Abstract
This longitudinal study explored the existence of, and the transition between, latent classes based on risk/need domains of the Structured Assessment of Violence Risk in Youth (SAVRY). The study included 4,267 male and 661 female justice-involved juveniles who had at least one SAVRY assessment completed between 2006 and 2011. A three-step approach was used for the latent class analyses (LCA): (1) A standard LCA estimated the classes; (2) the class-membership was determined; and (3) latent transition analyses estimated the likelihood of transition between the subgroups. For male adolescents, five latent classes were identified: (a) low risk/needs (36%); (b) low-moderate risk/needs (26%); (c) moderate risk/needs (11%); (d) moderate-high risk/needs (19%); and (e) high risk/needs (8%). For female adolescents, three subgroups were identified: (a) low risk/needs (30%); (b) moderate risk/needs (51%); and (c) high risk/needs (19%). Recidivism rates differentiated the subgroups, and the likelihood of transition within a 12-months timeframe was low.
Original language | English |
---|---|
Pages (from-to) | 350-364 |
Journal | International Journal of Forensic Mental Health |
Volume | 18 |
Issue number | 4 |
DOIs | |
Publication status | Published - 2019 |
Keywords
- Justice-involved juveniles
- Latent class analyses
- Gender
- Risk assessment
- SAVRY
- VIOLENCE RISK
- PREDICTIVE-VALIDITY
- STRUCTURED ASSESSMENT
- GENDER-DIFFERENCES
- CASE FORMULATION
- NEEDS
- PSYCHOPATHY
- RELIABILITY
- FRAMEWORK
- CRITERIA