A Re-Validation of the Security Reclassification Scale (SRSW) among federal women offenders for a period of at least six months

The Correctional Service of Canada (CSC) currently uses the Security Reclassification Scale for Women (SRSW) to ensure the use of gender-informed tools within women’s facilities across Canada. This actuarial tool, in conjunction with professional judgment, aids in the security classification of Federal Women Offenders after initial placement is determined from the Custody Rating Scale (CRS). To ensure the SRSW has remained valid within the Federal Women offender population, it was examined for a review period of at least six months.

This study was designed to validate the use of the SRSW in review periods of at least six months. A total of 610 security reviews of 397 women offenders were completed between June 27, 2007 to March 31, 2010. All women had a complete CRS and SRSW relating to the term in which the Offender Security Level decision was made. Of these, 525 reviews were specific to review periods of at least six months.

Analyses were conducted to examine the reliability of the scale through item-total correlations. Additionally, the convergent validity was assessed by examining the relationship between SRSW security level recommendations, final security level decision, and the measures of risk, need, and reintegration potential. Finally, the predictive ability was examined using several institutional and community outcomes.

Overall, the majority of SRSW recommendations were to medium security levels. Although few of the scale’s scores fell between discretionary ranges^{Footnote 1} (11%), the majority (53%) remained at the same level of security or was placed in a lower level of security (29%). Final decisions to override SRSW recommendations outside of discretionary ranges occurred relatively often (27%) and the majority was to higher levels (66%). The most common reason cited for the inconsistency related to the offender’s current behavior and attitude.

The scale proved to be reliable for the review period of six months or longer. As well, the removal of a specific scale item (ever unlawfully at large) increased the scale’s reliability.

In considering convergent validity, a moderate association was found between SRSW recommendations and risk, need, and reintegration potential. When examining predictive validity, the SRSW recommendations were predictive of all types of institutional incidents, charges and discretionary release. In each of these cases, the tool was more predictive than those of final security level decisions.

Evidence suggests that the SRSW remains valid for a review period of six months or longer, although there were a high proportion of inconsistencies found between caseworker recommendations and the SRSW recommendation. Furthermore, minimal differences were noted by Aboriginal self-identification; however, when differences were noted the tool tended to be more reliable and valid among Aboriginal women than Non-Aboriginal women.

Thompson, J., McConnell, A., & Paquin-Marseille, L. (2011). The Security Reclassification Scale (SRSW) for Shorter Review Periods among Federal Women Offenders. Ottawa: Correctional Service Canada

Prepared by: Ashley McConnell

Research Branch
(613) 995-3975
research@csc-scc.gc.ca