Difference-in-Differences (DiD)
Quasi-experimental method that estimates causal effects by comparing changes over time between treatment and control groups.
Difference-in-Differences estimates causal effects through double before-after comparison – the most important method for natural experiments in marketing.
Explanation
DiD compares: (Treatment-After - Treatment-Before) - (Control-After - Control-Before). The second difference eliminates time-invariant differences and common trends. Prerequisite: Parallel Trends Assumption.
Marketing Relevance
Ideal when A/B tests are impossible: Regional campaign launches, price changes, policy changes – estimating causal effects from observational data.
Common Pitfalls
Parallel trends assumption hard to verify. Spillover effects between groups. Staggered treatment timing requires modern DiD methods.
Origin & History
John Snow used an early form of DiD in 1854 (cholera study). Card & Krueger (1994) made DiD famous with their minimum wage study. Callaway & Sant'Anna (2021) solved problems with staggered DiD.
Comparisons & Differences
Difference-in-Differences (DiD) vs. A/B Testing
A/B testing randomizes; DiD uses natural variation and controls for trends – when randomization is impossible.
Difference-in-Differences (DiD) vs. Regression Discontinuity
RD uses thresholds; DiD uses before-after comparisons. Both are quasi-experimental but for different settings.