A three-year longitudinal study of the academic performance of students enrolled in 21st-Century Community Learning Center afterschool programs

Afterschool programs have changed and adapted in many ways since their inception. By the 1960s they moved from the purview of social programs into public education. The introduction of federally funded 21st-Century Community Learning Centers (21st-CCLC) in 1998 escalated the demand for afterschool programs to improve student academic performance and subsequently increased accountability measures.;This quasi-experimental study of a 3-year 21st-CCLC program in three sites (n = 539) analyzed changes in classroom grades and standardized test scores of regular attendees by program site, grade level, gender, and ethnicity to determine achievement. Student transiency, differing annual standardized tests, and other factors limit the data available for analysis. Dependent-t tests were used to measure change in grades and the Wilcoxon Signed Ranks was used to measure changes in ranks of standardized test scores. In the final year, two different standardized tests were administered so the Kruskal-Wallis H test was used to measure changes in ranks.;During each of the three years, without exception, all statistically significant changes (p < .05) as measured by classroom grades were negative. Conversely, all statistically significant changes as measured by standardized test scores were positive. There were no statistically significant classroom math grades changes in the first and second year, but in the third year, males, females and fourth graders had statistically significant declines.;In the second year, the all students category, females, and one site had statistically significant (p < .05) improvements on standardized reading test scores. In the third year, all students and fourth graders had statistically significant improvements in reading and females had similar improvements in math standardized test scores. Due to disparities in data reporting between sample and population standardized test scores (NCE scores vs. percentiles), there were no appropriate statistical methods for comparing the performance of program participants to non-participants in the three sites. A narrative describes the differences between sample and population academic performance as measured by standardized test scores.;The findings of this study differ somewhat from the summary findings of 21st-CCLC first year grantee evaluations by Mathematica. Additional longitudinal studies by experimental design with larger samples and comparison groups are suggested.