Unpacking Gender Differences in Big Six Self-Concept Across African Countries: An Item-Level Analysis

Gender differences in personality have been widely studied, particularly within Western cultural contexts. However, a notable gap exists regarding such investigations in African countries. Global comparative studies indicate a tendency for gender differences in personality traits to be more pronounced in Western cultures compared to other regions. Specifically, societies characterized by clearer, more traditional gender role divisions tend to exhibit fewer gender differences in self-reported personality. At the same time, initial findings from African countries suggest that gender differences in Big Six traits do emerge, even if results are not yet consistent across studies. Additionally, most existing research in African contexts has focused on the domain level, leaving the facet and item levels unexplored. Yet, some previous studies have shown that gender differences appear at the facet level, even when no consistent effects are observable at the domain level. This suggests that specific items within certain facets may be driving these differences. Consequently, it is currently unclear which specific items within these instruments are driving any observed gender differences in personality within African samples. The present study aims to address this research gap by examining the replicability of gender differences in self-reported Big Six personality traits within three African nations: Kenya, Namibia, and South Africa. If measurement invariance across genders is confirmed through Multigroup confirmatory factor analysis (MGCFA), we will proceed with item-level analyses and subsequently use Ant Colony Optimization (ACO) in an exploratory approach to explore item sets that minimize gender-related differences while maintaining adequate model fit. If measurement invariance across genders cannot be confirmed, we will employ ACO in an exploratory approach to identify a measurement-invariant item subset. Item-level analyses will then be conducted with the item subset to investigate which items contribute to the observed gender differences.