Regression Discontinuity Explained (with Examples)

Regression discontinuity design (RDD) identifies causal treatment effects by exploiting sharp thresholds in policy rules, delivering rigorous impact estimates that maximize ROI without the expense of randomized controlled trials. This comprehensive guide shows you how to implement RDD for data-driven business decisions that drive measurable cost savings.

What is Regression Discontinuity?

Regression discontinuity is a quasi-experimental design that estimates causal effects when treatment assignment is determined by whether an observed variable (the "running variable") crosses a known threshold. Unlike costly randomized experiments, RDD leverages existing administrative cutoffs to create natural comparisons between similar units just above and below the threshold.

The core insight is elegant: individuals just below a cutoff (like scoring 69 on a test) are nearly identical to those just above (scoring 70), except one group receives treatment and the other does not. By comparing outcomes in this narrow window around the threshold, we obtain credible causal estimates of treatment effects.

RDD comes in two primary forms:

• Sharp RDD: Treatment is deterministically assigned based on the threshold. Everyone above the cutoff receives treatment; everyone below does not.
• Fuzzy RDD: The threshold changes the probability of treatment but does not guarantee it. Some individuals above the cutoff may not receive treatment, or some below may receive it.

When to Use Regression Discontinuity

RDD is ideal when you face these conditions:

Clear Assignment Rules

Your organization must have an explicit, documented threshold that determines treatment. Common examples include:

• Age-based eligibility: Senior discounts at age 65, retirement benefits, minimum age requirements
• Performance thresholds: Bonus structures tied to sales quotas, promotion criteria based on evaluation scores
• Financial cutoffs: Credit limits based on income levels, subsidy eligibility tied to poverty lines
• Geographic boundaries: School district assignments, tax jurisdiction differences, zoning regulations
• Time-based rules: Policy implementation dates, deadline-driven program access

Sufficient Data Near the Threshold

You need adequate observations close to the cutoff. As a practical guideline, aim for at least 100-200 observations on each side of the threshold within your chosen bandwidth. Sparse data near the cutoff leads to imprecise estimates and reduces the business value of your analysis.

No Manipulation of the Running Variable

Individuals or administrators cannot precisely control the running variable to game the threshold. For example, if students can retake tests until they pass a cutoff, or if credit applicants can strategically adjust reported income, the RDD assumptions break down and estimates become biased.

Data Requirements for Regression Discontinuity

Successful RDD implementation requires four essential data elements:

1. Running Variable (Forcing Variable)

This continuous variable determines treatment assignment. It must be measured before treatment occurs and should be precisely recorded. Examples include test scores, age in days, income levels, or application timestamps. The running variable should have enough variation near the threshold to support local estimation.

2. Treatment Indicator

A binary variable indicating whether each unit received treatment. In sharp RDD, this perfectly corresponds to being above or below the threshold. In fuzzy RDD, the threshold affects treatment probability but does not determine it completely.

3. Outcome Variable

The metric you want to measure the treatment effect on. This could be revenue, customer retention, employee productivity, cost per acquisition, or any other business KPI. The outcome must be measured after treatment occurs and should be available for both treated and control units.

4. Baseline Covariates (Optional but Recommended)

Pre-treatment characteristics that help validate the RDD design and can improve precision. These variables should not be affected by treatment and should show no discontinuity at the threshold if the design is valid. Common examples include demographic characteristics, historical performance metrics, and geographic indicators.

Data Quality Checklist

• Running variable has no missing values near the cutoff
• Outcome data is available for at least 80% of units
• Treatment assignment follows documented rules
• Threshold value is precisely known and consistently applied
• Data collection predates your analysis to avoid selection bias

Setting Up Your Regression Discontinuity Analysis

A rigorous RDD implementation follows a structured workflow that balances statistical validity with computational efficiency.

Step 1: Visualize the Discontinuity

Begin with exploratory visualization. Plot the outcome variable against the running variable, marking the threshold with a vertical line. Divide the running variable into bins and calculate average outcomes within each bin. A clear visual discontinuity at the threshold provides initial evidence of a treatment effect.

Create separate plots for baseline covariates. These should show smooth trends across the threshold. Any discontinuities in pre-treatment variables suggest potential confounding or manipulation.

Step 2: Test for Manipulation

Examine the density of the running variable around the threshold. Create a histogram with narrow bins and look for unusual bunching or gaps exactly at the cutoff. Formal tests like the McCrary density test detect statistically significant manipulation.

If you detect manipulation, consider whether it invalidates the design or merely requires additional robustness checks. Minor manipulation may be addressed through donut-hole RDD designs that exclude observations immediately adjacent to the threshold.

Step 3: Select Bandwidth

The bandwidth determines which observations near the threshold are included in the analysis. Larger bandwidths increase sample size and precision but risk including dissimilar units. Smaller bandwidths improve comparability but reduce statistical power.

Data-driven bandwidth selectors like the Imbens-Kalyanaraman or Calonico-Cattaneo-Titiunik methods optimize the bias-variance tradeoff. These algorithms select bandwidths that minimize the mean squared error of the treatment effect estimator.

Always report results using multiple bandwidths (e.g., 0.5x, 1x, and 2x the optimal bandwidth) to demonstrate robustness. If estimates change dramatically across bandwidth choices, investigate further or acknowledge the sensitivity.

Step 4: Choose Estimation Method

Several estimation approaches are available:

• Local linear regression: Fits separate linear regressions on each side of the threshold using observations within the bandwidth. This is the current best practice and balances flexibility with boundary bias correction.
• Polynomial regression: Fits global polynomials of various orders. Modern research discourages high-order polynomials (cubic or higher) due to overfitting and sensitivity to specification choices.
• Local polynomial regression: Extends local linear regression to quadratic or higher orders within the bandwidth, useful when curvature is evident.

Step 5: Implement Robust Inference

Standard errors must account for potential heteroskedasticity and correlation structures in your data. Robust standard errors or cluster-robust standard errors (when data has natural groupings) provide valid inference.

For fuzzy RDD designs, use instrumental variables estimation where the threshold serves as an instrument for treatment receipt. The local average treatment effect (LATE) estimates the causal effect for compliers—those whose treatment status is changed by the threshold.

Maximizing ROI Through Proper Interpretation

Correct interpretation transforms statistical results into actionable business intelligence that drives cost-effective decisions.

Understanding the Local Average Treatment Effect

RDD estimates the treatment effect at the threshold, not the average effect across all individuals. This local effect applies to the marginal unit—the person or entity just meeting the eligibility criterion. For business decisions, ask whether the marginal case represents your target population or strategic interest.

Quantifying Business Impact

Translate statistical estimates into financial metrics. If RDD shows that a loyalty program increases customer lifetime value by 150 dollars for customers at the spending threshold, and you have 10,000 customers near that threshold, the program generates 1.5 million dollars in additional value annually. Compare this to program costs to calculate clear ROI.

Assessing Generalizability

Consider whether effects at the threshold generalize to other points in the running variable distribution. If treatment effects are heterogeneous, extrapolating results far from the cutoff becomes risky. Use economic theory or domain knowledge to reason about effect heterogeneity.

Communicating Results to Stakeholders

Present findings with clear visualizations showing the discontinuity, confidence intervals reflecting statistical uncertainty, and sensitivity analyses demonstrating robustness. Non-technical stakeholders respond well to simple before-after comparisons around the threshold accompanied by estimated dollar impacts.

Real-World Example: Customer Loyalty Program ROI Analysis

Consider an e-commerce company that offers free premium shipping to customers who spend over 500 dollars in a calendar year. Management wants to understand whether premium shipping increases customer retention and lifetime value, but running a randomized experiment would require complex infrastructure and risk customer dissatisfaction.

The Setup

The company has historical data on:

• Running variable: Annual spending in the year preceding program eligibility
• Treatment: Premium shipping access (assigned exactly at 500 dollars)
• Outcome: Customer lifetime value over the subsequent 24 months
• Sample: 50,000 customers with spending between 300 and 700 dollars

The Analysis

Visualization reveals a clear upward jump in lifetime value exactly at the 500-dollar threshold. Customers spending 495-499 dollars (just below) have an average 24-month lifetime value of 850 dollars. Customers spending 501-505 dollars (just above) average 1,020 dollars—a 170-dollar difference.

Formal RDD estimation using local linear regression with an optimal bandwidth of 50 dollars confirms a statistically significant treatment effect of 165 dollars (95% CI: 130-200 dollars). Baseline covariates like account age, product category preferences, and geographic region show no discontinuity at the threshold, validating the design.

Manipulation tests reveal no unusual bunching at exactly 500 dollars, suggesting customers do not strategically time purchases to reach the threshold.

The Business Decision

Premium shipping costs the company approximately 80 dollars per customer annually. With a 165-dollar increase in lifetime value and an 80-dollar cost, the net benefit is 85 dollars per customer. The company has 8,000 customers annually near the 500-dollar threshold.

ROI calculation: (8,000 customers × 85 dollars net benefit) / (8,000 customers × 80 dollars cost) = 106% return on investment. The program clearly generates positive ROI for customers at the margin.

Based on this analysis, management decides to expand the program and test lowering the threshold to 400 dollars, potentially capturing additional profitable customers. This decision is grounded in credible causal evidence obtained without expensive experimentation.

Best Practices for Regression Discontinuity

Follow these guidelines to ensure methodological rigor and business relevance:

Specification Robustness

• Report results across multiple bandwidth choices to demonstrate stability
• Compare local linear regression to polynomial specifications
• Test sensitivity to different kernel functions (uniform, triangular, Epanechnikov)
• Exclude observations very close to the threshold (donut-hole RDD) if manipulation is suspected

Validation Checks

• Verify no discontinuities in baseline covariates at the threshold
• Test for smooth density of the running variable (McCrary test)
• Conduct placebo tests at artificial thresholds where no treatment occurs
• Examine heterogeneity by subgroups to understand variation in treatment effects

Documentation and Transparency

• Preregister your analysis plan before examining the data when possible
• Document all specification choices, including bandwidth selection methods
• Report all sensitivity analyses, not just preferred specifications
• Make code and data available for replication when permitted

Practical Implementation

• Use established software packages (rdrobust in R or Stata) rather than custom implementations
• Automate diagnostic plots and tests for efficiency and reproducibility
• Build reusable analysis pipelines for recurring RDD applications
• Train multiple team members in RDD methods to avoid single points of failure

Related Causal Inference Techniques

Regression discontinuity is one of several quasi-experimental methods for causal inference. Understanding when to use alternatives maximizes analytical ROI.

Difference-in-Differences

When you have panel data and treatment timing varies across groups, difference-in-differences may be more appropriate than RDD. This method compares changes over time in treated groups to changes in control groups, controlling for time-invariant confounders.

Synthetic Control

For single or few treated units with rich time-series data, synthetic control methods construct weighted combinations of control units that closely match the treated unit's pre-treatment trajectory. This approach works well for case studies like policy changes affecting entire regions or countries.

Causal Impact Analysis

When analyzing the effect of a single intervention in time-series data without a natural threshold, Causal Impact with Bayesian structural time-series models provides an alternative framework. This method forecasts what would have happened without intervention and compares it to observed outcomes.

Propensity Score Matching

Without clear thresholds or panel structure, propensity score matching creates balanced treatment and control groups based on observed covariates. While matching requires stronger assumptions than RDD, it applies to broader settings where treatment assignment is not threshold-based.

Cost-Saving Implementation Strategies

Maximize the efficiency and business value of your RDD analyses with these practical strategies:

Leverage Existing Data Infrastructure

RDD analyses use administrative data already collected for operational purposes. Rather than building new data collection systems, identify existing thresholds in your organization (credit approvals, promotion criteria, program eligibility) and extract relevant data from current systems. This approach delivers insights at minimal incremental cost.

Automate Recurring Analyses

Many organizations have ongoing programs with stable thresholds. Build automated pipelines that update RDD estimates quarterly or annually as new data accumulates. Automation converts one-time analytical investments into continuous monitoring systems that track program effectiveness over time.

Prioritize High-Stakes Decisions

Focus RDD resources on decisions with substantial financial implications. Analyzing whether to expand a multi-million-dollar program justifies significant analytical effort. Evaluating minor operational changes may not warrant the same investment. Prioritization ensures analytical capacity aligns with business impact.

Build Internal Expertise

Training existing analysts in RDD methods costs less than repeatedly hiring external consultants. Develop internal capabilities through workshops, mentorship, and hands-on projects. Over time, in-house expertise enables rapid, cost-effective analysis of emerging questions.

Integrate with Decision Workflows

Embed RDD analyses into formal decision processes. Require evidence of program effectiveness at thresholds before budget renewals. Mandate impact evaluations for new threshold-based policies. Integration ensures that analytical investments translate directly into improved decisions rather than producing reports that sit unused.

Conclusion

Regression discontinuity design provides a powerful framework for extracting causal insights from threshold-based administrative rules. By comparing similar units just above and below policy cutoffs, organizations obtain credible treatment effect estimates without the expense and complexity of randomized controlled trials.

The ROI of RDD is compelling. Analyses leverage existing data, require no operational changes, and produce actionable evidence that informs high-stakes decisions. Whether evaluating customer loyalty programs, employee incentive structures, or public policy interventions, RDD delivers cost-effective causal inference that drives measurable business value.

Success requires attention to methodological details: testing for manipulation, selecting appropriate bandwidths, conducting robustness checks, and correctly interpreting local treatment effects. But when applied rigorously to appropriate settings, regression discontinuity transforms administrative thresholds from bureaucratic rules into valuable natural experiments that illuminate causal relationships and optimize resource allocation.

Start by identifying threshold-based policies in your organization. Gather data on running variables, treatment assignment, and outcomes. Apply the structured workflow outlined in this guide. And transform your analytical capabilities from descriptive reporting to causal inference that delivers sustained competitive advantage through evidence-based decision-making.