The factor structure of Big Five personality trait measures at the between- and within-person levels

The cross-sectional factor structure of Big Five personality trait measures is a weighted blend of the factor structure at the between-person level (i.e., the level of time-invariant interpersonal personality differences) and the factor structure at the within-person level (i.e., the level of intrapersonal personality fluctuations and changes). Thus, the Big Five might not represent the underlying factors at either of the two levels. To probe this issue, I fit multilevel exploratory and confirmatory factor analysis (EFA and CFA) models to longitudinal data from the Netherlands and Australia (N = 15,085 and 22,582, respectively). In both data sets, the canonical Big Five emerged at the between- but not at the within-person level. At the within-person level, a two-factor structure appeared to be most suitable. Factor 1 indicated positive attitudes towards the self. Factor 2 indicated neuroticism or negative attitudes towards the self. The two factors might reflect short-term fluctuations in positive and negative affect rather than long-lasting personality changes. The findings are tentative due to methodological limitations. We need more research on the within-person factor structure of personality trait measures to clarify what fluctuations and changes they assess and whether we need to revise the way we measure and analyze trait changes.