Figure - available from: Journal of Experimental Criminology
This content is subject to copyright. Terms and conditions apply.
Source publication
Objectives
We examine public attitudes towards false positives and false negatives in criminal justice risk assessment and how people’s choices differ in varying offenses and stages.
Methods
We use data from a factorial survey experiment conducted with a sample of 575 Americans. Respondents were randomly assigned to different conditions in the vig...
Citations
... Risk assessment instruments are designed to estimate the risk of future (violent) offending and to eventually, if personalized interventions are implemented based on the observed risk level of an individual, help prevent violent behavior. Using validated instruments within the prison setting is important, since false negative predictions may potentially harm societal and prison safety (Kang and Wu, 2022). For instance, when individuals with a high violence risk are granted unjustified leaves. ...
... From a deprivation perspective, this may in turn lead to an increased risk of violence. There is an ongoing debate about the ideal cost ratio of false positives (potential harm to the individual) versus false negatives (potential harm to others), which is also context dependent (Rice and Harris, 2005;Kang and Wu, 2022). Although criminal justice professionals and the general public opinion seem to prefer risk assessment models with a higher rate of false positives (Netter, 2007;Barnes and Hyatt, 2012;Kang and Wu, 2022). ...
... There is an ongoing debate about the ideal cost ratio of false positives (potential harm to the individual) versus false negatives (potential harm to others), which is also context dependent (Rice and Harris, 2005;Kang and Wu, 2022). Although criminal justice professionals and the general public opinion seem to prefer risk assessment models with a higher rate of false positives (Netter, 2007;Barnes and Hyatt, 2012;Kang and Wu, 2022). ...
Introduction
Physical and verbal violence toward staff or other detained individuals is a reoccurring problem within correctional facilities. Screening for violence risk within the prison setting could provide a valuable first step in the prevention of institutional violence. The brief and compact Risk Screener Violence (RS-V) has shown to be an efficient new method for assessing concerns regarding post-release violent offending for incarcerated persons. This study aimed to find out whether the RS-V is also able to predict future violent and aggressive incidents during imprisonment.
Methods
The predictive validity of the RS-V for future violent and aggressive incidents during a follow-up time of 4 months within prison was analyzed, using a file-based design. Violent incidents toward staff and other inmates (physical violence and violent threats), other aggressive incidents (aggression toward objects and verbal disruptive behavior), and both categories combined, were included as outcome measures based on disciplinary reports.
Results
The RS-V showed medium to large predictive values for both violent and aggressive behavior during prison stay. In particular, good predictive values of the RS-V were found for violence toward prison staff.
Discussion
This study shows that, besides post-release violent recidivism, the RS-V is able to accurately predict future violent and aggressive incidents during prison stay. By correctly differentiating between low concern and high concern individuals, the RS-V aims to contribute to more personalized interventions and risk management and, subsequently, to improved prison safety. Future studies using prospective prison practice data are needed to further support the validity of the RS-V regarding institutional violence.
