Regression Discontinuity Designs in Counseling Intervention Research and Program Evaluation: A Primer on a Strategy for Making Causal Inferences

Abstract

The move toward evidence-based practice within counseling intervention research continues to grow. However, there are many practical barriers to implementing research designs that can support causal claims regarding counseling intervention efficacy. Overcoming these hurdles serves a larger goal of improving client care and further legitimizing the counseling profession. This article outlines an underused research design called regression discontinuity (RD), one of the strongest quasi-experimental designs available. RD allows researchers to use a cutoff score to determine treatment group assignment in situations where a randomized control trial is not feasible or ethical, such as when assigning care based on need consistent with the social justice values of the field. Historically, RD is rarely employed in counseling research. Therefore, to facilitate use, we provide a primer on key RD design concepts along with a concrete demonstration of the design, data analysis, and interpretation. We review the causal logic of the basic RD design followed by two data outcomes for a hypothetical scenario. Data and SPSS syntax are provided to allow readers to follow along and conduct their own analyses.

Keywords:

• Regression discontinuity
• counseling intervention research
• causal inference
• quasi-experimental design

Disclosure Statement

We have no known conflict of interest to disclose.

Additional information

Notes on contributors

Bonnie L. Stice

Bonnie L. Stice, MA, is pursuing a PhD in Counseling at the University of North Texas. She holds a Licensed Professional Counselor (LPC) Associate license in the state of Texas. She is passionate about uplifting marginalized voices, in particular through research on intimate labor, relational-cultural theory, and the mental health effects of climate change.

Robin K. Henson

Robin K. Henson, DMin, PhD, is a Professor and Distinguished Teaching Professor of Educational Psychology in the Department of Educational Psychology at the University of North Texas. His research interests include applied behavioral science statistics and measurement and self-efficacy theory.