Risk Factors for Homelessness Among Recently Released Offenders

Abstract

The current study examined characteristics related to homelessness among released offenders to improve housing services for this group. Forty-one incarcerated Canadian prisoners were interviewed approximately 3 months prior to release about their former and anticipated accommodations. Homelessness prior to incarceration was positively related to violent institutional behavior, violence risk level, and criminogenic needs but not recidivism. Stable housing prior to incarceration was positively related to greater community support. Recommendations include creating more community and institutional housing programs for offenders more likely to experience housing difficulties in the community.

Acknowledgments

This research was funded by the Saskatchewan Housing Corporation and the Forensic Behavioural Sciences and Justice Studies Initiative, University of Saskatchewan. We thank the Correctional Service of Canada for permitting us to conduct this research. The views expressed in this article are those of the authors and not necessarily those of the Saskatchewan Housing Corporation, the Forensic Behavioural Sciences and Justice Studies Initiative, the Correctional Service of Canada, or the University of Saskatchewan.

Notes

¹Criminogenic risk factors are dynamic or changeable predictors of criminal behavior, also known as dynamic risk factors or criminogenic needs (Andrews & Bonta, 2006). The major risk/need factors are history of antisocial behavior, antisocial personality pattern, antisocial cognition, antisocial associates, family/marital circumstances, school/work, leisure/recreation, and substance abuse..

Note. One participant did not state what geographic setting he/she was living in prior to incarceration.

*p < .05.

²Responses were missing for two participants and were not applicable for the three participants who were homeless.

*p < .05.

Note. P.O. = Parole Officer.

^a 0 = no; 1 = yes;.

^b Responses to the question were recoded as “Suitable” (0) and “Unsuitable” (1) for analysis;.

^c Responses to the question were recoded as “Inadequate/Not Present” (0) and “Adequate/Present” (1) for analysis.

*p < .05, **p < .01.

³Unexpectedly, three participants were not released into the community between the time of the initial interview and the collection of recidivism data and thus were excluded from this calculation and any analyses involving recidivism.

Additional information

Notes on contributors

Karen Parhar

Karen Parhar is now at Kwantlen Polytechnic University, Surrey, British Columbia, Canada.