Dissertation/Thesis Abstract

The Inter-rater Reliability and Predictive Validity of a Structured Professional Judgment Violence Risk Assessment Tool in Workplace Environments
by Cawood, James S., Ph.D., Northcentral University, 2016, 150; 10092274
Abstract (Summary)

Violence risk assessment (VRA) is an essential forensic focus, driven by courts and workplaces tasked with identifying individuals who may pose a risk of harm to others, in the hope that identification and intervention will decrease the number of violent acts. Structured professional judgment (SPJ) VRA tools have improved the predictive validity of violence risk assessments in correctional and mental health populations. These tools are designed by selecting and categorizing risk factors that reflect the biological, psychological, sociological, contextual, and environmental (BPSCE) theory of human violence. However, although SPJ based VRA tools are used in everyday workplaces to make important risk and employment decisions, no tool has been tested for inter-rater reliability and predictive validity in these environments. Therefore, this quantitative postdictive study tested the extent of these properties to close a gap in the existing empirical literature related to SPJ VRA tools, while also providing insight into the degree of influence the environmental element of the BPSCE theory may have. Using the HCR-20v3, the latest version of the most utilized SPJ VRA tool in the world, a sample of 40 known outcome (violent and non-violent) case summaries, representing a range of untested workplace environments, were evaluated by coders blind to outcome. The results were statistically significant for both inter-rater reliability (ICC = .72, 95% CI: .58-.83, p < .001.) and predictive validity (AUC = .70, 95% CI: .61-.80, p < .001.), and demonstrated a good level of reliability and a moderate level of predictive validity, similar to levels of these measures found for the HCR-20v3 with other samples from more restrictive environments. These findings suggest that the environmental element of the BPSCE theory of human violence is not as influential as other elements of this theory, as operationalized in this SPJ VRA tool, while also providing new empirical evidence that the HCR-20v3 can be used ethically in standard workplace environments. It is recommended that further research replicating these results with larger samples and comparison of these findings with those from previously untested SPJ VRA tools that have been especially designed for use in common workplaces would be beneficial.

Indexing (document details)
Advisor: Remedios, Richard
School: Northcentral University
Department: Psychology
School Location: United States -- Arizona
Source: DAI-B 77/08(E), Dissertation Abstracts International
Subjects: Social psychology, Organizational behavior
Keywords: Hcr-20v3, Structured professional judgment, Threat management, Violence risk assessment, Workplace violence
Publication Number: 10092274
ISBN: 9781339597720
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy