Correlated scales can reveal tautological overlap rather than real psychological connections.

Evidence

Take those fish in the barrel above. It could be that the RWA associations we discovered with religious variables merely resulted from the religion-related items on the authoritarianism measure. We would then have merely discovered that if you ask people how religious they are on one scale, and then ask them how religious they are on another measure, the two sets of answers will covary, quite amazingly. (Do not laugh. This happened with at least one measure of conservatism [RWA, pp. 213-214].) — Bob Altemeyer, The Authoritarian Specter, 522

Categorization

• CANDY: That a published conservatism measure once ‘discovered’ religion correlates with itself is light, amusing methodological gossip worth sharing.
• XREF: connects to: The F Scale study’s coding allowed circular reasoning to creep into interpretations — Martin Jay, The Dialectical Imagination: A History…, 744; Overlapping measures of NPI and affective polarization make testing their theorized causal link impossible. — Sabrina J. Mayer and Luana Russo, What one is not: a new scale to measure…, 11