Regression Discontinuity Designs: Applications for School Psychologists

Abstract

Regression discontinuity designs offer school psychologists a powerful tool to evaluate the effects of programs on individuals in cases where: (a) selection into a program is based on a continuous variable (e.g., intelligence); (b) program outcomes may be meaningfully measured (e.g., achievement); and (c) it is not possible to assign candidates randomly to program/no program conditions. Procedures for conducting regression discontinuity designs are illustrated with actual and altered data. Strengths, weaknesses, and factors that influence the decision to use (or avoid) regression discontinuity designs are discussed.