Correctional Service Canada
www.csc-scc.gc.ca
Common menu bar links
FORUM on Corrections Research
Improving prediction of sex offender recidivism: A proposed study
|
Attention is increasingly being paid to sex offender management. Most Canadian penitentiaries now
have sex offender treatment programs. Increasingly broad support and control measures are being
established to help reintegrate these offenders into society, and the appropriate period of
incarceration for dangerous sex offenders is currently being debated. In short, there is a great need
for a better understanding of the variables associated with sex offender recidivism. This article sets out the framework for a proposed comprehensive study of sex offender recidivism. The first phase has already been approved by a Correctional Service of Canada Regional Research Committee (Quebec) and is under way at the Service's Regional Reception Centre (Quebec). Objectives The proposed study consists of two phases. The first phase will attempt to identify groups of factors that characterize various types of sex offenders, while the second phase will involve long-term offender follow-up to determine recidivism rates for each type of sex offender. All sex offenders who recidivate will also be reassessed to identify the factors that led to their recidivism. This study will be valuable because it will examine a wide range of variables, several of which have never before been considered in recidivism-related cluster analyses. The study will also examine several personal variables (such as personality disorders, psychopathy and deviant sexual preferences) that, unlike the static recidivism predictors used in the past, can be used as intervention targets within treatment programs.(2) Finally, we hope that this study will make it possible to develop a tool for predicting sex offender recidivism. Recidivism variables The variables associated with recidivism can be divided into four categories:
Yet, these variables (unlike criminal history variables) can be the focus of therapeutic intervention. Future studies should, therefore, give priority to these variables. Past research on personal characteristics has also focused on a limited number of factors. Further, several of these studies used instruments that do not meet current psychometric standards or tests that may not be appropriate for use with sexual aggressors.(3) It is, therefore, necessary to study the personal characteristics of sex offenders (in combination with other variables) to improve our ability to identify their risk of recidivism. The proposed study will deal primarily with four categories of personal characteristic variables that may have the potential to predict recidivism:
The offenders will be categorized according to the age and gender of their victims. Therefore, sex offenders who assaulted women older than 18 will be classified as rapists, those who assaulted pre-pubescent children will be classified as pedophiles, those who assaulted pubescent children younger than 18 will be classified as hebephiles (if there was at least a five-year age difference between the aggressor and the victim), and those who assaulted their own children will be classified as incest offenders. Measurement instruments Offender sexual preferences will be assessed using penile plethysmography.(5) The personality characteristics to be assessed relate to psychopathology and certain cognitive and affective variables. Finally, the research team has developed a computerized questionnaire to collect data relating to criminal history, current offence, and personal and family history. Procedure In Quebec, all sex offenders sentenced to two or more years are sent to the Regional Reception Centre. During their stay in this institution, they are assessed by members of the psychology section. All sex offenders who are assessed will be asked to participate in a study on recidivism variables and those who agree must sign a consent form. Each participating offender will then meet with a research assistant who will, after reviewing the offender's file, interview him to obtain any missing criminal history, current offence and personal/family history information. The research assistant will also give the offender psychometric tests that must be completed in a confidential setting. During the second phase of the study, recidivism data will be collected annually. Recidivism will be broken down into sexual recidivism, violent non-sexual recidivism, and non-violent non-sexual recidivism. Further, every offender who returns to the Regional Reception Centre after recidivating will be interviewed by a research assistant to gather information on the pre-crime phase and their new offence. Analytical strategy The first analytical step will be to perform descriptive and comparative studies of the offenders for all study variables. This should help to identify general sex offender characteristics. A variety of comparative analyses will then be carried out, using several variables at a time, to pinpoint the characteristics specific to particular sub-groups of sex offenders. The next step will involve analyzing the statistical relationships among criminal history, current offence, personal characteristics (personal/family history, psychometric and phallometric data, and pre-crime phase data), and past treatment. The first phase will conclude with a study of the development of sexual delinquency to identify the developmental processes in the criminal histories of sex offenders. After a few years, it will be possible to carry out the second phase of the study and examine recidivism. We will be able to examine the relationships between study variables and offender recidivism, the length of time offenders remain in the community without recidivating, the type of recidivism, and the seriousness of the recidivism (such as number of victims and the degree of violence used). Multiple and logistic regression analysis and survival analysis will be used at this time.(6) We hope that the results of this research will make it possible to create predictive tools for use with sex offenders. At the very least, this research should significantly increase our knowledge of the factors associated with sex offender recidivism and, therefore, improve treatment methods.  (1)P.O. Box 6128, Postal Station Centre Ville, Montreal, Quebec H3C 3J7. Please note that only the first phase of this study, the identification of factors characterizing various types of sex offenders, has been approved by the Regional Research Committee. Further approval must be obtained before the second phase of the study can begin. (2)R. K. Hanson, R. Steffy and R. Gauthier, "Long-term recidivism of child molesters," Journal of Consulting and Clinical Psychology, 61 (1993): 642-652. (3)R. K. Hanson, B. Cox and C. Woszczyna, "Assessing treatment outcome for sex offenders," Annals of Sex Research, 4 (1991): 177-208. (4)J. Aubut, Les agresseurs sexuels: Théorie, évaluation et traitement (Montreal: Les éditions de la Chenelière, 1993). (5)J. Proulx, "Sexual preference assessment of sexual aggressors," International Journal of Law and Psychiatry, 12 (1989): 275-280. (6)P. D. Allison, Event History Analysis [Regression for Longitudinal Event Data Series]: Quantitative Application in the Social Sciences, Paper 46 (Beverly Hills: Sage, 1984). |
