Abstract
Traditionally actuarials that have been used to predict recidivism have almost exclusively focused on historical or static factors. This limits their utility for measuring change on critical traits and identifying dynamic treatment targets. In 2016 Brouillette-Alarie, Babchishin, Hanson, and Helmus sought to address this limitation by factor analyzing combined items from the Static 99R/2002R. The goal was to identify the underlying factors that could then be looked at via the lens of latent psychological constructs. Three factors emerged: Persistence/Paraphilia, related to pedophilia; Youthful Stranger Aggression, related to youthfulness and serious offenses; and General Criminality. The present study endeavored to replicate the Brouillette-Alarie et al.’s (2016) factor analysis with a sub-sample of 533 sexual offenders (M age = 35, SD = 11) from the Massachusetts Treatment Center (MTC) who had been assessed for civil commitment between 1959 and 1984 and either been committed or released to finish their prison sentences. Brouillette-Alarie et al.’s (2016) three-factor solution was partially replicated, but the present studies’ results also identified a fourth factor, the Agonistic Continuum Factor, defined by non-contact sexual convictions and stranger victims.\r In order to further understand what each of the latent psychological constructs were measuring, external scales were created by factor analyzing items from diagnostic and typological measures, mechanical actuarials, dynamic risk assessment tools, structured clinical guidelines, and multiple scales developed as part of the MTC research program. The aim with each scale was to capture a particular psychological risk factor that was then correlated with the Static latent factors. The results of these correlations found that the Persistence/Paraphilia Factor was significantly correlated with the Pedophilia Scale, the Density of Sexual Offenses Scale, and negatively correlated with both the Offense Violence Scale and the Non-Sexual Violence Scale. The Youthful Antisociality Factor was found to significantly correlate with both impulsivity scales (Impulsivity in Offense and Lifestyle), as well as the Antisociality Scale and the Offense Violence Scale. The General Criminality Factor was found to have significant correlations with the Antisociality Scale, the Density of Sexual Offense Scale, and the Lifestyle Impulsivity Scale. Finally, the Agonistic Continuum Factor significantly correlated with the Offense Violence Scale, both impulsivity scales (Impulsivity in Offense and Lifestyle), the Antisociality Scale, and the Stranger Offenses Scale, while negatively correlating with the Pedophilia Scale. These findings suggest that the latent factor structure found within the Static99R/2002R not only can be used to predict recidivism, but have further credence for assessing not only treatment targets but change within those treatment targets as well.