|
Activity Number:
|
130
- Time Series Data, Trend Analysis, and Repeated Measures
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2018 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Government Statistics Section
|
|
Abstract #329904
|
Presentation
|
|
Title:
|
Repeated Measurement of a Criminal Risk Assessment on Individuals Over Time
|
|
Author(s):
|
David Schwager*
|
|
Companies:
|
Multnomah County DCJ
|
|
Keywords:
|
Criminal Justice;
Risk Assessment;
Changes over Time;
Recidivism;
Mixed Effects Model;
Bootstrapping
|
|
Abstract:
|
Many community justice agencies (those responsible for supervising adults on probation and parole) rely on criminal risk assessment tools to classify adults, triage workload, and assess service needs. These tools have repeatedly been shown to be accurate across populations and an improvement over unstructured professional judgments. However, there has been little research into how these tools perform at an individual level when multiple assessments are given over time. Do individuals exhibit similar risk trajectories as they progress through their criminal career? How similar are recidivism patterns among different individuals with the same risk score? Do changes in an individual's risk score accurately predict changes in their rate of recidivism? We leverage nearly a decade of LS/CMI and recidivism data in an attempt to answer these questions, using various statistical methods including mixed effects models and bootstrapping.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2018 program
|
