Rethinking indulgence: Cross-cultural measurement invariance and its association with impulsive buying in three...

Previous cross-cultural studies on impulsive buying (IB) often treated cultural values as external moderators and assumed that cultural dimensions were unidimensional and stable across nations. At the individual level, the measurement structure of a cultural-value measure may not be comparable across country samples. This study evaluates the unidimensional assumption of an individual-level indulgence measure and examines how its content dimensions are associated with impulsive buying in the three country samples. Data were collected from one cross-national survey of N = 644 participants in China, India, and the United States. The Individual-Level Indulgence Scale (IL-IDG) and Buying Impulsiveness Scale were administered; the IB score retained original items 1-7 and 9. Analytic Step 1 tested the cross-cultural measurement invariance of the IL-IDG scale through multigroup confirmatory factor analysis. The original one-factor model showed weak configural fit across the three country samples. Analytic Step 2 applied a hybrid ant colony optimization (hACO) model specification search to identify a candidate structure. It yielded a two-factor model of indulgence: psychological indulgence (general enjoyment and desire gratification) and physiological indulgence (represented narrowly by two sociosexual-permissiveness items). Analytic Step 3 used multigroup structural equation modeling (SEM) to examine how these two dimensions were associated with IB in the three country samples. Physiological indulgence was positively associated with IB in China and India but not in the United States, while psychological indulgence showed nonsignificant direct associations. The two-factor model added little explained variance (ΔR² = -0.011 to 0.015) beyond the whole IL-IDG score. The findings suggest that specific content domains within this individual-level measure differ in their associations with impulsive buying across the three country samples. Methodologically, the study shows how algorithm-assisted model search can support cultural measurement assessment when guided by item content and model fit. Practically, the findings indicate that indulgence measures should be validated before they are used in cross-cultural marketing.