Educators know that quality early childhood intervention programs servicing children at risk for early school failure have immediate or short-term positive effects. However, there are different opinions on the persistence of long-term program effects on participants' academic performances. This study compared the performance on standardized tests at grades one through three of children who participated in South Carolina child development programs for four-year-olds in the 1995-96 school year with that of their nonparticipant peers. The study then compared the students within the program to determine who among them benefited the most. All children (9,977) in the cohort group and a randomly selected comparable group of nonparticipants (7,889) were tracked to the third grade in the 1999-2000 school year. Findings indicated that program participants scored significantly higher on first grade and third grade standardized tests in both English language arts (ELA) and mathematics than did similar nonparticipants. Asian, Caucasian, and female students benefited significantly more from the program than did others. Although participants scored higher than nonparticipants on the second-grade standardized test, the difference was not large enough to be statistically significant. Comparisons between full-day and half-day program participants showed no significant differences on later academic performances. The estimated overall program effect size was 0.2. (Author)