Having these established risk assessment tools, the question then becomes how effectively the final measures have sampled and weighted the relevant variables. As such, if the evaluation is primarily concerned with violent recidivism, we recommend scales designed for that purpose e. Modern psychometrics: The science of psychological assessment 3 rd ed.
All three scales included in the current study are similar to each other in that they have the same purpose predicting sexual recidivism and are based on similar demographic and criminal history variables. Risk Assessment Instruments StaticR.
Criminal Justice and Behaviour, 37 If the selected scales are not sampling the same types of risk factors, then evaluators need a defensible model concerning 1 the latent constructs measured by the scales and 2 empirical evidence concerning how the constructs should be weighted and combined.
The factors are statistically weighted.
Conceptually, if an instrument provides new information to better understand an offender's risk, it provides incremental information. New York: Foundation Press. Having these established risk assessment tools, the question then becomes how effectively the final measures have sampled and weighted the relevant variables.
Analyses conducted separately for each comparison, with each sample entered as strata and the riskinstruments entered in Step 1. This suggests that the recidivism rates for each risk level in the table above are over-estimates of what recidivism rates would be today. E-mail: karl.
Archer, R. Specifically, Hanson and colleagues used Pearson correlation coefficients to compute the Hanley and McNeil test a less conservative test than the Kendall's Tau correlation coefficients whereas in the current study we used the Kendall's Tau.