Skip to main content

COMPEL Glossary / propensity-score-matching

Propensity-score matching

An observational-study method matching treated and control units on estimated probability of receiving treatment, then comparing outcomes across matched pairs.

What this means in practice

Reduces selection bias where randomisation is impossible; sensitive to unobserved confounders and to the quality of the propensity-score model.

Synonyms

PSM , propensity matching

See also

  • Synthetic control — A counterfactual constructed from a weighted combination of untreated donor units — the "synthetic" version of the treated unit.
  • Counterfactual outcome — The outcome that would have occurred without the AI intervention — the benchmark against which incremental AI value is measured.
  • Difference-in-differences (DiD) — A quasi-experimental design comparing treated and control trajectories over time — identifying causal effect from differential change.

Related articles in the Body of Knowledge