The SEM testing for strong invariance of the MTI psychopathy items across time (constrained age 17 Time 1 and age 23 Time 2 factor loading and item thresholds) indicated excellent model fit, TLI = .95, RMSEA = .05 (χ2(47) = 80.83). All item-to-factor loadings were significant (ps < .05–.001). Figure 1 displays the standardized structural parameters for these results, and indicates that there was moderate stability between the latent Detachment factors (.38) and strong stability between the latent Antisocial factors (.54). Also, there were strong concurrent correlations between the MTI psychopathy factors at each time point (ps < .05–.01). There were no significant cross-lag effects (ps > .05). Thus, the results highlight that males who reported more psychopathic emotional detachment also reported increased overt antisocial behavior at age 17 and similarly for the follow up at age 23. At the same time, emotional detachment at age 17 did not result in the temporal prediction of antisocial behavior at age 23 and visa versa.
