In order to determine whether a program of black literature read aloud to a group of black fourth-grade children would have a measurable effect on posttest scores of two self-concept measures, an experimental group of 23 black children was exposed to 36 sessions of children's literature about black figures--historical and fictional--who were strong, positive characters with whom the children could identify. A second group of 23 black children listened to children's literature without black characters, while a third group had no literature read to them. The children in the three groups were nearly equal in sex, age, grade, and intelligence, and all were pretested and posttested. Analysis of data shows that such a direct approach does promote a stronger self-concept and that the type of test used to measure self-concept may strongly influence the results. (JM)