Regression discontinuity design

Common card · No. 428848

Regression discontinuity designs (RDD) are a quasi-experimental pretest–posttest design that attempts to determine the causal effects of interventions by assigning a cutoff or threshold above or below which an intervention is assigned. By comparing observations lying closely on either side of the threshold, it is possible to estimate the average treatment effect in environments where random assignment to conditions is unfeasible. True causal inference using RDDs is still impossible, because the RDD cannot account for the potentially confounding effects of other variables without randomization. The RDD was originally applied by Donald Thistlethwaite and Donald Campbell (1960) to evaluate the effect of scholarship programs on student career plans. The RDD is used in disciplines like psychology, economics, political science, epidemiology, and other related disciplines. Recent comparisons of randomised controlled trials (RCTs) and RDDs have empirically demonstrated the internal validity of...

Read on Wikipedia