38978 research outputs found
Sort by
The mismeasure of comparability: Non-invariance and nomological validity in cross-cultural psychology—A commentary on Kusano et al. (2025)
No full textKusano et al. (2025) argue that strict measurement invariance (MI) requirements may be ill-suited for cross-cultural psychology because they can obscure theoretically meaningful differences between countries. We share their call to move beyond MI as a gatekeeper and to focus on the substantive relevance of non-invariance. However, we question whether correlating country-level scale scores with external indices separates construct-relevant factors from method biases. In a simulation calibrated to their empirical setting, we show that scalar non-invariance (threshold/intercept shift) (a) attenuates and destabilizes country-level correlations, (b) inflates correlations when bias aligns with the construct, and (c) suppresses associations when bias opposes it. We also outline a causal perspective on MI that treats non-invariance as theoretically informative and motivates explicit models of its causes. Using this causal view, we argue that adjusting for demographic covariates at the composite score level does not remedy item-level non-invariance
