Revalidation of the Custody Rating Scale for Indigenous
Women Offenders

What we did

The predictive validity of the CRS for Indigenous women offenders was assessed using a dataset of admissions to federal custody between 2013/2014 and 2018/2019. The dataset included Indigenous women offenders admitted to federal custody on a Warrant of Committal (WOC; N = 708) or following a revocation of conditional release (N = 290).

What we found

Concordance between the CRS and OSL was 70% for the WOC admission group and 66% for the revocation group, reflecting lower rates than previously reported (i.e., Barnum & Gobeil, 2012). Among new admissions, there was an equal percentage of security increases and decreases (see Table 1). A higher CRS designation was associated with a greater likelihood of involvement in an institutional incident and receipt of an institutional charge (see Table 2). Among new admissions, Area Under the Curve (AUC) values reached the threshold of ‘acceptable’ predictive accuracy for the CRS, and ‘good’ predictive accuracy in the case of OSL. Predictive accuracy was lower for the revocation group in the case of both the CRS and OSL.

What it means

Results reaffirm the predictive validity of the CRS for Indigenous women offenders at intake. CRS designation was predictive of institutional outcomes. Predictive accuracy was enhanced by OSL, reflecting the efficacy of professional judgement. However, predictive accuracy was weaker when it came to revocation admissions.

Table 1. CRS-OSL Relationship for Indigenous Women Offender Admissions, 2013/2014 to 2018/2019.


CRS-OSL Relationship

Admission Group

Warrant of Committal (WOC)

Revocation

Concordant

70%

66%

Security Increase

15%

21%

Security Decrease

15%

12%

Table 2. Institutional Outcomes by CRS Designation for Indigenous Women Offenders Admitted on a WOC or Revocation, 2013/2014 to 2018/2019.

Custody Rating Scale (CRS) Level

Outcome

Warrant of Committal (WOC)

Revocation

Min.

Med.

Max.

Min.

Med.

Max.

Incident a

55%

74%

91%

41%

49%

70%

Charge b

44%

72%

87%

32%

45%

65%

Table 3. Predictive Ability of CRS and OSL for Indigenous Women Offenders Admitted on a WOC or Revocation, 2013/2014 to 2018/2019.

Receiver Operating Characteristic (ROC) Analyses

Outcome

Warrant of Committal (WOC)

Revocations

AUC *

95% CI

AUC *

95% CI

Custody Rating Scale (CRS)

Incident a

.641

.602-.679

.572

.518-.626

Charge b

.675

.638-.711

.585

.531-.639

Offender Security Level (OSL)

Incidenta

.711

.675-.747

.589

.542-.635

Chargeb

.730

.696-.763

.586

.539-.632

Refers to involvement in at least one security or behavioural incident during the sentence for which the CRS was applied and following CRS administration.
Refers to an offender’s receipt of any serious or minor disciplinary charge during the sentence for which the CRS was applied and following CRS administration. Only charges resulting in an outcome of ‘guilty’ were included.
* The Area Under the Curve (AUC) is used to examine the predictive accuracy of a measure in ROC analyses; a value of 0.5 represents chance prediction, while a value of 1.0 represents perfect prediction.
Note: CI=Confidence Interval

References:

Barnum, G., & Gobeil, R. (2012). Revalidation of the Custody Rating Scale for Aboriginal and non-Aboriginal women offenders (Research Report R-273). Ottawa, ON: Correctional Service of Canada.


For more information

To obtain a PDF version of the full report, or for other inquiries, please e-mail the Research Branch.

You can also visit the Research Publications section for a full list of reports and one-page summaries.

Prepared by: The Special Projects and Data Management (SPDM) team.

Page details

Date modified: