This study examines the relationship between IQ and yearly academic growth rate in reading and mathematics for low income children in the primary grades involved in the Direct Instruction Follow Through program (DIFT). Low income children, from 20 communities in the United States, who engaged in the DIFT program between 1969 and 1977 for either 3 years (grades 1 through 3) or 4 years (kindergarten through grade 3), participated in the study. Upon entering the program, in either kindergarten or first grade, students were tested on the Wide Range Achievement Test (WRAT) and the Slosson Intelligence Test (SIT). Each spring students were tested on the WRAT, SIT, and (beginning in grade 1) the Metropolitan Achievement Test (MAT). For the purposes of the longitudinal analyses of variance, children's entry IQ score was used to classify them into one of six IQ blocks (70 and lower, 71-90, 91-100, 111-130, 131 and above), and then a mixed (split-plot) analysis of variance design was used to analyze results with one between-groups factor (IQ block) and one within-groups factor (time of test). The major finding of this study was that little, if any, relationship existed between entry IQ and yearly learning rate. Generally, the students entering the program at higher cognitive skill levels finished third grade at higher levels, but the growth rates (or amounts of materials learned) were not significantly different between IQ blocks. (Author/MP)