Abstract
The techniques of meta-analysis were used to provide a quantitative synthesis of the findings from 1077 studies investigating the nature of learning disability. Across four domains (linguistic, achievement, neuropsychological, and social/behavior) that included 38 variables, studies were summarized to quantify the level of differentiation between comparisons of learning-disabled and normal groups. In general, about 75% of learning-disabled subjects could be differentiated clearly from normal subjects, but no particular area emerged as a primary source of learning disability. Across 34 variables, only about 40% of the variance in learning disability functioning was explained. The greatest group differentiation was in the linguistic domain, but it did not account for a significantly greater proportion of the variance. Across the remaining domain and variables, the learning-disabled subjects revealed about the same level of differentiation from normal subjects. It was concluded that learning disability is a complex and multivariate phenomenon which involves a number of component deficits that all make an important contribution. Consequently, a multiple-syndrome paradigm is necessary to provide a comprehensive description of learning disability.