U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

LSI-R to LS/CMI Conversion: Analysis of Impact on Business Practices

NCJ Number
241402
Date Published
November 2012
Length
16 pages
Annotation
This study evaluated the impact on recidivism rates in Hennepin County when the Department of Community Corrections and Rehabilitation switched from the Level of Service Inventory - Revised to the Level of Service/Case Management Inventory for assigning offenders to appropriate levels of supervision.
Abstract
Key findings from the evaluation include the following: the Hennepin normative sample for the LS/CMI (Level of Service/Case Management Inventory) was different racially from the U.S. community normative sample; female offenders in the Hennepin normative sample had a lower risk for recidivism based on the LS/CMI than were females in the U.S. community sample; Black and American Indian offenders scored higher risk scores than did White and Asian offenders. The application of proposed cut points for Low, Moderate, and High Risk for the LS/CMI resulted in an increase in the number of offenders classified as Moderate Risk and a decrease in those classified as High and Low Risk compared to the LSI-R; and the 3-year recidivism rate for High Risk offenders based on the LS/CMI was 42 percent. When these scores were compared to those developed using the Level of Service Inventory-Revised (LSI-R), it was found that the LSI-R scores were highly related to the total converted scores on the LS/CMI, and that this correlation was observed for the U.S. normative sample. It was also found that the 1, 2, and 3-year recidivism rates were quite similar under both the LSI-R and the LS/CMI, lending support to the validity of the LS/CMI for predicting overall recidivism rates for offenders. Data for the study were downloaded from the Minnesota Statewide Supervision System for the period January 1, 2009, through October 12, 2012. The LSI-R scores for this period were converted to LS/CMI scores using a conversion algorithm. The findings from this evaluation provide preliminary confirmation on the validity of LS/CMI for determining recidivism risk among offenders. Tables, figures, and references