Menu
Home
Log in / Register
 
Home arrow Philosophy arrow Sex Crimes and Sex Offenders: Research and Realities
Source

Effectiveness of Sex-Offender Assessment Tools

A common method of assessing the effectiveness of predicting sexual recidivism is to test the tool with a sample of sex offenders. The assessment is scored for each person. Subsequently, the level of risk is compared to whether the offender actually recidivated. A statistical test is conducted, and the results are expressed as "area under the curve" (AUC). That number indicates the accuracy of the test in predicting sexual recidivism. For example, if a test places all those who do not actually recidivate in a low-risk category and places all those who do actually recidivate in a high-risk category, the measure is considered perfect. As indicated earlier, no test is perfect. Many of the tests have exceedingly high accuracy, as high as 0.9 AUC, where 1 corresponds to a perfect measure. The lowest AUC is 0, and any test that yields 0.5 or less is considered a failed test, or rather, no better than guessing. Thus, it does not accurately predict enough cases to warrant its use. Table 10.2 contains a summary of the results of a meta-analysis that assessed several sex-offender assessments (Tully et al., 2013). Most of the scales, on average, fall in the "moderate" range of predicting recidivism (Rice & Harris, 2005).

table 10.2 Summary of Actuarial Assessments Tools' Effectiveness

ASSESSMENT

NUMBER OF STUDIES CONDUCTED

AUC RANGE

AUC AVERAGE

Static-991

30

.57 - .92

.69

RRASOR1

13

.42 - .77

.69

SORAG1

9

.67 - .77

.68

RM2000/S1

8

.58 - .76

.67

Static-20021

7

.67 - .76

.70

SVR-201

5

.59 - .83

.70

MnSOST-R1

5

.59 - .71

.64

SACJ-Min2

1

.67

.67

Sources:

  • 1 (Tully, Chou & Browne, 2013)
  • 2 (Hanson and Thornton, 2000)
 
Source
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
 
Subjects
Accounting
Business & Finance
Communication
Computer Science
Economics
Education
Engineering
Environment
Geography
Health
History
Language & Literature
Law
Management
Marketing
Mathematics
Political science
Philosophy
Psychology
Religion
Sociology
Travel