Review:
Regression Discontinuity Design
overall review score: 4.2
⭐⭐⭐⭐⭐
score is between 0 and 5
Regression Discontinuity Design (RDD) is a quasi-experimental statistical method used in research to estimate causal effects of interventions or treatments. It relies on the assumption that units just above and below a specified cutoff score are comparable, allowing researchers to infer the effect of the treatment at the threshold by analyzing the discontinuity in the outcome variable.
Key Features
- Focuses on a specific cutoff point or threshold for assigning treatments
- Exploits local randomization around the cutoff to identify causal effects
- Requires a clear and well-defined running variable and cutoff
- Allows estimation of treatment effects with observational data
- Useful in policy evaluation, education studies, health research, and social sciences
Pros
- Provides credible causal estimates from observational data
- Applicable in real-world scenarios where randomized experiments are infeasible
- Effective for evaluating policy impacts at specific thresholds
- Relatively straightforward to implement and interpret
Cons
- Requires precise data around the cutoff, which may not always be available or accurate
- Assumes no manipulation around the threshold, which may not hold true in all cases
- Limited to estimating local effects near the cutoff, not generalizable to entire populations
- Sensitive to specification choices and bandwidth selection