Yes, But What's the Mechanism? (Don't Expect an Easy Answer)

Abstract

Psychologists increasingly recommend experimental analysis of mediation. This is a step in the right direction because mediation analyses based on nonexperimental data are likely to be biased and because experiments, in principle, provide a sound basis for causal inference. But even experiments cannot overcome certain threats to inference that arise chiefly or exclusively in the context of mediation analysis - threats that have received little attention in psychology. We describe three of these threats and suggest ways to improve the exposition and design of mediation tests. Our conclusion is that inference about mediators is far more difficult than previous research suggests, and best tackled by an experimental research program that is specifically designed to address the challenges of mediation analysis.

Keywords: Mediation, Experiments, Causal Inference, Indirect Effects

JEL Classification: C90