A tale of two general factors of personality in relation to intelligence and validity measures

Abstract

Four general factors of personality (GFP) were computed by analyzing two different personality measures using two different GFP extraction methods within a sample of job candidates. GFP scores were computed by both examining the first unrotated factor of the scales and by examining the higher order factor following a factor analysis of the scales allowing for multiple first-order factors. The two GFP scores within each personality scale had high correlations (although for one measure the correlation was negative). In general, the GFP scores from each personality measure had only moderate correlations with each other. Also examined were correlations with intelligence measures as well as the validity scales for faking (good and bad), infrequency, and social desirability. Similar to past findings, lower-order scales had slightly higher correlations with the intelligence measures than did the GFPs for both personality measures. Three of the four GFP scores had significant positive correlations with faking good and social desirability but non-significant correlations with faking bad and infrequency. The results are interpreted as suggesting that GFPs differ based on the method of factor extraction, that the GFP scores reflect positive impression management, and that the GFPs are less informative than the lower-order personality factors.