This longitudinal study examined two groups of children, boys in one sample and girls in another, to assess whether programs such as Head Start and Follow Through produce lasting measureable effects. Low-income children were examined separately from middle-income children and, within the low-income sample, a separate analysis was made for low-income black children (who comprise the largest single group in the final longitudinal sample). Effects of an extensive intervention program were seen upon the children's academic performance in mathematics, general information scores and Peabody Picture Vocabulary Test IQ Scores. Retesting, four to five years following completion of the intervention, showed these children still performing higher than nonintervention children. Reasons for sex differences are discussed; results from interviews with the children and analyses of their school attendance and grades are presented. (Author/BF)