We examined the psychometric properties of the Slovenian version of the Short Dark Triad questionnaire (SD3; Jones & Paulhus, 2014). The SD3, consisting of 27 items, measures the expression of narcissism, Machiavellianism, and psychopathy, three personality traits that constitute the dark triad construct. The study included a sample of 785 working individuals aged between 21 and 70 (67.3% women). The questionnaire’s internal structure validity was assessed using confirmatory factor analysis. The fit of the theoretically most supported three-factor model was evaluated, as well as the one-factor, two-factor, two hierarchical, and two bifactor models. Due to low and even negative inter-item correlations and unsatisfactory factor loadings, one narcissism item, two Machiavellianism items, and two psychopathy items were removed from the Slovenian version of the questionnaire. Based on the inspection of modification indices, residual covariances were freed between two psychopathy items, two Machiavellianism items, and two narcissism items. Taking these adjustments into account, the three-factor and the two-factor models demonstrated comparable fit to the data; the decision to retain the three-factor solution was based primarily on the theoretical justification for distinguishing the three dimensions of the Dark Triad. In addition, we examined measurement invariance of scores across gender and found support only for weak (metric) invariance. This indicates that mean score comparisons between men and women are not meaningful.
|
