The Effect of Research Methodology on Desegregation-Achievement Studies: A Meta-Analysis

A meta-analysis of 323 samples of black students experiencing desegregation,derived from 93 research studies, finds two methological factors correlated with the measured effect of desegregation on academic achievement. First,studies of students who received only a partial treatment (i.e., began desegregation after they had completed one or more years of segregated schooling) find weaker treatment effects. Second, the type of control group used was correlated with the measured effect of desegregation: studies which used random-assignment experimental designs found significantly stronger treatment effects, while studies which compared black achievement growth either to that of white students or to achievement test norms found much weaker (usually negative) treatment effects. We conclude that desegregation does enhance black achievement by about .3 standard deviations, but this effect seems to occur only in the earliest primary grades.