RT Article
T1 The predictive validity of the LS/CMI with Aboriginal offenders in Canada
JF Criminal justice and behavior
VO 42
IS 5
SP 481
OP 508
A1 Wormith, J. S.
A2 Hogg, Sarah M.
A2 Guzzo, Lina
LA English
YR 2015
UL https://krimdok.uni-tuebingen.de/Record/1885176678
AB This study examined the applicability of a general risk/need assessment tool, the Level of Service/Case Management Inventory (LS/CMI), to a large sample of Aboriginal offenders (n = 1,692) and compared the predictive validity with that of the rest of the cohort, a sample of non-Aboriginal offenders (n = 24,758). It examined the use of the clinical override with offenders. Aboriginal offenders had considerably higher scores and a greater recidivism rate than non-Aboriginal offenders. Internal consistency was high and virtually identical for both samples. The predictive validity for Aboriginal offenders on general recidivism was high, although slightly higher for non-Aboriginal offenders. The predictive validity was significant but low on violent recidivism for Aboriginal offenders, as were numerous subscales. Assessors used the override feature to change risk level less frequently on Aboriginal offenders. The implications of this study for policy (use on ethnic minority offenders) and practice (how to interpret potential recidivism) are discussed.
K1 Aboriginal offenders
K1 LS/CMI
K1 Level of Service
K1 override
K1 Recidivism
K1 Risk assessment
DO 10.1177/0093854814552843