Statistically small effects of the Implicit Association Test can have societally large effects.
Greenwald, Banaji and Nosek revisited the two meta-analyses of the test's predictive validity. They attribute most of the difference between them to which correlations were counted, argue that statistically small effects can be societally large when they apply to many people or recur for one person, and state that modest test-retest reliability and small effects make the test risky for classifying individuals.
Greenwald, Banaji and Nosek revisited the two meta-analyses of the test's predictive validity. They attribute most of the difference between them to which correlations were counted, argue that statistically small effects can be societally large when they apply to many people or recur for one person, and state that modest test-retest reliability and small effects make the test risky for classifying individuals.
Written from the abstract, vault paper record checked 2026-10-23.
What our sources record
Recorded when this source was checked, figures included, with page references and quotation marks removed.
Greenwald, A. G., Banaji, M. R., & Nosek, B. A. (2015). Statistically small effects of the Implicit Association Test can have societally large effects. Journal of Personality and Social Psychology, 108(4), 553-561. doi:10.1037/pspa0000016 https://doi.org/10.1037/pspa0000016