The First Grade Intervention Study
In the first major study pursing these objectives (Vellutino et al., 1996), we tracked literacy development of middle to upper middle class children, who were identified in mid-first grade as either struggling readers or normally achieving readers, for a period extending from their entry into kindergarten (before reader group membership was determined) through the end of fourth grade. These two groups represented sub-samples from the larger population of children we initially assessed (n = 1407) at the beginning of their kindergarten year. In order to evaluate the possible contribution of home and pre-school literacy experiences to early reading achievement, foundational literacy skills such as knowledge of the alphabet, knowledge of print concepts, print awareness, and phonological awareness were assessed at or shortly before the children entered kindergarten. Cognitive abilities underlying reading ability were also assessed in kindergarten and again in first and third grade: language and language-based abilities such as vocabulary knowledge, syntactic competence, rapid naming, and verbal memory; visual abilities such as visual memory and visual-spatial analysis and synthesis; and verbal and nonverbal intelligence. The possibility that early reading difficulties in some children may be the result of deficiencies in kindergarten literacy instruction was assessed through systematic observations of the kindergarten language arts programs in all participating schools (Scanlon & Vellutino, 1996; 1997). Finally, daily one-to-one tutoring was provided for most of the reading impaired children for up to two school terms (mid-first grade to mid-second grade), depending on the child's progress. We were especially interested in assessing possible differences between the cognitive profiles of impaired readers who were found to be difficult to remediate and impaired readers who were found to be readily remediated, relative to normally achieving readers.
In keeping with results obtained in previous intervention studies (Clay, 1985; Iversen & Tunmer, 1993; Pinnell, 1989; Wasik & Slavin, 1993), we expected that daily one-to-one tutoring would facilitate the acquisition of at least average level reading skills in most of the impaired readers. We also expected that the cognitive profiles of the readily remediated readers would be more like the cognitive profiles of the normally achieving readers than like those of the difficult to remediate readers. These expectations were generally confirmed.
First, the number of children who might have been classified as disabled readers prior to implementation of the intervention program represented approximately 9% of the first grade population from which our sample was selected (n = 1284 after attrition). However, the number of such children was appreciably reduced after only one semester of remediation and represented approximately 3% of the population when all children scoring below the 30th percentile on measures of basic word level skills were included in the count. This figure dropped to 1.5% when only the lowest performing and most difficult to remediate children were included in the count (i.e. those scoring below the 15th percentile). Moreover, the children who were found to be the most difficult to remediate generally performed below the level of both the normally-achieving readers and the children who were found to be readily remediated on tests of phonologically-based skills such as phonological awareness, verbal memory and fluency in name retrieval. In contrast, the readily remediated children, in many cases, performed at levels comparable to those of the normal readers on the same measures and could no longer be classified as disabled readers. And, although there were substantial reductions in the percent of tutored children who scored at least within the average range (relative to normative standards) on measures of basic word level skills after tutoring was discontinued, such reductions occurred primarily among children who showed the least amount of initial growth in response to remediation (Vellutino & Scanlon, 2002; Vellutino, Scanlon, & Jaccard, 2003).
The possibility that early reading achievement may be due, in part, to the quality of the child's home and pre-school literacy experiences was given some support by the finding that all children identified as problem readers in mid-first grade were found to be deficient in emergent literacy skills such as letter identification and phonological awareness at the beginning of kindergarten. The possibility that early reading achievement may be due, in part, to the quality of early literacy instruction was given some support by the finding that a child was less likely to be identified as a poor reader if he or she had been exposed to a comprehensive and balanced language arts program in kindergarten, that included both word level and text processing activities in the instructional program (Scanlon & Vellutino, 1996, 1997).
Finally, we found that the tests of intelligence did not differentiate the impaired readers who were difficult to remediate and the impaired readers who were readily remediated, nor did they differentiate children in the tutored groups and normal readers of average intelligence. At the same time, normal readers of average intelligence and normal readers of above average intelligence did not differ appreciably on measures of basic word level skills (i.e. word identification and phonological decoding). In addition, IQ-achievement discrepancy scores were not significantly correlated with growth in reading (Vellutino et al, 2000). These findings add to the growing body of evidence questioning the validity of the IQ-achievement discrepancy to define specific reading disability (Aaron, 1997; Fletcher et al, 1994; Francis, Shaywitz, Steubing, & Shaywitz, 1996; Siegel, 1988, 1989; Stanovich & Siegel, 1994).
We concluded, from the results obtained in this study, that early reading difficulties in most impaired readers can be successfully remediated and that experiential and instructional deficits are more likely to be the primary cause of such difficulties than are basic cognitive deficits of biological origin. However, we also concluded that reading problems in some children may well be due primarily to basic cognitive deficits, especially phonological deficits that impede the child's ability to acquire functional use of the alphabetic code. Thus, in the interest of more effectively distinguishing between these two populations, we suggested that the use of the IQ-achievement discrepancy and other psychometric/exclusionary criteria for diagnosing specific reading disability may be usefully supplanted by response to remedial intervention as a "first-cut diagnostic" and that categorical classification, when necessary, be deferred until such diagnostic intervention has been implemented.
However, results from the kindergarten assessment and classroom observation components of our study also led us to conclude that children's pre-first grade literacy experiences and instruction may be critically important determinants of early reading achievement. Thus, it seemed that a logical and potentially useful extension of our first grade intervention study would be to evaluate the utility of identifying children at risk for early reading difficulties on entry into kindergarten and implementing intervention in kindergarten to prevent long-term reading difficulties. We were also interested in further evaluating the utility of using remedial intervention to aid in distinguishing between cognitive versus experiential factors as primary causes of reading difficulties. To accomplish these objectives, we conducted a second longitudinal study that incorporated both kindergarten and first grade intervention components to prevent long-term reading difficulties in kindergarten children found to be at risk for early reading difficulties. We discuss initial findings from this study in the remainder of this paper.
Previous Page | Next Page
Introduction | Kindergarten Intervention
