As standards and accountability have become increasingly prominent features of the educational landscape, educators have relied more on remedial programs such as summer school and grade retention to help low-achieving students meet minimum academic standards. Yet the evidence on the effectiveness of such programs is mixed, and prior research suffers from selection bias. However, recent school reform efforts in Chicago provide an opportunity to examine the causal impact of these remedial education programs. In 1996, the Chicago Public Schools instituted an accountability policy that tied summer school and promotional decisions to performance on standardized tests, which resulted in a highly nonlinear relationship between current achievement and the probability of attending summer school or being retained. Using a regression discontinuity design, we find that the net effect of these programs was to substantially increase academic achievement among third-graders, but not sixth-graders. In addition, contrary to conventional wisdom and prior research, we find that retention increases achievement for third-grade students and has little effect on math achievement for sixth-grade students.

We introduce a test of the rank invariance or rank similarity assumption common in treatment effects and instrumental variables models. The test probes the implication that the conditional distribution of ranks should be identical across treatment states using a regression-based test statistic. We apply the test to data from the Tennessee STAR class-size reduction experiment and show that systematic slippages in rank can be important statistically and economically.