This study evaluated the hypothesis that schools in Connecticut's Alliance Districts (lowest-performing districts) with higher-quality school improvement plans (SIPs) would have higher levels of student achievement. An exploratory research question evaluated whether SIPs predicted achievement of particular subgroups. SIPs were obtained and scored for 108 schools. Consistent with the hypothesis, SIP total scores correlated positively with student school-level student achievement. However, in multiple regression analysis controlling for demographic factors and prior school-level student achievement, SIP total score's coefficient did not reach the 0.05 level of significance. None of the subgroups with reasonably large N's (Hispanic, students eligible for free or reduced price lunch, and "high needs" students) had significant regression coefficients even though all three subgroups had small, yet significant correlations with SIP total score. We recommend that schools develop high-quality SIPs that focus efforts directly on subgroup performance to close current gaps in achievement.