ABSTRACT
This chapter describes the process used by one institution in evaluating outcomes of a new tiered and renewable merit program on first-year matriculation and second-year retention for a very high-achieving subset of first-year resident students. The use of regression discontinuity (RD) design for analysis of financial aid is an increasingly common approach, offering robust analyses of aid policy changes at the federal and state levels and at the institutional levels. Using the RD, this analysis capitalized on the natural experiment that occurred when the policy caused an abrupt change in how dollars were awarded and allowed decision makers to understand the impact on intended outcomes and unintended consequences. The RD design used to evaluate the effectiveness of the new merit program allowed comparisons of the enrollment decisions of merit recipients and non-recipients at three different merit award thresholds.
