Evaluating the Effectiveness of Developmental Mathematics by Embedding a Randomized Experiment Within a Regression Discontinuity Design

Abstract

Using a novel design approach, a randomized experiment (RE) was embedded within a regression discontinuity (RD) design (R-RE-D) to evaluate the impact of developmental mathematics at a large midwestern college ( n = 2,122). Within a region of uncertainty near the cut-score, estimates of benefit from a prospective RE were closely comparable with those found using an RD design, based on retrospective data. Parametric and nonparametric analyses were conducted, yielding increases that generally ranged from one quarter to one third of a grade point. The within-study findings from the RE further established the credibility of RD to produce unbiased estimates. Qualifications of this embedded design strategy were discussed along with its numerous strengths, including elimination of additional validity threats.