The regression discontinuity design is based on the seminal work by Thistlethwaite and Campbell (1960). It builds on the observation that treatments in educational, policy, and business settings are often granted based on whether subjects are below or above a certain threshold. For instance, a government may decided to provide subsidies only to companies with less than 250 employees at a given point in time. The key idea of the regression discontinuity design is that near the (often arbitrary) threshold, the assignment process is almost random (Dunning, 2012). That is, it is almost random whether a company has 248 or 252 employees. Therefore, we expect that subjects who are below, but close to the threshold, do not substantially differ from subjects who are above, but close to the threshold. Consequently, we can estimate the causal effect of the treatment by comparing the score on an outcome variable between the treament group and the control group near the threshold. Indeed, simulations showed that the regression discontinuity design “recovered the true causal parameters almost precisely” (Antonakis, Bendahan, Jacquart, & Lalive, 2010, p. 1108).