Cross-cultural comparisons and the presumption of equivalent measurement and theoretical structure - A look beneath the surface

The purpose of this article is to demonstrate, paradigmatically, the extent to which item score data can vary across cultures despite measurements from an instrument for which the factorial structure is equivalently specified in each group. Based on item scores from the Beck Depression Inventory, the authors first tested for the group invariance of factorial structure and then examined the patterns of item score distributions and response frequencies across Canadian(n = 658), Swedish(n = 1,096), and Bulgarian (n = 691) high school adolescents. Findings carry important implications for multigroup comparisons research in general, and for cross-cultural research in particular, where the primary interest focuses on testing for mean group differences.