reading comprehension. Articulation accuracy may be taping both speech production rate and the quality of phoneme representations.
Purpose and Method: The purpose of this study was to explore the procedure of having students orally pronounce unknown words encountered during otherwise silent reading to determine whether this contributes to their learning of new vocabulary words. A matched-pair random assignment experimental design was utilized. Sixty-four fifth-grade students silently read passages in which low-frequency concrete nouns were embedded and underlined; half of the participants orally decoded target words when they came to them, and half engaged in a lexical decision task. A second purpose of this study was to inquire about students' self reported normal strategy use for figuring out unknown words during independent reading. A subset of students (n=32) was asked open-ended questions about what they normally do when they encounter new or difficult words in text. Results: Findings indicate that phonological recoding of new words enhances vocabulary learning. Target word decoding accuracy predicted significant unique variance in students' memory for words' pronunciations and spellings. However, participants, especially poorer word readers, were likely to report skipping over unknown words rather than attempting to decode them during independent reading. Conclusions: Implications for instruction include that readers should be taught to attempt to pronounce unknown words during reading, rather than skipping and guessing as many reported doing.
Although many students with Autism Spectrum Disorders (ASD) show evidence of reading comprehension difficulties, they seem to be able to have some understanding of what they read (Nation et al., 2006). The purpose of this study was to conduct a first test of the effectiveness of a small-group reading comprehension intervention designed specifically for ASD students. The intervention combined instruction on decoding of multi-syllabic words, word meaning (vocabulary), reading of connected text, main idea identification (Carnine et al., 2004) and story schema (Williams et al., 2001). Thirteen ASD students (7-11 year old) with end of first-grade or better word recognition ability participated in the study. They were randomly assigned to the intervention condition or to a control condition (regular instruction only). Results demonstrated substantial near-transfer effects on researcher-developed measures of instructed vocabulary (d = 1.06) and main idea identification skills (d = 1.3). No differences were observed on far-transfer measures of comprehension. These results suggest that with adequate support, ASD students make significant gains in reading comprehension.
Abstract This study evaluated the research-based vocabulary strategies used in fourth grade social studies instruction and evaluated the efficacy of professional development on the quantity and quality of strategy use. Research questions addressed: 1. What research-based vocabulary strategies do fourth-grade teachers use during social studies instruction and how do these compare to Durkin's (1978-79) findings? 2. Does professional development in research-based vocabulary instruction result in differential quantity and quality of strategy use? Twenty-six teachers were randomly assigned to either a typical practice (n=8) or a professional development (n=17) group (PDG). The PDG group received XXX hours of distributive training focused on research-based vocabulary strategies. Three lessons 20-30 minutes each, for each teacher were audio record. Audios were coded and analyzed. Multivariate analysis of covariance will be used to analyze the impact of professional development on teacher strategy use. Results Preliminary results seem to corroborate the findings of Durkin (1978-79) that minimal instruction in reading comprehension strategies, including vocabulary, occurs. Additionally, professional development situated within the content appears to positively impact practice. Conclusions A study of teachers' use of research-based vocabulary strategies is important because: * Documents changes in vocabulary practices in fourth grade social studies classrooms * Identifies vocabulary strategies appropriate for social studies classrooms * Identifies future directions for professional development in vocabulary.
Purpose: The study investigated predictors of reading skill gains resulting from intervention programs administered to low literate adult learners. Method: Low literate adults are defined as scoring between 2nd and 6th grade ability on an English word recognition (naming) task. A sample of 133 completed one of three intervention programs; one focused on decoding; one on reading fluency; and one a mixture of the two. Learners were pre and post tested in word attack (WA), real word recognition (WREC), reading fluency (RFL), and passage comprehension (PC). Multiple regression analyses were conducted on each post-test gain, with pre-test scores entered first, followed by cognitive, linguistic, instructional and demographic variables. Results: Over and above pre-test levels, gains were predicted by variables consistent with theoretical and practical hypotheses. WA gain variance was predicted by a combination of pre-test WA, WREC, spelling for sounds, and elision skill; WREC gains was predicted by a combination of WREC, DEC, PC, and vocabulary; PC gain, by PC and vocabulary; RFL by RFL, age, and instructional sessions. Conclusion: The results will be interpreted in terms of models of reading development/disability, as well as how such models might be useful in planning instructional interventions for low level adults.
Purpose Embedded picture mnemonics for learning letters involves embedding a letter within a picture whose name begins with an associated phoneme (e.g., the small case letter "a" in a picture of an apple). Our experiment investigated whether using an embedded picture mnemonic would promote learning letter names and associated phonemes more effectively than using more traditional disassociated letter-picture mnemonics (i.e., the small case letter "a" next to a picture of an apple). Method A sample of 25 low SES kindergarten students was identified with no knowledge of eight letters and little phoneme isolation ability. They were randomly assigned to one of two training conditions: embedded picture mnemonics or dissociated picture mnemonics. Training sessions were held for each of the eight letters over two weeks with a review session each week. Results Post-treatment results after one week and after two weeks showed that embedded picture mnemonics significantly increased letter name knowledge but not letter sound knowledge. Little letter-sound learning occurred in either group. Conclusions Embedded picture mnemonics is an effective technique for learning letter names even with children that have little phoneme knowledge. However, learning letter names does not necessarily transfer to letter-phoneme associations where phoneme knowledge is absent.
In three ERP experiments, morphology-based decomposition of words and pseudowords is explored in Spanish. Subjects were asked to perform a lexical decision on morphologically simple- or complex-word-strings, such as 'alerg+ic' and 'alerg+ist', while family size for both lexemes (LFS) and morphemes (MFS) varies. In Experiment I, former results (Schreuder & Baayen, 1997) were replicated: Monomorphemic High-LFS targets required less brain activity and produced faster responses than monomorphemic Low-LFS targets. In Experiment II, when lexemes are used as primes, responses to High-LFS targets were inhibited more than Low-LFS. In Experiment III, High-LFS targets are processed faster than Low-LFS when FS is controlled for both, lexemes and morphemes. The temporal course of brain activity of the word recognition process is discussed. Both, behavioral results and Evoked-Related Potentials endorse the view that morphological decomposition emerges after lexical access.
Maria Thereza Mazorra Santos (Associate researcher of the Department of Physiotherapy, Communication Sciences and Disorders and Occupational Therapy, School of Medicine - USP);Befi-Lopes, Debora Maria - Vocabulary, Phonological Awareness and Rapid Naming: Contributions for Spelling and Written Production.
This research aimed at understanding the linguistic processes involved in the learning of written language, and developing a protocol to analyze written text. It included eighty-two 3rd graders (ages 9 - 10) and residents of the suburbs of São Paulo, Brazil. Three studies composed this investigation. The first analyzed student performance on tests of vocabulary knowledge, phonological awareness and objects rapid naming. All students performed as expected for their ages. The second study focused on the analysis of written production, as evaluated through a word and pseudoword spelling task and a written composition. The spelling errors observed were compatible with the development of written skills of the subjects, however their classification is not enough; it is necessary to understand which are the adopted spelling-strategies. Regarding to text-writing skills, although the children have not yet mastered the narrative scheme, they displayed enough linguistic knowledge to elaborate original written stories. The proposed protocol was revealed as a practical tool to evaluate several aspects involved in written narrative. The third study focused on the correlations among the different performance variables analyzed, including vocabulary, phonological awareness, rapid naming tasks, spelling and production of narratives. Vocabulary correlated to all levels of written composition analysis, however phonological awareness and rapid naming correlated mostly to the syntax and grammar structure of the text generation.
Purpose: With the passage of the latest Individuals with Disabilities Education Act (IDEA), the federal government no longer requires an IQ-Reading Achievement discrepancy to classify students as having a reading disability. Instead, a number of alternative classifications systems based on a students Response to Instruction (RTI) have been proposed. To evaluate various operational definitions on a large sample of students in the state of Florida, data from 34,000 first and second grade students were used to identify students using various RTI classification schemes to investigate their degree of overlap and predictive utility on future reading performance. Method: Data from DIBELS Oral Reading Fluency, Peabody Picture Vocabulary Test-III and SAT10 Reading Comprehension were analyzed using the following six classification methods: discrepancy, normalization, dual discrepancy, benchmark, median split on growth, and quartile split on growth. The sample was restricted to children who had data for all time points (including the outcome measures from the following year) and children who came from classrooms that had at least 12 students (to stabilize the Dual Discrepancy definition). Results: Classification of reading disabilities varied by ethnicity. Overall, the classification schemes performed differently in terms of predicting reading performance one year later. While there was some overlap among the classification schemes, others varied widely in the number and types of students identified. The schemes also varied in terms of their overall sensitivity and specificity to predicting future reading performance. Conclusions: Variability among the number and type of students identified using a set of different proposed classification schemes exists. Researchers and educators need to try and come to a consensus regarding the classification of students with reading disabilities.
Purpose: To evaluate technical adequacy of a parent report measure for assessing the home literacy environment (HLE) of primary aged children in New Zealand. Method: A newly developed HLE questionnaire was administered to parents of 274 children in Years 1 - 5 as part of a study of developing literacy. All children were assessed on grade appropriate measures of developing literacy three times per year using the Dynamic Indicators of Early Literacy Skill (DIBELS, Kaminski et al., in press). Children in Years 1, 3, and 5 were also administered the Peabody Picture Vocabulary Test (PPVT). Results: Factor analyses suggest four distinct factors, with generally good internal consistency, assessing child home reading, parent support of child language/literacy, limited TV viewing, and family literacy environment. Correlational analyses suggest differential patterns of associations of factors with other family status variables (parent education, occupation) and children's performance on literacy and language assessments over development. Preliminary analyses controlling for PPVT performance or mother's education found aspects of the HLE assessed by this measure contributed significantly to the variance in children's performance on DIBELS, although patterns of association differed by grade. Conclusions: Findings suggest HLE is multidimensional and should continue to be considered across reading development.
Purpose: The purpose of this study was to examine the development of reading prosody and its impact on later reading skills, in particular: (a) how the key features of oral reading prosody unfold with development; and (b) the extent to which the development of reading prosody is predictive of later reading fluency and comprehension outcomes beyond word reading skills alone. Method: Suprasegmental features of oral reading (in particular, the number of pausal intrusions within sentences and adult-child intonation contour match) and word reading skills (TOWRE) were measured for 92 children at the end of grades 1 and 2 along with outcome assessments of oral reading fluency (GORT) and reading comprehension (WIAT) at the end of third grade school year. Results: Path model tests found evidence of a relationship between the presence of few pausal intrusions during oral reading in first grade and subsequent development of an adult-like intonation contour in second grade. Outcome model tests indicated that the intonation contour was a significant predictor of later fluency once word reading skills were taken into account, but both the cumulative effect of decreases in the number of pausal intrusions and early acquisition of an adult-like intonation contour predicted later comprehension skill. Conclusion: Changes in reading prosody between first and second grade predict reading skills in third grade beyond that accounted for by word reading skills. Development of reading prosody is an important element of reading fluency and should be considered a key aspect in any definition of it.
Purpose: With the percentage of adults who report reading for pleasure at an all time low, the need to understand the factors contribute to an individual's early interest in pursuing reading activities in their leisure time seems increasingly important. The purpose of the current study was to examine the factors that influence preschool children's participation in reading activities during their leisure time. Method: Thirty preschool children and their parents participated in a study in which observations of preschooler's participation in reading activities during their free play were collected along with measures of children's early literacy competence, and objective checklist measures of parental reading habits and children's exposure to shared reading at home. Results: Parental reading habits, but not children's early literacy competence or exposure to shared book reading at home, explained significant and unique variance in the amount of time children were observed engaging in reading activities, even after controlling for the influences of parental education and child age. Conclusions: These results suggest that parental modeling of literate behavior in the home is an important component of the home literacy environment and that it may be an effective way to help develop an interest in reading in young preschool children.
This report is part of a larger longitudinal study focusing on the development of writing skills in EL1 and EL2 children between grades 4 and 6. Participants were 103 normally developing children recruited from 12 different schools. Children were matched on language status (EL1 and EL2), gender, and non-verbal cognitive ability (Raven Test percentiles). Stories were written in grades 4 and 6 as part of the Test of Written Language (TOWL-3). Children were given 15 minutes to write the story in response to the picture stimulus (futuristic scene). For the analysis we counted introductory phrases, verb phrases, subordinate clauses, total number of clauses, simple sentences, compound sentences, complex sentences, compound-complex sentences, run-on sentences, not capitalized/unpunctuated sentences, T-units and number of words (story length). Repeated measures ANOVA showed no differences between the EL1 and EL2 groups and no gender differences except for the story length (girls wrote more). However, there were differences based on cognitive ability with higher cognitive ability children performing better. In grade 4, they wrote more simple and compound sentences with less run-on sentences; in grade 6, they produced more compound and complex sentences. Raven scores correlated with memory, reading comprehension, decoding skills, auditory analysis, vocabulary, and arithmetic.
Eliane Segers (Behavioural Science Institute, Radboud University Nijmegen); Davids, Nina; Verhoeven, Ludo - Origins of Phonological Awareness in Children with Specific Language Impairment: A Neurocognitive Perspective
Purpose Children with Specific Language Impairments (SLI) often have problems in phonological awareness, leading to reading problems. We tested whether these problems have speech and/or non-speech specific origins. This has been studied before in behavioral experiments, but drawbacks of these are that such factors as attention and motivation to do the test can influence the results. Furthermore, linguistic and non-linguistic stimuli in earlier studies differed in frequency and acoustic complexity. By using EEG measures, the first problems were overcome. To avoid the latter, non-linguistic stimuli were created by rotating the spectrum of the linguistic stimuli. Method Forty five-year-old kindergartners (20 normal language, 20 SLI, matched on age and non-verbal IQ) were first tested on rhyming, auditory discrimination, letter knowledge, attention, non word repetition and non-verbal IQ. In the behavioral experiment, the Tallal repetition task was used to test discrimination in linguistic and non-linguistic stimuli. In the EEG experiment mismatch negativity was elicited in an oddball paradigm. Results The data show behavioral and neural differences in auditory perception between the two groups. The differences were larger for speech as compared to non-speech. Conclusions The results provide further insight into the nature of the problems in phonological awareness in children with SLI.
The aim of this study is to examine how Spanish code complexities influence the dyslexic children's literacy abilities. It was used a reading-level match design. Participants were 31 dyslexic children, 31 children with the same chronological age and 31 children with the same reading level, thereby younger. Two reading and spelling tasks of 127 words and 127 pseudowords, respectively were carried out. Items were chosen according to their complexity - simple, consonant cluster, digraph, contextual and position influence, silent letter, inconsistency, and written accent-. It is observed a dissimilar result patter in reading and spelling regarding complexities influence. The effect of type of complexity is more evident in Spelling. Written accent is the type of complexity which shows dramatically lower results. Findings are discussed regarding implications for dyslexia assessment and intervention in Spanish.
This multi-year project used a multidimensional, theoretical model of reading development and failure in order to advance our understanding of the critical instructional components (decoding, fluency, and text comprehension) needed by students with mild mental retardation to acquire reading skills. Two different multi- focus intervention programs (Phonological Analysis and Blending (PHAB); PHAB + RAVE-O) were compared to a control program (MATH). The instructional programs were equal in duration (1 hour) and in length (120 sessions). Instruction was delivered in small groups (n=4) by experienced teachers who are members of the research project staff. Each child met state criteria for mental retardation and district eligibility criteria for special education services and were assigned randomly to a program (n= 80). Students were assessed at three different time points: pre- instruction, following 60 hours of instruction) and post intervention (after 120 hours of instruction). A battery of both standardized and experimental tests were administered. Significant interaction between instructional group and time point was found on measures of phonological awareness and word identification. Children with mild intellectual disabilities can learn to read and can benefit from instruction utilizing similar content and procedures to those used with typically developing children.
Purpose: This study investigated English-Hebrew bilingual children's emergent inflectional morphological awareness (MA) in Hebrew(their L2; typologically different from their L1). Examined was whether knowledge of the specific word being inflected, versus a more general vocabulary measure, impacted the children's abilities to properly inflect words. Expressive and receptive measures were used in order to further determine whether type of task matters. Method: Fifty primary students were tested in spring,2007. Measures: receptive vocabulary, experimental vocabulary task specific to the test items; two expressive MA analogies tasks- items that tested experimental vocabulary words and items that did not; two receptive MA choice-based analogies tasks- one with items that were experimental vocabulary items and one with nonwords; a control nonverbal intelligence task. Analyses: partial correlations, hierarchical linear regressions, and ANOVAs/MANCOVAs with post- hoc contrast analyses. Results: Familiarity with specific word meanings affected children's ability to inflect words in the expressive task condition, but not in the receptive task. Familiarity with specific word meanings accounted for unique variance in both types of tasks. Conclusions: Findings are directly relevant to the development of metalinguistic skills and expressive versus receptive MA measures. Results are extending the relationships between vocabulary and MA within an emergent bilingual literacy context.
Donald Shankweiler (University of Connecticut - Haskins Labs);Mencl,W.Einar; Braze,David; Pugh,Kenneth; Tabor,Whitney; Fulbright,Robert - Skill differences in convergence of auditory and visual information in sentence processing: an fMRI study
Purpose - Functional magnetic imaging was used to investigate the impact of literacy skills in young adults on the distribution of cerebral activity during comprehension of matched sentences in spoken and printed form. The aim was to test hypotheses about the convergence of speech and print streams during sentence processing, and to discover whether convergence is affected by the level of reading skill. Method - Brain activity was monitored while participants read or heard short sentences that were well-formed or contained anomalies of verb agreement or pragmatic content, and while they were engaged with a comprehension task unrelated to the anomalies. Regions of interest were defined based on locations of neural responses to the anomalies. Results - Each region was activated by both print and speech versions. The degree of speech-print convergence was measured for each participant. This quantity varied continuously as a function of skill level, with strongest correlation at the left inferior frontal region (IFG), where only the more highly skilled readers showed strong convergence. Conclusion - It is proposed that IFG, a region which is also implicated in dyslexia, is a binding site, active in integrating sentence-level abilities triggered by speech and print.
Purpose: Examined the diagnostic validity of two screening measures of risk as well as the degree to which adding a screening measure of reading comprehension enhanced the prediction of Oral Reading Fluency to outcomes of student reading performance on the state high stakes assessment. Method: Participants: 1,000 students in grades 3 (n= 401), 4 (n= 394) and 5 (n= 205) from a total of 6 elementary schools across 3 districts. Material: DIBELS Oral Reading Fluency Passages (DORF) obtained during the fall and winter. 4Sight Benchmark Assessment (successforall.net) obtained fall and winter. Pennsylvania System of School Assessment (PSSA) obtained in later winter. Analysis: Diagnostic Validity. Prediction of PSSA risk was tested by generating receiver-operating characteristic (ROC) curves. Differential Probability Analysis. A series of logistic regression analyses was used to determined relative prediction. Results: Results showed that the addition of a measure of reading comprehension (4Sight) to DORF enhanced the decision making process for identifying students at risk for reading difficulties, especially for those students at higher elementary grades. Although DORF alone showed a good level of prediction to the statewide assessment, the combination of the DORF plus 4Sight measures resulted consistently in the best predictive outcomes.
Purpose The purpose of this study is to examine executive functions performance in adult university students with AD/HD, dyslexia, and AD/HD with comorbid dyslexia. Method A series of analyses of variance with post-hoc tests of group differences was employed comparing performance of four groups of university students (23 AD/HD, 21 dyslexic, 10 AD/HD + dyslexia, and 19 controls) on executive functions (response inhibition, visual and verbal working memory, and set shifting), processing speed, vigilance, and phonological processing. Results The AD/HD group exhibited slight performance weaknesses pertaining to response inhibition, set shifting, vigilance, and processing speed measures. The dyslexic group showed significant deficits in phonological processing, verbal working memory, and processing speed measures. The AD/HD with comorbid dyslexia group exhibited significant impairments on all cognitive measures except visual working memory. Conclusion Study findings substantiate research indicating subtle weaknesses in executive functions in AD/HD adults (Wilcutt, Doyle, Nigg, Farone, & Pennington, 2005) and significant verbal working memory deficits in adult dyslexics (Swanson & Ashbaker, 2000). The findings also indicate the phonological processing deficits are specific to the dyslexic and AD/HD with comorbid dyslexic groups and dissociated from the AD/HD group (Wilcutt et al., 2001).
Lori Skibbe (University of Michigan); Kevin Grim; Frederick Morrison - Academic versus Summer Literacy Development: Evidence for Differential Effects of Instruction on Children's Literacy Achievements
Differences in literacy growth over the summer versus the school year were examined in order to isolate how varying amounts of instruction impacts children's literacy development. Children (n = 383) were tested individually twice each year starting when they entered the study in their first year of preschool until they finished second grade. Thus, children had the potential to be tested up to 10 times over the course of the five year study. Four subtests from the Woodcock-Johnson III Tests of Cognitive Abilities (WJ-III; Woodcock, McGrew, & Mather, 2001) were used: sound awareness, letter-word identification, passage comprehension, and picture vocabulary. Using a latent basis growth curve model, results indicated that differential patterns of literacy growth were observed during the summer and school year, starting as early as the second year of preschool, with children growing less than expected during summer months. Patterns of growth differed based on the grade examined, with the largest difference in expected versus actual growth occurring during the summer before kindergarten. Effects also varied dramatically in magnitude. Effects were greatest for decoding skills and reading comprehension and were less evident for phonological awareness and vocabulary. Educational implications and future directions will be discussed.
Purpose: The objective was to evaluate the impact of individual, computer-mediated guided oral reading sessions on participants' reading growth using a sample of adult prisoners. Method: 30 adult male offenders selected from the ABE classes in a large state prison participated. All were intermediate-level readers (G.E. 4.0-7.0) and participated in a 25 hour intervention project using software designed primarily to improve fluency. The software presented text on screen, using speech-recognition technology to provide assistance and trigger comprehension queries during readings and recorded progress and difficulties in the background. Pre- and post-test contrasts of participants' oral reading fluency and reading comprehension were used to evaluate the intervention's impact. Results: Relative efficacy of the approach was gauged through a historical control design, comparing participants' prior growth across outcomes from prior ABE coursework, where the median number of ABE course hours was 903 or more than 7 instructional quarters. At the end of the intervention, significantly large gains were found in reading comprehension and fluency using the Woodcock Reading Mastery Test and Tests of Adult Basic Education, respectively. Conclusions: Results showed fluency training supports improved comprehension among this challenging population and that speech recognition-enhanced software offers promising means of individual instructional support.
Purpose The current study investigated the amount of growth that can be expected in oral reading fluency (ORF) based upon initial ORF proficiency. Typically, students' growth rates are compared to average growth rates, but research has demonstrated that ORF growth in grades 2-6 is slower for the lowest and highest students, especially in lower grades (Silberglitt & Hintze, 2007). The current study extends the research to include first grade students while replicating analyses for grades 2 and 3, where the most significant ORF growth differences were found in a previous study (Silberglitt & Hintze, 2007). Method Data from 15,455 students were collected four times per year through Florida's Progress Monitoring & Reporting Network. For each grade, students were grouped by deciles based upon initial ORF performance. After individual growth curves were calculated, the average slope for each decile was compared to the 50th percentile to determine if growth was significantly different. Results ORF growth differences were significantly different for each decile in first grade whereas in second and third grades only the lowest and highest deciles were significantly different from the 50%ile. Conclusions Oral reading fluency is frequently used to assess the effectiveness of instruction and interventions, but these results suggest that the outcome criteria for effectiveness vary based upon ORF proficiency, especially in first grade. By understanding the expected growth rate based upon students' proficiency levels, more accurate evaluations of student progress and effectiveness of interventions can be determined. Results should also be considered for Response to intervention models where service decisions are based upon student outcomes.
This prospective study examined early L1 predictors of later L2 reading (word decoding, comprehension) and spelling skills by conducting a series of multiple regressions. Measures of L1 word decoding, spelling, reading comprehension, phonological awareness, receptive vocabulary, and listening comprehension administered in first through fifth grades were used as predictors of L2 reading (word decoding, comprehension) and spelling skills in high school. The best predictor of L2 decoding skill was a measure of L1 decoding, and the best predictors of L2 spelling were L1 spelling and L1 phonological awareness. The best predictor of L2 reading comprehension was a measure of L1 reading comprehension. When L2 word decoding skill replaced L1 word decoding as a predictor variable for L2 reading comprehension, results showed that L2 word decoding was an important predictor of L2 reading comprehension. The findings suggested that even several years after students learned to read and spell their L1, word decoding, spelling, and reading comprehension skills transfer from L1 to L2.
Purpose: This study investigates children's vowel spelling in multisyllabic words. It compares the spelling of overt vowels in stressed syllables with vowels in unstressed syllables that reduce to schwa. Prosodic sensitivity (the ability to detect linguistic stress patterns) and derivational morphological awareness (the ability to recognize and manipulate morphemes) are examined as possible contributors to vowel spelling. Method: Ninety children in grades 3, 5, and 7 will complete tasks measuring spelling, reading, and oral language proficiency, morphological awareness, verbal and nonverbal IQ, and short-term memory. A mixed-model ANOVA will compare the spelling of reduced and overt vowels at different grade levels. Hierarchical regression analyses will identify skills predictive of successful spelling. Results: I expect that at all grade levels, spelling accuracy will be higher for overt vowels than reduced vowels, and that an interaction will exist between vowel type and grade. Furthermore, I expect that prosodic sensitivity and morphological awareness will predict spelling success after controlling for age, reading ability, and IQ. Conclusions: Educators will benefit from an understanding of strategies that children employ when spelling; this knowledge can be applied to improve spelling curricula and to aid the early identification of children who are likely to experience spelling difficulties.
Purpose: This study examined relationships among two print-exposure measures, an Author Recognition Test (ART) and a self-report of reading habits (RSR), with component reading measures and two reading-comprehension measures. Our main questions were: 1) How highly would the two print exposure measures correlate with each other and various reading measures? 2) Which print exposure measure would best predict reading achievement? Method: Participants were 85 sixth graders from three demographically varying schools. Participants were individually tested on the component reading measures (PPVT and subtests from the Woodcock-Johnson), the RSR, and ART; their scores on a state-mandated reading test were also obtained. Results: The RSR and ART correlated significantly (r = .394, p < .01), but the ART had a much stronger, more consistent relationship with reading abilities than the RSR, which correlated only with vocabulary and comprehension, not word reading. Relationships between individual RSR items and reading abilities involved only fiction, not nonfiction, books. There were large, statistically significant differences between students with high vs. low passage comprehension in ART score but not total RSR score or self-reported pleasure reading. Conclusions: Checklist author recognition measures have a stronger relationship to reading achievement than does a self-report measure of reading habits.
Yi-Fen Su (Dept. of Educational Psychology and Counseling, National Taiwan Normal University)Chen, Ju-Ling - The Role of Phonetic and Orthographic Cues for Taiwanese Beginning Readers to Identify Chinese Regular and Irregular Compound Characters
In Chinese writing system, characters are the basic orthographic units, which correspond directly to morphemic meanings and syllables. Among modern Chinese characters, about 82 percent of the characters are compound characters, which contain two parts: a radical, which provides information about meaning, and a phonetic, which carries information about pronunciation. According to the consistency of pronunciation between compound characters and their phonetics, two types of compound characters are defined. One type has a regular pronunciation, and the other type has an irregular pronunciation. The purpose of this research is to investigate the role of phonetic and orthographic cues for Taiwanese beginning readers to identify Chinese regular and irregular compound characters. The subjects included three groups of average readers selected from the first, second, and third grades, and one group of second grade poor readers, whose decoding abilities were equivalent to the first grade average readers. A task developed by Khomsi (1985) was administered to each subject in order to know which cues they used to identify regular and irregular compound characters. The results showed that the phonetic cues played a more important role than the orthographic cues for beginning readers to identify regular compound characters, which pronunciations were consistent with their phonetics. However, regarding the irregular compound characters, which pronunciations were inconsistent with their phonetics, both the phonetic and orthographic cues played equally important roles for beginning readers to identify the characters. Keywords: Phonetic Cues, Orthographic Cues, Chinese Compound Characters, Pronunciation Regularity, Beginning Readers.
M. Kendra Sun-Alperin (University of Maryland, College Park);Sun-Alperin, M. Kendra; Wang, Min - Cross-language transfer of phonological and orthographic processing skills in Spanish-speaking children learning to read and spell in English
Purpose The current study aimed to determine whether Spanish L1 phonological and orthographic processing skills contributed a unique amount of variance to English L2 word reading and spelling, over and above these skills in English. We hypothesized that L1 phonological processing skills would contribute a significant amount of unique variance to L2 word reading and spelling, but that L1 orthographic processing skills would contribute only a limited amount of variance to L2 reading and spelling, given the difference in orthographic depth in the two languages. Method Eighty-nine Spanish-English bilingual children in grades 2 and 3 were given phonological and orthographic processing, reading, and spelling tasks in both Spanish and English. Hierarchical regression analyses were conducted to determine the contribution made by Spanish phonological and orthographic processing skills to English word reading and spelling over and above within-English variables. Results and Conclusions We found that cross-language phonological transfer occurs from Spanish to English reading and spelling, supporting previous research (Durgonoglu et al., 1993; Gottardo, 2002; Manis et al, 2004). However, Spanish orthographic processing only predicted English reading, not spelling. We suggest that this non-significant orthographic transfer to spelling could be due to several factors. First, if orthographic patterns are language specific (Durgunoglu, 2002), orthographic knowledge in an L1 may not benefit spelling in an L2. Also, English spelling is inherently more difficult than reading, due to the assymetrical relationship between reading and spelling (Kessler & Treiman, 2001).
Purpose Character learning is the central task for Chinese reading acquisition. The relations between naming, comprehension, and phonetic accessibility of characters are fundamental to understanding Chinese children$B!G(Js reading performance. Chinese characters not only contain varying degrees of phonetic information, but also represent the phonological and semantic cues separately. This study examined whether reading Chinese characters requires separate abilities for each of these aspects of reading, depending on the linguistic features of the characters being read. Method 836 children at fourth to sixth grades completed ten different Chinese character reading tasks. Confirmatory factor analysis was employed to test four a priori models in which abilities for naming and comprehension were crossed with phonetic and non-phonetic character reading abilities. The fit indices of these models were compared to determine which model fit the data best. Results The internal consistency coefficients were .61-.96. The model with naming and comprehension as distinct latent factors for character reading provided the best fit compared with the single general factor model and two models differentiating reading characters with or without useful phonetic cues. Conclusions The results suggest that reading characters of various phonetic accessibilities is driven by the two abilities of naming and comprehension.
Purpose: Frequency and consistency influence word and nonword reading in both adults and children. Ratings of frequency and consistency have been problematic in previous investigations, especially those exploring reading development, as accurate lexical statistics are difficult to obtain. To overcome this, the current series of studies used an artificial language learning paradigm to control exposure to the statistics of the language. Method: English-speaking adults learnt to read new words written in novel characters. Vowel pronunciations varied in frequency and were dependent on the surrounding consonants (consistency). Sensitivity to these variables was assessed by examining learning trajectories and performance on lexical decision and nonword reading tasks post-training. Results: Speed and proficiency of learning varied greatly. Despite this, consistency, frequency and interaction effects were observed both in training and testing. Also of interest was participants' poor lexical decision performance despite showing good generalisation in nonword reading at post-test. Conclusions: The paradigm made it possible to control people's exposure to the language. This made conclusions as to how the statistical properties of a language affect learning to read more certain. Future research will focus on individual difference analyses and examine how the addition of semantics influences performance, particularly on the lexical decision task.
Purpose: This study investigates the relationship between phonological awareness, reading and rhythmic sensitivity. It is the first to look systematically at non-linguistic rhythm, linguistic rhythm, phonological and literacy skills in the same sample of children. Method: Fifty elementary school children, both with and without reading difficulties took part in the study. Children were tested using standardized measures of phonological awareness (CTOPP), vocabulary (PPVT), reading and spelling (WRMT, GORT). In addition, children were assessed on their non-linguistic rhythm sensitivity (discriminating between sounds with different amplitude rise times) as well as their linguistic rhythm sensitivity (distinguishing between noun phrases with different stress patterns). Results: Initial analyses confirm associations between non-linguistic and linguistic rhythm task performance, with more complex relationships present between the rhythm tasks, phonological awareness and literacy skills. Profiles for children with and without reading difficulties will also be presented. Conclusions: These results provide preliminary evidence for a developmental pathway from non-linguistic rhythm skills to literacy. Because rhythm awareness can be tested even before children learn to read, increased inclusion of rhythm assessment in pre-literacy screening tools could prove important for successful early intervention.
In this study, we examined 5th grade readers' cognitive processes, and comprehension outcomes when assigned a specific reading purpose. Equal groups of good and struggling readers (N = 40) read expository texts in two conditions: general comprehension and for a specific purpose. The following questions were examined from think aloud statements, sentence highlighting tasks, and text retells: 1) Are readers aware of the reading purpose during and after reading? 2) Does having an assigned reading purpose influence the cognitive processes readers' engage in when reading? 3) Does the reading purpose influence the amount and quality of information in text retells? After reading, both groups accurately highlighted text sentences related to the reading purpose. When reading for a specific purpose, good readers increased and struggling readers decreased monitoring statements in think alouds. In both conditions, good readers used more inferential statements in text retells, received higher cohesion ratings, and had higher latent semantic analysis scores than struggling readers. No group or condition differences were observed in the number of inferential statements in think alouds or the number of idea units in text retells. Results suggest that awareness of reading purpose alone is not sufficient for improving 5th graders' comprehension outcomes.
Shelley Xiuli Tong (Division of speech and hearing science )McBride-Chang, Catherine - Cues of Radicals and Positional Regularity: Strategy Change in Coding Pseudo-Chinese Characters among Hong Kong Chinese children
A new measure of orthographic-semantic awareness, designed on the basis of structural and functional properties of radicals in pseudo-character Chinese compounds, was used to test children's understanding of structural and functional principles of radicals. In it, semantic and phonetic radicals cued the meanings and sounds, respectively, of compound characters in relatively stable positions within each character. Semantic and phonetic radicals were mutually exclusive--children could select only one or the other as a cue for each item. Among 198 kindergarten, 172 second-grade, and 165 fifth-grade Hong Kong Chinese children, strategies in use of semantic radicals, phonetic radicals, and positional regularity changed by grade, wilk Λ = .53, F (4, 1062) = 99.56, p < .001, partial η2 = .27: Older children more readily used semantic radical and positional regularity, rather than phonetic radicals. In a regression analysis, the contribution of semantic radical awareness to Chinese character reading increased with grade. In contrast, positional regularity awareness decreased by grade level. These variables explained 69% of the total variance in Chinese character recognition. Findings suggest a process of learning Chinese characters in which children progress from sound and positional cues in early grade levels to meaning cues in later grade levels.
Minna Torpa (University of Jyväskylä);Tolvanen, Asko; Poikkeus, Anna-Maija; Eklund, Kenneth; Lerkkanen, Marja-Kristiina; Leskinen,Esko; Lyytinen, Heikki - Reading Development Subtypes and Their Early Characteristics
Purpose. The study was designed to examine whether subgroups of developmental paths can be identified based on profiles of word recognition and reading comprehension in young children. Secondly, we studied what kind of early language and literacy skill profiles and reading experiences characterize children with differing reading development. Method. The present findings are drawn from the Jyväskylä Longitudinal Study of Dyslexia (JLD) in which approximately 100 children with familial risk of dyslexia and 100 control children have been followed from birth. In this paper we report data on the reading development of the JLD children and their classmates, a total of 1750 children from four measurement points during the first two school years. Results. The mixture modeling procedure resulted in five subtypes: (1) poor readers, (2) slow decoders, (3) poor comprehenders, (4) average readers, and (5) good readers. The children with familial risk for dyslexia performed on average at a lower level in all reading tasks than both their classmates and the controls and they were over-represented in slow decoders subtype. Conclusions. Differences between the subtypes were found in the early language and literacy skill development as well as in the reading experiences of the reading subtypes.
Purpose Fluent reading would seem to depend on (a) accurate, automatic word recognition that enables comprehension (Adams, 1990; Perfetti, 1985, 2007; Rayner, 1997; 2006), yet others argue that fluency involves more than rate or speed, and include prosody-knowledge of the distinctive rhythms and structure of written language-in their definitions of fluency (Kuhn & Stahl, 2003; Rasinski, 2003; Schreiber, 1991). In this paper, we present results from a longitudinal study in which the same 50 children were assessed at fourth and fifth grades in the following areas: isolated word recognition; oral reading accuracy, comprehension, and rate; and spelling. The purpose was to ascertain the relationships between prosody and other measures of reading fluency. Method and Results Prosody was measured by listened carefully to 150 oral reading protocols (fifty 4th graders reading 4th and 5th grade passages and those same students in 5th grade reading 5th grade passages), rating the readings for fluency and marking where the children paused when reading the text (inter-rater reliability = .94). Results include: high correlations between prosody ratings and pauses (-.90), oral reading accuracy (-.89), and reading rate (-.91). Students' ratings dropped and pauses increased as students read harder text, but the correlations remained high. The full set of results will be presented and developmental trends in the data discussed. In summary, our findings indicate that reading rate is an appropriate proxy for one aspect of reading fluency, prosody.
Purpose - The optimal viewing position (OVP) of words refers the fixation location in a word where the recognition time is minimized. Many studies consistently showed the OVP is a bit left to the word center. One of the sources to account for the asymmetry of the OVP is lexical constraint, since word-beginning contains less ambiguous information than word-ending. The current study investigated the influence of word combinability of constituent characters in determining the OVP of Chinese words. Method - Twenty-five Chinese students participated in a lexical decision task and their gaze positions were monitored by an eye-tracker. Five fixation positions in two-character words and two types of word combinability were manipulated. One set of words has large combinability in the first character but has small combinability in the second character (LC1) and the other set of words has the opposite pattern (LC2). Results - The results showed the OVP was on the first constituent characters for the LC1 words. However, the OVP was on the word center for the LC2 words. Conclusions - The findings suggest that word combinability is an important factor in determining the OVP. Moreover, the influence of word combinability in word-beginning is more prominent than that in word-ending.
Purpose This study examines representations of vowels in American urban kindergartners' phonemic segmentations and invented spellings in regard to possible influences (e.g., complexity of vowel sound, position of articulation, orthography). Method Data collected from 90 kindergartners in both fall and spring included word identification, letter name knowledge, phoneme segmentation, invented spelling, and an orthographic scoring of vowel pattern spellings. Segmentation and spelling of the same 24 words was scored developmentally using a system in which close articulatory proximity of substituted vowels received higher scores than more distantly positioned vowels. Results Overall, spelling was significantly stronger than segmentation. The unconventional choice of vowel graphemes in spellings was often associated with proximity of articulation. Consistent with Treiman (1993), analyses of vowel spellings revealed patterns in which letter names were used to represent phonemes (e.g., AM for aim), but also in which diphthongal tendencies were represented with extra letters (e.g., AEP for ape, SIY for sigh). Invented spelling was the strongest correlate of word reading in the fall, whereas the orthographic scale was strongest in the spring. Conclusions Findings have implications for teaching as well as for developing research instruments that are sensitive to both the phonological and orthographic knowledge of emergent readers.
Purpose: Investigate whether teachers vary in the extent to which they utilize culturally-responsive practices, and examine whether students whose teachers use culturally-responsive practices with greater frequency are more likely to demonstrate stronger reading skill and vocabulary growth than are students whose teachers use culturally-responsive practices less frequently. This randomized control field trial investigated the effect of a culturally-responsive vocabulary intervention on first grade students' vocabulary and early reading skills. Method: First grade students, randomly assigned within classrooms at ethnically and socioeconomically diverse schools, to treatment or alternate treatment conditions (n=2 schools, 7 teachers and 140 students). Student groups assigned to the treatment condition received small group vocabulary instruction using culturally-responsive teaching practices conducted by researchers. Student groups assigned to the alternate treatment condition received small group vocabulary instruction using business as usual teaching practices. Results: Preliminary results suggest that student groups in the treatment condition demonstrated stronger early reading skills (WJ-III Letter-word and passage comprehension) compared to student groups in the alternative treatment condition. Conclusions: Results suggest that explicit vocabulary instruction incorporating culturally-responsive teaching practices may be associated with stronger student outcomes. Implications for research and practice will be discussed.
Purpose: This longitudinal study examines the contributory role of home literacy environments (HLE) in fostering emergent literacy (EL) skills in the under-researched low-income Indian context. It explores a) the dimensionality of HLE for children learning to read and write in Hindi and b) the effects of HLE and parental schooling on growth in EL skills across the final year of preschool and EL skills at preschool year-end. Method: The 160 participating children from low-income communities in Mumbai, India attended Hindi preschools and were assessed at the beginning, mid-year, and year-end in expressive and receptive vocabulary, concepts about print, and knowledge of the Hindi alphasyllabary. Parent interviews provided information about the home supports for language and literacy development. Results: Cluster analysis revealed four HLE dimensions: storytelling, book-reading, parental teaching, and adult literacy activities. Individual growth modeling indicated that parental schooling had a positive effect on growth in children's knowledge of the Hindi alphasyllbry. Year-end EL skills were positively associated with storytelling and adult literacy activities and negatively associated with parental teaching controlling for the effects of relevant child and background characteristics. Conclusions: HLE is multidimensional and its effect on children's EL development demonstrates consistencies and inconsistencies with findings in other languages/contexts.
Purpose: The notion of visual attention span (VAS) was recently introduced to account for variability in developmental dyslexia. We have reported data showing that a VAS disorder relates to the poor performance of dyslexic children independently of their phonological skills. The aim of the current research was to provide evidence in support of a causal relationship between VAS and reading performance. Method: We conducted a longitudinal study examining the role of phoneme awareness and VAS in the development of reading and spelling abilities. A battery of cognitive and linguistic tasks was administered to 130 French children in kindergarten (T1) and was repeated twice at the end of the 1st (T2) and 2nd Grade. Results: Both phoneme awareness and VAS abilities at Time 1 related to reading and spelling performance at Time 2. VAS abilities predicted later reading/spelling performance independently of the child phoneme awareness and after control for other abilities known to influence reading acquisition. Conclusions: The results suggest that, in addition to phoneme awareness, VAS abilities might play a causal role in learning to read and spell. Like phoneme awareness, a VAS disorder might be responsible for the poor reading outcome of at least a subset of dyslexic children.
Purpose. European Portuguese is characterized by a strong vowel decoding ambiguity. The purpose of this study was to examine reading progression on vowel decoding. Method. We tested 140 children from 1rst to 4th grade. Children were asked to read frequent and rare bisyllabic words, stressed in the first and in the last syllable. Words varied in the consistency of the vocalic grapheme-to-phoneme conversion and in the syllabic complexity. Results. Results indicate that children's errors were mainly originated by the vocalic grapheme to phoneme decoding, whereas only a few of the errors was originated by consonants. An analysis of the errors within the vocalic decoding revealed that children perform worse on reading CV.CV than CVG.CV words, respectively containing simple oral vowels and diphtongs. Also, although pre-final syllable stress is the predominant stress pattern in Portuguese, children read the first vowel on CV.CVC words worse when they were stressed on the first syllable than when they were stressed on the last syllable, respectively, with an irregular grapheme-to-phoneme mapping and a regular grapheme-to-phoneme mapping. Conclusions. These results indicate that the vocalic inconsistency plays a more important role than the syllable structure on reading accuracy being that this difference is kept until the fourth grade.
Purpose. This study investigated how well a one-minute timed reading fluency measure (wcpm), used in so many classrooms, predicted overall reading performance of English Language Learners (ELLs) and their English-only (EO) counterparts. We also examined how rate, accuracy, expression, and comprehension interacted to produce fluent reading for both groups. Our goal was to elucidate the nature of ELLs' reading and to investigate the use of wcpm for screening these students. Method. Over 200 grade 2, 4, and 6 students, half of whom spoke a home language other than English, orally read one narrative and one expository grade level passage (rate, accuracy, wcpm, passage comprehension) and took the Woodcock Johnson II word identification and ITBS. ANOVA, SEM, and regression analyses were used. Results. Results indicated no significant differences on fluency-related measures between ELLs and EOs. However, there were significant differences on comprehension measures and the percent of false negatives, and substantial differences on the wcpm-comprehension correlations. Qualitative analyses of reading errors and comprehension responses support important differences between the groups. Conclusions. Fluency measures using a wcpm metric operate differently for ELLs and EOs. Furthermore, wcpm as a screening tool is likely to missing a large percent of students experiencing reading difficulty.
Purpose. A paper-and-pencil version of the lexical-decision (LD) procedure has been proposed as an alternative for testing the word identification skill of primary school students by means of reading aloud (e.g. van Bon et al., 2000). Some questions about the validity of this LD procedure have been answered in earlier studies: how is the LD score related to the naming score? do children achieve high scores through guessing? Relevant results will be summarized by way of background for a comparison of lexical decision and reading aloud as 'predictors' of text comprehension. Method. 120 Students from Dutch regular and special primary education (RPE viz. SPE) did the LD test, read the LD items aloud, and participated in a standardized test of reading comprehension that was appropriate for their level of word identification skill. Results. The LD score appeared to be a (slightly) better predictor of text comprehension than the naming score, and naming did not explain additional text comprehension variance, which relation was not different for RPE and SPE. Conclusions. This and other validity studies show that the LD procedure is at least as adequate as naming, perhaps even more adequate for children with speech production problems.
Spelling and reading are related in monolingual English speakers (Ehri, 2000; Frith, 1985). In addition, first language and second language reading are related in bilinguals (Durgunoglu, 2002). Therefore, L1 and L2 spelling should be related to each other. L1 and L2 spelling has been studied in across disparate orthographies, Chinese and English (Wang & Geva, 2003; Wang, Koda & Perfetti, 2003). However, relationships between spelling in two alphabetic languages with similar phonologies and many cognates in oral language but different orthographic depths (i.e. English & Norwegian) have rarely been tested. We examined L1 and L2 spelling in children who were English-Norwegian and Norwegian-English bilinguals in a longitudinal design. Errors specific to L1 were expected on specially designed spelling lists. Relationships between L1 and L2 spelling were examined. Results will be discussed in relation to L1 models of reading-spelling development and orthographic depth of the L1 and L2.
Purpose: In the present study, children's development of the print lexicon was examined by following a longitudinal design. Unidimensionality of the print lexicon was hypothesized. A second hypothesis was that the development of the print lexicon is predicted by children's word decoding and oral vocabulary. Method: Print lexicon and word decoding and oral vocabulary scores were longitudinally studied in a representative group of Dutch children in grade 3 to 6. Item Response Theory was used to evaluate the dimensionality of children's print lexicon. Structural equation modeling was used to uncover the relation between print lexicon and word decoding and oral vocabulary. Results: IRT results showed reading vocabulary to be unidimensional across different sets of items and grades. The order of complexity of word items could partly be predicted from word length as index of orthographic complexity and word frequency as index of word familiarity. Childrens's reading vocabulary scores could also be related to their word decoding and oral vocabulary at the start they learned to read. Conclusions: The results show that the children's lexical quality, i.e., the precision and redunancy of word representations can be seen as crucial predictor of the development of tehir print lexicon.
Purpose Recent research has demonstrated that sensitivity to prosodic information in speech may be related to reading skills. This study investigates whether sensitivity to prosodic information that is carried by morphological rules of English (e.g., Jarmulowicz, 2000) is predictive of word-level reading skills in 60 children aged 8 - 13. Method Sensitivity to morpho-prosodic information was measured by a task in which children heard a pseudoword (e.g., FROsure) and then had to choose which of two alternatives made a better derivation for that pseudoword (e.g., FROsureful or froSUREful). Responses were judged correct or incorrect according to what theory would predict, based on the type of suffix (neutral or stress-shifting). The dependent variable in this study was a composite variable measuring word level decoding skills, composed of the TOWRE Words and Nonwords and the Woodcock Word Identification subtest. Control measures included the Matrix Analogies Test and the PPVT to assess nonverbal and verbal ability, RAN numbers and Phoneme Elision from the CTOPP. Results Children's performance showed a high degree of correspondence with theoretical predictions, demonstrating that they are sensitive to the fact that morphological information (i.e., suffix type) drives stress assignment, even in nonwords. In a regression analysis using age-adjusted scores, we observed that, after verbal and nonverbal ability (accounting for 16% of variance) and phoneme elision and RAN (another 22%), morpho-prosodic sensitivity explained a unique 11% of variance in word reading ability. Conclusions Long words in English can be difficult for children to decode, but morphological information can assist in finding the appropriate stress assignment and therefore the correct pronunciation. Developing readers possess an implicit linguistic competence that can assist them in this task.
Purpose: The lexical restructuring theory hypothesizes that phonological awareness emerges as a result of growth of the mental lexicon. While children start out storing words as units, they proceed by learning words that overlap in sound with stored items leading to phonologically specified representations. To test the lexical restructuring hypothesis, the emergence of rhyme awareness in Dutch kindergarteners was studied as a function of phonological similarity and lexical status. Method: To examine the effect of phonological similarity children were asked to make rhyme judgments on auditory clue-target pairs. The presented targets were either (1) rhyming, (2) phonologically overlapping or (3) phonologically unrelated to the clues. To investigate lexical effects the words and pseudowords were used in two separate blocks. Results: In the real word condition only, we found children to respond much faster to rhyming word pairs as compared to unrelated pairs. Furthermore, in the word and pseudoword condition, overlapping pairs were processed more slowly and less accurate than rhyming and unrelated items. Conclusions: Faster responses to rhyming word targets seem to be the result of activation at the lexical level through prelexical activation of the clue words. This can be seen as support of the lexical restructuring hypothesis.
Purpose: Semantic transparency has been shown to be a critical factor in determining whether a compound word is decomposed into constituent morphemes. We examined the effect of semantic transparency in processing compound words among adult Chinese-English bilinguals. Method: A lexical decision task in English was administered. A factorial design of 2 (semantic transparency in English: transparent vs. opaque) X 2 (lexicality of the translated compounds in Chinese: real word vs. nonword) was employed. The English target items contain two free constituent morphemes which map onto the desired translations of the Chinese compounds. Combination of the translated constituents in English forms a translated compound in Chinese, which can be either a real word or nonword. The English transparent and opaque compounds are divided into two equal groups depending on the lexicality of their translated compounds in Chinese. Results: We predict a significant interaction between semantic transparency of English compound words and lexicality of the translated Chinese compounds. The semantic transparent compounds in English are affected more by the lexicality of Chinese compounds in comparison to the opaque counterparts. Data collection will be completed by early spring (February). Conclusion: Cross language activation of compound words in bilinguals is affected by semantic transparency of the target compounds.
This paper will describe the results of a multi-site study investigating the impact of small group reading tutoring on students identified as having emotional/behavioral disorders. The study was a part of the larger Vanderbilt Behavior Research Center Project which is focused on reducing problem behavior in schools. Two hundred forty students, kindergarten through fourth grade were randomly assigned to intervention and comparison groups during the 2006-2007 school year. Students were recruited from 3 metropolitan school districts (Nashville, TN, Minneapolis, MN, and Richmond, VA). Intervention group received small group tutoring 3 times per week for approximately 6 months. The tutoring program consisted of a modified version of the Horizons Direct Instruction program. Data regarding student performance, both reading and social behavior, will be presented. Summaries include issues in implementation, revisions and replication of procedures, and suggestions for future studies.
Purpose: Resilient readers show poor phonological processing skills but do not show a deficit in reading comprehension. In this study, resilient readers' orthographic and semantic processing skills were examined and compared with those of poor and proficient readers in order to examine possible mechanisms of compensation. Method: 21 resilient readers (below 25th percentile on word attack and above 45th percentile on passage comprehension), 22 proficient readers (above 45th percentile on both subtests) and 12 poor readers (below 25th percentile on both subtests) were recruited from a college population. Participants completed a battery of tasks including phoneme deletion, orthographic choice, semantic priming, verbal working memory, and pseudoword reading. Results: Resilient and poor readers were less accurate on phoneme deletion and pseudoword reading tasks than proficient readers. Resilient and poor readers were also less accurate on the orthographic choice task and showed a smaller verbal working memory span than proficient readers. Resilient readers showed greater semantic priming than either poor or proficient readers. Conclusion: Resilient readers showed no evidence of enhanced orthographic processing. Instead, resilient readers differed from other readers primarily on a semantic priming task, indicating that resilient readers may rely on semantic information as a form of compensation for poor phonological decoding.
Purpose: The study is the first phase of a short-term longitudinal study examining the development of prosodic skills (sensitivity to the tempo, rhythm and stress of language) in primary school children and its contribution to listening and reading comprehension. Prosodic development continues until at least 12 years of age, and primary-school children rely more on prosody and its cues to understand spoken language than adults. Moreover, written language with its paucity of prosodic information may place particular demands on prosodic sensitivity, above that required for spoken language. Method: Phase 1 of the research assesses a cross-sectional sample of children in grades 3 and 5, for prosodic skills at the syllable, word, and phrase level to examine the relationship to listening comprehension, reading and other linguistic skills, whilst accounting for phonological awareness and working memory. Results: Hierarchical regression analysis examined the prosodic development at this crucial stage and its larger role in both language and reading development. Conclusions: The examination of prosody (suprasegmental phonology) provides insight into the nature and loci of phonological representations known to be required for reading development and also shows relationships to other linguistic domains such as syntax, semantics and morphology which, in turn, also impact on both spoken and written language development.
Title: Unexpected Poor Comprehenders: A comparison across different measures of reading comprehension Purpose: Unexpected Poor Comprehenders (UPCs) demonstrate accurate and efficient word reading skill, but perform poorly on global measures of reading comprehension (RC). We investigated UPC and control group performance on (a) a range of RC measures, (b) predictors of component word and RC skills, and (c) educational outcomes. Method: Participants were 140 grade 10 students. Measures of word reading, word processing (phonological, orthographic, morphological), reading comprehension, vocabulary, nonverbal ability, and working memory were administered. Performance on school system RC tests at grades 3, 6, and 10 will also be collected. Groups were identified by regression, after controlling word reading (accuracy and rate), vocabulary, and non-verbal ability. Results: UPCs performed worse on a variety of RC measures than groups whose RC was 'as expected' or 'greater than expected' from control measures. UPCs also performed worse on measures of orthography, morphology, and working memory. Performance on school-system tests is being collected and will be analysed. In further analyses we will test alternative means of identifying UPCs and evaluate the effect of controlling other variables. Conclusions: The performance of UPCs is consistent across a range of reading comprehension measures. Further conclusions must await the additional analyses.
Purpose: To present the implications of the SERIOL model of orthographic processing for letter perceptibility patterns in the right and left visual fields, and to evaluate these predictions for dyslexic and unimpaired readers. Method: We perform a new analysis of published data on trigram identification, from an in-depth study of one dyslexic and seven unimpaired seventh-graders (Dubois, Lafaye de Micheaux, Noël & Valdois, 2007). The effect of a letter's position within a string (first versus third) is evaluated at retinal locations -3 to -1 (LVF) versus 1 to 3 (RVF). Results: Unimpaired readers show an asymmetric influence of position - within a given retinal location in the LVF, accuracy decreases with increasing position, while position has no effect in the RVF. In contrast, the dyslexic subject, MT, shows a symmetric pattern, with no effect of position in the LVF or RVF. At each LVF retinal location, the size of the positional effect separates MT from all unimpaired readers. Conclusions: This dyslexic subject has not learned normal orthographic processing. Further research is required on a large cohort of dyslexic and unimpaired subjects to evaluate the generalizability of this finding.
Several properties of a text affect how difficult it will be to understand. Readability statistics consider surface level features like word length, syntactic complexity, and word frequency. We explored whether the number of situation models in a text should be considered as an index of text difficulty. Following Barsalou (1999) we defined situation models as analog representations built up by the reactivation of perceptual information. We grouped a set of narrative passages into one of two categories: multiple situation models or single situation model. We hypothesized that the creation of a new situation model should be more difficult than the process of updating a pre-existing situation model. College students read six passages, which all contained several infrequently occurring words (e.g., pseudowords) whose meanings were inferable from the surrounding context. After reading each passage, reading comprehension was assessed by the quality of the definitions that participants provided for each pseudoword. Our analyses indicated that participants were significantly better at inferring the meaning of pseudowords when the text contains a single-situation model. This finding occurred despite the fact that single situation model passages had higher average readability statistics (e.g. Lexiles and Flesch-Kincaid) than passages with multiple situation models.
Shauna Wilson (Florida State University); Puranik, Cynthia; Lonigan, Christopher; Sims, Darcey; Hume, Laura - Preschool Phonological Awareness: Children's Developing Capacities by Age and SES-linked Risk Status
The negative effect of SES on phonological awareness (PA) has been well documented. However, McDowell, Lonigan, and Goldstein (2007) found that age moderates the relationship between SES and PA. They state that "young children may display lower PA because they lack capacity, exposure, or both, whereas in older children, a display of lower PA skills is more likely to represent capacity, signifying risk for reading problems." The specific aim of this study was to further explore the causes of lower PA in younger versus older children controlling for SES-linked risk status. We assessed the PA of 214 children from low-income families (80% African American, 54% male) at two time points; all children were enrolled in a PA intervention. Preliminary analyses suggest that both hypotheses offered by McDowell et al. were partially supported. Although the reliabilities of blending measures were consistently high, reliabilities for rhyming measures, and to some degree, elision measures, increased for all three age groups, with the most dramatic increase seen in the youngest children. These results suggest that capacity (i.e., age) and exposure (i.e., intervention) impact PA, which has implications for the age at which children should be taught PA as well as required intensity of intervention.
Purpose The first purpose of this study was to examine whether different measures of oral reading fluency relate differentially to reading comprehension in two groups of students: 1) students evidencing improvement in reading comprehension, and 2) students evidencing little improvement in reading comprehension. The second purpose of the study was to examine whether these relationships changed over time. Method Participants were 756 2nd grade students. Students were administered measures of nonword, real word, and connected text oral reading fluency, and reading comprehension at fall and spring time points. Reading comprehension groups were formed by conducting a median split on improvement on reading comprehension scores. Results Three hierarchical multiple regression analyses controlling for age were conducted with each measure of oral reading fluency predicting reading comprehension at the two time points. Analyses indicated that real word reading fluency accounted for the largest amount of unique variance at both time points and for reading comprehension groups (r2 = .68 and r2 = .63, improvement group; r2 = .64 and r2 = .52, little improvement group). Conclusions Results of this study suggest that real word oral reading fluency is the best predictor of reading comprehension. Results further suggest that fluent, automatic word recognition importantly influences reading comprehension abilities.
Purpose: While lexical access is the primary function ascribed to serial naming, little attention has been given to the phonological features of stimuli used in rapid automated picture naming tasks. The purpose of this study was to examine the effects of high vs. low rime neighborhood density on RAN time among individuals with and without developmental dyslexia. Based on the theory of lexical restructuring, high rime neighborhood density was expected to enhance speed of access to sub-lexical (phonological) information in a rapid automated picture naming task. Method: From DeCara & Goswami's (2202) lexical data base, two experimental RAN plates controlled for high and low rime neighborhood density were created. RAN times from 17 dyslexic subjects and 11 controls (aged 8-15) were compared. Results: For typically developing readers, high vs. low rime neighborhood density did not show significant effects on RAN speed. Among dyslexic readers, however, RAN speed was significantly faster (p<.008) for pictures from neighborhoods of high rime density compared to those from neighborhoods of low rime density. Conclusion: These findings suggest that for individuals with dyslexia, RAN is affected by the sub-lexical (phonological) features of stimulus items and that underspecified phonological representation impairs rapid lexical access.
The role of semantic knowledge has been relatively neglected in the reading disabilities (RD) field and its intervention, despite its increasingly known importance for fluent comprehension. The current study investigates the contributions of semantic knowledge not only to fluent text comprehension, but also to accuracy in single word reading in second and third grade children with RD. Our questions are: 1) Can an intervention that emphasizes the formation of high quality semantic representations and their explicit connections to other reading subskills (i.e., RAVE-O program) facilitate semantic depth of knowledge and semantic flexibility in children with RD? 2) Will such an intervention increase accuracy in trained and untrained words and fluent comprehension? 3) Will there be differential treatment effects on the latter measures for the RAVE-O intervention, compared to phonological-decoding interventions without semantic emphases? Results from multiple regression analyses indicate first, that children with RD can increase the depth and flexible use of their semantic knowledge, and second, that this semantic knowledge significantly contributes to word identification and fluent comprehension. Finally, differential treatment effects were found. Results will be explained within both the broad context of the importance of multidimensional intervention approaches and the more specific context of semantic emphases within interventions for reading fluency.
The purpose of this study was to examine which of the 3 formulae (R=DxL, R=D+L, R=DxL+D+L) derived from the Simple View of Reading would explain the relation between decoding, language comprehension and reading comprehension the most in Chinese. The other purpose was to investigate the fitness of the addition of word fluency in the Simple View of Reading.There were 931 students from the third to eighth grades in Taiwan participating in this study. They took a battery of tests including decoding, listening comprehension and reading comprehension tests. The data was analyzed by Pearson product-moment correlation, simple regression analysis, multiple regression analysis and moderated regression analysis.The results of this study were summarized as follows: 1)All of the three formulae accounted for the variance of reading comprehension of all students significantly. The results support R=DxL to be the best formulae of three.2)The correlations between decoding, listening comprehension and reading comprehension increase by grade.3)The correlation between fluency and reading comprehension is rather high, but fluency itself was not a significant independent predictor. Thus, fluency is unable to be a single variable in the formula of Simple View of Reading.According the findings, the limitations of this study and the recommendations to further research and practical implementations were made.
Purpose This study tired to develop a Chinese Readability Formula to provide parents, teachers, educators, and researchers an objective and convenient way to get difficulty information of a text. Methods This study created a database which includes all articles in the Chinese textbook from the elementary level and used for teaching reading from different publishers. Then, a computer program was developed for calculating the text characteristics of an article, such as the average of word frequency, the average length of a sentence, the length of an article…and so on. Several regression analyses were used to get regression formulas, which are the readability formulas, and these regression models were used to identify the text difficulty level of each article. Results The results showed that the text characteristics influencing the Chinese readability can be classified into three categories: the number of strokes, the word (character) frequency, and the length of the text. Among several regression models, the results showed that the accuracy of the best readability formula to provide the text difficulty information was 79%. Conclusion The prediction of the readability formula for grade 4 and 5 are more accurate than other grades. Although the accuracy is not perfect, this formula does provide a simple way to measure the text readability in Chinese.
Zohreh Yaghoub Zadeh (Canadian Council on Learning)Farnia, Fataneh; Chan, Eric; Thompson, Terri; Cohen, Nancy; Ungerleider, Charles - Phonological Awareness and Reading Comprehension: Meta Analyses of Cross-Sectional, Longitudinal, and Intervention Studies
We report the results of three meta-analyses that included cross-sectional (n=12), longitudinal (n=18), and intervention (n=12) studies that examined the relationship between phonological awareness and reading comprehension between1995 to 2007. We used Comprehensive Meta-Analysis version2 to analyze the data. Correlations were used as the effect size measure for cross-sectional and longitudinal studies for both independent and dependent variables were continuous. Hedge's g was used as the effect size for the intervention studies. The effect sizes associated with cross-sectional (r=.46), longitudinal (r=.42), and intervention (Hedges' g=.45; which is equivalent to r=.23) studies were then compared. Analysis of Variance indicated that the effect sizes across three research design are significantly different (F (2) = 5.42, p<.01). Tukey Post Hoc indicated that the effect size of intervention studies is significantly lower than those of cross-sectional (p<.01) and longitudinal (p<.05) studies. Furthermore, there was no significant difference between the effect sizes associated with longitudinal and cross-sectional studies. Q test and I2 values indicated that the effect sizes were highly heterogeneous within cross-sectional and longitudinal studies but not for the intervention studies. Fail-safe Ns suggest that the results are highly unlikely to be affected by publication bias. The implication of these findings will be discussed.
Purpose: We investigated whether marked differences between the phonological structures of Mandarin and English influence pre-literate bilingual children's development of syllable, onset-phoneme and rime awareness. Method: A 60-item English task requiring syllable deletion, onset-phoneme and rime report, was administered twice to two groups of 4-5 year-old bilingual children (n=40 Mandarin L1-English L2; n=49 English L1-Mandarin L2), with a 5-month interval between T1 and T2, during their first year in an English-medium kindergarten. Results: There were unexpected floor effects for rime awareness at T1 and T2 for both groups, but a 2 (Mandarin L1/English L1) x 2 (syllable/onset-phoneme) x 2 (T1/T2) mixed ANOVA revealed a significant 3-way interaction. At T1, planned comparisons showed that English L1 children performed significantly better at both syllable and onset-phoneme levels than their Mandarin L1 classmates. By T2, the Mandarin L1 group's performance was equivalent for syllable deletion, but remained significantly weaker for onset-phoneme awareness. Conclusion: For pre-literate bilingual children, the course of phonological awareness development is related to the phonological structure of their first language. This has implications for ESL children's reading and spelling acquisition.
Purpose: Phonological awareness (PA) and vocabulary skills are widely assumed to be essential for reading acquisition and strong predictors for success. We addressed the question of the nature of the underlying cognitive resources, such as phonological memory (PM) and working memory (WM) capacity that should be available to pre-readers for achieving success in reading. We hypothesized that there is a strong correlation between PM and WM capacity and different aspects of PA skills, as well as vocabulary. We also hypothesized that this correlation should be seen regardless of language of the input, i.e. transparent vs. opaque language. Method: 20 English- and 20 Croatian-speaking typical language pre-readers (mean age 5;3) participated in this study. All children were administered Nonword repetition task (NWR), measure of WM capacity, tasks addressing all components of PA, and PPVT as vocabulary measure. Results: Our results indicate significant interactions between the measures of PM, WM capacity and different aspects of PA, as well as vocabulary, regardless of language of exposure. Summary: This finding indicates that the underlying cognitive processes required for reading are unified and independent of language of input. This line of research is essential for application to the disordered population at risk for reading failure.
Purpose Due to the pictorial features of Chinese words, it is important for children to recognize the visual features of words (visual processing) and correctly map the visuals to the sounds (paired associate learning). However, their abilities in paired associate learning have not been explored. This research was aimed to investigate whether Chinese children with developmental dyslexia were deficiency of paired associate learning skills. Method 22 children with Chinese developmental dyslexia and 22 children without dyslexia participated in this study. Children of the two groups were matched according to their ages and intelligent scores. They were from third to fifth grade of an elementary school in urban area of southern China. Both groups were asked to associate non-sense visual shapes with randomized spoken monosyllables. This task required children to activate and retain a temporarily a visual and a verbal presentation, and to form a new association between the visual and verbal presentation. Results A paired-t test showed a significant difference between the dyslexia group and the control group. Conclusions Chinese children with dyslexia are short of the association skills, which are particularly important in learning rote associations between the sound and visuals of Chinese characters.