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

Journal of Personality and Social Psychology, Vol. 98, pp. 550-58, April 2010

45 Pages Posted: 29 Jan 2010 Last revised: 21 Jun 2010

See all articles by John G. Bullock

John G. Bullock

University of Texas at Austin

Donald P. Green

Columbia University

Shang E. Ha

Brooklyn College - CUNY

Date Written: January 28, 2010

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

Suggested Citation

Bullock, John G. and Green, Donald P. and Ha, Shang E., Yes, But What's the Mechanism? (Don't Expect an Easy Answer) (January 28, 2010). Journal of Personality and Social Psychology, Vol. 98, pp. 550-58, April 2010. Available at SSRN: https://ssrn.com/abstract=1543925

John G. Bullock (Contact Author)

University of Texas at Austin ( email )

2317 Speedway
Austin, TX 78712
United States

Donald P. Green

Columbia University ( email )

7th Floor, International Affairs Bldg.
420 W. 118th Street
New York, NY 10027
United States

Shang E. Ha

Brooklyn College - CUNY ( email )

2900 Bedford Avenue
Brooklyn, NY 11210
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
361
rank
80,152
Abstract Views
1,996
PlumX Metrics