Impact of Environmental Factors on Components of Reading and Dyslexia, The

Impact of Environmental Factors on Components of Reading and Dyslexia, The

Samuelsson, Stefan

Data from 123 male adults were analyzed to estimate environmental influences on components of literacy skills and to explore the impact of environmental factors in different approaches to define reading difficulty. Literacy skills were decomposed into general cognitive function, reading comprehension, spelling, word reading, and phonological ability. Environmental factors examined were related to home conditions, school conditions, and literacy environment. Results suggest that there is a substantial social-cultural bias in the delineation of literacy skills and in the definitions of reading disabilities. Results also suggest that phonological ability constitutes the only measure relatively unaffected by environmental influences. This study has brought forward a new argument for emphasizing phonological deficits as the core component in defining dyslexia.

In literacy research, the impact of socioeconomic status, home environment, and early print exposure have been thoroughly documented (cf. Molfese, Modgline, & Molfese, 2003; OECD, 2000; Senechal & LeFevre, 2002). In these studies, no attempts have been made to disentangle subcomponents of literacy and the relative impact of environmental factors on these components.

However, behavioral genetics have paid attention to the decomposition of literacy skills in attempts to examine the relative influence of genetic and environmental factors on reading subskills. For example, it has been demonstrated that environmental influences is stronger for reading comprehension as compared to spelling and word recognition, that environmental influences are less salient for phoneme awareness and phonological decoding compared to word recognition, and that shared environment has a stronger impact on intelligence compared to word reading ability (Gayan & Olson, 2001, 2003; Olson, Forsberg, & Wise, 1994). These findings suggest that environmental influences vary across components of literacy skills. The analytical tools used in behavioral genetics, however, are only designed to give information on environmental influences in general terms. Thus, the environment is not specified, characterized, or measured in any sense. One aim of this study was not only to decompose literacy skills, but also to provide a more detailed account of environmental factors that might be expected to influence literacy. The major aspects of environmental influences examined in this study were related to home conditions, school conditions, and literacy environment. In order to maximize the variance of the environmental factors, we selected a sample of prison inmates where the expected variance is particularly large. Sweden is a rather homogeneous society, and it is difficult to find groups of adult with a sufficient variance in socioeconomic and cultural background. However, a representative sample of prison inmates is expected to be heterogeneous in terms of socioeconomic background. Each of these environmental components was assessed by a large number of indicators. Literacy skills were decomposed into general cognitive function (intelligence), reading comprehension, spelling, word reading, and phonological ability. Based on behavioral genetics research (e.g. Gayan & Olson, 2003), it is hypothesized that more specific functions are less vulnerable to environmental influences than more general cognitive functions. Thus, it is expected that the encapsulated module of phonological ability should be most resistant to environmental influences (Stanovich & Siegel, 1994), whereas reading comprehension is more influenced by environmental factors (Olson, Forsberg, & Wise, 1994).

Evidence for the importance of phonological ability in reading has been provided by an abundance of research among children (Fletcher, Shaywitz, Shankweiler, Katz, Liberman, Stuebing, Francis, Fowler, & Shaywitz, 1994; Foorman, Francis, Fletcher, & Lynn, 1996; Liberman & Shankweiler, 1985; Stanovich, & Siegel, 1994). Perhaps more important, the long-term stability of the phonological module has been documented in studies of adults (Bruck, 1990, 1992; Byrne & Ledez, 1983; Decker, 1989; Elbro, Nielsen, & Petersen, 1994; Felton, Naylor, & Wood, 1990; Gottardo, Siegel, & Stanovich, 1997; Paulesu, Frith, Snowling, Gallagher, Morion, Frackowiak, & Frith, 1996; Pennington, Van Orden, Smith, Green, & Haith, 1990; Snowling, Nation, Moxham, Gallagher, & Frith, 1997). For example, Svensson and Jacobson (2003) have demonstrated that phonological skills as indicated by nonword reading accuracy and speed is a very persistent trait. The correlation between non-word reading from the age of nine to 19 amounted to .73 for the reading disabled group, whereas the correlation concerning word recognition across the corresponding period was relatively modest (.48).

A further aim was to focus on the definition of reading disability in the light of the decomposition of environment and literacy skills. The traditional way of defining reading disability or developmental dyslexia relates to a discrepancy between general intelligence and reading ability (American Psychiatric Association, 1994; World Health Organization, 1993). More recently, this approach has been challenged (for a review see Sternberg & Grigorenko, 2002). Instead, the error-prone and dysfluent word identification has been emphasized as the marker symptom of reading disability. This problem is assumed to be based on phonological deficits (e.g., Frith, 1999; Høien, & Lundberg, 2000; Lyon, 1995; Stanovich, 1996). A more ecologically oriented alternative to define reading disability departs from the lower end of a normal continuum of literacy skills. In this approach, literacy skills are often measured by composite scores of reading comprehension, word recognition, and spelling. The purpose of a definition may vary. In some cases, the purpose is to provide specific accommodation for reading disabled individuals (e.g., extended testing time, technical support), obtain guidelines for remedial intervention, give the basis for causal interpretations, or to specify a useful definition for research purposes. Whatever definition is chosen, one should be aware of the purpose of the definition as well as the potential environmental influences on the condition defined. One purpose of the present study is to explore to what extent there might be a social-cultural bias introduced in different approaches to defining reading disabilities.

METHOD

PARTICIPANTS

The study was conducted on adults (with 67 percent of the adults being prison inmates) with the purpose of including a larger range of environmental factors with a possible impact on literacy skills. The adults participating in this study were part of a larger study of reading and writing skills among prison inmates conducted in the southeast region of Sweden (Samuelsson, Herkner, & Lundberg, 2003). Participants were 82 male prison inmates and 41 additional male adults, originally recruited as a comparison group. The mean age of the inmates was 34 years (SD = 10.9) and 38 years (SD = 11.1) of the nonincarcerated. To ensure a large range of different backgrounds, the inmates were recruited from different prisons ranging from high security to local open units. The average term of punishments was 4.9 years with 14 percent being sentenced for murder, 20 percent for robbery, 21 percent for assault, 16 percent for drug offences, 11 percent for sexual assault, and 18 percent for economic crimes. The participants were all of Swedish ancestry and had Swedish as their first language. Although adult controls were selected randomly, we sought out individuals where the formal level of education, on average, was expected to be relatively low. For these reasons, most controls were recruited among male students (n = 21) attending municipal adult education (Komvux in Swedish) with the purpose of improving their educational level. An additional 20 adults were recruited either from one rescue center or from the staff working at the prisons. This selection procedure was assumed to minimize differences in environmental conditions between prison inmates and adult controls, and potential differences in literacy skills. In fact, there were no differences observed on factors such as socioeconomic status, educational level, and reading habits between the groups, nor any differences between the groups on measures of reading and writing skills (Samuelsson, Herkner, & Lundberg, 2003). It should be noted, however, that there were still large variances obtained within the sample on environmental factors such as educational level (5 to 22 years of education), socioeconomic status (5 to 18 points on a 20-point scale), and truancy in school (0 to 180 days per year on average). This wide variation is exactly what is needed to study the impact of environmental factors on literacy skills.

MEASURES

Intelligence. General cognitive ability was estimated by the Raven’s Standard Progressive Matrices (Raven, Court, & Raven, 1983) combined with a standardized vocabulary test (Johansson, 1992). The vocabulary test contained 40 target words, each followed by four words of which one was a synonym. These two tasks were used to construct a composite score of general cognitive ability.

Reading Comprehension. Reading comprehension was measured by subtests developed by the International Association for the Evaluation of Educational Achievement (Elley, 1992). The subtests selected included six different text passages, each followed by four to six multiple-choice questions. The length of each text passage varied between 100 to 500 words and there were a total of 30 questions. The task was not timed with the purpose of emphasizing reading comprehension rather than word reading efficiency.

Spelling. The spelling test consisted of 50 target words (Johansson, 1992). In the test, each target word was pronounced by the experimenter in a sentence and then pronounced again in isolation to dictate the word to be spelled.

Word Reading. Word reading ability was measured with four different tasks: two word-chain tests, a word-reading task, and an orthographic choice task. In the word-chain test, three words of concrete nouns were presented on paper without inter-word spaces. There were 80 such chains of words in each test and the task was to segment by pencil marks as many chains as possible into their constituent words within three minutes (Jacobson, 1994). The words were semantically unrelated in the first version of the test and semantically related in the second version. The number of word-chains correctly segmented in each test was recorded as a dependent variable.

In the word-reading test, the participants were asked to read aloud 48 Swedish words, varying between three and seven letters in length and in complexity regarding spelling/sound correspondences (cf. KOAS; a test battery designed by Høien and Lundberg, 2000). The words were presented on a computer screen one at a time and the number of correctly read words per minute was used as a measure of word reading efficiency.

The final word reading ability test employed was an orthographic choice test in which participants were asked to decide which letter string out of two was a correctly spelled word (see Olson, Forsberg, Wise, & Rack, 1994). There were a total of 36 pairs of words, and the nonwords in each pair were pseudo-homophones. To emphasize word reading efficiency, the number of correct decisions per minute was taken as a dependent measure.

Phonological Ability Three different tasks were used to measure phonological ability: a nonword reading test, a phonological choice test, and a spoonerism task. The nonword reading test comprised 48 nonwords, varying between three and seven letters in length. The non words were constructed to avoid orthographic similarities with real Swedish words (cf. KOAS; H0ien & Lundberg, 2000). The non words were presented one at a time on a computer screen and the number of correctly read nonwords per minute was taken as a dependent variable. All phonological tasks basically involve a mixture of various cognitive and linguistic subcomponents. In the case of non word reading, the phonological demands are mixed with letter identification and the utilization of orthographic structures. However, by deliberately avoiding orthographic similarities with real Swedish words, the significance of phonological processing should be more salient. Also, letter knowledge among adults is expected to be at the ceiling level, and thus what remains of the performance variance can mainly be attributed to phonological processes. The same line of arguments can be applied with regard to the phonological choice task.

In the phonological choice test, 36 pairs of nonwords were presented one at a time on a computer screen and the participants were asked to decide which nonword sounded like a real Swedish word. Again, the number of correct responses per minute was used as a dependent measure.

The spoonerism task is assumed to measure phonological ability without contamination by reading experience (Perin, 1983). In the test, the participants are asked to perform a transposition of the initial sounds of two orally presented words and then repeat the transposed nonwords back aloud. A rather pure measure of phonological awareness is given by the task demand of explicitly manipulating the constituent phonemes of words in the spoonerism task. Again, the number of correct transpositions per minute was registered as dependent variable.

ENVIRONMENTAL FACTORS

A total of 19 environmental variables were employed in the study, covering three main areas: home background, literacy environment, and school condition (statistical details in constructing these three factors are provided in the Results section).

Home Condition. Home background was determined using different questionnaires covering socioeconomic status (Graffar, 1956), parental rearing behavior (Arrindell, Perris, Perris, Eisemann, Van der Ende, & vonKnorring, 1986), number of siblings, and early literacy socialization. Socioeconomic status of the family in childhood was based on four questions focusing on occupational criteria, level of education, main source of family income, and quality of dwelling. Each question was assessed on a 5-point scale, providing a range of scores from 4 to 20 points. Parental rearing styles were assessed by a self-report inventory focusing on rejection and emotional warmth as two main aspects of parental rearing behavior (Arrindel, et al., 1986). The questionnaire comprised 43 questions (25 concerning rejection and 18 focusing on emotional warmth) and the participants rated both their father and mother with respect to all questions. Number of siblings as well as early literacy socialization were captured by questions included in a structured interview performed individually with each participant.

Literacy Environment. Literacy environment was indexed by a composite of book reading habits, number of books as home in childhood, cultural activity related to theater and arts, and educational level. These aspects of literacy environment were operationalized by five different questions included in the interview and by an author recognition test. Book reading and cultural habits were measured by three separate questions measuring, on a 5-point scale, the frequency with which participants normally read books or visit cultural events (each question was assessed on a 5-point scale ranging from never to everyday). Book reading habits were also indirectly assessed by an author recognition test in which participants were asked to indicate whether they were familiar with the name of 20 well-established authors by putting a check mark next to the name (cf. Cunningham & Stanovich, 1997). The list of authors also included foils, that is, names of people not known as popular writers. Furthermore, number of books at home as a child was assessed using a 6-point scale (range from 0 to 10 to more than 200 books) and educational level was determined by the total number of years attending different educations.

School Condition. The assessment of school condition was also based on six variables. Truancy (days per year on average), number of different schools and teachers, amount of special education (hours per week on average), homework commitment (a 5-points scale ranging from always to never), and adjusted arrangements for schooling (hours per week on average) were all used to indicate school condition.

RESULTS

COMPOSITE SCORES OF ENVIRONMENTAL FACTORS AND LITERACY SKILLS

Environmental Factors. In order to bring a simple structure into the wealth of background data in the present study, a series of principal component analyses were conducted. We distinguished between home background, literacy environment, and school condition. The home background factor included parental rearing styles, socioeconomic status, number of siblings, and early informal literacy socialization. More than 50 percent of the total variance was explained by one principal component based on seven indicators of home characteristics. Literacy environment was indicated by six variables such as book reading habits, number of books at home, cultural habits related to theater and arts, and educational level. One principal component explained 44.3 percent of the total variance. School condition was indicated by six variables including number of different schools and teachers, amount of special education, truancy, homework commitment, and adjusted arrangements for schooling. The total amount of explained variance was 42.1 percent. In all principal component analyses, only one major component with eigen-values greater than 1 could be extracted.

To establish comprehensive scales without outliers for each of these three background conditions, factor scores were computed. The factor scores were then categorized into 20 score groups. The choice of 20 score groups was, of course, arbitrary, but was considered practically useful (the correlations between this categorized scale and the original factor scores were, in fact, all above .95, demonstrating that the data reduction did not introduce any real change). The intercorrelations between the three factors varied from .20 to .31, indicating that the factors had only 4 percent to 9 percent of common variance. The three factors were used as the primary independent variables in the further analysis of how variance in components of literacy skills can be explained.

Abilities and Literacy Skills. Five major dependent variables were examined. Intelligence was indicated by scores on the Raven’s progressive matrices and scores on the vocabulary test. These scores were combined on the basis of a principal component analysis where the factor scores were grouped into 20 score categories. The principal component explained 76 percent of the variance. Reading comprehension had only one indicator, the scores on the reading comprehension test. Spelling also had only one indicator; the spelling-to-dictation test. Word reading skills had four indicators; scores on the word-chain tests (semantically unrelated words, semantically related words), word reading efficiency (number of correct per minute), and scores on the orthographic choice test. The word reading factor explained 65.8 percent of the variance. The factor scores were again grouped into 20 score categories. Phonological abilities were indicated by nonword reading, phonological choice, and the spoonerism task, all variables measured as number of correct per minute. The principal component explained 73.0 percent of the variance. Table I displays the correlations between variables of intelligence and components of literacy skills. Although all correlations reached significance, we will especially draw attention to the rather impressive correlation obtained between word reading efficiency and phonological ability (i.e., r = 0.78).

ENVIRONMENTAL FACTORS AND COMPONENT SKILLS IN LITERACY

Each of the five factors of abilities or literacy skills was used as a dependent variable in five separate multiple regression analyses with the three background factors as independent variables. The main results of the analyses are summarized in table II.

The second and third columns clearly indicate that the amount of explained variance by experiential and environmental factors is reduced as one moves from broad aspects of cognitive functioning to a more specific linguistic module (phonological ability) with very limited malleability. This trend is further highlighted in figure 1. Notable is that the environmental factors explained 26 percent of the variance of a composite measure of literacy based on reading comprehension, spelling, and word reading ability.

As the phonological choice task and the nonword reading task involve a component of reading, there is a risk that our measure of phonological ability is confounded. Thus, a reanalysis was made with spoonerism as a pure language measure and a word-attack factor based on nonword reading and the phonological choice task. The common principal component explained 82 percent of the variance, and the correlation between nonword reading and phonological choice was .64. The correlation between the spoonerism task and the composite score of word attack was .66. The revised multiple regressions with the environmental factors as independent variables yielded R^sup 2^ = .29 for word attack as dependent variable and .27 for spoonerism. Thus, no real change of the results could be observed in the reanalysis.

THE PHENOTYPE OF DYSLEXIA

Despite the variation in clinical practice concerning the definition of dyslexia, there seems to be a rather general consensus that dyslexia is a constitutional dysfunction with a strong genetic influence, implying that experiential and environmental factors might play a minor role. From the results so far, it seems as if the use of intelligence or more general aspects of literacy such as reading comprehension in delineating the phenotype of dyslexia, tends to invite the influence of environmental factors to operate. Thus, one would expect that an inclusion of these factors might obscure the picture of dyslexia.

As a first step, we will study the participants with low performance (one standard deviation below the mean) on each of the ability or skill factors. As these skills are continuously distributed in the population, no natural cutoff point is available for defining dyslexia. One procedure to categorize individuals as dyslexies is, then, to set up a rather arbitrary statistical cutoff point on the skill continuum.

However, the question is which continuum should be used? If the individuals within the critical region of a specific skill differ significantly from the remaining individuals on the background factors, one would not expect that particular skill to be a good candidate for defining the phenotype of dyslexia. Obviously, the individuals in the critical region suffer from an environmental deficiency seriously confounded with their literacy problems.

In figure 2, the comparisons between low performing participants and the remaining group on the background factors for each of the five abilities or skills are presented. In addition, a composite measure of reading comprehension, spelling, and word reading is examined. This combined literacy index has often been used as a basis for delineating individuals with reading difficulties.

Figure 2 clearly indicates that the difference in background factors is large, and significant between the groups for intelligence, reading comprehension, literacy, and spelling. Significant differences in all three environmental conditions were also found between the groups for the composite score of literacy skill. For word reading, only literacy environment yielded a significant difference between the groups. Phonological ability did not display any difference between the individuals in the critical low-performing region and the remaining group. Thus, the reasonable conclusion is that phonological ability meets the criterion of being resistant to environmental influences.

THE DISCREPANCY CRITERION OF DYSLEXIA

One traditionally applied procedure is to define dyslexia in terms of the discrepancy between the actual reading performance and the performance predicted from intelligence scores. In this study, we established a literacy index based on a composite of reading comprehension, spelling, and word reading. All individuals with a literacy score one standard deviation below the mean were selected as poor readers. Some of these individuals fulfilled a discrepancy criterion whereas others did not. The literacy index was regressed on the intelligence score. Those individuals who had a literacy achievement below one standard deviation from the regression line fulfilled the criterion of discrepancy dyslexia (N = 9). Individuals with low literacy scores and with a performance level within the range predicted from the intelligence scores are labelled as garden-variety poor readers (N = 21). This garden-variety group was compared on the background factors with the discrepancy group and with remaining individuals denoted “normal” readers.

Table III shows comparisons between dyslexics, garden-variety poor readers, and normal readers on the background factors. The ANOVAs demonstrated that the garden-variety poor readers had significantly lower scores on school conditions and literacy environment than the other groups. The difference for home condition was close to significance, F(2,120) = 2.68, p = .07. There were no differences between the dyslexic discrepancy group and the “normal” readers on any background factor. Thus, if a discrepancy criterion is applied excluding the garden-variety poor readers from the dyslexia category, then one should be aware of the fact that a social-cultural bias is introduced in the phenotypic delineation of dyslexia.

DISCUSSION

This study examined the extent to which environmental factors influence different aspects of literacy skills. The purpose was also to examine the extent to which different phenotypic criteria for reading disabilities interact differently with environmental factors. Three main findings have been reported. First, environmental influences decreased gradually as one moves from broad aspects of cognitive functioning and literacy skill to a more specific phonological module. Second, large and significant differences in environmental factors were observed between low performing participants (i.e., defined as the lower end of a normal continuum for each component of literacy skill) and the remaining group for intelligence, reading comprehension, and spelling, as well as for a composite score of literacy skill. Importantly, the same division for phonological ability did not reveal any differences on environmental factors. Third, garden-variety poor readers displayed significantly lower scores on environmental conditions compared to IQ-discrepancy defined dyslexies and normal readers.

Together, these findings suggest that environmental factors influence most measures used to establish literacy skill, and that there might be a substantial social-cultural bias in the phenotypic delineation of literacy skills and in the definitions of reading disabilities. More precisely, tasks emphasizing reading comprehension, spelling, or composite scores based on several reading and spelling components invite environmental factors to play a substantial role in establishing levels of literacy skill. Between 17 percent and 26 percent of the variance in spelling, reading comprehension, and composite measures of literacy skills (i.e., including reading comprehension, spelling, and word decoding skills) were accounted for by environmental factors. It should be noted that some part of the correlation between reading and environmental factors might well be genetically mediated (Plomin, DeFries, & Loehlin, 1977; Scarr & McCartney, 1983). However, this does not necessarily affect the relative comparisons of environmental influences on various subcomponents of literacy.

The data also indicate that definitions based on either cutoff scores at the lower end of a normal continuum of literacy skills or on IQ-achievement discrepancies are all substantially confounded by environmental factors. For research purposes, these findings point to the problem that environmental factors interact with different behavioral and cognitive phenotypes of reading disability. For clinical purposes, the findings point to the risk that environmental factors might have a substantial impact on diagnoses of reading disabilities, guiding decisions made by school authorities about the amount and type of remediation efforts.

One way around this problem would be to emphasize specific phenotypic components largely unaffected by environmental influences and also strongly related to literacy skill, especially word recognition skills. Our findings indicate that one such approach would be to emphasize phonological processing skill as the main underlying component associated with word decoding deficits, and also in the definition of reading disabilities. Despite overwhelming support that poor phonological processing skills are the core and defining deficit in dyslexia among both children and adults, there are still few studies defining reading disabilities mainly based on phonological processing deficits.

References

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed. rev.). Washington, DC: American Psychiatric Press.

Arrindell, W. A., Perris, C., Perris, H., Eisemann, M., Van der Ende, J., & von Knorring, L. (1986). Cross-national invariance of dimensions of parental rearing behavior: Comparison of psychometric data of Swedish depressives and healthy subjects with Dutch target ratings on the EMBU. Bristish Journal of Psychiatry, 148, 305-309.

Bruck, M. (1990). Word decoding skills in adults with childhood diagnosis of dyslexia. Developmental Psychology, 26, 439-454.

Bruck, M. (1992). Persistance of dyslexics” phonological awareness deficits. Developmental Psychology, 28, 874-886.

Byrne, B., & Ledez, J. (1983). Phonological awareness in reading-disabled adults. Australian Journal of Psychology, 35, 185-197.

Cunningham, A. E., & Stanovich, K. E. (1997). Early reading acquisition and its relation to reading experience and ability 10 years later. Developmental Psychology, 33, 934-945.

Decker, S. N. (1989). Cognitive processing rates among disabled and normal reading young adults: A nine year follow-up study. Reading and Writing: An Interdisciplinary Journal, 2, 123-134.

Elbro, C., Nielsen, I., & Petersen, D. K. (1994). Dyslexia in adults: Evidence for deficit in non-word reading and in the phonological representation of lexical items. Annals of Dyslexia, 44, 205-226.

Elley, W. B. (1992). How in the world do students read? The Hauge: International Association for the Evaluation of Educational Achievement.

Felton, R. H., Naylor, C. E., & Wood, F. B. (1990). Neuropsychological profile of adult dyslexies. Brain and Language, 39, 485-197.

Fletcher, J. M., Shaywitz, S. E., Shankweiler, D. P., Katz, L., Liberman, I. Y., Stuebing, K. K., Francis, D. J., Fowler, A. E., & Shaywitz, B. A. (1994). Cognitive profiles of reading disability: Comparison of discrepancy and low achievement definitions. Journal of Educational Psychology, 86, 6-23.

Foorman, B. R., Francis, D. J., Fletcher, J. M., & Lynn, A. (1996). Relation of phonological and orthographic processing to early reading: Comparing two approaches to regression-based, reading-level-matched designs. Journal of Educational Psychology, 88, 639-652.

Frith, U. (1999). Paradoxes in the definition of dyslexia. Dyslexia, 5, 192-214.

Gayan, J., & Olson, R. K. (2001). Genetic and environmental influences on orthographic and phonological skills in children with reading disabilities. Developmental Neuropsychology, 20, 483-507.

Gayan, J., & Olson, R. K. (2003). Genetic and environmental influences on individual differences in printed word recognition. Journal of Experimental Child Psychology, 84, 97-123.

Gottardo, A., Siegel, L. S., & Stanovich, K. E. (1997). The assessment of adults with reading disabilities: What can we learn from experimental tasks? Journal of Research in Reading, 20, 42-54.

Graffar, M. (1956). Une metode de classification sociale d’echantillons de population (A method of social classification of sample populations). Courier, 6, 8.

Høien, T., & Lundberg, I. (2000). Dyslexia: From theory to intervention. Dordrecht: Kluwer Academic Publishers.

Jacobson, C. (1994). Word recognition index (WRI) as a quick screening marker of dyslexia. The Irish Journal of Psychology, 16, 260-266.

Johansson, M-G. (1992). Diagnostic tests in reading and writing for students in grade 7 to grade 10. Stockholm: Swedish Psychological Corporation.

Liberman, I. Y., & Shankweiler, D. (1985). Phonology and the problems of learning to read and write. Remedial and Special Education, 6, 8-17.

Lyon, G. R. (1995). Toward a definition of dyslexia. Annals of Dyslexia, 45, 3-27.

Molfese, V. J., Modglin, A., & Molfese, D. L. (2003). The role of environment in the development of reading skills: A longitudinal study of preschool and school-age measures. Journal of Learning Disabilities, 36, 59-67.

OECD. (2000). Literacy in the information age: Final report of the international adult literacy survey. Paris: OECD.

Olson, R. K., Forsberg, H., & Wise, B. (1994). Genes, environment, and the development of orthographic skills. In V. W. Berninger (Ed.). The varieties of orthographic knowledge I: Theoretical and developmental issues (pp. 27-71). Dordrecht, The Netherlands: Kluwer Academic Publishers.

Olson, R. K., Forsberg, H., Wise, B., & Rack, J. (1994). Measurement of word recognition, orthographic, and phonological skills. In G. R. Lyon (Ed.). Frames of reference for the assessment of learning disabilities (pp. 243-277). Baltimore: Paul H. Brookes Publishing Company.

Paulesu, E., Frith, U., Snowling, M., Gallagher, A., Morton, J., Frackowiak, R. S. J., & Frith, C. D. (1996). Is developmental dyslexia a disconnection syndrome? Evidence from PET scanning. Brain, 119, 143-157.

Pennington, B. F., Van Orden, G. C., Smith, S. D., Green, P. A., & Haith, M. M. (1990). Phonological processing skills and deficits in adult dyslexics. Child Development, 61, 1753-1778.

Perin, D. (1983). Phonemic segmentation and spelling. British Journal of Psychology, 74, 1129-1144.

Plomin, R., DeFries, J. C., & Loehlin, J. I. (1977). Genotype-environment interaction and correlation in the analysis of human variation. Psychological Bulletin, 84, 309-322.

Raven, J. C., Court, J. H., & Raven, J. (1983). Manual for Raven’s progressive matrices and vocabulary scales, part three, section 7, research. London: H. K. Lewis.

Samuelsson, S., Herkner, B., & Lundberg, I. (2003). Reading and writing difficulties among prison inmates: A matter of experiential factors rather than dyslexic problems. Scientific Studies of Reading, 7, 53-73.

Scarr, S., & McCartney, K. (1983). How people make their own environments: A theory of geneotype-environment effects. Child Development, 54, 424-435.

Senechal, M., & LeFevre, J. A. (2002). Parental involvement in the development of children’s reading skill: A five-year longitudinal study. Child Development, 73, 445-460.

Snowling, M., Nation, K., Moxham, P., Gallagher, A., & Frith, U. (1997). Phonological processing skills of dyslexic students in higher education: A preliminary report. Journal of Research in Reading, 29, 31-41.

Stanovich, K. E. (1996). Toward a more inclusive definition of dyslexia. Dyslexia, 2, 154-166.

Stanovich, K. E., & Siegel, L. S. (1994). Phenotypic performance profile of children with reading disabilities: A regression based test of the phonological-core difference model. Journal of Educational Psychology, 86, 24-53.

Sternberg, R. J., & Grigorenko, E. L. (2002). Difference scores in the identification of children with learning disabilities: It’s time to use a different method. Journal of School Psychology, 40, 65-83.

Svensson, I., & Jacobson, C. (2003). How persistent are phonological difficulties? A longitudinal study of reading disabled children. Manuscript submitted for publication.

World Health Organization. (1993). The ICD-10 classification of mental and behavioral disorders: Clinical descriptions and diagnostic guidelines. Geneva: World Health Organization.

Manuscript received February 3, 2003.

Accepted May 29, 2003.

Stefan Samuelsson

Centre for Reading Research, Stavanger University College, Norway Department of Behavioral Sciences, Linkoping University, Sweden

Ingvar Lundberg

Department of Psychology, Goteborg University, Sweden

ACKNOWLEDGMENTS

This research was supported by a grant to Stefan Samuelsson and Ingvar Lundberg from the Swedish Council for Social Research (SFR: F0022/1999).

Address correspondence to: Stefan Samuelsson, Stavanger University College, Centre for Reading Research, P.O. Box 8002, 4068 Stavanger, Norway. Phone: 47-5183 3240; Fax: 47-5183 3250. E-mail: Stefan.samuelsson@slf.his.no.

Copyright International Dyslexia Association 2003

Provided by ProQuest Information and Learning Company. All rights Reserved