Comparer des méthodes

	Ecological Inference	Analyse causale de médiation (effets directs et indirects naturels)
Domaine	Political Science	Inférence causale
Famille	Regression model	Regression model
Année d'origine	1997	2010
Auteur d'origine	Leo Goodman (ecological regression); Gary King (statistical EI solution)	Pearl (2001); general framework by Imai, Keele & Tingley (2010)
Type	Aggregate-data model inferring individual-level rates from grouped totals	Counterfactual causal decomposition
Source fondatrice	King, G. (1997). A Solution to the Ecological Inference Problem: Reconstructing Individual Behavior from Aggregate Data. Princeton: Princeton University Press. ISBN: 9780691012414	Pearl, J. (2001). Direct and Indirect Effects. In Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence (UAI), 411-420.
Alias	EI, Ecological regression, King's ecological inference, Aggregate-to-individual inference	natural direct effect, natural indirect effect, NDE / NIE decomposition, counterfactual mediation
Apparentées	5	5
Résumé	Ecological inference is the problem of learning about individual behavior — such as how Black and white voters cast their ballots — when only aggregate data are available, like precinct-level turnout and racial composition. Because individual-level data are missing, the within-group rates are not directly observed; ecological inference recovers them by combining the deterministic accounting constraints that each precinct must satisfy with a statistical model of how the unobserved rates vary across precincts. Gary King's 1997 solution unified the deterministic method of bounds with Leo Goodman's classic ecological regression, sharply reducing the long-standing risk of the ecological fallacy.	Causal mediation analysis is a counterfactual framework that splits a treatment's total effect into a Natural Direct Effect (NDE) and a Natural Indirect Effect (NIE) that runs through a mediator. The modern general approach was formalised by Pearl (2001) and Imai, Keele and Tingley (2010), giving the decomposition a precise causal interpretation.