Research Reports

The Changing Profile of the Federal Inmate Population: 1997 and 2002

This report is also available in French. Ce rapport est également disponible en français. Veuillez vous adresser à la direction de la recherche, Service Correctionnel du Canada, 340 avenue Laurier ouest, Ottawa (Ontario) K1A 0P9. Should additional copies be required they can be obtained from the Research Branch, Correctional Service of Canada, 340 Laurier Ave., West, Ottawa, Ontario, K1A 0P9.

 

2003 N^o R-132

The Changing Profile of the Federal
Inmate Population: 1997 and 2002

Roger Boe
Mark Nafekh
Ben Vuong
Roberta Sinclair
Colette Cousineau

Research Branch

Correctional Service of Canada

January 2003

Table of Contents

Executive Summary*

Men in Custody*

Women in Custody*

A: Has there been a Change in the Federal Prison population*

B: Methodology*

1). In-custody snapshots:*

2). Admission cohorts:*

3). OIA Indicators:*

4). OIA assessment under-coverage estimates*

C: Overview of Federal Inmate Population*

D: Profiles of Federal Inmates "In-Custody"*

D1: Institutional Profile of Federally Sentenced Men*

1.Initial Custody Rating Score*

2.Risk Factors*

3.Prior Sentence History*

4.Previous Breaches of Trust*

5.Length of Sentence*

6.Major Offence (Current Sentence)*

7.Global Rating – Static Factors*

8.Global Rating – Dynamic Factors*

9-a.Domain Rating - Dynamic Factors (Some Need)*

9-b.Domain Rating - Dynamic Factors (Considerable Need)*

10.Employment / Education Indicators*

11.Marital/Family Indicators*

12.Associates / Social Interaction Indicators*

13.Alcohol/Drug Abuse Indicators*

14.Cognition Indicators*

15.Mental Health Indicators*

Summary*

D2: Institutional Profile of Federally Sentenced Women*

1.Initial Custody Rating Scores*

2.Risk Factors*

3.Prior Sentence History*

4.Previous Breaches of Trust*

5.Length of Sentence*

6.Major Offence (Current Sentence)*

7.Global Rating - Static Factors*

8. Global Rating – Dynamic Factors*

9-a.Domain Ratings - Dynamic Factors (Some need)*

9-b.Domain Ratings -Dynamic Factors (Considerable need)*

10.Employment / Education Indicators*

11. Marital / Family Indicators*

12. Associates / Social Interaction Indicators*

13.Alcohol/Drug Abuse Indicators*

14.Cognition Indicators*

15. Mental Health Indicators*

Summary*

 E. Profiles of Federal (Warrant of Committal) Admissions.*

E1: Admission Profiles of Federal Men*

1.Initial Custody Ratings*

2.Risk Factors*

3.Prior Sentence History*

4.Previous Breaches of Trust*

5.Length of Sentence*

6.Major Offence (Current Sentence)*

7.Global Rating - Static Factors*

8.Global Rating – Dynamic Factors*

9a.Domain Rating – Dynamic Factors (Some Need)*

9b.Domain Rating – Dynamic Factors (Considerable Need)*

10.Employment / Education Indicators*

11.Marital / Family Indicators*

12.Associates / Social Interaction Indicators*

13.Substance Abuse Issues*

14.Cognition Indicators*

15.Mental Health Indicators*

Summary*

E2. Admission Profiles of Federal Women*

1.Initial Custody Ratings*

2.Risk Factors*

3.Prior Sentence History*

4.Previous Breaches of Trust*

5.Length of Sentence*

6.Major Offence (Current Sentence)*

7.Global Rating - Static Factors*

8.Global Rating – Dynamic Factors*

9a.Global Rating – Dynamic Factors (Some Need)*

9b.Domain Rating – Dynamic Factors (Considerable Need)*

10.Employment / Education Indicators*

11.Marital / Family Indicators*

12.Associates / Social Interaction Indicators*

13.Substance Abuse Issues*

14.Cognition Indicators*

15.Mental Health Indicators*

Summary*

Appendix A: References*

Appendix B: OIA Coverage Issues*

Estimates of admission under-coverage*

Estimates of Snapshot Under-coverage*

Executive Summary

This report compares the profile of the federal inmate population at two points in time – March 1997 and March 2002 – and addresses the general issue of whether there have been significant changes in the federal offender population over the past several years.

Profiles were developed both for in-custody snapshot populations and for admission cohorts using a standard template. A separate analysis is provided of the trends for federal women and men. The report compares sets of profiles at two different dates:

1. First, two federal snapshots, representing the in-custody federal population counts as of March 31^st 1997 and 2002, and;
2. Second, two federal admission cohorts, representing all Warrant of Committal admissions during Fiscal Year (FY) 1996/97 and 2001/02.

The data for the study comparisons were obtained from a new Climate Indicator and Profile (CIPS) System that is currently being developed for the CSC, as part of the toolkit for institutional threat-risk assessment. The profiles compared criminal history and risk/need assessment information on offenders, collected through various offender assessment processes such as the Offender Intake Assessment (OIA), the CRS (Custody Rating Scale), the SIR scale, and other standard assessment tools. The report compares offender information (both static and dynamic factors) covering the following indicators: 1- Initial Custody Rating Score; 2-

Risk Factors; 3-Prior Sentence History; 4- Previous Breaches of Trust; 5- Length of Sentence; 6- Major Offence (Current Sentence); 7- Global Static Factor Assessment; 8- Global Dynamic Factor Assessment; 9- Domain Dynamic Factors ("Some" and "Considerable need"); 10- Employment/Education Indicators; 11- Marital/Family Indicators; 12- Associates; 13- Substance Abuse; 14- Cognition; 15- Mental Health.

Men in Custody

The average number of men in federal custody at year-end has decreased since our base comparison year (1997). This population decrease reflects the general decline in crime rates that began in 1991. However, the composition of the federal prison population has changed in ways that may make inmates today more "correctionally challenging" for federal institutions.

We first examined the indicators of risk for increasing severity. We found that more men in the custody snapshot were designated Maximum-Security in 2002 than in 1997 — the proportion increased from 14% to 21% (which represents a 50% jump between 1997 and 2002). This increase was driven by a comparable 50% increase in the proportion of men designated as requiring Maximum-Security in the 2001-02 admission cohort (up from 7% to 14% or also a 50% gain since FY 1996-97).

The average age of men in federal custody is increasing and this is revealed in the two snapshots (the proportion Under 30 decreased from 32% to 30% between 1997 and 2002). This is partly due to an aging of the admission cohorts (admissions under age 30 also decreased, from 43% to 41%). From a risk assessment standpoint, younger men are generally at higher risk for institutional incidents and post-release re-offending.

Although the population of men, in general is getting older in federal prisons, the proportion of men Under-30 in Maximum Security institutions is generally increasing: for example, the proportion Under-30 Years in Kingston Pen increased from 29% to 32%; in Atlantic Institution from 45% to 48%; in Donnacona Institution from 41% to 42%; and in Edmonton Institution up from 49% to 51%. In Kent Institution, however, the proportion decreased from 41% to 39%. The population of some (but not all) Medium-Security institutions has also been getting younger.

There has been a substantial increase in the proportion of men rated as CRS-High Institutional Adjustment Risk (increasing from 7% to 13% - nearly a 50% increase over 1997), and CRS-High Security Risk (up from 37% to 42% - a 14% increase). Further, the number of men in prison with a gang affiliation also increased (a 17% gain in five years).

Once again, the trends found on these risk indicators in the custody snapshots are also reflected in the admission cohorts: the proportion of admissions rated CRS-High Institutional Adjustment Risk increased from 5% to 11% (a 120% gain); the proportion of admissions rated High Security Risk increased from 19% to 23% (a gain of 22%); and the proportion of admissions with a Gang Affiliation increased from 11% to 14% (a gain of 27%); and finally,. the proportion of men rated as "low reintegration potential" also increased from 33% to 41% (an increase of 25%).

 

More federal men in custody in 2002 had prior Youth and Provincial Adult records than before, and a greater proportion had previous breaches of trust, including: previous disciplinary segregation (up from 26% to 34% - a 30% gain), prior escapes (up from 23% to 28% - a 22% gain), and more prior failures on conditional release. These higher rates of failure, coming mainly from prior Youth and/or Provincial Adult convictions, can be expected to carry forward into the federal system since past behavior is generally a good predictor of future behaviour. As the proportion of federal admissions who are already recidivists from prior Youth and Provincial sentences increases, so too will the challenge of reducing federal recidivism rates.

Other indicators of higher-risk characteristics have also increased. Although the proportion of men serving a sentence for homicide in 2002 increased substantially over 1997, the average sentence was bi-modal — a greater proportion of men were either serving shorter sentences than previously, or they were serving a life or indeterminate sentence.

The proportion of men with an indicator identified needs on the dynamic factor assessments ratings was generally lower. However, a larger proportion had "considerable need" for improvement in the area of personal/emotional orientation. The proportion of men identified with problems in the five key areas of employment and education, marital and family functioning, criminal associates, substance abuse, cognitive skills or mental health has also increased since 1997.

In summary, the evidence reviewed here does support the contention that there has been a general "hardening" of the population of men in federal prisons since 1997.

Women in Custody

The average number of women in-custody on any given day has increased significantly since our base year of 1997. The population has not only been growing, which poses challenges in its own right, but the composition has also changed in ways that may pose greater challenges for correctional managers and staff.

More custodial women have had a Maximum or Minimum Security rating at intake, while a smaller proportion have a Medium rating. Moreover, the proportion of women rated CRS-High Institutional Adjustment at intake has increased from 6 to 9% (a gain of 50%). There has also been an increase in gang affiliation among women prisoners.

Federal women in 2002 also tended to have had greater prior contact with the Youth and Provincial Adult correctional systems than before, and a greater proportion had previous breaches of trust, especially disciplinary segregations and more failures on conditional release.

In general the women were serving shorter sentences, while the proportion with robbery or drug convictions was greater than previously.

Overall Dynamic Factor domain assessments ratings were generally were down but at the same time a larger proportion had "considerable need" for improvement in two areas -- substance abuse and personal/emotional issues. There was also an increase in the proportion of women with an identified problem in one or more of the five specific indicator areas (of employment and education, marital, associates, substance abuse, cognitive and mental health issues).

The examination of admission cohorts was found to generally reflect and reinforce the trends and patterns found for the custody snapshots.

In summary, the widespread belief amongst operational managers and staff that there has been a "hardening" of the federal offender population is supported by the changes in the risk and needs profiles we have identified in this report.

Standard offender assessments are indicating a general increase in criminal history and associated static (risk) factor indicators. This was true, both for men and women, both between the two snapshot profiles as well as the two admission cohorts.

A: Has there been a Change in the Federal Prison population

This report addresses whether there has been a change "of the federal offender population over the past five years. To address this issue, we have developed profiles of both the admission and the in-custody populations, from admission records and inmate snapshots at two suitable intervals.

B: Methodology

The study reports on the profiles of two types of federal populations, each compared against each other at two standard points in time:

1. Federal inmate in-custody snapshots are compared for March 31 1997 and 2002, and
2. Warrant of committal admission cohorts are compared for FY 1996/97 and 2001/02.

1). In-custody snapshots:

The snapshots are used to compare the composition of the federal inmate populations on March 31^st 1997 with those on March 31^st 2002, while admissions are examined comparing Fiscal Year 1996/97 and 2001/02 cohorts. The first set of comparisons represents changes to the prison stocks, while the latter represent trends in prison flows. Generally, flow characteristics change faster than stock populations, but persistent changes in flows are almost always indicative of the direction the stock characteristics are heading.

March 31, 1997 was chosen as the base for comparing the two prison snapshots. This choice is based on several considerations: first, 1997 represents a reasonable period (5-years counting back from 2002) over which to adequately gauge longer-term trends. Second, the year 1997 represents a turning point in Correctional Service of Canada's (CSC) inmate and admission population growth trajectory. Prison and admission counts had reached their maximum size by 1997, after a period of rapid growth from 1990. Since about 1997, the federal population has been experiencing a gradual but significant decrease in size. It is important to know whether the decrease in population size has also been accompanied by a decrease in the average level of federal offender risk and needs.

Finally, there is one other technical considerations: it would be difficult to analyze a prison snapshot of inmates prior to March 1997, as the proportion of offenders without an OIA assessment increases rapidly in earlier years. For federal offenders with a determinate sentence, the average sentence length is about 4 years. Because the Offender Intake Assessment (OIA) process was first implemented in November 1994, offenders admitted before November 1994 will not have OIA assessments. Only a very small proportion of the offenders in-custody on March 31^st 1995 would have been admitted and assessed after November 1994, but of course the proportion with an OIA assessment increases quite rapidly, and after four years, most offenders in custody will have had an OIA assessment. Moreover, in 1996, the Service conducted a "backfill" to assign OIA indicators to the custodial population admitted prior to OIA implementation. Thus, by March 1997, about two-thirds of the in-custody population had either a current admission date after November 1994 and therefore have either an OIA indicator-level assessment or have a domain-level assessment from the 1996 backfill operation.

2). Admission cohorts:

To examine changes in the federal admission population, we selected admission cohorts of all federal Warrant of Committal admissions for fiscal years (FY) 1996/97 and 2001/02. The admission cohorts provide greater insight into the flow of offenders into the stock population each year from Canadian courts. In general, the proportion of inmates with shorter sentences is greater amongst admission cohorts than in the in-custody snapshots, where the turnover of inmates with longer sentences is much slower (thus they tend to cumulate in the stock population whereas the turnover for short-sentenced inmates is fairly rapid).

3). OIA Indicators:

As noted, the profiles compare information on each offender, collected through the various assessment processes that comprise the Offender Intake Assessment (OIA) process including pre-OIA assessment tools such as: 1) the Custody Rating Scale (CRS) and; 2) the Statistical Information on Recidivism - Revised 1 (SIR-R1) scale. The CSC initiates a standard OIA assessment process for each new offender at the beginning of every federal sentence, with universal assessment information being acquired and logged into OMS (the Offender Management System).

Each federal inmate is assessed at intake to determine the risks s/he poses to the public, to the security of the institution, to staff or other inmates, and to themselves. As well, new admissions undergo a battery of assessments along standard needs domains (dynamic factors) which research has shown have criminogenic potency. These latter items especially, are used to develop a correctional plan that may include correctional treatment programs to enhance the likelihood of the offender being successfully returned to the community when his/her sentence has expired or they are granted a conditional community release.

These dynamic factors include assessments of the contribution to criminogenic behavior likely attributable to problems related to education and employment, marital/family issues, associates, substance abuse, community functioning, personal/emotional issues and attitude (these seven areas constitute the criminogenic needs "domains"). Within each of these domains are a battery of standard checklist indicators that collectively provide a gauge of the specific problems and issues that the offender might have. Based on this information, a correctional plan is developed with each offender to identify treatment or programs that might assist in addressing these issues.

For the purpose of this profile, we report separately on the proportion of offenders for whom the OIA dynamic factor assessment indicated a "considerable need for improvement" on an item (i.e., indicating the possible need for an intensive or moderate level of programming) and in an additional table for those where "some need for improvement" is indicated (i.e., suggesting the potential need for lower-intensity programs).

The following offender assessment information (covering both OIA static and dynamic factors as well as standard pre-OIA assessment indicators) has been used to construct the inmate profiles:

Table 1: List of Standard Assessment Characteristics

1. Initial Custody Rating Score	9a. Domain Rating – Dynamic Factors (Some Need)
2. Risk Factors	9b. Domain Rating – Dynamic Factors (Considerable Need)
3. Prior Sentence History	10. Employment / Education Indicators
4. Previous Breaches of Trust	11. Marital / Family Indicators
5. Length of Sentence	12. Associates Indicator
6. Major Offence (Current Sentence)	13. Alcohol / Drug Abuse Indicators
7. Global Rating – Static Factors	14. Cognition Indicators
8. Global Rating – Dynamic Factors	15. Mental Health Indicators

 

Note that the "Risk Factors" included in number 2 above are selections from some of the indicators that were tested in a multivariate statistical model developed by the Research Branch to predict major incidents in Maximum-security institutions, and which in have proven most potent for that purpose.

4). OIA assessment under-coverage estimates

Complete OIA assessments, especially at the indicator level, are not always available for federal offenders. For offenders in the admitting cohorts since November 1994, coverage for OIA indicators is more than 95% and generally approaches 99%, as we get closer to today’s date.

OIA indicator-level assessments (i.e. tables 10 through 14) are missing for about 40% of the population who were in-custody as of March 31, 1997. The missing OIA assessments arise – as has been mentioned - because OIA was not implemented until November 15, 1995 (inmates in-custody on the March 1997 snapshot date, who were admitted prior to November 15^th 1994, will not have complete indicator-level OIA assessments).

For institutional snapshots, the proportion of historical cases with a missing OIA indicator-level assessment score in 1997 represents 40% of the March-31^st snapshot.⁹ This reflects the fact that a significant proportion of inmates in the March 1997 snapshot had been admitted prior to OIA implementation (November 1994) and had therefore never received an OIA intake assessment. There are also some 19% of offenders who are missing OIA indicator-level assessments for the 2002, because by then there were far fewer inmates in-custody who had originally been admitted before November 1994. Under-coverage issues are discussed more fully in Appendix B.

  

C: Overview of Federal Inmate Population

Correctional Service of Canada (CSC) operates 42 52 federal facilities or penitentiaries consisting of four different security levels across Canada. A further 17 Community Correctional Centres are also operated by the CSC, but these are administered by community corrections and are covered in separate profiles that are being developed for the community supervision population.

Table 2: Federal Correctional Facilities by Security Level.

Minimum Security	Medium Security	Maximum Security	Multi-Level Security
15	20	9	8
Source: Adult Correctional Services in Canada, 1999-2000 (Canadian Centre for Justice Statistics)

 

Of these 42 52 federal facilities, five are regional Federal Women’s Institutions, one is a women’s healing lodge, and the remainder are facilities for men. At the time of the two snapshots, federal women offenders in British Columbia are were accommodated in a Provincial women’s institution (Burnaby Correctional Centre for Women) under an Exchange of Services Agreement with the Provincial authority. The federal women’s institutions are classified as multi-level, as are several regional facilities for offenders with psychiatric needs.

Currently, aA small number of federal women offenders classified as maximum-security are currently housed separately within the existing facilities designated primarily for men. The Service is currently presently adding capacity to its federally sentenced women’s facilities to accommodate maximum-security federal women.

Among federal men’s institutions, three broad security levels are identified: minimum, medium, and maximum. Within the three broad security levels, the Service also operates specialized regional facilities for offenders with psychiatric needs and a Special Handling Unit (SHU) to manage offenders who must be segregated from the general inmate populations. The security level of federal correctional facilities is shown in Table 3.

The total number of federal inmates in-custody on March 31^st 2002 was significantly lower than the comparable number in 1997. Therefore it is more appropriate to compare the proportional distributions (which are shown along with the aggregate numbers) of the earlier and current periods.

The population of federal men shown as in-custody in OMS has decreased from 1997 to 2002 which is in contrast with the increase shown for women in federal custody during the same period.

Table 3: Size and Composition of the Snapshot Population

	April 2002 Snapshot		April 1997 Snapshot
Men total	12,285	976.29%	13,819	97.67%
Women total	351	2.87%	331	2.3%
Total	12,636	100%	14,150	100%

Source: Weekly Population records extracted from OMS.

• The daily count of federal men decreased by approximately 1,534 inmates (13,819 to 12,285) reflecting an 11% overall decrease from the count of males in March 1997.
• The daily count of federal women increased during the period from 331 to 351 inmates (a gain of 20 inmates or approximately +6% over the period).

The number — hence the percentage — being reported for each profile table (below) is representative only of those cases with a valid assessment on that indicator, thus not the total population in that group. It may be misleading to interpret the numerical changes between the two snapshots as an indication of the actual increase or decrease in the number of offenders with that characteristic.

 D: Profiles of Federal Inmates "In-Custody"

The Research Branch has developed Institutional snapshots to cover all federal in-custody inmate populations on two comparable days, March 31^st 2002 and March 31^st 1997. To facilitate analysies, separate profiles are shown for men and women. In the following tables, the characteristics of federal inmates are shown for the in-custody population at each of the two population snapshots.

D1: Institutional Profile of Federally Sentenced Men

Note: The statistical significance of a change reported in an indicator between 1997 and 2002 is indicated by showing the results of a Chi-square test of significance (associated either with each item, or with the overall table where items form part of a scale): the level of significance is indicated as follows: *** (p< .001); ** (p< .01); * (p< .05); ^(n.s.) Not significant.

1.Initial Custody Rating Score

1. Initial Custody Rating Score ***
	March 2002		March 1997
Men	Number	%	Number	%
Minimum	1,958	18	1,715	16
Medium	6,846	61	7,365	70
Maximum	2,347	21	1,458	14

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

Significant (p< .001) overall changes were found in the Custody Ratings scores for men in institutions, between 1997 and 2002. The proportion of federal inmates who were rated Maximum security at intake increased from 14% to 21% (a 50% gain) and the proportion of federal men with a Minimum-security rating also increased slightly, from 16% to 18% between 1997 and 2002. These increases were offset as the population rated as Medium-security decreased from 70% to 61%.

2.Risk Factors

2. Risk Factors
	March 2002		March 1997
Men	Number	%	Number	%
Inmates Under Age 30 ***	3,666	30	4,434	32
SIR Score - High Risk **	2,732	27	2,959	25
CRS - High Institutional Adjustment Risk ***	1,416	13	701	7
CRS - High Security Risk***	4,673	42	3,883	37
Reintegration Potential Low***	4,391	41	3,300	33
Gang Affiliation***	1,218	14	821	12

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

The proportion of men in the snapshots that were Under 30 Years of Age decreased significantly (p< .001), from 32% in 1997 to 30% in 2002.

The proportion of the maximum-security population rated as SIR Score - High Risk increased moderately (p< .01) from 25% to 27%.

The proportion that was were rated asCRS - High Institutional Adjustment Risk on the Custody Rating Scale (CRS) nearly doubled, increasing significantly from 7% to 13% (p< .001). The proportion that was CRS -assessed as High Security Risk on the CRS also increased significantly (p< .001), from 37% to 42%.

The proportion rated as having Low Reintegration Potential Low increased significantly (p< .001) from 33% in 1997 to 41% in 2002.

The proportion indicated with a Gang Affiliation increased significantly (p< .001) from 12% in 1997 to 14% in 2001.

3.Prior Sentence History

3. Prior Sentence History
	March 2002		March 1997
Men	Number	%	Number	%
Served a prior federal sentence***	3,048	34	2,012	29
Served Prior Youth and/or Adult Sentence***	8,149	91	6,141	88
Served a prior Youth Sentence***	4,359	49	2,912	42
Served prior Adult Provincial Sentence***	6,558	73	4,844	70

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

The proportion of inmates in 2002 who had Served a prior federal sentence increased from 29% to 34% (p< .001).

The proportion that had Served a prior Youth Sentence also has increased significantly (p< .001), from 42% to 49% as has the proportion who Served prior Adult Provincial Sentences, from 70% to 73% (p< .001).

Thus, the proportion having Served Prior Youth and/or Adult Sentence combined has also increased (p< .001) from 88% to 91%.

 

4.Previous Breaches of Trust

4. Previous Breaches of Trust
	March 2002		March 1997
Men	Number	%	Number	%
Previous Breaches of Trust (n.s.)	1,946	22	1,482	22
Segregation (disciplinary)***	2,941	34	1,727	26
Escapes/UAL***	2,494	28	1,620	23
Re-classified to higher secrity***	1,809	21	1,108	17
Failure on Conditional Release***	3,812	43	,460	36

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

The proportion of men in maximum-security with some Previous breaches of trust remained unchanged at 22%. Those showing a prior Segregation (disciplinary) increased significantly (p< .001), from 26% to 34% (a gain of 31%) during this period, as have those with an Escape/UAL history (p< .001) from 23% to 28%; those Re-classified to higher security increased from 17% to 21% (p< .001) and those with Failures on Conditional Release increased from 36% to 43% (p< .001).

 

5.Length of Sentence

5. Length of Sentence***
	March 2002		March 1997
Men	Number	%	Number	%
Under three years	2,640	21	2,713	20
Three to six years	3,676	30	4,294	31
Six to ten years	1,662	14	2,194	16
Ten years or more	1,589	13	2,096	15
Life or indeterminate	2,718	22	2,522	18

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

Overall, there were found to be significant (p< .001) changes in the Length of Sentence being serves served by inmates in 2002 versus 1997. However, most of this change involved an increase in the proportion of inmates serving a Life or indeterminate sentence from 18% to 22% (a 22% gain). The changes in other categories was much more modest: the proportion serving a sentence Under three years, increased only slightly (from 20% to 21%) while the proportion serving Ten years or more decreased from 15% to 13%.

 

6.Major Offence (Current Sentence)

6. Major Offences (current sentence)
	March 2002		March 1997
Men	Number	%	Number	%
Homicide***	3,071	25	2,944	21
Sexual Offences***	2,255	18	2,942	21
Robbery*	4,396	36	4,792	35
Drugs**	2,729	22	2,889	21

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

The proportion with a conviction for Homicide showed a significant (p< .001) increased from 21% to 25% (a 19% gain).; Tthe proportion with a Sexual Offence also decreased significantly (p< .001), from 21% to 18%; while those with a sentence for Robbery increased from marginally (p< .05) (from 35% to 36%). The increases in the those with Drug offences population was modest (p< .01), up from 21% to 22%.

 

7.Global Rating – Static Factors

7. Global Rating – Static Factors *
	March 2002		March 1997
Men	Number	%	Number	%
Low Risk	728	6	883	7
Moderate Risk	3,635	32	4,136	33
High Risk	7,045	62	7,549	60

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

The overall Global Rating – Static Factors scores for the population of federal men in maximum-security underwent only modest changes (p< .05 overall) from 1997 to 2002. The proportion with a low-risk or moderate-risk assessment decreased (from 7% to 6%, and from 33% to 32% respectively), and there was a corresponding increase in the proportion rated High Risk (from 60% to 62%).

8.Global Rating – Dynamic Factors

8. Global Rating – Dynamic Factors ***
	March 2002		March 1997
Men	Number	%	Number	%
Low Need	491	4	744	6
Moderate Need	3,103	27	3,923	31
High Need	7,814	68	7,901	63

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

The overall Global Rating – Dynamic Factors profile for men changed significantly (p< .001) over the five years. The proportion of men rated High Need, gained nearly 8%, increasing from 63% to 68%. This increase was offset by smaller decreases in both the proportions rated as Moderate and Low need (down from 31% to 27%, and from 6% to 4% respectively, p<.001. 001).

9-a.Domain Rating - Dynamic Factors (Some Need)

9a. Domain Rating – Dynamic Factors (Some Need for Improvement)
	March 2002		March 1997
Men	Number	%	Number	%
Employment ***	4,188	37	5,045	40
Marital/Family***	3,074	27	4,186	33
Associates (n.s.)	4,319	38	4,785	38
Substance Abuse (n.s.)	2,347	21	2,603	21
Community Functioning***	3,486	31	5,486	44
Personal/Emotional***	2,409	21	3,154	25
Attitude (n.s.)	3,214	28	3,624	29

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.)

 

Of the seven Domain Rating - Dynamic Factors, we found the proportion rated as requiring "some need for improvement" decreased significantly (p< .001) for the offollowing domains: Employment; Marital/Family; Community Functioning, and; Personal/Emotional.

 

 

9-b.Domain Rating - Dynamic Factors (Considerable Need)

The three exceptions were the Associates, Substance Abuse and Associates domains, which remained largely unchanged.

 

9a. Domain Rating – Dynamic Factors
(Considerable Need for Improvement)*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

9b. Domain Rating – Dynamic Factors (Considerable need for Improvement)
	March 2002		March 1997
Men	Number	%	Number	%
Employment***	2,115	19	4,515	36
Marital/Family***	2,645	23	3,916	31
Associates***	3,094	27	4,226	34
Substance Abuse***	5,975	52	6,866	55
Community Functioning***	1,361	12	2,741	22
Personal/Emotional***	8,163	72	8,535	68
Attitude***	4,107	36	4,906	39

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.)

 

Examining the proportion of men rated as requiring "considerable need" for improvement in the seven Domain Rating - Dynamic Factors, we found significant decreases were found for the domains of Employment,; Marital/Family,; Associates,; Substance Abuse,; Community Functioning,; and, Attitude a significant decrease (p< . 001). The sole exception was the Personal/Emotional domain, which showed a significant (p< .001) increase from 68% to 72% (p< .001).

 

10.Employment / Education Indicators

10. Employment / Education Indicators
	March 2002		March 1997
Federal Men	Number	%	Number	%
Has an Unstable Job History ***	6,684	73	5,004	70
Has No High School Diploma (n.s.)	7,149	79	5,659	79

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

A significantly larger proportion of men in March 2002 Had an Unstable Job History (increasing from 70% to 73%, p< .001). The education indicator remained unchanged at 79% for both periods.

11.Marital/Family Indicators

11. Marital / Family Indicators
	March 2002		March 1997
Men	Number	%	Number	%
Currently Single***	5,975	65	4,332	60
Had Dysfunctional Parents **	4,595	51	3,415	49

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

Both of the chosen Marital / Family Indicators showed an increases, with the proportion of men who were Currently Single increasing significantly (p< .001) from 60% to 65%, and the proportion who Had Dysfunctional Parents increaseding moderately (p< .01), from 49% to 51%.

12.Associates / Social Interaction Indicators

12. Associates / Social Interaction Indicators
	March 2002		March 1997
Federal Men	Number	%	Number	%
Gang Affiliation ***	1,218	14	821	12
Mostly Criminal Friends ***	3,985	44	2,806	41
Criminogenic Living ***	2,524	29	1,789	26

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

All three of the Associates / Social Interaction Indicators showed a significant (p< .001) increase: Gang Affiliation increased from 12% to 14%; the proportion that had Mostly Criminal Friends increased from 41% to 44%; and the proportion who had been engaged in mostly Criminogenic Living also increased significantly (from 26% to 29%).

 

13.Alcohol/Drug Abuse Indicators

13. Alcohol/Drug Abuse Indicators
	March 2002		March 1997
Men	Number	%	Number	%
Abuses Alcohol*	5,411	59	4,339	61
Abuses Drugs **	6,144	67	4,662	65
Abuses Drugs and Alcohol *	7,720	79	5,887	78

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

Offenders with substance abuse issues continuesOffenders with substance abuse issues continue to poses serious challenges., not least for staff in institutions because of the continuous attempts by inmates to obtain illicit drug and alcohol in prisons. The proportion of the men at intake that Abuses Alcohol decreased slightly (from 61% to 59%: p< .05), whereas the proportion that Abuses Drugs increased modestly (p< .01) from 65% to 67%. The proportion that Abuse Drugs and Alcohol combined therefore increased slightly (p< .05) from 78% to 79%.

14.Cognition Indicators

14. Cognition Indicators
	March 2002		March 1997
Men	Number	%	Number	%
Poor Problem Solving***	7,261	80	5,116	72
Unable to Generate Choices***	6,536	72	4,261	61
Impulsive***	7,120	78	4,924	70

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

Managing offenders who are Poor at Problem Solving, Unable to Generate Wise Choices, and Impulsive, also pose programming and re-integration challenges. The proportion of men rated as having specific the above specifiedabove-specified cognitive issues all increased significantly (p< .001) during this period; from 72% to 80%; from 61% to 72%; and from 70% to 78% respectively.

15.Mental Health Indicators

15. Mental Health Indicators
	March 2002		March 1997
Men	Number	%	Number	%
Has a Past Diagnosis***	1331	15	725	10
Has Current Diagnosis***	875	10	492	7
Prescribed Medication Currently***	1449	16	667	9

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

Lastly, the proportion of men with serious mental health issues is a growing challenge for the CSC. The proportion of inmates in 2002 with: a Past Diagnosis increased significantly (p< .001) from 10% to 15% (a 50% gain in just five years); as was did the proportion with a Current Diagnosis from 7% to 10% (a 43% gain in this period: p< .001); and the proportion Prescribed Medication Currently from 9% to 16% (a 78% gain: p< .001). Other research currently being conducted by the Branch, examining the total population using an even wider array of mental health indicators, has found that similar trends to exist for most indicators.

Summary

The average number of federal men in-custody has decreased since the base year of 1997

Note:Individuals could appear in more than one category, therefore numbers won't add up.

. However, those that remain definitely constitute a higher risk and needs group. The composition of the population has changed in ways that likely make it more difficult to manage – and to reintegrate back safely into the community – than previously.

More men had a Maximum Security rating at intake, the proportion increased from 14% to 21% (representing a gain of 50%), and the absolute number in our the sample increased from 1,500 to nearly 2,350 inmates. Moreover, the proportion of men rated CRS-High Institutional Adjustment and High Security Risk has also increased substantially. There were also significant increases in the proportion of men with a gang affiliation (as identified at intake).

Federal men in 2002 also had significantly more prior contact with the Youth and Provincial Adult correctional systems, and a greater proportion had previous breaches of trust, including previous disciplinary segregation, escapes, and more with a prior failure on conditional release., etc.

On average, the male inmates were either serving shorter sentences than previously, or they were serving a life or indeterminate sentence and the proportion with a homicide conviction has increased substantially.

Dynamic factor assessments ratings were lower overall, however but in the areas of personal/emotional issues, there were showed a larger proportions that had been assessed as having "considerable need for improvement". The proportion of men with identified issues in the five specific areas examined — education and employment, marital/family, associates, substance abuse, cognition skillspersonal emotional orientation and mental health issues — has also increased significantly since 1997.

 

D2: Institutional Profile of Federally Sentenced Women

Note: Statistical tests of significance for changes in the composition of women offenders often return "non-significant" results, even though the proportional changes appear to be quite large. This is largely due to the smaller base population being tested. Thus, even a 10% change for women may not show as statistically significant, but of course may have a significant impact on the correctional system (significance then becomes a matter of judgment).

 

1.Initial Custody Rating Scores

1. Initial Custody Rating Score (n.s.)
	March 2002		March 1997
Women	Number	%	Number	%
Minimum	103	32	62	26
Medium	174	53	149	62
Maximum	49	15	28	12

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

The Initial Custody Rating Scores for women inmates show there has been a rising proportion of women rated at intake as Maximum-security: up from 12% in 1997 to 15% in 2002 (a gain of 20%). The proportion with an initial Minimum-security rating has also increased (from 26% to 32% — a 24% gain). Therefore, the proportion rated medium-security has been eroding, decreasing from 62% to 53% of the woman inmate population.

 

2.Risk Factors

2. Risk Factors
	March 2002		March 1997
Women	Number	%	Number	%
Inmates Under Age 30 (n.s.)	118	34	124	37
SIR Score - High Risk	-	-	-	-
CRS - High Institutional Adjustment Risk(n.s.)	30	9	14	6
CRS - High Security Risk (n.s.)	84	26	69	29
Reintegration Potential Low (n.s.)	82	27	49	24
Gang Affiliation (n.s.)	31	10	16	8

* Note: The SIR Scale is traditionally not applied to women offenders.

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

Between 1997 and 2002 the proportion of women who were Under 30 Years of Age has decreased,: declining from 37% in 1997 to 34% in 2002.

The proportion of women who were rated as having CRS - High Institutional Adjustment Risk on the Custody Rating Scale (CRS) increased during between the two the periods, from 6% to 9% (a gain of 50%), but the proportion that were CRS - High Security Risk decreased, falling from 29% to 26% (down 10%).

The proportion of federally sentenced women that were rated at intake as having Low Reintegration Potential-Low increased, from 24% to 27%.

The number of women affiliated with gangs increased by 20%, as the proportion indicating a Gang Affiliation rose from 8% to 10%.

 

3.Prior Sentence History

3. Prior Sentence History
	March 2002		March 1997
Women	Number	%	Number	%
Served a prior federal sentence (n.s.)	33	11	32	16
Served Prior Youth and/or Adult Sentence (n.s.)	231	76	146	72
Served a prior Youth Sentence (n.s.)	100	33	57	28
Served prior Adult Provincial Sentence (n.s.)	169	56	100	50

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

Between 1997 and 2002 the proportion of female women inmates who had Served a prior federal sentence decreased from 16% in 1997 to 11% in 2002.

The proportion that had Served a prior Youth Sentence increased from 28% to 33% (a gain of 18%), as did the proportion that had Served prior Adult Provincial Sentence - 50% to 56% (a gain of 12%). Combined, the number of women recidivists in federal facilities has grown as the proportion that had Served Prior Youth and/or Adult Sentences increased from 72% to 76% over the period.

 

4.Previous Breaches of Trust

4. Previous Breaches of Trust
	March 2002		March 1997
Women	Number	%	Number	%
Previous Breaches of Trust*	26	9	9	4
Segregation (disciplinary) (n.s.)	74	26	40	20
Escapes/UAL (n.s.)	52	17	35	17
Re-classified to higher security (n.s.)	23	8	18	9
Failure on Conditional Release (n.s.)	79	26	43	21

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

The proportion of women that had Previous Breaches of Trust more than doubled (from 4% to 9%: p< .05) between 1997 and 2002.

The proportion that had has a prior Segregation (disciplinary) increased from 20% to 26% (a gain of 30%). The proportion with a previous Escapes/UAL remained unchanged (17%).

The proportion of women offenders Re-classified to higher security decreased slightly (9% to 8%).

The proportion of women offenders with a previous Failure on Conditional Release increased from 21% to 26%).

 

5.Length of Sentence

5. Length of Sentence *
	March 2002		March 1997
Women	Number	%	Number	%
Under three years	129	37	100	30
Three to six years	97	28	91	27
Six to ten years	42	12	45	14
Ten years or more	13	4	24	7
Life or indeterminate	70	20	71	21

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

There were some significant changes (p< .05) in sentence length, with the proportion of women serving a sentence Under Three Years increasing from 30% in 1997 to 37% in 2002 (a 5-year gain of 23%), and the proportion sentenced to Three to six years, which also increased, from 27% to 28%. All the longer sentenced categories showed a decrease during this period (the largest decrease was in the proportion serving Ten Years or More, which declined from 7% to 4% or by 43%).

 

6.Major Offence (Current Sentence)

6. Major Offences (current sentence)
	March 2002		March 1997
Women	Number	%	Number	%
Homicide (n.s.)	103	29	114	34
Sexual Offence (n.s.)s	8	2	8	2
Robbery (n.s.)	87	25	68	21
Drugs (n.s.)	104	30	88	27

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

Among women inmates, the principle change in the Major Offence was a large decrease in Homicide (decreased from 34% to 29%, or by 15%), with a commensurate increase in Robbery (up from 21% to 25%), and Drugs (increase from 27% to 30%).

7.Global Rating - Static Factors

7. Global Rating - Static Factors *
	March 2002		March 1997
Women	Number	%	Number	%
Low Risk	75	23	72	28
Moderate Risk	156	48	100	39
High Risk	94	29	85	33

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

Changes to the overall Global Rating - Static Factors scores for women were significant (p< .05): the population of federal women rated Moderate Risk showed a large increase, from 39% to 48% (a 5-year gain of 23%). This increase was offset by smaller decreases, almost equally divided, in the proportion of women rated High and Low Risk (28% to 23%, and 33% to 29% respectively). However, because the overall number of women inmates increased during this period, the number of women inmates rated as Low or High Risk in 2002 was actually greater than it had been in 1997.

8. Global Rating – Dynamic Factors

8. Global Rating – Dynamic Factors *
	March 2002		March 1997
Women	Number	%	Number	%
Low Need	38	12	49	19
Moderate Need	126	39	97	38
High Need	161	50	111	43

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

The Global Rating – Dynamic Factors profile for women changed significantly (p< .05) over the five years. The proportion of women rated High Need, gained 16%, increasing from 43% to 50%. This increase was offset by a decreasesby decreases in the proportion as Low Need (down from 19% to 12%, while women assessed as Moderate Need remained relatively unchanged.

 

9-a.Domain Ratings - Dynamic Factors (Some need)

9a. Domain Rating – Dynamic Factors (Some Need for Improvement)
	March 2002		March 1997
Women	Number	%	Number	%
Employment**	133	41	136	53
Marital/Family *	125	38	123	48
Associates ***	135	42	149	58
Substance Abuse *	49	15	56	22
Community Functioning ***	102	31	154	60
Personal/Emotional ***	94	29	124	48
Attitude **	85	26	45	18

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

Analysis of Domain Ratings - Dynamic Factors found significant decreases in the proportion rated as requiring "some need" for improvement in the domains of Employment (p< .01), Marital/Family (p< .05); Associates (p< .001); Substance Abuse (p< .05); Community Functioning (p< .001), and Personal/Emotional (p< .001). The only domain to show an increase (up from 18% to 26%) was in the Attitude domain (p< .01).

 

9-b.Domain Ratings -Dynamic Factors (Considerable need)

9b. Domain Rating – Dynamic Factors (Considerable Need for Improvement)
	March 2002		March 1997
Women	Number	%	Number	%
Employment **	50	15	62	24
Marital/Family (n.s.)	87	27	75	29
Associates (n.s.)	58	18	45	18
Substance Abuse ***	181	56	109	42
Community Functioning	22	7	21	8
Personal/Emotional ***	205	63	110	43
Attitude (n.s.)	45	14	32	12

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

The proportion of women in federal custody in 2002 with ratings of "considerable need" on the various dynamic factor domains has mainly decreased. However, a detailed analysis of Domain Ratings -Dynamic Factors found significant increases in the dynamic factor domains of Substance Abuse (p<. 001) and Personal/Emotional (p< .001). There was also a decrease in the Employment (p< .01) domain.

 

 

10. Employment / Education Indicators

10. Employment / Education Indicators
	March 2002		March 1997
Women Offenders	Number	%	Number	%
Has an Unstable Job History (n.s.)	221	71	140	66
Has No High School Diploma (n.s.)	222	71	143	68

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

An examination of the two Employment / Education Indicators found both had increased modestly since 1997. There was less movement in the education than in the employment variable: the proportion of women that Has an Unstable Job History increased from 66% to 71% (an 8% gain), while the proportion who Has No High School Diploma increased from 68% to 71% (a 4% gain).

11. Marital / Family Indicators

11. Marital / Family Indicators
	March 2002		March 1997
Women Offenders	Number	%	Number	%
Currently Single *	173	55	135	64
Had Dysfunctional Parents (n.s.)	167	55	109	53

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

Changes in the two Marital / Family Indicators were modest: The proportion of women Currently Single decreased moderately (p< .05) from 64% to 55% while the proportion who Had Dysfunctional Parents increased but the change was not significant.

12. Associates / Social Interaction Indicators

12. Associates / Social Interaction Indicators
	March 2002		March 1997
Women Offenders	Number	%	Number	%
Gang Affiliation (n.s.)	31	10	16	8
Mostly Criminal Friends (n.s.)	99	32	64	31
Criminogenic Living (n.s.)	86	28	65	32

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

There were no significant changes in the three Associates / Social Interaction Indicators for women.

13.Alcohol/Drug Abuse Indicators

13. Alcohol/Drug Abuse Indicators
	March 2002		March 1997
Women	Number	%	Number	%
Abuses Alcohol (n.s.)	166	53	109	51
Abuses Drugs **	215	69	125	59
Abuses Drugs and Alcohol **	263	79	172	70

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

A majority of women in federal custody have identified substance abuse problems, and this proportion has been increasing quite significantly. The proportion of women inmates who Abused Alcohol increased slightly (from 51% to 53%, n.s.), however the proportion that Abused Drugs increased substantially (p< .01), from 59% in 1997 to 69% (a 17% gain) in 2002.

14.Cognition Indicators

14. Cognition Indicators
	March 2002		March 1997
Women	Number	%	Number	%
Poor Problem Solving (n.s.)	187	60	118	56
Unable to Generate Choices **	164	53	134	64
Impulsive (n.s.)	213	68	125	59

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

A majority of women in federal custody also have identified cognitive skills issues, which also poses challenges. There were increases in the proportion of women inmates who were rated to be Poor at Problem Solving (56% in 1997 versus 60% in 2001), and Impulsive (59% to 68%:). However, a significant decrease (64% to 53% a 17% decline) was found among women inmates identified as Unable to Generate Choices.

15. Mental Health Indicators

15. Mental Health Indicators
	March 2002		March 1997
Women	Number	%	Number	%
Has a Past Diagnosis *	71	23	41	20
Has Current Diagnosis (n.s.)	48	16	28	13
Prescribed Medication Currently (n.s.)	105	34	68	32

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

Finally, a smaller but not insignificant number of federal women inmates have mental health issues, and both the number and proportion has been increasing quite significantly. The proportion of admissions with a Past Diagnosis increased from 20% to 23% (a gain of 15%). Increases were also found among the proportion of women inmates with a Current Diagnosis (13% in 1997 to 16% in 2002), and with a currently Prescribed medication (from 32% in 1997 to 34% in 2002.

Summary

The average number of women in-custody on any given day grew significantly from the 1997 base year

* Sex offences have been presented separately from the schedule 1 category to illustrate their proportion in the inmate population.

Note:Individuals could appear in more than one category, therefore numbers won't add up.

. The population has not only been growing, which poses challenges in its own right, but the composition has also changed in some ways that portend greater challenges for correctional managers and staff.

A larger proportion of custodial women have had a Maximum security rating at intake, while a smaller proportion have a Medium rating. Moreover, the proportion of women rated CRS-High Institutional Adjustment at intake has increased from 6 to 9%, or from 14 to 30 inmates. There has also been an apparent increase in gang affiliation.

Federal women in 2002 also had more prior contact with the Youth and Provincial Adult correctional systems, and a greater proportion had previous breaches of trust, especially disciplinary segregations and more failures on conditional release.

In general the population were serving shorter sentences than previously, and the proportions in for robbery and drug offences was greater than previously.

Dynamic factor assessments showed a larger proportion had "considerable need" for improvement in the areas of substance abuse and personal/emotional issues. The proportion of women with identified employment and education, substance abuse, cognition and mental health issues also was up substantially.

 

 

E. Profiles of Federal (Warrant of Committal) Admissions

Admission profiles cover all Warrant of Committal admissions to federal custody for two complete fiscal year periods, FY 2001/02 and FY 1996/97. Separate profiles are shown for men and women. This section provides general background and highlights changes that are occurring in the whole federal population through the admission of offenders with a federal Warrant of Committal sentence from the courts.

Inmates enter into a federal penitentiary for the first time on a Warrant of Committal following sentence by a Court or by way of an international transfer or an exchange of service agreement with another Canadian jurisdiction. They may also be admitted into the penitentiary during their sentence if, while on community supervision, their conditional release is revoked by the National Parole Board for the commission of a new offence or for a breach of a condition of their release.

The profiles created for admissions refer to the population admitted by way of a Warrant of Committal and does not include re-admissions of offenders already serving a federal sentence. Profiles are also restricted to federal admissions (i.e., offenders with a sentence to federal jurisdiction), thus excluding those initially admitted to prison under a sentence to Provincial jurisdiction (i.e., a prison sentence of less than two years) but later transferred (under an exchange of service agreement) to a federal penitentiary.

The total number of Warrant admissions in 2001/02 was significantly lower than the number admitted in 1996/97. Thus, the analyst must take relative population size into account when making comparisons between the two profiles. It is typically easier to understand changes over this period if the proportional distributions (percentages) are considered from one period to the other rather than the raw numbers.

Although the majority of federal admissions continue to be comprised of men offenders, the number of women admitted has been increasing for the past half-decade. This increase in women admissions is contrasted with an admission decline in men offender admissions. The total number of admissions by way of a Warrant of Committal in 1996/97 was 4,652, of which 4,461 were men and 191 were women (the proportion of women admitted was 4%). This compares to the total of 4,128 Warrant admissions in 2001/02, of which the number of men was 3,925, and the 203 women admitted now accounted for 5%.

Admission Under-coverage Estimates.

• It is generally not necessary to take into account data under-coverage for the CSC’s admission population. Data coverage is very good for admission data. The CSC policy is to complete intake assessments within 8 weeks of admission. Thus there will be under-coverage issues only if a sufficient time lapse is not allowed between the end-of-year and the data collection date (for more details, see Appendix B).

 

E1: Admission Profiles of Federal Men

Overall, the number of men admitted into federal jurisdiction by way of a Warrant of Committal decreased between FY 1996/97 and FY 2001/02 (4,461 to 3,925). This represents a reduction of 536 admissions or 12% fewer admissions during FY 2001/02 than in 1996/97.

1.Initial Custody Ratings

1. Initial Custody Rating Score ***
	FY 2002		FY 1997
Men	Number	%	Number	%
Minimum	1,227	31	1,179	27
Medium	2,190	55	2,821	66
Maximum	546	14	301	7

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

There were significant changes (p< .001 overall) in the security placement ratings from the Initial Custody Ratings for men admitted in 1996/97 and 2001/02. For example, the proportion of men rated Maximum at intake increased from 7% to 14% (a 100% jump) while the proportion rated minimum-security increased from 27% to 31%. As a consequence of these increases, the proportion rated medium-security fell from 66% in 1996/97 to 55% in FY 2001/02.

2.Risk Factors

2. Risk Factors
	FY 2002		FY 1997
Men	Number	%	Number	%
Inmates Under Age 30 *	1,653	41	1,926	43
SIR Score - High Risk **	832	25	879	22
CRS - High Institutional Adjustment Risk ***	438	11	212	5
CRS - High Security Risk ***	903	23	819	19
Reintegration Potential Low ***	1,303	33	1,080	26
Gang Affiliation ***	523	14	465	11

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

The overall proportion of men who were Under Age 30 at the time of admission decreased somewhat (p< .05) from 43% in 1996/97 to 41% in 2001/02.

There was a moderate (p< .01) increase in the proportion with a rating SIR Score - High Risk, from 22% to 25%.

The proportion of CRS - High Institutional Adjustment Risk and CRS - High Security Risk both increased significantly (up from 5% to 11%, and from 19% to 23% respectively: p< .001).

The proportion rated as Reintegration Potential Low also increased significantly (p< .001), from 26% to 33%.

Finally, the proportion of admitted men with a Gang Affiliation also increased significantly (p< .001) from 11% to 14%.

3.Prior Sentence History

3. Prior Sentence History
	FY 2002		FY 1997
Men	Number	%	Number	%
Served a prior federal sentence***	1,138	29	1,064	25
Served Prior Youth and/or Adult Sentence **	3,518	89	3,650	87
Served a prior Youth Sentence***	1,867	48	1,653	40
Served prior Adult Provincial Sentence**	2,762	70	2,803	67

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

The proportion of 2001/02 admissions that has Served a prior Federal Sentence increased significantly (p< .001) from 25% to 29% in 2001/02.

Offenders admitted in 2001/02 were proportionately more likely to have had prior Young Offender and/or Provincial Adult sentences than in 1996/97. The proportion that had Served a prior Youth Sentence increased significantly (p< .001), from 40% (1996/97) to 48% (2001/02), while there was a moderate increase (p< .01) in the proportion that had Served a prior Adult Provincial Sentence - 67% to 70%. The proportions that had Served Prior Youth and/or Adult Sentence increased only modestly (p< .01), from 87% to 89%.

While these trends in prior sentence history do not in themselves represent large changes, it is clear that a very large — and growing — proportion of new admissions are prior "recidivists" from the Youth and/or Provincial Adult correctional systems before they even get in the federal penitentiary doors. The best predictor of future behaviour is usually past behavior, and these higher rates of prior Youth and/or Provincial Adult recidivism may anticipate increases in CSC’s rates of federal recidivism in the future.

4.Previous Breaches of Trust

4. Previous Breaches of Trust
	FY 2002		FY 1997
Men	Number	%	Number	%
Previous Breaches of Trust (n.s.)	794	20	795	19
Segregation (disciplinary)***	1,091	29	921	23
Escapes/UAL***	988	25	875	21
Re-classified to higher security***	646	17	562	14
Failure on Conditional Release***	1,577	41	1,379	34

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

The proportion of men admitted with Disciplinary Segregation, Escape/UAL, Re-classified to Higher Security and Failure on Conditional Release all registered a significant (p< .001) increase between 1996/97 and 2001/02. The largest increases are in the proportions with Segregation (disciplinary) (up from 23% to 29%), and in the proportion with a Failure on Conditional Release (increasing from 34% to 41%).

 

5.Length of Sentence

5. Length of Sentence***
	FY 2002		FY 1997
Men	Number	%	Number	%
Under three years	1,888	47	1,516	34
Three to six years	1,528	38	1,956	44
Six to ten years	328	8	567	13
Ten years or more	96	2	204	5
Life or indeterminate	154	4	190	4

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

Changes in the Length of Sentence distribution were significant (p< .001). Men admitted with a sentence Under Three Years accounted for most of the change, increasing from 34% to 47% (a gain of 38%), while the proportion admitted with Three to six years sentenced decreased from 44% to 38% (down 14%). All of the longer sentence categories also showed a decrease except Life or indeterminate which remained unchanged. Incidentally, since the number of lifers being admitted annually has not grown, the increase in lifers in the stock population must be due to retention.

 

6.Major Offence (Current Sentence)

6. Major Offences (current sentence)
	FY 2002		FY 1997
Men	Number	%	Number	%
Homicide *	287	7	275	6
Sexual Offences ***	523	13	783	18
Robbery (n.s.)	876	22	977	22
Drugs (n.s.)	881	22	953	21

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

There were few significant changes in the Major Offence profile, with the exception of a significantly decrease (p< .001) — from 18% to 13% — in the proportion convicted for a Sexual Offence. There was also a small (p< .05) change in the proportion admitted for a Homicide offence (a 1%-point increase).

 

7.Global Rating - Static Factors

7. Global Rating - Static Factors **
	FY 2002		FY 1997
Men	Number	%	Number	%
Low Risk	634	16	590	14
Moderate Risk	1,605	41	1,727	40
High Risk	1,713	43	1,998	46

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

For federal men, there was a moderate (p< .01) overall shift in the Global Rating - Static Factors scores. This included a small decrease in the proportion of admissions rated as High Risk, down from 46% to 43% (a 7% decrease). There were corresponding increases in the proportion rated as moderate-risk (from 40% to 41%) and Low Risk (14% to 16%) during this period.

8.Global Rating – Dynamic Factors

8. Global Rating – Dynamic Factors *
	FY 2002		FY 1997
Men	Number	%	Number	%
Low Need	401	10	507	12
Moderate Need	1,233	31	1,624	38
High Need	2,318	59	2,184	51

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

For federal men, there was a small (p< .05) overall shift in the Global Rating - Dynamic Factors scores. This included a significant increase in the proportion of admissions rated as High Risk, up from 51% to 59% (a 16% increase). There were offsetting decreases in the proportion rated as moderate-risk (from 38% to 31%) and Low Risk (12% to 10%) during this period.

9a.Domain Rating – Dynamic Factors (Some Need)

9a. Domain Rating – Dynamic Factors (Some Need for Improvement)
	FY 2002		FY 1997
Men	Number	%	Number	%
Employment ***	1,389	35	2,012	47
Marital/Family ***	884	22	1,394	32
Associates ***	1,677	42	1,646	38
Substance Abuse *	830	21	980	23
Community Functioning ***	943	24	1,779	41
Personal/Emotional ***	1,002	25	1,257	29
Attitude *	1,172	30	1,197	28

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

For most domains of the Domain Rating - Dynamic Factors table, a significant decrease in the proportion of men identified as having "some need" of improvement was found. The exception was the Associates domain (which showed a significant (p< .001) increase from 38% to 42%) and Attitude, which increased more modestly (p< .05) from 28% to 30%.

 

9b.Domain Rating – Dynamic Factors (Considerable Need)

9b. Domain Rating – Dynamic Factors (Considerable Need for Improvement)
	FY 2002		FY 1997
Men	Number	%	Number	%
Employment ***	355	9	1,198	28
Marital/Family ***	614	16	979	23
Associates ***	887	22	1,225	28
Substance Abuse (n.s.)	1,896	48	2,081	48
Community Functioning ***	180	5	557	13
Personal/Emotional (n.s.)	2,456	62	2,680	62
Attitude ***	1,196	30	1,450	34

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

Fewer men were rated as having a "considerable need" for improvement on the domains in the Domain Rating - Dynamic Factors table. An analysis of the scores indicates significant decreases (p< .001) in the proportions for the following domains: Employment; Marital/Family; Associates; Community Functioning, and; Attitude. Although they continue to represent a large proportion of male admissions, the Substance Abuse and Personal/Emotional domains remained largely unchanged.

 

10.Employment / Education Indicators

10. Employment / Education Indicators
	FY 2002		FY 1997
Men	Number	%	Number	%
Has an Unstable Job History **	2,766	70	2,877	67
Has No High School Diploma ***	2,970	76	3,351	79

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

There was a moderate (p< .01) increase in the proportion who Has an Unstable Job History (67% to 70%) but a significant (p< .001) decrease in the proportion who Has No High School Diploma (down from 79% to 76%).

11.Marital / Family Indicators

11. Marital / Family Indicators
	FY 2002		FY 1997
Men	Number	%	Number	%
Currently Single ***	2,466	63	2,484	58
Had Dysfunctional Parents (n.s.)	1,832	47	1,974	47

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

There was a significant (p< .001) increase in the proportion who were Currently Single (58% to 63%) but no-significant change in the proportion who Has Dysfunctional Parents.

 

12.Associates / Social Interaction Indicators

12. Associates / Social Interaction Indicators
	FY 2002		FY 1997
Men	Number	%	Number	%
Gang Affiliation ***	523	14	465	11
Mostly Criminal Friends ***	1,628	42	1,527	37
Criminogenic Living **	1,007	27	985	24

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

All three Associates / Social Interaction Indicators were found to have increased significantly (p< .001).

13.Substance Abuse Issues

13. Alcohol/Drug Abuse Indicators
	FY 2002		FY 1997
Men	Number	%	Number	%
Abuses Alcohol (n.s.)	2,256	57	2,475	58
Abuses Drugs **	2,639	67	2,724	64
Abuses Drugs and Alcohol (n.s.)	3,081	78	3,281	77

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

The proportion of men at admission who were assessed as having Abused Drugs increased moderately (from 64% to 67%, p< .01), but the proportion who Abused Alcohol decreased slightly (from 58% to 57%), resulting in very little change overall in the proportion who Abuses Drugs and Alcohol combined from 77% to 78% (n.s.).

14.Cognition Indicators

14. Cognition Indicators
	FY 2002		FY 1997
Men	Number	%	Number	%
Poor Problem Solving ***	3036	77	2856	68
Unable to Generate Choices ***	2823	72	2466	59
Impulsive ***	2910	74	2829	67

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

 

There was a significant increase in the proportion of men who were assessed as having cognition problems at admission. The proportion of men who are Poor at Problem Solving, Unable to Generate Wise Choices, and Impulsive have all increased significantly (p< .001): 68% to 77%; 59% to 72%; and 67% to 74% respectively.

15.Mental Health Indicators

 

 

 

15. Mental Health Indicators
	FY 2002		FY 1997
Men	Number	%	Number	%
Has a Past Diagnosis ***	513	13	353	8
Has Current Diagnosis ***	335	9	243	6
Prescribed Medication Currently ***	669	17	378	9

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

There was a significant increase (p< .001) in the proportion of men who were assessed with mental health problems at admission. The proportion of admissions in 2001/02 with a Past Diagnosis nearly doubled, rising from 8% to 13% (a 63% gain). The proportion with a Current Diagnosis also increased rapidly, from 6% to 9% (a 50% gain); and the proportion Prescribed Medication Currently increased from 9% in 1996/97 to 17% in 2001/02 (an 89% gain).

Summary

Admission trends for men generally reinforce what we have already observed for the custodial population. To summarize admission trends, the major patterns include a small decrease in the proportion of men under thirty years of age at admission; a significant increase in the proportion of men admitted with a Maximum security recommendation; higher CRS Institutional Adjustment and Security Risk scores; more men have prior gang involvement; they have more prior exposure to the Youth and Adult justice system; and more disciplinary segregation, escape and release failures than previously. The proportion serving shorter sentences (under 3 years) significantly increased while the proportion with a sexual conviction or serving longer sentences decreased. A significantly greater proportion of male admissions have criminal associates, substance abuse (drug), cognition, and mental health issues than only five years previously.

Finally, a very large — and growing — proportion of men at admission have previously served a Youth and/or Provincial Adult sentence. They are thus already "recidivists" before they even get in the penitentiary door. Since the best predictor of future behaviour is past behavior, these high rates of prior recidivism will greatly increase the challenge that the CSC faces, to reduce federal recidivism rates.

 

 

E2. Admission Profiles of Federal Women

Overall, there was an increase in the number of women admitted by way of a Warrant of Committal during the same period from 191 in FY 1996/97 to 254 in FY 2001/02. This represents an increase of 63 additional women, or nearly a 33% increase. Because of the smaller population of women who are admitted, standard statistical tests typically will not indicate significant differences where cell counts are small (such as chi-square, which is sensitive to cell counts).

Note: Care must be taken in interpreting test results for women that are non-significant as their population size is quite small and this influences the statistical test. Indeed, where for men a 2% change is usually significant, for women sometimes even a 10% or larger change may not show statistically. Of course, changes of this proportion may still have quite a significant operational impact even if not statistically significant (and this is clearly a matter of judgment).

1.Initial Custody Ratings

1. Initial Custody Rating Score **
	FY 2002		FY 1997
Women	Number	%	Number	%
Minimum	121	57	70	44
Medium	78	37	82	51
Maximum	14	7	8	5

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

The Initial Custody Ratings for women at admission has changed enough to test statistically significantly (p< .01) between 1996/97 and 2001/02. The proportion rated Minimum-security at intake increased from 44% to 57% while the proportion rated Medium-security decreased by a corresponding amount (51% to 37%). On the other hand, the proportion rated maximum-security at intake increased from 5% to 7%.

2.Risk Factors

2. Risk Factors
	FY 2002		FY 1997
Women	Number	%	Number	%
Inmates Under Age 30 (n.s.)	79	37	85	45
SIR Score - High Risk t	-	-	-	-
CRS - High Institutional Adjustment Risk *	11	5	2	1
CRS - High Security Risk *	15	7	20	12
Reintegration Potential Low (n.s.)	38	18	25	16
Gang Affiliation (n.s.)	16	8	9	5

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

t Note: It is CSC policy not to administer the SIR scale to women.

 

There was a small (not significant) decrease the proportion of women Under Age 30 at admission, 45% in 1996/97 to 37% in 2001/02.

The proportion of women admissions CRS - High Institutional Adjustment Risk increased (from 1% to 5%: p< .05) but the proportion of women CRS - High Security Risk decreased significantly (down from 12% to 7%: p< .05).

The proportion of women who were Reintegration Potential Low at admission increased from 16% in 1996/97 to 18% in 2001/02.The proportion of women with a Gang Affiliation at the time of admission also increased from 5% in 1996/97 to 8% in 2001/02 (although neither of these changes showed statistical significance).

3.Prior Sentence History

3. Prior Sentence History
	FY 2002		FY 1997
Women	Number	%	Number	%
Served a prior federal sentence (n.s.)	13	6	14	8
Served Prior Youth and/or Adult Sentence (n.s.)	156	75	120	72
Served a prior Youth Sentence (n.s.)	60	29	39	24
Served prior Adult Provincial Sentence (n.s.)	112	54	92	55

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

The proportion of women having Served a prior federal sentence decreased from 8% to 6%. The majority of woman admissions have not been in federal custody before.

However, the proportion of women who Served a Prior Youth Sentence increased from 24% to 29% (a gain of 21%), while the proportion who had Served a prior Adult Provincial Sentence decreased slightly (from 55% to over 54%), resulting in a overall increase in the combined proportion of women offenders who Served Prior Youth and/or Adult Sentences (from 72% to 75%).

 

4.Previous Breaches of Trust

4. Previous Breaches of Trust
	FY 2002		FY 1997
Women	Number	%	Number	%
Previous Breaches of Trust (n.s.)	13	6	5	3
Segregation (disciplinary) (n.s.)	37	18	27	17
Escapes/UAL (n.s.)	30	14	22	13
Re-classified to higher security (n.s.)	9	4	10	6
Failure on Conditional Release *	44	21	34	20

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

The proportion of women admitted with some prior Breech of Trust nearly doubled, from 3% to 6%, although the absolute numbers remained quite small.

Small proportional increases were shown for all the remaining (Segregation, Escapes), except for Re-classified to higher security (down from 4% to 6%) and Failure on Conditional Release the latter which increased significantly (p< .05) more in number (up from 34 to 44 admissions) than in proportion (up from 20% to 21%)

 

5.Length of Sentence

5. Length of Sentence **
	FY 2002		FY 1997
Women	Number	%	Number	%
Under three years	128	60	78	41
Three to six years	66	31	83	43
Six to ten years	13	6	16	8
Ten years or more	2	1	3	2
Life or indeterminate	6	3	11	6

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

There have been significant changes (p< .01) in the sentence distribution of women admitted during the five-year intervals. The proportion of women admissions serving a sentence Under Three Years, increasing significantly from 41% to 60% while the proportion with longer sentences all showed some decrease. The proportion entering the federal system in 2001/02 with a Life or indeterminate sentence actually fell by half in 2001/02 (down from 6% to 3%). Earlier research conducted by the Branch has found a very rapid increase in the number of women sentenced to federal custody from 1994/95 to 1998/99. Although overall prison sentences for women in Canada were down during this period, sentences to federal terms (i.e., sentences of "two years and over") actually increased. The number of women sentenced to "two years or more" was 55 cases in 1994/95 but increased by 1.7 times, to 148 cases by 1998/99. This suggests judges may be stiffening the penalties for women offenders generally, since the was not a corresponding change in the number of "major’ offences tried in Canada’s adult criminal courts during that period. The report also noted that the rapid growth of women admissions during this period coincided with the opening of new Federal Women’s Facilities throughout the nation, and especially institutions to serve the Atlantic and Prairie Regions where there previously had been no federal regional establishments. This it was suggested may have influenced the number of federal sentences in those regions.

6.Major Offence (Current Sentence)

6. Major Offences (current sentence)
	FY 2002		FY 1997
Women	Number	%	Number	%
Homicide *	14	7	22	12
Sexual Offences (n.s.)	5	2	5	3
Robbery (n.s.)	41	19	28	15
Drugs (n.s.)	94	44	88	46

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

In terms of Major Offence (Current Sentence), there were relative modest overall changes in the women admission profiles. Homicide showed the largest decrease (p< .05), from 12% to 7%, while Sexual Offences decreased from 3% to 2%. Robbery offences increased from 15% to 19%. Although Drug Offences remained by far the largest single offence category for women, there was actually a small decrease in the proportion admitted in 2001/02 versus 1996/97 (down from 46% to 44%).

 

7.Global Rating - Static Factors

7. Global Rating - Static Factors *
	FY 2002		FY 1997
Women	Number	%	Number	%
Low Risk	91	44	71	42
Moderate Risk	84	40	54	32
High Risk	34	16	44	26

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

For federal women, the Global Rating - Static Factors scores indicate a significant (p< .05) shift: the proportion of admissions rated as high-risk: decreased from 26% in 1996/97 versus 16% in 2001/02, and there was a corresponding increase in the proportion rated as moderate- and low-risk (from 32% to 40% and from 42% to 44% respectively) during this period.

8.Global Rating – Dynamic Factors

8. Global Rating – Dynamic Factors (n.s.)
	FY 2002		FY 1997
Women	Number	%	Number	%
Low Need	49	23	41	24
Moderate Need	81	39	68	40
High Need	79	38	60	36

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05).

For federal women, the Global Rating - Dynamic Factors scores indicate no significant shift: the proportion of admissions rated as High, Moderate or Low-needs: although the proportion rated High Need increased from 36% in 1996/97 versus 38% in 2001/02, and there were proportional decreases in the proportion rated as moderate- and low-needs during this period.

9a.Global Rating – Dynamic Factors (Some Need)

9a. Domain Rating – Dynamic Factors (Some Need for Improvement)
	FY 2002		FY 1997
Women	Number	%	Number	%
Employment ***	93	44	117	69
Marital/Family *	87	42	89	53
Associates ***	89	43	106	63
Substance Abuse (n.s.).	26	12	32	19
Community Functioning ***	58	28	102	60
Personal/Emotional ***	73	35	95	56
Attitude ***	51	24	19	11

 *** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

An analysis of Domain Dynamic Factor Assessment scores found significant (p< .05 to p< .001) decreases in the proportion rated as requiring "some need" for improvement except Substance Abuse which decreased but not significantly. Conversely, the Attitude domain increased significantly (up from 11% to 24%; p< .001).

 

9b.Domain Rating – Dynamic Factors (Considerable Need)

9b. Domain Rating – Dynamic Factors (Considerable Need for Improvement)
	FY 2002		FY 1997
Women	Number	%	Number	%
Employment *	26	12	36	21
Marital/Family **	26	12	40	24
Associates (n.s.).	31	15	24	14
Substance Abuse **	106	51	64	38
Community Functioning (n.s.).	7	3	11	7
Personal/Emotional **	102	49	57	34
Attitude (n.s.)	20	10	15	9

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

Analysis of Domain Dynamic Factor Assessment scores ("considerable needs") found significant increases for women admissions in only two Substance Abuse and Personal/Emotional domains: up from 38% to 51% and from 34% to 49% respectively. Assessments for Associates (14% to 15%) and Attitude (9% to 10%) also increased slightly. The Employment and Marital/Family domains decreased significantly (p< .05 to p< .01) - 21% to 12% and 24% to 12% respectively.

 

 

10.Employment / Education Indicators

10. Employment / Education Indicators
	FY 2002		FY 1997
Women	Number	%	Number	%
Has an Unstable Job History (n.s.)	137	66	105	62
Has No High School Diploma (n.s.)	136	65	115	68

 

A larger proportion of women admitted in 2000/01 Has an Unstable Job History than in 1996/97 (62% to 66%), but the proportion that Has No High School Diploma decreased (68% to 65%). Neither change was found to be statistically significant.

11.Marital / Family Indicators

11. Marital / Family Indicators
	FY 2002		FY 1997
Women	Number	%	Number	%
Currently Single **	107	51	109	64
Had Dysfunctional Parents (n.s.)	103	50	84	52

 

The proportion of admitted women who were Currently Single decreased significantly (64% to 51%, p< .01) but there was no significant change in the proportion that Had Dysfunctional Parents.

 

12.Associates / Social Interaction Indicators

12. Associates / Social Interaction Indicators
	FY 2002		FY 1997
Women	Number	%	Number	%
Gang Affiliation (n.s.)	16	8	9	5
Mostly Criminal Friends (n.s.)	57	27	45	28
Criminogenic Living (n.s.)	57	28	51	32

 

There were no significant changes in the proportion of admitting women with positive Associates / Social Interaction Indicators.

13.Substance Abuse Issues

13. Alcohol/Drug Abuse Indicators
	FY 2002		FY 1997
Women	Number	%	Number	%
Abuses Alcohol (n.s.)	90	43	69	41
Abuses Drugs *	127	61	88	52
Abuses Drugs and Alcohol	144	69	105	62

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

 

A significant proportion of federal women offenders have a substance abuse problem. Among women admissions, the proportion of admissions who Abused Alcohol has increased, from 41% in1996/97 to 43% in 2001/02; while the proportion who Abused Drugs increased from 52% to 61%.

 

14.Cognition Indicators

14. Cognition Indicators
	FY 2002		FY 1997
Women	Number	%	Number	%
Poor Problem Solving (n.s.)	108	52	80	47
Unable to Generate Choices ***	86	41	100	60
Impulsive (n.s.)	132	63	101	60

*** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

A majority of federal women admissions also indicate Cognitive Skills issues. Managing offenders who are Poor at Problem Solving, Unable to Generate Wise Choices, and Impulsive also pose distinct challenges. The only significant finding was among women offenders who were identified as Unable to Generate Choices, which evidenced a decrease (60% in 1996/97 to 42% in 2001/02, p<.001).

15.Mental Health Indicators

15. Mental Health Indicators
	FY 2002		FY 1997
Women	Number	%	Number	%
Has a Past Diagnosis (n.s.)	40	19	27	16
Has Current Diagnosis (n.s.)	19	9	23	14
Prescribed Medication Currently (n.s.)	56	27	50	30

 *** Chi-square test of significant (p< .001); ** (p< .01); * (p< .05); Not sig. ^(n.s.).

Finally, managing offenders with a mental disorder(s) is also challenging. The proportion of admissions in 2001/02 with a Past Diagnosis increased slightly, while those with a Current Diagnosis as well as the proportion Prescribed Medication Currently decreased slightly. However, none of these changes proved to be significant.

Summary

The trends we have observed in women admitted in FY 2001/02 reinforce what has been happening in the custodial population. The population of women under age thirty at admission is decreasing slightly. More custodial women now have a Maximum or Minimum Security designation at intake, while a smaller proportion have a Medium rating. Moreover, the proportion of women rated CRS-High Institutional Adjustment Risk at intake has increased. There has also been an increase in the proportion with a gang affiliation.

Federal women admitted in 2002 already had more prior contact with the Youth and Provincial Adult correctional systems than before, and a greater proportion had previous breaches of trust, especially disciplinary segregations and more prior failures on conditional release.

In general a significantly greater proportion of the women admissions in 2002 were for shorter sentences than previously, and the proportion sentenced for robbery was much greater.

Dynamic factor assessments show that a larger proportion of the women admitted had "considerable need" for improvement in the key areas of substance abuse and personal/emotional issues. The proportion of women with identified cognitive and mental health issues also increased substantially.

 

 

Appendix A: References

2002Larry Motiuk, Roger Boe and Mark Nafekh. The Safe Return of Offenders to the Community --Statistical Overview, April 2002. Research Branch, the Correctional Service of Canada.

2002Mark Nafekh and Larry Motiuk: The Statistical Information on Recidivism – Revised 1 (SIR-R1) Scale: A Psychometric Examination. Research Branch, the Correctional Service of Canada. Report R-126.

2002Mark Nafekh and Jillian Flight: A review and Estimate of Time Spent in Prison by Offenders Sentenced for Murder. Research Branch, the Correctional Service of Canada. Report R-27.

2002 Roger Boe and Ben Vuong: Mental Health Trends Among Federal Inmates. Forum on Corrections Research, (forthcoming). Research Branch, the Correctional Service of Canada, (Forthcoming).

2000Correctional Service of Canada. Report of the Task Force on Security.

2000Roger Boe, Cindy Lee Olah and Colette Cousineau, Federal Imprisonment Trends for Women 1994-95 to 1998-99. Research Branch, the Correctional Service of Canada. Report R-93.

1997Larry Motiuk: "Classification for correctional programming: The Offender Intake Assessment (OIA) process". Forum on Corrections Research, Volume 9, Number 1, January 1997 (pp. 19-22).

1996Frederick P. Luciani, Laurence L. Motiuk and Mark Nafekh: An Operational Review of the Custody Rating Scale: Reliability, Validity and Practical Utility. Research Branch, the Correctional Service of Canada. Report R-47 (July).

 

 

Appendix B: OIA Coverage Issues

The following indicators were examined in OIA:

• All indicators are taken from the master file of the Offender Intake Assessment (OIA) module in the Offender Management System (OMS).
• Definitions for the indicators can be found on the Infonet, "Offender Intake Assessment and Correctional Planning: Standard Operating Practices (Interim) 700-04". See: "Annex 700-4C Dynamic Factor Analysis".

Estimates of admission under-coverage

• The information on admission represents "Warrant of Committal" admissions, for federal offenders. Data are the admission counts for each fiscal year (April 1^st to March 31^st of the following year).
• Admission coverage for OIA indicators is generally near 99% for current admission data. Under-coverage rates generally occur if a waiting period is not observed, since CSC policy requires that intake assessments be completed within 8 weeks of offender admission.

Table B-1: Admission Assessment Coverage

Admission Fiscal Year	WOC Admissions	OIA * Assessments	Not Assessed	% Not Assessed
1996/97	4,554	4,406	148	3.3%
1997/98	4,418	4,230	188	4.3%
1998/99	4,640	4,449	191	4.1%
1999/00	4,352	4,292	60	1.4%
2000/01	4,280	4,229	51	1.2%
2001/02	4,129	4,085	44	1.1%
* This analysis is based on an examination of data responses for the indicator "Hospitalized Current".

 

Estimates of Snapshot Under-coverage

• The information on institutional population represents all federal offenders. Data are snapshot counts for the fiscal year-end (defined as the Tuesday weekly OMS data extract closest to March 31^st of each year).

• Snapshot under-coverage for the in-custody population is much higher than for admission populations, in the earlier years, and decreases as we move forward in time. For example, under-coverage was about 47% in 1997, and this under-coverage improves to less than 20% by 2002.
• However, this under-coverage arises because OIA assessments were begun for all new admissions, starting November 15^th 1994. Hence, anyone admitted prior to that date would obviously not have an intake assessment.
• There is no reason to expect a systemic difference in the profile of the populations with and without an OIA assessment, however, since the lack of an assessment is due solely to the chance date of the offender’s admission.
• Moreover, a back-fill operation was undertaken in 1996 to cover the assessment gap for offenders admitted prior to OIA implementation. So, there is no under-coverage of OIA domain-level assessments.
• However, this backfill assessments were conducted at the domain level only (i.e., overall risk and need domains, etc.), and did not include assessments of individual indicators. Thus, under-coverage of OIA assessments is limited to the individual indicator-level assessments.

 

 

Table B-2: Snapshot Assessment Coverage

Snapshot for Mar. 31st	Custodial Count	OIA * Assessment	Not Assessed	% Not Assessed
1997	14,082	7,449	6,633	47.1%
1998	13,340	8,439	4,901	36.7%
1999	13,057	9,244	3,467	27.3%
2000	12,756	9,483	3,273	25.7%
2001	12,688	9,890	2,778	21.9%
2002	12,636	10,188	2,448	19.4%
* The illustration is based on data for the indicator: "Hospitalized Current".