Elaborating on the construct validity of the antisocial process screening device in Chinese children and adolescents: Across-informants and across-samples

Abstract

The Antisocial Process Screening Device (APSD) is a widely used measure to assess psychopathic traits and antisocial behavior in youth, although controversy exists with respect to its factor structure. The aim of the current study was to further examine the factor structure and construct validity of the APSD using data from multiple informants (i.e., parent-report and self-report) and varied samples: community adolescents (N = 1073), male prisoners (N = 613), and schoolchildren (N = 473). Confirmatory factor analyses showed that the existing factor structures (i.e., two-factor, three-factor, and four-factor models) did not yield adequate fit in any of the three samples. Exploratory structural equation modeling approach did not yield a consistent solution among samples, either. The APSD total scores showed acceptable internal consistency (Cronbach’s alpha) and mean item correlations (MIC), while the subscale scores showed low to moderate consistency and MICs. The convergent and criteria validity of both the APSD total and factor scores were supported, exhibiting the expected correlations with other measures of psychopathy and aggressive behavior (r ranging from −.01 to .62 for adolescent sample; from −.16 to .69 for prisoner sample; and from −.05 to .70 for children sample). These findings suggest that the APSD total score is a reasonable measure of psychopathy in the Chinese youth population, although the use of APSD factor scores should be avoided in future studies with Chinese youth since no structural model provided adequate fit with acceptable reliability.