The external-validity transfer matrix

Each line names a claim related to this exhibit, its source, and how it was checked.

• A EH23-C1 "we propose a formal framework that encompasses four dimensions of external validity x t y and c validity populations treatments outcomes and contexts"Egami & Hartman. (2023)
• A EH23-C2 Two goals, not one: "effect generalization generalizing the magnitude of causal effects" and "sign generalization generalizing the direction of causal effects", each with its own estimators and assumptionsEgami & Hartman. (2023)
• A EH23-C3 "effect generalization generalizing the magnitude of causal effects", for which the authors "introduce three estimators of the target population causal effects"Egami & Hartman. (2023)
• A EH23-C4 "sign generalization generalizing the direction of causal effects", for which they "propose a novel multiple testing procedure under weaker assumptions": a different target needs a different toolEgami & Hartman. (2023)
• B R01-OWN-C1 External validity is target-relative; a result does not possess universal transfer by itselfAuthor framework
• B R01-OWN-C3 Changing one transfer field changes the target claim boundary before a verdict is offeredAuthor boundary
• B R01-OWN-C4 A self-serve small-business experiment does not automatically answer an enterprise implementation targetAuthor synthetic illustration
• B R01-OWN-C7 Replication, transport, and analogy are different decisions with different evidentiary burdensAuthor framework

Grades: A, verified against the printed page of the primary source · B, primary source, text layer only · C, authoritative secondary · D, reported.

This exhibit belongs to an essay: the numbers live here, the argument lives there.