Correctional Service Canada
Symbol of the Government of Canada

Common menu bar links

Research Reports

Warning This Web page has been archived on the Web.

Profile of Federally Sentenced Women Drug Offenders

2009 No R-204

Renée Gobeil
Correctional Service of Canada
May 2009

ACKNOWLEDGEMENTS

Appreciation is extended to Kelley Blanchette (Senior Director of Correctional Research, Correctional Service of Canada) and Nicole Crutcher (Senior Research Officer, Department of Justice Canada) for their role in planning the contents of this report. The author is also grateful to Tammy Cabana (Research Manager, Correctional Service of Canada) and Kelley Blanchette for their editorial expertise.

EXECUTIVE SUMMARY

In the 25 year period from 1981 to 2006, the proportion of Canadian women serving federal sentences for drug offences increased notably. Nonetheless, only limited basic research has been conducted with this group – most research in the area has focused on relatively complex topics, such as the impacts of the “war on drugs” policies in the U.S. Moreover, what research does exist on women drug offenders is frequently marked by a confounding of drug offenders and drug users, without any underlying research to justify this grouping. For these reasons, a profile of women drug offenders was both timely and necessary to fill existing gaps in research. It was intended that this profile serve to identify characteristics associated with drug offending among women, thereby informing policies and interventions specific to this group.

The profile was based on retrospective reviews of women’s computerized case files, and included all women admitted to federal custody on a new Warrant of Committal in the five year period between January 1, 2000, and December 31, 2004. Based on the offences for which they were convicted, women were divided into groups of drug offenders (n = 373) and non-drug offenders (n = 666). For follow-up analyses, the drug-offender group was divided into sub-groups of women whose most serious drug offence was importing / exporting (n = 132) and trafficking (n = 224). Low base rates precluded the examination of sub-groups of women whose most serious drug offence was production / cultivation (n = 2) and simple possession (n = 15).

Altogether, women in the drug offence group generally, and those in the importing / exporting sub-group specifically, had less extensive criminal histories, lower levels of risk and need, and higher levels of motivation. Women in the importing / exporting sub-group were also less likely than their counterparts in the trafficking sub-group to have a history of substance abuse or to have previously participated in substance abuse treatment. Finally, they were also more likely to be granted discretionary release and were less likely to return to custody, either due to a breach of supervision conditions or to a new offence. These findings underscore the heterogeneity among women whose most serious offences were drug offences. They also serve to reinforce the importance of differentiating between women who have been convicted for drug offences and those who have been identified as having substance abuse problems. In other words, drug offender and drug user are not interchangeable concepts for women offenders, despite previous researchers having examined these two groups together.

The information in this profile may be useful to those involved in the development of programs for women offenders. With awareness of the differences between and among groups of women comes the possibility of tailoring programs and other correctional interventions, thereby potentially increasing their ability to promote change.

Table of Contents

List of Tables

List of Appendices

Introduction

In recent decades, there has been substantial growth in the proportion of Canadian women offenders serving federal sentences for drug offences. Whereas only 16% of federally incarcerated women were serving sentences for drug offences in 1981 (Sinclair & Boe, 2002), this proportion increased to about 28% (a 175% increase) by 2007 (Correctional Service of Canada Research Branch, 2008). Given this growth, a profile of women offenders convicted of drug offences is timely. Profiles allow for the recognition of differences among groups of offenders, and can therefore eventually inform relevant policies and interventions. They can also act as a foundation for further, more detailed, research.

Women Drug Offenders

Given that drug offenders represent a growing proportion of federally incarcerated women, it is surprising how little basic research has been conducted with this group. One relevant recent study was a brief overview of men and women drug offenders under federal jurisdiction at the end of 2005 (Motiuk & Vuong, 2006). Though many statistics in this overview (e.g., sentence length, proportions of offenders in the community and in the institution) were calculated for men and women together, criminal history and criminogenic needs variables were examined separately for women. In this study, the researchers found that women drug offenders had extensive criminal history profiles, and considerable proportions of these women had identified criminogenic needs, primarily in the area of antisocial associates. Motiuk and Vuong (2006) recognized that women drug offenders differed significantly from their male counterparts in terms of criminal history and needs. Though these researchers provided statistics by type of drug offence (i.e., trafficking, importation, cultivation, and possession), they did not conduct statistical comparisons of these sub-groups. Nonetheless, some differences were evident. For example, only 14% of the women in the importation group were identified as having needs in the substance abuse area, while about 65% of the women in the trafficking and possession groups had substance abuse problems. Similarly, 48% of women in the importation group were identified as having needs in the employment domain, compared to nearly 70% of their counterparts in the trafficking and possession groups.

Differences among types of women drug offenders have also emerged elsewhere. In examining 630 women drug offenders incarcerated in a maximum security institution in Virginia, one researcher found that those convicted of possession reported lower levels of internal distress and of interpersonal conflict than did those convicted of trafficking or of non-drug offences (Loper, 2002). Together, these findings suggest that there is considerable heterogeneity among women drug offenders, both in terms of criminogenic need and of institutional adjustment.

Research Limitations

An important factor contributing to the scarcity of research specific to women drug offenders is that in examining this group, many researchers seem to use the terms “drug offenders” and “drug users” interchangeably (e.g., Fischer, Geiger, & Hughes, 2007; Hser, Evans, Teruya, Huang, & Anglin, 2007). However, as Loper (2002) has argued, “women drug offenders may vary in terms of their substance abuse history” (p. 1034). For this reason, it is important that researchers explicitly examine differences in the rates of substance abuse in their samples prior to using such terminology in this way.

Another factor is that in the U.S., much research has focused on the impacts of policies related to the “war on drugs.” Though these studies have been very informative, most have failed to provide a basic overview of the characteristics of women drug offenders. Instead, they report rates of arrest and incarceration (e.g., Harrison & Beck, 2006), and focus on issues such as the impact of mothers’ incarceration on children (Mumola, 2000). Altogether, then, only limited basic research exists on women drug offenders. Much of what does exist, moreover, is marked by methodological weaknesses such as grouping heteregoneous types of drug offenders or confounding substance abusers and drug offenders. Fundamental information on women drug offenders is required.

Current Study

Within this context, the objective of this study was to develop a profile of federally sentenced Canadian women whose most serious conviction was for a drug offence. The profile focused on demographic variables, criminal history variables, levels of risk and need, and recidivism. In order to provide context, findings specific to women convicted of drug offences were contrasted with those of their counterparts convicted of other offences. Given that previous researchers have found that women with various drug convictions differ importantly, women whose most serious drug offence was importing or exporting were also contrasted with their counterparts whose most serious drug offence was trafficking or possession for the purposes of trafficking.1

Method

Sample

The sample for this study was all women admitted to federal custody on a new Warrant of Committal (that is, admission for a new conviction) in the five year period between January 1, 2000, and December 31, 2004, for whom the most serious offence (of the current conviction) was a drug offence.2 Offences in this category are possession, trafficking, possession for the purpose of trafficking, production, and importation of drugs. In cases where the same woman had more than one admission which met this criterion, the first was included in the sample and the other(s) omitted.3 A total of 388 women were identified for inclusion in the drug offender sample. Fifteen women were eliminated as their files did not include complete intake assessment data, leaving 373 women in the drug offence group. For the purposes of follow-up analyses, the 373 women in this group were categorized according to their most serious drug offence: importing or exporting (n = 132); production or cultivation (n = 2); trafficking or possession for the purposes of trafficking (n = 224); and, simple possession (n = 15).

The comparison sample comprised all other women admitted to federal custody on a new Warrant of Committal within the study period. Again, the first admission was used for each woman. The non-drug offence sample originally included 716 women. The files of 48 women were removed due to missing intake assessment data. An additional two women were excluded from this sample as they died prior to warrant expiry. This left a total of 666 women in the non-drug offence group.

Data

Archival data were retrieved from the Offender Management System, Correctional Service Canada’s automated offender data system. For each offender, data pertaining to the following areas were retrieved: demographic characteristics; offence and criminal history; global assessments of risk, need, and motivation; specific assessments of need in the areas of employment, marital / family relations, associates / social interaction, community functioning, attitude, substance abuse issues, and personal / emotional orientation; history of substance abuse; release date and type; and post-release outcome.

Other than information related to release and to post-release outcome, data were recorded on intake for each woman as part of the standardized Offender Intake Assessment. During this process, overall risk, overall need, the seven specific assessments of need (in the domains listed above), and history of substance abuse were assessed for each offender by caseworkers through interviews and reviews of official documents, including police reports, criminal history records, and judge’s comments. These data were obtained retrospectively from offender files for this study.

Data regarding release and post-release outcome were then recorded for the first release on each woman’s sentence. In other words, if a woman was released, had her release revoked, and was subsequently re-released, only the first release was considered in the present analyses. Two indices of post-release outcome were considered: (1) any revocation of conditional release (i.e., due to breaching a condition of supervision or due to a new offence), and (2) revocation due to a new offence or outstanding charge.

Analyses

First, a series of descriptive analyses were completed in order to examine the profile of federal women drug offenders. Analyses were also conducted in order to examine sub-groups of drug offenders according to their most serious drug offence.4 Next, chi-square tests of independence were undertaken in order to discern any differences in the profiles of women according to their most serious offence. Due to the number of comparative analyses conducted, a Bonferroni-like correction was applied to the detection of statistical significance, and only findings reaching a probability level of .002 or less were considered in these analyses. Finally, Cox regressions were used to contrast the rate and speed of return to custody for the offender groups.

Results

Offence Profile

The group of interest in this study was women whose most serious offence was a drug offence. On average, these women were convicted of 1.3 drug offences on the current sentence. Table 1 presents the numbers and percentages of women convicted of each type of drug offence. These numbers are presented in two ways – first as a function of all the drug convictions on the current sentence, and secondly, considering only the most serious drug conviction on the current sentence. In both cases, trafficking and possession for the purposes of trafficking was the most common category, followed by importation / exportation. Given the numbers of women whose most serious drug convictions fell into the importation / exportation and trafficking / possession for the purposes of trafficking categories, subsequent analyses will also consider these subgroups individually. Similar analyses will not be conducted for those whose most serious offences were production / cultivation and simple possession due to insufficient numbers.

Table 1
Drug Offence Profile
  All drug convictions a Most serious drug conviction b
Offence % (n / 373) % (n / 373)
Importation / exportation 35.4 (132) 35.4 (132)
Production / cultivation 0.5 (2) 0.5 (2)
Trafficking / possession for the purposes of trafficking 63.5 (237) 60.1 (224)
Simple possession 16.9 (63) 4.0 (15)

a Proportions and counts for this column will sum to more than 100 as many women were convicted of more than one drug offence on the current sentence.
b The ranking of seriousness for drug offences, from most to least serious, is as follows: importation/ exportation, production/ cultivation, trafficking/ possession for the purposes of trafficking, simple possession.

Most serious offences were also considered among women whose most serious offences were non-drug offences. In this group, robbery was the most serious offence for the largest proportion of women (see Table 2). Murder / manslaughter, assault, and ‘other’ property offences (i.e., other than break and enter) were also relatively common.

Table 2
Most Serious Offences, Non-Drug Offence Group
Most Serious Offence  % (n / 666)
Murder / manslaughter 14.9 (99)
Attempted murder / manslaughter 1.8 (12)
Sexual offences 2.1 (14)
Robbery 29.4 (196)
Assault 18.2 (121)
Other violent offences 3.2 (21)
Impaired offences 2.0 (13)
Break and enter 4.7 (31)
Other property offences 19.2 (128)
Other offences 4.7 (31)

Though only most serious offences are considered in subsequent analyses, all offences on the current conviction were also examined for both the drug offence group and the non-drug offence group. As Table 3 demonstrates, most women in the non-drug offence group (about 90%) were convicted only of non-drug related offences. Conversely, many offenders convicted of drug offences were convicted of only drug offences, or, if convicted of other offences, were often convicted of ‘other’ offences. This category includes primarily administration of justice offences, such as failing to appear to court. Altogether, there was limited overlap between the two categories of offenders.

Table
Offences on Present Conviction
  Drug offence group Non-drug offence group
Offence % (n / 373) % (n / 666)
Murder / manslaughter - - 14.9 (99)
Attempted murder / manslaughter - - 2.4 (16)
Sexual offences - - 2.3 (15)
Robbery - - 29.6 (197)
Assault - - 26.6 (177)
Other violent offences - - 12.2 (81)
Drug offences 100 (373) 9.5 (63)
Impaired offences 0 (0) 2.1 (14)
Break and enter 1.1 (4) 10.5 (70)
Other property offences 17.2 (64) 41.9 (279)
Other offences 51.5 (192) 63.7 (424)

Note. Proportions sum to more than 100 as many women were convicted multiple offences on the current sentence.

Demographic Profile

Table 4 presents an overview of the demographic characteristics of both the drug offence group and non-drug offence group. Significant differences were identified in the women’s ethnicities, χ2 (3, N =1039) = 95.59, p < .0001, with a greater proportion of the women in the drug offence group being Black, and a greater proportion of their counterparts in the non-drug offence group being Caucasian or Aboriginal. Marital status did not differ statistically for the two groups; most women were single, widowed, divorced, or separated, though a considerable proportion were married or in common-law relationships. Finally, the mean age at admission for women in the drug and non-drug offence groups was 34.0 (SD = 10.0) and 33.5 (SD = 9.04) years respectively; these were also not found to differ by group.

Table 4
Demographic Characteristics
  Drug offence group Non-drug offence group
Variable % (n / 373) % (n / 666)
Ethnicity *  
Caucasian 50.9 (190) 62.3 (415)
Aboriginal 16.4 (61) 28.2 (188)
Black 20.9 (78) 4.7 (31)
Other / unknown 11.8 (44) 4.8 (32)
Marital Status  
Single, widowed, or divorced 58.2 (217) 62.0 (413)
Married or common-law 40.0 (149) 36.3 (242)
Unknown 1.9 (7) 1.7 (11)

* p < .0001.

Follow-up analyses, detailed in Table 5, revealed considerable differences in the demographic profiles of women whose most serious drug offences were importing / exporting and trafficking / possession for the purposes of trafficking. Relative to their counterparts in the trafficking group, a much greater proportion of the women in the importing / exporting group were Black, while fewer in this group were Caucasian or Aboriginal. An appreciably larger proportion of those in the importing / exporting group were also single, widowed, or divorced. Lastly, women in the importing / exporting group were also significantly younger (M = 31.4; SD = 8.9) than those in the trafficking group (M = 34.7; SD = 9.0), t(354) = 3.42, p < .001. Seemingly, the absence of significant differences in marital status and age in the comparisons involving women whose most serious offences were and were not drug offence (above) served to mask considerable variability on these variables within the drug offence group.

Table 5
Demographic Characteristics by Most Serious Drug Offence
  Import / Export Trafficking
Variable % (n / 132) % (n / 224)
Ethnicity *  
Caucasian 32.6 (43) 58.9 (132)
Aboriginal 2.3 (3) 25.0 (56)
Black 50.0 (66) 5.4 (12)
Other / Unknown 15.2 (20) 10.7 (24)
Marital Status *  
Single / Widowed / Divorced 72.7 (96) 48.7 (109)
Married / Common-law 24.2 (32) 50.0 (112)
Unknown 3.0 (4) 1.3 (3)

* p < .0001.

Criminal History

As shown in Table 6, there was considerable variability in the amount of previous contact with the criminal justice system had by the women in each group. Women in the drug offence group had significantly fewer convictions both as youth, χ2 (5, N =1039) = 32.97, p < .0001, and as adults, χ2 (5, N =1039) = 22.88, p < .001, than did those in the non-drug offence group. In comparing sub-groups of women from the drug offence group, those whose most serious offence was importing / exporting were found to have fewer convictions as adults than those whose most serious offence was trafficking, χ2 (5, N = 356) = 90.65, p < .0001. In fact, the previous adult conviction profiles of women in the trafficking group were very similar to those in the non-drug offence group. Turning to convictions as youth, however, no differences were found between the women whose most serious offences were importing / exporting and trafficking.

Table 6
Criminal History
  Drug Offences Non-Drug Offences
Variable % (n / 373) % (n / 666)
Previous Youth Convictions * 16.6 (62) 30.0 (200)
1 youth offence 7.0 (26) 6.3 (42)
2-4 youth offences 4.3 (16) 9.0 (60)
5-9 youth offences 3.2 (12) 6.2 (41)
10-14 youth offences 1.1 (4) 4.1 (27)
15 or more youth offences 1.1 (4) 4.5 (30)
Previous Adult Convictions * 60.6 (226) 72.7 (484)
adult offence 8.6 (32) 7.2 (48)
2-4 adult offences 15.0 (56) 14.7 (98)
5-9 adult offences 11.8 (44) 14.3 (95)
10-14 adult offences 5.1 (19) 9.2 (61)
15 or more adult offences 20.1 (75) 27.3 (182)
Gang Affiliation 4.8 (18) 2.1 (14)

* p < .0001.

Notably, though a slightly larger proportion of the women in the drug offence group (4.8%) than in the non-drug offence group (2.1%) were affiliated with a gang or with organized crime, this difference was not significant. Similarly, no significant differences were detected in the proportions of women whose most serious drug offences were importing / exporting and trafficking who were affiliated with a gang or with organized crime. Finally, sentence length was also examined. None of the women in the drug offence group were serving an indeterminate sentence; conversely, 4.2% (28) of the women in the non-drug offence group were. Among the women serving determinate sentences, the mean aggregate sentence length was significantly longer in the non-drug offence sample (M = 3.1 years; SD = 1.5) than for the drug offence sample (M = 2.7 years; SD = 1.0), t(1037) = 4.29, p < .0001. Considering sub-groups within the drug offence sample, sentence length was longer for those convicted of importing / exporting (M = 3.1 years; SD = 1.2) than it was for those convicted of trafficking, (M = 2.5 years; SD = 0.8), t(354) = 5.27, p < .0001.

Risk, Need, and Motivation

Women in the drug offence and non-drug offence categories were then compared in terms of assessed levels of risk, need, and motivation. As presented in Table 7, relative to women in the non-drug offence group, those in the drug offence group had lower ratings of risk, χ2 (2, N =1039) = 131.30, p < .0001, and need, χ2 (2, N =1039) = 167.24, p < .0001. Women in the drug offence group also tended to have higher ratings of motivation, though this difference was not statistically significant.

Table 7
Ratings of Risk, Need, and Motivation
  Rating
Variable Low Moderate High
Risk *  
Drug offences 67.0% 27.4% 5.6%
Non-drug offences 27.3% 46.0% 26.7%
Need *  
Drug offences 37.3% 45.8% 16.9%
Non-drug offences 13.1% 38.6% 48.4%
Motivation  
Drug offences 3.5% 37.3% 59.3%
Non-drug offences 6.3% 44.7% 49.0%

Note. Drug offence group: N = 373. Non-drug offence group: N = 666.
* p < .0001.

Highly significant differences also emerged when analyses focused on sub-groups of women in the drug offence group. As can be seen in Table 8, women whose most serious drug offence was importing / exporting had much lower ratings of risk, χ2 (2, N =356) = 55.32, p < .0001, and need, χ2 (2, N =356) = 105.86, p < .0001, than did their counterparts whose most serious drug offence was trafficking. The importing / exporting group also had higher ratings of motivation, χ2 (2, N =356) = 43.46, p < .0001. Altogether, these data demonstrate that while women whose most serious offences were drug offences, as a group, had more promising risk, need, and motivation profiles than women whose most serious offences did not involve drugs, there was considerable variability among women in the drug offence group. Those whose most serious drug offence was importing / exporting had more promising profiles than did those whose most serious drug offence was trafficking. Again, there were similarities in the profiles of women in the trafficking group and those in the non-drug offence group, most notably as regards motivation.

Table 8
Ratings of Risk, Need, and Motivation by Most Serious Drug Offence
  Rating
Variable Low Moderate High
Risk *  
Import / Export 92.4% 7.6% --
Trafficking 54.9% 36.2% 8.9%
Need *  
Import / Export 72.7% 23.5% 3.8%
Trafficking 18.3% 58.5% 23.2%
Motivation *  
Import / Export 2.3% 16.7% 81.1%
Trafficking 4.5% 50.0% 45.5%

Note. Import / Export drug offence sub-group: N = 132. Trafficking drug offence sub-group: N = 224.
* p < .0001.

The intake process also includes assessments of seven specific areas of criminogenic need: employment, marital / family relations, associates / social interaction, community functioning, attitude, substance abuse, and personal / emotional orientation.5 A series of chi-square tests of independence revealed significant differences in ratings in four of these domains (see Table 9). Perhaps ironically, women in the drug offence group were found to have fewer difficulties regarding substance abuse issues and personal / emotional orientation. Significant differences were also identified in the associates / social interaction and marital / family relations domains, with women convicted of drug offences being more likely to receive a rating of some difficulty for associates and less likely to receive a rating of factor seen as an asset for marital / family relations.

Table
Specific Ratings of Criminogenic Need
  Rating
Domain Asset No Diff. Some Diff. Cons. Diff.
Employment  
Drug offences 4.8% 32.4% 50.4% 12.3%
Non-drug offences 4.5% 39.0% 39.8% 16.7%
Marital / Family *  
Drug offences 9.1% 50.9% 33.5% 6.4%
Non-drug offences 22.4% 37.4% 35.3% 5.0%
Associates / Social Interaction *  
Drug offences 1.9% 34.1% 50.1% 13.9%
Non-drug offences 5.1% 41.7% 34.8% 18.3%
Community Functioning  
Drug offences 3.0% 64.4% 29.3% 3.3%
Non-drug offences 3.8% 67.6% 25.7% 3.0%
Attitude  
Drug offences 16.9% 56.8% 17.7% 8.6%
Non-drug offences 10.7% 56.3% 20.6% 12.5%
Substance Abuse *  
Drug offences -- 49.1% 13.9% 37.0%
Non-drug offences -- 29.3% 12.3% 58.4%
Personal / Emotional *  
Drug offences -- 33.2% 42.4% 24.4%
Non-drug offences -- 10.8% 34.4% 54.8%

Note. Drug offence group: N = 373. Non-drug offence group: N = 666.
* p < .0001.

Follow-up analyses contrasting women whose most serious drug offences were importing / exporting and trafficking revealed that those in the trafficking sub-group had more problematic profiles in six of the seven domain areas: employment, marital / family relations, associates / social interaction, attitude, substance abuse issues, and personal / emotional orientation. Women in the importing / exporting sub-group, on the other hand, had slightly more problematic profiles in the community functioning domain.

Finally, a number of substance use variables were also explored. Analyses revealed that significantly greater proportions of women in the non-drug offence group than in the drug offence group had histories of alcohol abuse (49.1% vs. 22.0%), χ2 (1, N =1039) = 73.65, p < .0001, drug abuse (65.3% vs. 55.2%), χ2 (1, N =1039) = 10.29, p < .001, and substance abuse treatment (50.8% vs. 37.3%), χ2 (1, N =1039) = 17.51, p < .0001. Turning to the sub-groups of women in the drug offence group, significantly greater proportions of women in the trafficking group were found to have histories of alcohol abuse (31.7% vs. 6.1%), drug abuse (74.1 vs. 20.5%), and substance abuse treatment (50.8% vs. 9.8%). Again, the broad analyses conducted first may have concealed the magnitude of some of the differences within offender groups, in that women whose most serious drug offence was trafficking generally had more serious substance use profiles than women in the non-drug offence group, while those whose most serious offence was importing / exporting had more promising profiles in this area than those in the non-drug offence group.

Release

Release and post-release outcome analyses included only the women released by August 1, 2007. At this time, all of the women in the drug offence group (n = 373) and 93.7% of their counterparts in the non-drug offence group (n = 624) had been released. Notably, women in the drug offence group were significantly more likely to be released on day or full parole (94.6%) than were their counterparts in the non-drug offence group (71.2%), χ2 (1, N =1039) = 80.29, p < .0001. Women in the latter group were more often released at statutory release or at warrant expiry. There were no significant differences in type of release granted to the sub-groups of women whose most serious offence was a drug offence.

When released on day parole, full parole, or statutory release, offenders can be assigned supervision conditions which they must respect while on conditional release. On average, women in the non-drug offence group were assigned a greater number of supervision conditions than were their counterparts in the drug offence group (2.2 vs. 1.8), t (944 [Satterthwaite-corrected df]) = 4.68, p < .0001. No differences in the number of conditions assigned were apparent for the sub-groups within the drug offence group

Post-Release Outcome

Post-release outcome was examined only for those women who were not considered deportable, as deported women would have greatly diminished opportunity to have their release revoked or to re-offend while in Canada.6 Thirty-six women were therefore excluded from these analyses7; five were from the non-drug offence group (leaving 619 released women in this group, 29 were from the importing / exporting sub-group (leaving 103), and two were from the trafficking sub-group (leaving 222). A total of 342 women remained in the drug offence group. It is notable that while less than one percent of the women convicted of non-drug or trafficking offences were deportable, more than one in five in the importing / exporting group met this criterion.

The follow-up period for post-release outcome analyses encompassed the time elapsed from each woman’s date of release until warrant expiry date or August 1, 2007, whichever occurred first. On average, the length of this period was of 2.8 years (SD = 1.0 year, ranging from 1.7 to 7.4 years). The mean time period was significantly shorter for women in the drug offence group than for those in the non-drug offence group (2.6 vs. 2.9 years), t (959) = 4.01, p < .0001, which is not surprising, considering that this group was also found to have shorter sentences. Also consistent with the findings on sentence length was the fact that women whose most serious drug offence was importing / exporting had slightly (but non-significantly) longer sentences than did those who most serious drug offence was trafficking (2.8 vs. 2.5 years).

In contrasting rates of return to custody, women in the non-drug offence group were found to receive any type of revocation of conditional release (i.e., for technical reasons or due to a new offence) more frequently than their counterparts in the drug offence group (49.5 vs. 37.7%), χ2 (1, N =1003) = 12.55, p < .001. Though women in the non-drug offence group were also more likely to have their conditional release revoked due to a new offence or outstanding charge (14.8% vs. 9.9%), this difference was not statistically significant. A similar pattern was found in contrasting sub-groups of women in the drug offence group, with women convicted of trafficking having their conditional release revoked for any reason more frequently than women convicted of importing / exporting, (48.2% vs. 11.7%), χ2 (1, N =325) = 40.49, p < .0001, but having their release revoked due to new offences or charges at statistically similar rates (12.1% vs. 3.9%).

As mentioned, however, follow-up periods were not equivalent for the different groups. It is therefore possible that the results found are partially attributable to the fact that those in the non-drug offence group and the importing / exporting drug offence sub-group, having longer follow-up periods, had more opportunity to have their release revoked or to be involved in a new offence. For this reason, Cox regression analyses were also computed. This technique assesses whether there are differences, associated with group membership, in the average length of time elapsed prior to returning to custody. The technique also compensates for differing follow-up periods. A final advantage is that Cox regression allows for the inclusion of covariates. This point is particularly important given that the groups were found to differ on risk, with women in the drug offence group being assessed as lower risk than those in the non-drug offence group, and those in the trafficking sub-group being assessed as higher risk than those in the importing / exporting sub-group. Since assessed risk is also associated with post-release outcome (CSC, 1989), including risk as a covariate allows for the examination of whether any differences in the average length of time before return to custody remained after accounting for this known association.

Table 10 presents the results of Cox regressions conducted both with and without risk as a covariate for the drug offence and non-drug offence groups. As can be seen, the time elapsed prior to revocation (either any revocation or revocation with a new offence or charge) was found to differ significantly by group when risk was not statistically controlled, with women convicted of non-drug offences having higher rates of revocation. These differences, however, were no longer present in the analyses which examined group differences after accounting for risk.

Table 10
Cox Regression of Time to Revocation on Risk and Offence Group
Regression Model B SE Wald H.R χ2
Any Revocation  
Model 1: Offence group         23.23***
Group membership a 0.50 0.11 22.76*** 1.65  
Model 2: Offence group added to effect of risk b         2.02
Risk 0.59 0.07 79.71*** 1.81  
Group membership a 0.16 0.11 1.99 1.17  
Revocation with New Offence / Charge  
Model 1: Offence group         5.14*
Group membership a 0.46 0.21 5.05* 1.58  
Model 2: Offence group added to effect of risk b         0.60
Risk 0.51 0.13 15.14*** 1.67  
Group membership a 0.17 0.22 0.59 1.18  

Note. Drug offence group: N = 342. Non-drug offence group: N = 619.
a The reference category for group membership is the drug offence group.
b Regressions of time to revocation on offence group addedto the effect of risk represent a test of the change in predictive ability over a model including only risk.
* p < .05. *** p < .0001.

Survival graphs were also plotted to complement these analyses (see Appendix A). These confirm that offenders convicted of non-drug offences had their conditional releases revoked more rapidly and at higher rates than did those convicted of drug offences. However, this difference narrowed substantially after accounting for the effect of risk. The same patter was visible regarding revocations with new offences or charges.

In considering sub-groups of women whose most serious offence was a drug offence, however, the pattern of findings was somewhat different (see Table 11). Again, before controlling for risk, the time elapsed prior to revocation (any revocation or revocation with a new offence or charge) differed for women whose most serious drug offence was importing / exporting and women whose most serious offence was trafficking. After controlling for risk, however, the between-group difference persisted for any revocation. In other words, even after accounting for the different risk profiles of women convicted of importing / exporting and trafficking, those in the trafficking group had their conditional releases revoked both at higher rates and more rapidly than did those in the importing / exporting group. No such difference, however, was found when considering only revocations due to a new offence or charge.

Table 11
Cox Regression of Time to Revocation on Risk and Offence Group by Most Serious Drug Offence
Regression Model B SE Wald H.R χ2
Any Revocation  
Model 1: Offence group         44.38***
Group membership a -1.80 0.31 34.51*** 0.17  
Model 2: Offence group added to effect of risk b         33.97***
Risk 0.54 0.13 16.78*** 1.71  
Group membership a -1.57 0.32 24.73*** 0.21  
Revocation with New Offence / Charge  
Model 1: Offence group         7.33**
Group membership a -1.35 0.54 6.33* 0.26  
Model 2: Offence group added to effect of risk b         3.32
Risk 0.75 0.26 8.03** 2.10  
Group membership a -0.95 0.57 2.82 0.39  

Note. Import / export drug offence sub-group: N = 103. Trafficking drug offence sub-group: N = 222.
a The reference category for sub-group membership is the import / export sub-group.
b Regressions of time to revocation on offence group added to the effect of risk represent a test of the change in predictive ability over a model including only risk.
** p < .01. *** p < .0001.

Again, the results of these analyses graphed for visual examination (see Appendix B). In examining any revocation, it is clear that women convicted of trafficking offences had their releases revoked at higher rates and more rapidly than did those convicted of importing / exporting, even after accounting for the effect of risk. After considering risk, however, no significant difference remained when considering revocation with a new offence or charge.

Discussion

This report provided a profile of federal women offenders whose most serious offence was a drug offence. In order to provide context, this profile was contrasted with that of women whose most serious offence was not a drug offence. Follow-up analyses were also conducted to compare women whose most serious drug offence was importing / exporting to those whose most serious drug offence was trafficking or possession for the purposes of trafficking.

Turning first to demographic characteristics, it was clear that there were more differences between the sub-groups of drug offenders than between women whose most serious offences were related to drugs and those whose most serious offences were not related to drugs. A much greater proportion of women in the importing / exporting sub-group were identified as being deportable. Also, a larger proportion of the women in this group were Black, while larger proportions of those in the trafficking sub-group and the non-drug offence group were Caucasian or Aboriginal. This difference may be due to the racial composition of many countries from which the importation of drugs is common (e.g., certain Caribbean islands, certain African countries; Royal Canadian Mounted Police, 2007). Women convicted for importing and exporting were also generally younger and more likely to be single than those convicted for trafficking.

Overall, women in the drug offence group generally, and those in the importing / exporting sub-group specifically, had less extensive criminal histories, represented lower risk, had fewer criminogenic needs, and were assessed as more motivated than their counterparts in the non-drug offence group. In these areas, the profiles of women in the trafficking sub-group tended to resemble those of the women whose most serious offence was not a drug offence. In contrasting women whose most serious offences were and were not drug offences, differences in the global assessment of criminogenic need were mainly attributable to women in the drug offence group having fewer difficulties in the areas of substance abuse and of personal / emotional orientation. Among the women in the drug offence group, those whose most serious offence was trafficking had more problematic need profiles than those whose most serious offence was importing or exporting in six of seven domain areas: employment, marital / family relations, associates / social interaction, attitude, substance abuse issues, and personal / emotional orientation. Women in the trafficking sub-group were also more likely than their counterparts in the importing / exporting sub-group and the non-drug offence group to have a history of alcohol abuse, drug abuse, and substance abuse treatment.

These findings underscore the importance of differentiating between women who have been convicted of drug offences and those who have substance abuse problems. Indeed, in this sample, only half of the offenders in the drug group were identified as having criminogenic needs in the area of substance abuse. Though there is certainly overlap between drug offenders and drug users, these findings, together with those reported by Motiuk and Vuong (2006), suggest that it is relatively common for women to become involved in importation and exportation of drugs in the absence of strong ties to criminal culture and substance abuse problems. In other words, women may have motivations other than drug use for their involvement in such offences. Specifically, these motivations might be financial. Indeed, Adler (1993) found that for many drug offenders involved in importation, importing was viewed as an opportunity to make relatively quick and effortless money. For instance, it is estimated that 100 tonnes of heroin are imported into Canada each year (Kennedy, 2002, as cited in Nolin & Kenny, 2002). With a street value of $201.60 US per gram in 2005 (United Nations Office on Drugs and Crime, 2008), such a quantity has a value of over $200 million per year.

Notably, however, some researchers have suggested that women are often not committed to criminal careers (Adler, 2003); instead, they are involved only tangentially and at relatively low levels in the importation process (i.e., as couriers or “mules”; e.g., Caulkins, Johnson, Taylor, & Taylor, 1999; Davis, Johnson, Randolph & Liberty, 2005).8 In such roles, they likely access only relatively small proportions of the money made from importation.

Conversely, women involved in trafficking may be more entrenched in a criminal lifestyle. For example, Adler (2003) found that those involved in trafficking drugs were more likely than those involved in importation or exportation to view this activity as part of a criminal career, to have developed networks of criminal acquaintances, and to be at least partially motivated by drug consumption. Exploration of motivation and other factors contributing to offending behaviour (e.g., financial gains versus supporting substance abuse problems) among women drug offenders may be fruitful areas for future research.

Turning to release and post-release outcome, women whose most serious offences was a drug offence were more likely to be granted discretionary release, and were less likely to return to custody after release. Among the women in the drug group, those whose most serious offence was trafficking were more likely to have their conditional release revoked than those whose most serious offence was importation or exportation. Findings regarding rates of return to custody, however, were partially attributable to the difference in levels of risk between the groups of women. After adjusting statistically for assessed risk, no differences remained in the rate or speed at which women in drug and non-drug offence groups returned to custody, either for any reason (i.e., technical revocation or revocation due to a new offence or charge) or due to a new offence or charge. Notably, however, even after adjusting for differing risk profiles, women in the trafficking sub-group had their releases revoked more frequently and more rapidly than their counterparts in the importing / exporting group. Given that this was not the case for revocations attributable to new offences or charges, it can be inferred that these revocations were technical in nature, arising from breaches of supervision conditions.

Supervision conditions are used to manage risk associated with criminogenic need. For example, a “non-association” condition might be applied to someone who has more problems in the area of criminal associates, while abstinence from drugs, alcohol, or both and random urinalysis testing conditions might be used to manage risk associated with substance use. As such, it might seem logical that women whose most serious drug offence was trafficking, who had more problematic need profiles, also were assigned a greater number of supervision conditions. If this were the case, there may have been more opportunity for these women to breach the conditions of their supervision than would be the case for women convicted of importing / exporting or non-drug related offences. However, analyses showed that there was no difference in the number of supervision conditions assigned to women in each of the drug offence sub-groups, which demonstrates that this explanation is not adequate.

Another possibility is that the women in the trafficking sub-group’s more problematic need profiles may explain the differences in revocation rates more directly. Research indicates that high levels of criminogenic need are associated with higher levels of return to custody and of re-offending (Gendreau, Little, & Goggin, 1996). Indeed, this is why criminogenic needs are sometimes termed dynamic risk factors.In this case, the failure to detect differences in rates of re-offence may mean that these women’s conditional releases were revoked for technical reasons before the women had the opportunity to re-offend.

Limitations and Future Directions

The present profile compared women whose most serious offences were drug offences to those for whom this was not the case. In doing so, however, it was not possible to fully disaggregate the group of women convicted of drug offences. While sufficient numbers of women were convicted of importation / exportation and trafficking to allow for investigation of their profiles, low base rates prevented examination of the profiles of women whose most serious offences were drug production / cultivation and simple possession. It is therefore important to use caution in applying findings from the current profiles to women for whom these were the most serious offences.

This profile reinforces the heterogeneity among types of women drug offenders, especially with regard to their histories of substance abuse and their levels of criminogenic need in this area. It was notable that those women whose most serious offence was trafficking had need profiles similar to those of women whose most serious offence was not a drug offence. These findings underscore the importance of discriminating both among types of drug offenders and between drug offenders and substance abusers in research and practice.

Conclusion

This report aimed to provide a basic profile of federally-sentenced Canadian women whose most serious offence was a drug offence. Sizeable differences were found in comparing drug offenders to non-drug offenders as well as among drug offenders, according to the type of drug offence. Indeed, one type of drug offender, those whose most serious offence was importation or exportation, was conspicuous in its differences from both the remaining drug offenders (traffickers) and non-drug offenders. In this sample, women whose most serious conviction was importation or exportation stood out as representing a lower level of risk, having lower assessed levels of need, and having less problematic histories and needs in the areas of substance abuse.

Results are important for those who are involved in the development of programs for women offenders. With awareness of the differences between and among groups of women comes the possibility of tailoring programs and other correctional interventions, thereby potentially increasing their ability to promote change. For instance, the present findings suggest that women convicted for importation may benefit in only a limited fashion from participation in substance abuse programming. Indeed, to be assigned to substance abuse treatment, it is expected that women reach threshold levels on measures of alcohol and / or drug dependency; in practice, however, this requirement is sometimes disregarded based on caseworkers’ clinical impressions. These results therefore underscore that as defined in policy, women convicted for importation who do not meet these thresholds are likely best excluded from substance abuse treatment. Given findings among male offenders that drug importation is often undertaken as a means of making money (Adler, 1993), such women may be better served by participation in either programs focused on changing their pro-criminal attitudes or education and employability programs, with the ultimate goal of helping women earn better wages legitimately.

References

Adler, P.A (1993). Wheeling and dealing: An ethnography of an upper-level drug dealing and smuggling community (2nd ed.). New York: Columbia University Press.

Caulkins, J.P., Johnson, B.D., Taylor, A., & Taylor, I. (1999). What drug dealers tell us about their costs of doing business. Journal of Drug Issues, 29, 323-340.

Correctional Service of Canada, Research Branch (2008). [The changing federal offender population yearly snapshot data].Unpublished raw data.

Correctional Service Canada, Research and Statistics Branch (1989). The Statistical Information on Recidivism scale. Brief B-01. Ottawa, ON: Author.

Davis, W.R., Johnson, B.D., Randolph, D., & Liberty, H.J. (2005). Gender differences in the distribution of cocaine and heroin in Central Harlem. Drug and Alcohol Dependence, 77, 115-127.

Fischer, M., Geiger, B., & Hughes, M.E. (2007). Female recidivists speak about their experience in drug court while engaging in appreciative inquiry. International Journal of Offender Therapy and Comparative Criminology, 51, 703-722.

Gendreau, P., Little, T., & Goggin, C. (1996). A meta-analysis of the predictors of adult offender recidivism: What works! Criminology, 34, 575-607.

Harrison, P.M. & Beck, A.J. (2006). Prisoners in 2005. Washington, DC: US Department of Justice, Bureau of Statistics.

Hser, Y.-I., Evans, E., Teruya, C., Huang, D., & Anglin, M.D. (2007). Predictors of short-term treatment outcomes among California’s Proposition 36 participants. Evaluation and Program Planning, 30, 187-196.

Loper, A.B. (2002). Adjustment to prison of women convicted of possession, trafficking, and nondrug offences. The Journal of Drug Issues, 32,1033-1050.

Motiuk, L.L. & Vuong, B. (2006). Re-profiling the drug offender population in Canadian federal corrections. Forum on Corrections Research, 18, 24-29.

Mumola, C.J. (2000). Incarcerated parents and their children. Washington, DC: US Department of Justice, Bureau of Statistics.

Nolin, P.C. & Kenny, C. (2002). Cannabis: Our position for a Canadian public policy: Report of the Senate Special Committee on Illegal Drugs. Retrieved March 4, 2009 from http://www.parl.gc.ca/37/1/parlbus/commbus/senate/com-e/ille-e/rep-e/repfinalvol1part1-e.htm

Royal Canadian Mounted Police (2007). Drug situation report 2005. Ottawa, ON: Author.

Sinclair, R.L. & Boe, R. (2002). Canadian Federal Women Offender Profiles: Trends from 1981 to 2002 (Revised). Research Report R-131. Ottawa, ON: Correctional Service Canada.

United Nations Office on Drugs and Crime (2008). World drug report 2008. Retrieved March 4, 2009 from http://www.unodc.org/unodc/en/data-and-analysis/WDR-2008.html

United States Sentencing Commission (2007). Cocaine and federal sentencing policy. Washington, DC: Author.

Appendices

Appendix A. Survival Curves: Drug and Non-Drug Offenders

Table A1
Survival Curves for Drug and Non-Drug Offenders: Time until Any Revocation
Survival Curves for Drug  and Non-Drug Offenders: Time until Any Revocation
Survival Curves for Drug  and Non-Drug Offenders: Time until Any Revocation

Drug Offenders - Non-Drug Offenders

Table A2
Survival Curves for Drug and Non-Drug Offenders: Time until Revocation with New Offence / Charge

Survival Curves for Drug  and Non-Drug Offenders: Time until Revocation with New Offence / Charge
Survival Curves for Drug  and Non-Drug Offenders: Time until Revocation with New Offence / Charge

Drug Offenders - Non-Drug Offenders

Appendix B. Survival Curves: Importing / Exporting and Trafficking Offenders

Table B1
Survival Curves by Drug Offender Subgroup: Time until Any Revocation

Survival Curves by Drug  Offender Subgroup: Time until Any Revocation
Survival Curves by Drug  Offender Subgroup: Time until Any Revocation

Importing / Exporting Offenders - Trafficking Offenders

Table B2
Survival Curves by Drug Offender Subgroup: Time until Revocation Due to New Offence / Charge

Survival Curves by Drug  Offender Subgroup: Time until Revocation Due to New Offence / Charge

Survival Curves by Drug  Offender Subgroup: Time until Revocation Due to New Offence / Charge

Importing / Exporting Offenders - Trafficking Offenders


1 Other sub-groups of drug offences could not be included for analysis due to low base rates.

2 Most serious offence was used rather than any offence in order to ensure comparability with other publications in the area, many of which use the most serious offence criteria. Most serious offence was determined according to a hierarchy of offences, with the following categories ranked from most to least serious: murder / manslaughter, attempted murder / manslaughter, sexual offences, robbery, assault, other violent offences, drug offences, impaired offences, break and enter, other property offences, other offences.

3 This procedure facilitated the examination of recidivism, as considering earlier convictions ensured the women in question had more time in which to be released, and, if applicable, have their release revoked.

4 Importation / exportation was considered most serious, followed by production or cultivation, trafficking and possession for the purposes of trafficking, and simple possession. Trafficking and possession for the purposes of trafficking were grouped together both due to their conceptual similarity and to be consistent with procedures followed in previous research (Motiuk & Vuong, 2006).

5 The first five of these domains each receive a rating of factor seen as an asset, no current difficulty, some difficulty, or considerable difficulty. Substance abuse issues and personal / emotional orientation receive one of the three ratings of difficulty, but cannot be rated as an asset.

6 This is relevant because follow-up analyses of revocation and re-offending were based exclusively on records from Canadian sources.

7 Of these 36 women, 11 were Caucasian, 19 were Black, none were Aboriginal, and 6 were in the ‘other / unknown’ ethnicity category.

8 American research demonstrated that only one out of every 13 defendants in drug-related cases was classified as a high-level importer (United States Sentencing Commission, 2007). Though this finding might not be fully generalizable to the Canadian context, it suggests a considerable proportion of importers in Canadian institutions may be classified at a relatively low level in the importation hierarchy.