This study attempted to determine whether attendance at a traditional preschool resulted in significant cognitive gains for young children over a 3-year period. The experimental group consisted of 59 children at four preschools. Each was matched with a nonattender on such variables as age, sex, intelligence, and family characteristics. The mean Stanford-Binet IQ was approximately 119. Testing was done before the experimental group began preschool, at the end of their preschool year, at the end of kindergarten, and at the end of first grade. The Stanford-Binet (S-B) and the Peabody Picture Vocabulary Test (PPVT) were the main instruments used in the testing. These were supplemented by tests on reading, mathematics, spelling, and Piagetian concept attainment. The results indicated: (1) significant differences between the two groups on the S-B at the end of preschool, kindergarten, and first grade, with the experimental mean higher than the control; (2) significant PPVT increases by the experimental group but not by the control group; (3) no significant difference between groups on the Piagetian test of operativity at the end of kindergarten, but a significant difference in favor of experimental children by the end of first grade; and (4) no significant differences between the two groups on tests of school performance. (JMB)