Typology of Relational Crime Perpetrator : A Latent Class Analysis Approach

With the growing importance of promptly assessing the risk of recidivism in relational crimes, in 2024 Korean National Police Agency initiated efforts to streamline case processing procedures by integrating two risk assessment used by police officers at the scenes of intimate partner violence and stalking. This move reflects an organizational commitment to actively respond to relational crimes. However, prior studies have identified various subgroups of offenders depending on the type of relational crime, suggesting that heterogeneous characteristics may become more pronounced when domestic violence, dating violence, and stalking are treated collectively as relational crimes. Accordingly, this study aimed to identify heterogeneous perpetrators profiles and examine whether these profiles differ by type of relational crime. To this end, latent class analysis(LCA) and chi-square tests were conducted on a total of 5,586 cases of domestic violence, dating violence, and stalking reported between September 13 and October 11, 2024, to police stations under the jurisdiction of five metropolitan/provincial police agencies: Seoul, Northern Gyeonggi, Southern Gyeonggi, Incheon, and Daegu. As a result, four types were identified: Low-Risk, Acute High-Risk, Latent Risk, and Stress-Triggered. Significant differences in perpetrator type distribution were found across relational crime types. In domestic violence and dating violence cases, the most common perpetrator types were Low-Risk and Latent Risk. In contrast, stalking cases showed a high proportion of Low-Risk offenders, followed by the Stress-Triggered type. Based on these findings, the study discusses its implications, limitations, and future directions for research on typologies of relational crime perpetrators.