Background
Regression discontinuity design (RDD) is a sophisticated econometric technique used primarily for causal inference. This methodological approach allows researchers to estimate the effect of a treatment or intervention by exploiting a predetermined cutoff point or threshold within the assignment variable.
Historical Context
The concept of regression discontinuity design was first introduced by Donald Thistlethwaite and Donald Campbell in 1960. Originating from the field of educational psychology, RDD has since evolved and found broad application across various disciplines including economics, political science, and epidemiology.
Definitions and Concepts
Regression discontinuity design involves the application of a treatment based on whether an observed variable, known as the assignment or forcing variable, crosses a specific threshold. In doing so, researchers can create a clear division between treated and untreated groups, facilitating the estimation of causal effects.
For instance, consider a merit award granted to students scoring above a certain percentile on an exam. By comparing academic outcomes of students just above and just below this cutoff, one can infer the causal impact of receiving the award.
Major Analytical Frameworks
Classical Economics
Classical approaches often focus on theoretical and qualitative aspects rather than specific empirical methodologies like RDD.
Neoclassical Economics
RDD fits well into the neoclassical framework by providing a rigorous method for identifying causal relationships and quantifying the effects of policy interventions.
Keynesian Economics
While Keynesian analysis predominantly addresses macroeconomic phenomena, RDD can be applied to micro-level policy evaluations which may in turn have macro-level implications.
Marxian Economics
RDD is less commonly used within the Marxian paradigm, but can still offer valuable empirical insights into the effects of class and economic inequalities.
Institutional Economics
Institutions often enact policies with cutoff-based eligibility criteria, making RDD a valuable tool for examining the effectiveness of institutional interventions.
Behavioral Economics
RDD can capture behavioral responses to discrete changes in economic incentives and can thus contribute to studies examining human behavior under different experimental conditions.
Post-Keynesian Economics
Post-Keynesian scholars might utilize RDD to analyze how policies affect income distribution, savings, investment behaviors, and broader economic variables.
Austrian Economics
Although Austrian economics often emphasizes theoretical over empirical work, RDD can be utilized within this school to enhance understanding of intervention impacts.
Development Economics
RDD is extensively applied in development economics as it helps assess the impact of policies like education programs, healthcare interventions, and cash transfers in low-income settings.
Monetarism
While monetarism elders focus on money supply effects on macro variables, RDD can be engaged to understand the micro-level impacts of monetary policy interventions.
Comparative Analysis
Regression discontinuity design stands out from other causal inference methods such as randomized controlled trials (RCTs) or difference-in-differences (DiD) due to its ability to exploit real-world policy cutoffs. It offers a robust alternative to randomization by providing quasi-experimental conditions to identify causal effects.
Case Studies
-
Impact of Merit-Based Scholarships on Academic Achievement: Researchers utilize RDD to compare students scoring just above and below scholarship thresholds to evaluate the award’s impact on academic performance and future opportunities.
-
Health Policy Evaluations: Examination of the effects of health interventions provided to individuals or groups crossing eligibility thresholds.
Suggested Books for Further Studies
-
“Mastering Metrics: The Path from Cause to Effect” by Joshua Angrist and Jörn-Steffen Pischke
A comprehensive introduction to modern econometric techniques, including a full chapter on RDD. -
“Mostly Harmless Econometrics: An Empiricist’s Companion” by Joshua D. Angrist and Jörn-Steffen Pischke
An essential resource for grasping the empirical strategies central to regression discontinuity analysis.
Related Terms with Definitions
- Causal Inference: Methodologies and strategies used to estimate the causal effect of one variable on another.
- Assignment Variable: The variable that determines the assignment to treatment or control in RDD.
- Threshold or Cutoff: The specific point of the assignment variable at which the treatment is initiated.
By understanding the application and implications of regression discontinuity design, researchers can robustly infer causal relationships in policy contexts and beyond.
