7 152 CHAPTER 7 interrelationships among dynamic risk factors, it can be noticed that although the first network study had a relatively small sample size, the findings were replicated in an independent sample, of adequate size (resp. N = 805, and N = 4,511). The case studies, however, were clearly limited in number and, as a result, these findings should be interpreted with caution (N = 5). Second, most if not all North American and European research on dynamic risk factors predictive of sexual reoffending has been conducted predominantly in heterosexual men adjudicated for sexual offenses. This also applies to the studies in this dissertation, limiting the generalization of the findings to other populations (e.g., women, transgender and gender diverse individuals, juveniles, women and men from different cultures or cultural contexts, with different sexual orientations, and those who have not entered the criminal justice system). Third, findings on the interrelationship of dynamic risk factors in chapters 2 and 3 are based on two North American routine correctional samples of adult males charged or convicted for a sexual offense. The North American legal system and approach to law, referred to as common law, however, differs from jurisdictions that adopted the inquisitorial system of law, or the civil law approach (Gunn & Mevis, 2018). In the common law tradition, fact-finding is the responsibility of defense and prosecution through equality of arms, meaning that both the accused and the prosecutor must be given the opportunity to present their case under conditions that do not place them at a disadvantage vis-à-vis the opponent. In contrast, in the inquisitorial system of law, a fair trial is attained by the judges’ role of independent and “incorruptible” inquirer (van der Wolf et al., 2022). Differences in legal systems impact criminal law and criminal procedure, including the supervision of men with a history of sexual offenses (Gunn & Mevis, 2018). Therefore, caution should be taken in generalizing results from Chapters 3 and 4 across continents and countries. Fourth, studies in this dissertation did not differentiate based on offense type (e.g., collection of child sexual exploitation material, voyeurism, indecent exposure, child sexual abuse, rape, sexual violence in intimate relationship, and sexual murder). Men perpetrating these different crimes may share dynamic risk factors but may also differ in dynamic risk factors more typically associated with their specific type of offense (e.g., hostility towards women vs. emotional congruence with children). Moreover, network structures and metrics might differ across subsamples of individuals even when they include the same dynamic risk factors (Borsboom et al., 2017). Fifth, the networks in the original and replication study in Chapters 3 and 4 were estimated based on data obtained from the STABLE-2007. This dynamic risk assessment instrument developed for adult men with a history of sexual offenses is empirically and theoretically driven and previous studies revealed no differences in predictive propensities between North America and several European countries (Brankley et al., 2021). However, the STABLE-2007 captures a limited number of dynamic risk factors. Adding an additional dynamic risk factor to a network analysis may impact the network construction and strength centrality of all dynamic risk factors. A sixth and last limitation concerns the network construction. That
RkJQdWJsaXNoZXIy MjY0ODMw