RANDOM AND CUTOFF - BASED ASSIGNMENT 1 Running Head : RANDOM AND CUTOFF - BASED ASSIGNMENT A Randomized Experiment Comparing Random to Cutoff - Based Assignment

Regression discontinuity designs (RDD) assign participants to conditions using a cutoff score, with those above the cutoff going to one condition and those below to another. Statistical theory shows that a correctly implemented and analyzed RDD gives unbiased effect estimates, just as in a randomized experiment. This study tests that theory by randomly assigning 588 participants to be in a randomized experiment or a regression discontinuity design, and then comparing results. Participants were tested on vocabulary and mathematics pretest and outcomes, and on many covariates, and were given a mathematics or vocabulary intervention. Results suggest that estimates from regression discontinuity designs approximate the results of randomized experiments reasonably well, but also raise the issue of what constitutes agreement between the two estimates.