Contradictory findings signal weak effects rather than mere measurement differences

Evidence

With enough ingenuity, one could probably trace this extreme confusion to different uses of concepts and measures. But if the relationships between the variables were very strong, mere differences in preferred measures ought not to produce such wildly divergent findings. — John T. Jost and Jim Sidanius, Political Psychology, 4344

Categorization

• DEFINE: A general epistemic criterion: result instability across measurement choices warns that the underlying relationship is weak.
• XREF: Classic robustness-check logic from metascience; echoes the replication-crisis argument that fragile findings across specifications indicate small true effects. — connects to: Null racial-priming results may reflect weak messages rather than real effects. — Rachel Wetts and Robb Willer, Who Is Called by the Dog Whistle?…, 26; Measurement instrument choice drastically changes forgiveness findings — Nyla R. Branscombe and Bertjan Doosje (eds.), Collective Guilt: International…, 1375; Failing to control for population density produces spurious no-disparity findings in waste-siting studies.