This paper describes validity of quality improvement system (QIS) performance measures and longitudinal change over four years in Oklahoma 21st Century Community Learning Centers (CCLC) programs. In general, our findings indicate that the Oklahoma Afterschool Improvement Process is performing in accordance with its purposes: using accurate performance data to incentivize improvement in the quality of services. Findings for the reliability and validity of the measures indicate that all of the 23 measures demonstrated acceptable levels of reliability. There is evidence for construct validity at each time point and factorial invariance across time points. Findings for performance improvement indicated that nearly all measures incrementally improved during a four year (2010-2013) period, while a subset demonstrated statistically significant growth. For nearly all measures, lower-performing sites at the baseline year (2010-2011) improved most. A subset of models demonstrated statistically significant effects. The indicator with the largest increase over four years was Targeting At-Risk Students, suggesting that even though the students served became more challenging, service quality was also generally improving. Findings for intervention fidelity indicate that higher fidelity of Youth Program Quality Intervention (YPQI) implementation is positively associated with growth on nearly all performance measures at over half of all year-to-year time increments, in line with the YPQI theory of change.