Allowing for these significant deviations from measurement invariance across cohorts by applying the Bayesian model, Tables 3 and 4 show uncorrected means and variances per cohort, as measured by the NEO-FFI and EPQ-R items. Note that we included all cohorts with NEO data (NEO-PI-R or NEO-FFI), but using only the 12 items that are part of both the NEO-PI-R and the NEO-FFI. NESDA shows the highest mean Neuroticism score (which is expected given that it concerns a sample selected for depression and anxiety) and PAGES the lowest mean for NEO data. For NEO Extraversion, the QIMR adolescents show the highest mean (as expected based on their age), and CILENTO the lowest mean. Based on the EPQ data, the Croatian samples have the highest Neuroticism and Extraversion scores, and ORCADES the lowest. Some variance differences across cohorts are also observed, which can partly be explained by differences in age distribution, birth cohort and inclusion criteria. Note that for the NEO, the variances for Neuroticism are larger than for Extraversion, which is explained by the higher reliability of the Neuroticism scale. This
