Correlates of achievement in learning disabled children

This study was designed as a preliminary investigation seeking to discover some of the correlates of scholastic achievements in learning disabled children and to determine the effects of educational therapy as it is practiced in this region. The investigation was made ex post facto employing data developed by psychologists and psychiatrists who had performed the psychological appraisals and by school personnel who routinely administered the SRA Achievement Series to the fourth, fifth, and seventh grades in September and/ or October each year. A multiple regression program was used to determine which combination of variables best accounted for criterion variance when remediation was given and when it was not administered. The individual hypotheses were tested by correlational analysis by first computing the achievement test scores for each student and then correcting for age variance at the time of achievement testing. Analysis indicated that the experimental and control groups were not significantly different except with regard to age at time of testing. (The experimental group averaged being 8.6 months older than the control group.) This factor was therefore controlled statistically to remove this source of variance. The first hypothesis stated that experimental children would excel their learning disabled peers who had not received educational therapy in all criterion areas-language arts, arithmetic, and reading. This proved to be true only with regard to language arts scores. Arithmetic and reading scores were not significantly different for the two groups. An expectation that all girls would exceed boys in academic achievements, and that those attending remedial classes would require shorter periods of remediation, was not confirmed by these data. With age controlled, the data indicated no significant differences between males and females. A third hypothesis anticipated that children who were placed in remediation when younger would require shorter periods of remediation. The data indicated, however, that the converse was true for the subjects of this study. A corollary expectation that the younger children would later prove to have higher achievement scores was not supported either. The younger children did relatively less well on arithmetic and did not differ significantly on language arts and reading from these whose remedial period had started at later ages. The expectation that children who required shorter periods of remediation would later have better achievement scores proved to be accurate in this case for language arts and arithmetic. Reading scores tended in the same direction but did not reach the .05 level of confidence. The proposition that a child's academic abilities would accelerate as the period between the end of remediation and SRA testing increased was not supported by the data. It was anticipated that VIQ scores were significantly related to the achievement criteria for both groups. This proved correct with the single exception of an insignicant correlation between VIQ and reading achievement for the control group. For the remedial group, PIQ scores were significantly related to all criterion measures. For the control group, PIQ was not significantly related to any criteria. The expectation that VIQ would evidence a stronger degree of relationship to the criteria than PIQ was met only for the experimental group for language arts and for the control group for language arts and arithmetic. In all other instances, PIQ showed a stronger degree of relationship with the criteria. The proposal that FSIQ scores would be related to achievement scores was met in every instance but one. There was not a significant degree of correlation between FSIQ and reading achievement scores for the control group. Multiple regression equations for the remedial group revealed age at SRA testing, FSIQ, and length of remedial period to account for a significant amount of the variance of language arts scores, A combination of age at time of SRA testing, PIQ, length of remedial period, VIQ, and FSIQ was most potent in accounting for variance of arithmetic scores for the experimental group. For reading scores for this group, only PIQ proved to be a significant function. For the control group, multiple regression equations showed VIQ as the only variable with a significant function, in language arts score variance. For arithmetic scores for this group, a combination of age at SRA, VIQ, and time between diagnosis and SRA testing proved to account for a significant proportion of the variance in this criterion. For reading scores of the control group, no individual predictor variables had a significant beta weight, but a combination of age at SRA, PIQ, and FSIQ accounted for a significant amount of the variance.in this criterion area. Given the totality of expectations set forth in the hypotheses for this study, results were largely negative. Many of the findings made for small, select groups of children did not hold true for this sample drawn from the broadbased population attending public schools. It seems now imperative for psychologists and educators to re-examine more closely some of the assumptions based on popular beliefs which have failed to find support in this study.