2014
DOI: 10.1037/a0035080
|View full text |Cite
|
Sign up to set email alerts
|

Thirty years of research on the Level of Service Scales: A meta-analytic examination of predictive accuracy and sources of variability.

Abstract: We conducted a comprehensive meta-analysis of the Level of Service (LS) scales, their predictive accuracy and group-based differences in risk/need, across 128 studies comprising 151 independent samples and a total of 137,931 offenders. Important potential moderators were examined including ethnicity, gender, LS scale variant, geographic region, and type of recidivism used to measure outcome. Results supported the predictive accuracy of the LS scales and their criminogenic need domains for general and violent r… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

30
271
6
7

Year Published

2017
2017
2022
2022

Publication Types

Select...
5
3

Relationship

0
8

Authors

Journals

citations
Cited by 234 publications
(321 citation statements)
references
References 110 publications
30
271
6
7
Order By: Relevance
“…The SAVRY is a structured professional judgment tool designed to assess violence risk in youth (Borum et al, 2006), whereas the YLS/CMI is designed to assess general reoffense risk in youth (Hoge & Andrews, 2002, 2012. Both tools are widely used (Viljoen, McLachlan, & Vincent, 2010), and predict violent and any reoffending with moderate effect sizes (Olver, Stockdale, & Wormith, 2009;Olver, Stockdale, & Wormith, 2014;Singh, Grann, & Fazel, 2011).…”
Section: Does Reassessment Of Risk Improve Predictions? a Framework Amentioning
confidence: 99%
“…The SAVRY is a structured professional judgment tool designed to assess violence risk in youth (Borum et al, 2006), whereas the YLS/CMI is designed to assess general reoffense risk in youth (Hoge & Andrews, 2002, 2012. Both tools are widely used (Viljoen, McLachlan, & Vincent, 2010), and predict violent and any reoffending with moderate effect sizes (Olver, Stockdale, & Wormith, 2009;Olver, Stockdale, & Wormith, 2014;Singh, Grann, & Fazel, 2011).…”
Section: Does Reassessment Of Risk Improve Predictions? a Framework Amentioning
confidence: 99%
“…However, the correlation between total scores for the two versions is very high (r = .99, n = 21 cases; Gray, Viljoen, & Douglas, 2015). In a recent meta-analysis, the YLS/CMI demonstrated moderate associations with any and violent reoffending (r w = .32 and .26, respectively; Olver, Stockdale, & Wormith, 2014). In the present study, interater reliability was excellent for the YLS/CMI Risk Total Score (ICC = .82 for a two-way random effects model, single raters, absolute agreement, n = 28; Cicchetti, 1994;McGraw & Wong, 1996), fair to excellent for subscales (ICCs = .90,.54,.79,.75,.58,.60,.87,and .60 for Prior and Current Offenses, Family Circumstances/Parenting, Education/ Employment, Peer Associations, Substance Abuse, Leisure/Recreation, Personality/Behavior, and Attitudes/Orientation, respectively), and good for the summary risk rating (ICC = .71).…”
Section: Methodsmentioning
confidence: 99%
“…The SAVRY and YLS/CMI have demonstrated good predictive validity (Olver, Stockdale, & Wormith, 2009;Olver, Stockdale, & Wormith, 2014;Singh, Grann, & Fazel, 2011), and include an emphasis on dynamic or modifiable factors, such as anger management difficulties, rather than solely historical factors, such as past offending. However, remarkably little research has directly examined the ability of these tools to measure changes in risk.…”
Section: Are Adolescent Risk Assessment Tools Sensitive To Change? a mentioning
confidence: 99%
“…La evidencia actual muestra la eficacia de intervenciones basadas en el modelo RNR que focaliza factores de riesgo dinámicos, como desenganche escolar progresivo, abuso de drogas y alcohol (Guillén, Roth, Alfaro, & Fernández, 2015), amigos con alto compromiso delictivo, Olver, Stockdale, & Wormith, 2014); y su impacto se eleva cuando la intervención se ajusta a variables personales, como lo propone el MIID (Fréchette & Le Blanc, 1998).…”
unclassified
“…La primera que Morizot (2015) describe como (i) predisposición, plantea que los rasgos de personalidad tienen un efecto en el inicio del comportamiento delictivo (Jolliffe & Farrington, 2009;Moffitt, 2015;Tremblay, Côté, Salla, & Michel, 2017); la segunda, (ii) plasticidad, señala a la personalidad como un factor de activación, pero también de agravamiento del comportamiento antisocial (De Bolle, Beyers, De Clercq, & De Fruyt, 2012;Fréchette & Le Blanc, 1998;Le Corff & Toupin, 2010); y la tercera (iii) de remisión (Le Blanc & Loeber, 1998) que enfatiza el efecto de la personalidad en procesos de desistimiento del comportamiento delictivo. Además, se ha encontrado en un metanálisis reciente (Olver, Stockdale & Wormith, 2014), que la tendencia personal antisocial, asumida en el modelo RNR, es uno de los 6 factores de riesgo de mayor tamaño de efecto en la predicción de reincidencia. Esta evidencia la respaldan estudios de seguimiento que coinciden al señalar que las variables personales juegan un rol diferenciador de trayectorias delictivas persistentes (Farrington, Ttofi, & Piquero, 2016;Fréchette & Le Blanc, 1998;Loeber et al, 2003).…”
unclassified