Over the past 15 years, increased attention has been directed at social skills and their relationship to learning disabilities. Using the methods of meta-analysis, this investigation explores the nature of social skill deficits among students with learning disabilities. Across 152 studies, quantitative synthesis shows that, on average, about 75% of students with learning disabilities manifest social skill deficits that distinguish them from comparison samples. Approximately the same level of group differentiation is found across different raters (teachers, peers, self) and across most dimensions of social competence. Although social skill deficits appear to be an integral part of the learning disability experience, a number of questions about the relationship between learning disability and social skill deficits remain unanswered. Until these questions are answered, social skill deficits are best viewed as one among many elements of the learning disability constellation, and no significant definitional changes related to social skill deficits appear warranted.

In this article, we analyze critically available definitions of learning disability (LD). The general problem of definition is discussed first followed by a discussion of individual LD definitions from the earliest to the latest. We conclude that LD definitions fail to provide substantive insight into the nature of the condition. The reasons for this failure are discussed in relation to the nature of definition and the difficulties in providing operational definitions of LD that are meaningful and significant. Finally, means for resolving the problem of definition are discussed.

Many programs designed for children and youth with emotional or behavioral disorders (EBD) include a social skill training component. Using quantitative methods of meta-analysis, the findings from 35 studies investigating the effects of social skill interventions for students with EBD were synthesized. The pooled mean effect size (ES) was 0.199, from which the average student with EBD would be expected to gain a modest eight percentile ranks on outcome measures after participating in a social skill training program. Studies were further grouped and analyzed according to different variables (e.g., similarities of the intervention, participants, and assessment procedures). Slightly greater ESs were found for interventions that focused on teaching and measuring specific social skills (e.g., cooperating, or social problem solving) compared to more global interventions. Several pertinent issues for reviewing the results of this research synthesis are addressed.

Prevalence of children with emotional or behavioral disorders (EBD) is a critical component in the discussion of underidentification of children served in special education. This discussion has previously focused almost exclusively on point prevalence or the number of children with EBD presumably needing services at any single point in time. Cumulative prevalence, on the other hand, is the number of children who have had EBD at some point in their lives before high school graduation. In the authors’ review of both types of prevalence, they found that estimates of the latter far exceed those of the former, significantly highlighting the service gap that exists between prevalence estimates and special education identification. Even when point prevalence is limited just to children with moderate or severe disorder, special education identification in the emotional disturbance category appears restricted to less than the bottom tenth of all children in need. Implications for special education are discussed, including issues around underidentification, misidentification, underservice, and related issues concerning children with EBD.

Sixty-four single-subject studies examining the effectiveness of social skills interventions with students with emotional or behavioral problems were included in this synthesis. The results of quantitative synthesis procedures using percentage of nonoverlapping data suggest that social skills interventions have limited empirical support for their overall effectiveness. Implications for future social skills research and quantitative analysis methodology are discussed.