Survey correlations between prejudice scales may be inflated by response style, not real relationships

Evidence

The third January student study was prompted by John Duckitt, who wondered if the “nest” of SDO correlations was actually assembled by social desirability effects. Suppose some people are willing to admit bad things about themselves on a survey, but others are relatively reluctant to admit such, or are even unaware of them. If so, then responses to a collection of scales that raise sensitive issues will be glued together to some extent by this “willing-to-look-bad versus try-to-look-good” response style, whether the traits involved really are related. In this light, high SDOs may not be more prejudiced than others, nor more PP-Mad, E-MAD, accepting of rape myths, and so on, but merely less defensive. — John T. Jost and Jim Sidanius, Political Psychology, 1044

Categorization

• QUESTION: The author tests and appears to reject this, but the response-style artifact critique applies to any self-report correlation study and is worth holding onto as a general methodological caveat.
• XREF: connects to: Pre-1962 survey phrasing ambiguously mapped opposition to integration. — Ilyana Kuziemko and Ebonya Washington, Why Did the Democrats Lose the South?…, 1330; Vague survey wording can make racial prejudice undetectable in response data; Measuring stereotypes indirectly reveals prejudice explicit surveys miss — Anthony G. Greenwald and Mahzarin R. Banaji, Implicit Social Cognition: Attitudes,…, 194