Martijn van Teffelen

Chapter 3 60 Next, Pearson’s correlation analyses were run on all study variables (Table 3). These bivariate analyses were explorative in nature. The results showed a positive association between provoked aggression and negative affective change ( p = .015) only, while all other correlations between the diverse outcome variables were non-significant ( p ’s > .056). The data further revealed a positive relationship between psychopathy and narcissism ( p < .001). The results also showed that narcissism was linked with increased threat perception ( p = .020). All other relations were found to be non-significant (p ’s > .060) . Table 3 Pearson’s correlations aggression, negative affect change, ego-threat, and psychopathic and narcissistic traits UA PA a NA change b Threat perception Psychopathy Narcissism UA 1 PA a .02 1 NA change b -.09 .29** 1 Threat perception -.20 -.01 .17 1 Psychopathy .08 -.04 -.20 .19 1 Narcissism .03 .18 -.07 .25* .53*** 1 Note. a 7 cases excluded as they did not believe the aggression manipulation. b Spearman’s ranked correlations were calculated due to non-normality. * significant at p <.05; ** significant at p <.01; *** significant at p <.001. NA = negative affect; UA = unprovoked aggression; PA = provoked aggression. Then, multiple linear regression analyses were run to closer inspect the unique (i.e., residualized) multivariate relationships between residualized psychopathic and narcissistic traits and unprovoked aggression, provoked aggression, change in negative affect, and threat perception. Results are shown in Table 4. The findings suggest that residualized psychopathic traits were unrelated to aggression, change in negative affect, and threat perception. 2 The results also show that residualized narcissistic personality traits were positively related to provoked aggression, but unrelated to unprovoked aggression, change in negative affect, and threat perception. Closer inspection of multicollinearity diagnostics reveals a variance inflation factor of 1.47, not exceeding a critical threshold of 10 (Hair et al., 1995). In short, residualized narcissistic traits were positively related to aggressive responding after provocation. 2 For completeness, when people who were suspicious of the CRTT manipulation were retained in the model psychopathic traits did negatively predict provoked aggression at β = -.27, p = .033.