Difference-in-differences identification and estimation
Specify comparison assumptions and estimands
With staggered adoption and heterogeneous effects, a conventional two-way fixed-effects coefficient can combine unsuitable comparisons. Define group-time effects, eligible controls and aggregation, with assignment-level inference.
Flow diagram
Steps
Define treatment-effect estimands
Specify group-time, event-time or aggregate effects and target weighting.
State identification assumptions
Document parallel untreated trends, no anticipation and interference assumptions.
Choose timing-compatible estimator
Use methods suitable for staggered adoption and heterogeneous effects when applicable.
Specify adjustment and uncertainty
Specify covariates measured before treatment, any sampling weights and clustering at the assignment level. With few policy units, choose inference that addresses the small number of clusters.
Plan aggregation and interpretation
Prespecify comparison groups, event-time support and reported contrasts.
Difference-in-differences identification and estimation
End flow
Tables and supporting information
Comparative change estimands
The simple two-group two-period contrast is illustrative, not a universal estimator.
Illustrative model specification
The four means describe treated (T) and comparison (C) groups before and after intervention. The contrast subtracts the comparison group’s change from the treated group’s change.
Method and protocol evidence
Attach the protocol, methodological sources and versioned software specification supporting these decisions. Record citations and provenance; do not enter participant identifiers.
Model assumptions
Two-way fixed effects can mix inappropriate comparisons under staggered adoption and heterogeneous effects. Causal interpretation needs parallel untreated trends and other design assumptions. This expression is not a universal staggered-adoption estimator; specify group-time effects, valid controls and aggregation when adoption varies.