Research Reports

An Analysis of Temporary Absences and the People Who Receive Them

¹ This report was originally prepared for the Temporary Absence Review Panel.

1992, N°. R-25

Research Report No. 25

Prepared by:

Brian A. Grant
R. L. Belcourt

Research and Statistics Branch
Correctional Service of Canada

January, 1992

Table of Contents

Executive Summary

Forward

List of Tables

List of Figures

Introduction

Analysis of Temporary Absences

16 Year Trends

5 Year Trends

1990/91 Data

All TAs

Group TAs

Single ETAs

Failed TAs

Summary

Analysis of those Receiving Temporary Absences

5 Year Trends

1990/91 Data

Summary

Special Analyses

Race

Summary

Gender

Summary

Sex Offenders

Summary

List of Tables

Table 1: Number of TAs and Failure by Type of Escort

Table 2: Number of Group and Single TAs, and Failures

Table 3: Number of TAs and Failures by Purpose of TA

Table 4: Detailed Completion Information for All TAs

Table 5: Number of TAs and Failures by Granting Authority

Table 6: Number of TAs and Failures by Region

Table 7: Number of TAs and Failures by Security of Institution

Table 8: Number of TAs and Failures by Major Admitting Offence

Table 9: Number of TAs by Escort and Type (Group or Single)

Table 10: Number of TAs by Purpose of TA and Escort Type

Table 11: Number of TAs by Region and Escort Type

Table 12: Size of Groups Used for Group TAs

Table 13: Number of Group TAs by Purpose and Escort Type

Table 14: Number of Group TAs by Region and Escort Type

Table 15: Type of Escort for Single ETA's by Success

Table 16: Type of Escort for Single ETA's by Purpose of ETA

Table 17: Type of Escort for Single ETA's by Region

Table 18: Major Admitting Offence for Offenders Receiving TAs

Table 19: Percentage of Sentence Served Before First UTA or Family and Community ETA by Major Admitting Offence

Table 20: Term of Incarceration for Offenders Receiving TAs

Table 21: Sentence Type for Offenders Receiving TAs

Table 22: Category 1 Designation for Offenders Receiving TAs

Table 23: C67 Offence Designation for Offenders Receiving TAs

Table 24: Percentage of Sentence Served Before First UTA and First Family and Community ETA

Table 25: Regional Distribution for Offenders Receiving TAs

Table 26: Percentage of Sentence Served Before First UTA and First Family and Community ETA by Region

Table 27: Gender of Offenders Receiving TAs

Table 28: Race

Table 29: Marital Status of Offenders Receiving TAs

Table 30: Religion

Table 31: TAs by Race and Type of TA

Table 32: Number of TAs by Race and Purpose

Table 33: Number of TAs by Race and TA Region

Table 34: Distribution of Racial Groups by Region

Table 35: Number of Offenders Receiving TAs by Race and Gender

Table 36: Offenders Receiving TAs by Number of Terms in Federal Institutions and by Race

Table 37: Number of Offenders Receiving TAs by Race and Major Admitting Offence

Table 38: Number of Offenders receiving TAs by Race and Designation as Category 1 and C67

Table 39: Percentage of Sentence Served before First UTA and First Family and Community ETA

Table 40: Number of Offenders Receiving TAs by Race and Region

Table 41: Number of TAs which were Group or Single by Gender

Table 42: Number of TAs by Gender and Purpose

Table 43: Number of TAs by Gender and Type of Escort

Table 44: Number of Offenders Receiving TAs by Gender and Major Admitting Offence

Table 45: Number of Offenders receiving TAs by Gender

Table 46: Number of Offenders Receiving TAs by Gender and Designation as Category 1 or C67

Table 47: Percentage of Sentence Served Before First UTA and First Family and Community ETA by Gender

Table 48: Percentage of Sentence Served Before First UTA and First Family and Community ETA by level of Physical Injury to Victim

Table 49: Percentage of Sentence Served Before First UTA and First Family and Community ETA by Level of Force Used

Table 50: Percentage of Sentence Served Before First UTA and First Family and Community ETA by Alcohol and Drug Involvement in Offence

Table 51: Percentage of Sentence Served Before First UTA and First Family and Community ETA by Treatment within the Correctional Service

List of Figures

Figure 1: Number of (in 1000's) of Temporary Absences: 1975/76 - 1990/91

Figure 2: Relative Frequency of Escorted and Unescorted Temporary Absences

Figure 3: Percentage of Temporary Absences Failures

Figure 4: Percentage of Failures for Escorted and Unescorted Temporary Absences

Figure 5: Relative percentage of Reasons for Escorted Absence

Figure 6: Relative Percentage of Reasons for Unescorted Temporary Absences

Figure 7: Reasons for Temporary Absences: Average Percentages over 5 Years

Figure 8: Escorted and Unescorted Temporary Absence Failures National - 1986/87 to 1990/91

Figure 9: Types of Major Offences for those on Temporary Absences

Figure 10: Major Admitting Offence for those Released on Temporary Absences (average of past 5 years)

Figure 11: Distribution of Major Admitting Offence for those Receiving Temporary Absences and Institution Population

Figure 12: Distribution by Race of those Released on Temporary Absences

Figure 13: Race of those Released on Temporary Absences (Average of past 5 years)

Figure 14: Race Distribution of those Receiving Temporary Absences and Institution Population

Figure 15: Gender Distribution for those Receiving Temporary Absences and Institution Population

Executive Summary

The report was prepared for the Temporary Absence Review Panel which was appointed by the Solicitor General to provide an independent review of the temporary absence program. The report provides a descriptive analysis of temporary absences (TAs) and the offenders who have received them. In addition to basic analyses the report contains a section looking at racial and gender differences, as well as, a section concerned specifically with sexual offenders.

The data for TAs indicate that a large number of TAs are taken by offenders in the federal correctional system. The success rate for TAs has been very high for the 16 years covered in this report (1976-1991), and has been improving during the past few years. Group TAs and TAs with resocialization escorts are used frequently and these allow for the most efficient use of correctional officers for the supervision of TAs. The majority of single TAs requiring a security escort are for medical reasons.

The analyses of offenders indicates that those convicted of the most serious offences are less likely to receive TAs and that they serve a larger percentage of the their sentence before receiving an escorted temporary absence (ETA) for family and community contact or an unescorted temporary absence (UTA). Multiple term offenders and those classified as Category 1 by the National Parole Board are also less likely to receive TAs. The data on TA failures indicate that failures for those offenders convicted of murder and other serious crimes is lower than the already very low failure rate for other offenders. This suggests that caution is exercised in determining when an offender can safely be exposed to the community. The data also suggest that there are regional differences in the percentage of people who are granted TAs and in the percentage of sentence that must be served before being permitted a UTA or family and community ETA. These regional differences may be due to differences in the offender profile in each region.

The analysis of racial data shows that, in terms of both the number of TAs granted and the number of offenders who receive TAs, Native peoples are under-represented. However, the analyses also show that Native peoples are more likely to have been convicted of more serious offences involving violence and are more likely to have served multiple federal prison terms. Further analyses are required to explain the results. In addition, it was shown that in the case of compassionate TAs, Native peoples are more likely to be granted a TA than are other racial groups.

The comparison of male and female offenders suggests that females are not treated differently from males in terms of the number of TAs received. However, there is evidence that females receive less family and community contact TAs than males. The offence data suggest that females are more likely to be serving time for crimes involving violence and this may explain the relatively fewer family and community TAs. Another possible reason for fewer family and community TAs may be the centralization of prison services for women in Kingston.

The data for sexual offenders seems to indicate that those who have committed more serious offences must wait longer for the opportunity to receive TAs that are discretionary (UTA or family and community ETA). Given that the percentage of time served is influenced by the length of the original sentence, and that more serious assaults are likely to result in longer sentences, it is probable that those offenders committing the most serious sexual assaults are spending considerably more time in institutions before receiving the opportunity for a temporary absence.

Forward

Two preliminary reports were prepared and provided to the Review Panel. The data that were presented in those reports have been incorporated into this report. Minor discrepancies in data presented in the previous reports and the final report should not be considered problematic as the earlier reports were based on interim data. These minor differences do not affect the interpretation of the results.

As the data are subdivided into smaller groups for detailed analyses the size of the sample decreases quite dramatically in some cases. Analyses based on small numbers of cases may be less accurate than those based on large samples and caution is required when interpreting the results obtained with small samples.

Introduction

The purpose of this report is to present an analysis of data on temporary absences and the people who receive temporary absences. The temporary absence program is not so much a program, as an element in the correctional process which offenders are exposed to while they are under the authority of the Correctional Service of Canada. Temporary absences provide an opportunity for the correctional system (consisting of case management officers, offenders, treatment personnel, parole officers, and administrators) to test the offenders ability to behave appropriately in non-prison settings under varying degrees of supervision.

As the level of supervision is reduced (from full security escort, possibly including restraint devices, to unescorted absences) offenders must take on greater and greater responsibility for their behaviour and thereby accept the risks that they may fail. Society, implicitly, accepts these risks too. However, while the offender is under the control of the Correctional Service it is much easier to provide controls on behaviour which will reduce the chances of failing to behave in acceptable ways. If failure occurs during this phase additional controls can easily be put in place, or the offender can be removed from potentially dangerous situations with relative ease. Once the offender has left the control of the Correctional Service the additional rights returned to the offender provide less opportunity for supervising behaviour and remedial action can only be taken when another offence has been committed.

Temporary absences provide a means for the system to test the offenders acceptance of the responsibilities of good citizenship. For the offender, they provide an opportunity for gradual reintegration into the community and the opportunity to review problems at an early stage, before they represent a threat to the offender or society.

Each year there are approximately 50,000 temporary absences from the Correctional Service prisons. The absences are for periods ranging from a few hours to 3 days. More than 99.5% of temporary absences are completed successfully, and in recent years this has risen to 99.9%. About 7,000 individuals are granted one or more temporary absences each year.

In analyzing data about temporary absences (TAs) it is necessary to make a distinction between the number of people who receive temporary absences and the number of temporary absences. As noted above, there are approximately 50,000 temporary absences (TAs) each year, but only 7,000 people receive them.

The data analyses which follow are presented separately for people and for absences. Some variables can only be analyzed on the basis of the people who received temporary absences and others can only be meaningfully analyzed on the basis of the number of TAs. A few variables may be analyzed for both people and TAs.

The report is divided into three main sections. The first section presents analyses based on TAs, the second section presents analyses based on the people who received TAs, and the third section presents data for analysis of special groups of offenders. The analysis of TAs is subdivided into three parts. The first presents the data for a period of 16 years from 1975/76 to 1990/91. These data provide information on long term trends. The second subsection, which contains more detailed analyses, presents the data for the five year period from 1986/87 to 1990/91. These analyses provide an overview of the data for the five year period and include more detailed analyses. The third subsection presents analyses of data for the year 1990/91. This section contains the most detailed analyses of the groups of offenders most likely to succeed and fail while on a temporary absences. The second section contains analysis of 5 year trend data, and detailed analyses of 1990/91 data. The third section presents analyses of TAs for racial groups (Caucasian, native peoples and other racial groups), gender and offenders convicted of sexual offences.

Analysis of Temporary Absences

16 Year Trends

The number of TAs for each year, from 1975/76 to 1990/91, are presented in Figure 1. These data indicate that the number of TAs has varied from 41,000 to 60,000, with the peak in 1987 and the lowest level of TAs in 1981. Since 1987 there has been a decline in the number of TAs, although there was an increase in the number in 1990/91. Superimposed on the Figure is the number of 'on register' offenders for each year. These data show that as the offender population grew so did the number of TAs until 1987. After that date the offender population continued to grow,but the number of TAs began to decline. The change in the patterns of TAs may be due to changes in the nature of the offender population, changes in the use of day parole, and closer scrutiny of TAs resulting from concern of public safety.

There are two main types of TAs, escorted TAs (ETAs) and unescorted TAs (UTAs). During an escorted TA the offender, or offenders, are accompanied by either a Correctional Service staff member or a private citizen volunteer. Figure 2 presents data on the relative frequency of escorted and unescorted TAs for the past 16 years. The data indicate that there has generally been a decline in the relative number of unescorted TAs. In 1975/76 over 40% of all TAs were unescorted, whereas in recent years unescorted TAs have represented only 15% of all TAs. Unescorted TAs are the most likely to result in failure (see Figure 4), and the decline in relative use may be, in part, due to greater concern about the risk to society.

The failure rates for TAs has consistently been very low. Figure 3 presents the percentage of TAs which have been classified as failures for the past 16 years. Failures are defined as cases where an offender has failed to return to the institution and has been charged with being unlawfully at large, where an offender has been detained by the police (they may be detained without being charged with committing a new offence), or where the TA has been terminated by the National Parole Board. The data in the figure show a general decline in the percentage of failures to a level of 0.13% for the two years 1989/90 and 1990/91. The highest percentage of TA failures occurred in 1979 when 0.49% of TAs failed, but even this figure represents only one half of one percent of all TAs. More detailed analyses of failures are presented for the 1990/91 data.

Figure 1: Number (in 1000's) of Temporary Absences and On-register Population: 1975/76 - 1990/91

Figure 2: Relative Frequency of Escorted and Unescorted
Temporary Absences

Figure 3: Percentage of Temporary Absences Failures

The failure rate for escorted and unescorted TAs is different and Figure 4 presents the failure rates separately for the two types of TAs. Escorted TAs have been extremely unlikely to fail over the years with less than 0.10% resulting in failure. Although more likely to result in failure, only about 1% of unescorted TAs result in failures, and the data in Figure 4 suggest that the percentage of failures has been declining over the past 10 years.

5 Year Trends

TAs are granted for four main purposes, medical, compassionate, administrative, and family and community contact. Medical TAs are used when an offender requires medical treatment which is not available in the institution. Compassionate TAs are granted when a family member dies or is seriously ill. Administrative TAs are used when an inmate needs to make contact with agencies in the community to make arrangements for other types of release (day parole, parole, mandatory supervision). Family and community contact TAs are granted to allow the offenders to participate in community activities which will assist in their adjustment to life after release from an institution. Family and community contact TAs range from meeting ones family outside the institution to participating in recreational activities designed to show the offender alternative ways of using free time.

Figures 5 and 6 present the relative percentages of the reasons for escorted and unescorted TAs. Figure 5 presents the data for escorted TAs, the most common type. The relative percentages have remained fairly constant over the 5 years, except for administrative TAs which have declined in use. Figure 6 presents the relative percentages of each type of unescorted TA. These data indicate that about 95% of all UTAs are for family and community contact, Figure 7 presents a summary of the 5 year data for all types of TAs. The data in the figure indicate that 71% of TAs are for community and family contact, about 22% are for medical reasons, and a small percentage, 6% are for administrative reasons. Compassionate TAs represent approximately 1% of all TAs.

Figure 8 presents the number of failures for escorted and unescorted TAs for the 5 year period. In absolute numbers escorted TA failures have declined from 39 to 13, and as shown in Figure 4, there was also in decline in the percentage of failures. Also, in absolute terms, unescorted TA failures have declined over the 5 year period from 137 in 1986/87 to 55 in 1990/91. Percentage failure rates for unescorted TAs have also shown a decline during this period.

Figure 4: Percentage of Failures for Escorted and Unescorted Temporary Absences

Figure 5: Relative Percentage of Reasons for Escorted Temporary Absence

Figure 6: Relative Percentage of Reasons for Unescorted Temporary Absences

Figure 7: Reasons for Temporary Absences: Average Percentages over 5 years

Figure 8: Escorted and Unescorted Temporary Absence Failures National - 1986/87 to 1990/91

1990/91 Data All TAs

In this section data are presented which describe TAs which occurred in 1990/91. Most tables provide information on the number of TAs for each analysis variable, and the number of failures which occurred for these types of TAs. Where appropriate the percentage of failures is also calculated.

Table 1 presents the number of TAs for each type of escort. TAs are classified into three escort types. Maintaining an acceptable level of risk for the community requires that some offenders receive greater supervision while on a temporary absence. Therefore, those offenders who pose the greatest risk are permitted TAs only with a security escort. A security escort is a correctional officer, restraint devices may be used. Other offenders, who represent less of a threat to the community , have a resocialization escort. These escorts may be a correctional officer, another Correctional Service staff member, or a private citizen volunteer. Those offenders requiring the least amount of supervision receive unescorted temporary absences. Almost 50% of all TAs have a resocialization escort. Another 37% of TAs have a security escort, and 15% of TAs have no escorts. Of the TAs which result in failure, 81% were unescorted. Those TAs with security escorts were the least likely to result in failure. The percentage of TAs which failed was less than 1.0% of all TAs and escorted TAs had failures less than 0.1% of the time.

Table 1: Number of TAs and Failure by Type of Escort

Type of Escort	All TAs		Failed TAs		Percentage of Failures
	Percent	Number	Percent	Number
Security	36.8	19,575	8.8	6	0.03
Resocialization	48.6	25,811	10.3	7	0.03
Unescorted	14.6	7,736	80.9	55	0.71
Total	100	53,122	100	68

Slightly more than 50% (54%) of all TAs occur with participants as members of a group (see Table 2). Of the 28,601 offenders who were on group TAs in 1990/91 only 3 failed to return, that is, 99.99% of all group TAs are successful. Failure is more likely with single TAs where individual offenders are responsible for the success, but even in these cases 99.63% of TAs are successful.

Table 2: Number of Group and Single TAs, and Failures

Type	All TAs		Failed TAs		Percentage of Failures
	Percent	Number	Percent	Number
Group	53.8	28,601	4.4	3	0.01
Single	46.2	24,521	95.6	65	0.27
Total	100	53,122	100	68

Data in Table 3 show the distribution of TAs by the purpose for which they are granted. As was seen in the 5 year trend data, the most common reason for granting TAs is for family and community contact. Relative to the percentage of these TAs granted they have more failures, but this is due in part to the fact that almost all unescorted TAs are for community and family contact. Medical TAs are the ones least likely to result in failure, but they are also the ones most likely to have a security escort.

Table 3: Number of TAs and Failures by Purpose of TA

Purpose	All TAs		Failed TAs		Percentage of Failures
	Percent	Number	Percent	Number
Compassionate	1.0	521	1.5	1	0.19
Medical	23.0	12,194	8.8	6	0.05
Administrative	2.8	1,479	0	0	0
Family and Community Contact	73.3	38,928	89.7	61	0.16
Total	100.1	53,122	100	68

TAs are normally classified as either successful or unsuccessful, but a more detailed breakdown of information on the completion of TAs is presented in Table 4. TAs are classified as successful because the offender returns to the institution, but the offender may return late, or may seek an extension because of unforseen circumstances. Failures may occur because the offender fails to return and is declared unlawfully at large (77%), or because they are detained by police (15%), or the TA is terminated by the National Parole Board because of a violation of conditions, or other concerns about the behaviour of the offender.

Table 4: Detailed Completion Information for All TAs*

Type of Completion		All TAs
		Percent	Number
Success	On time	94.2	48,959
	Extension	1.8	958
	Late	3.7	1,921
	Total	99.9	51,838
Failures	Unlawfully at large	76.5	52
	Detained by police	14.7	10
	Terminated by NPB	108.8	6
	Total	100	68

* A small percentage of TAs are used to permit offenders early release when a release date falls on a weekend or holiday, and these are not included in this table.

Table 5 presents the number of TAs which were approved by the National Parole Board, the Institution Director with the approval of the National Parole Board, the Institution director alone, and the Commissioner. The latter is used in very rare cases. The majority of TAs (93%) are approved by the Institution Director, and about half of these receive confirmation from the Parole Board. A smaller percentage of TAs receive their approval from the Parole Board. Failures are highest for those TAs approved by the National Parole Board, but these represent the most difficult cases.

Table 5: Number of TAs and Failures by Granting Authority

Granting Authority	All TAs		Failed TAs		Percentage of Failures
	Percent	Number	Percent	Number
National Parole Board	6.9	3,667	19.1	13	0.35
Inst. director with approval of NPB	46.2	24,566	33.8	23	0.09
Institutional director	46.8	24,869	47.1	32	0.13
Commissioner	0.03	17	0	0	0
Total		53,119		68

The number of TAs granted in each region is presented in Table 6. Relative to the size of the offender population in their region, the Pacific region grants the most TAs while Quebec and the Prairie regions are least likely to grant TAs. Those regions which are least likely to grant TAs are also the ones with the highest failure rates.

Table 6: Number of TAs and Failures by Region

Region	All TAs		Failed TAs		Percentage of Failures	Offender Population (March 31, 1991)
	Percent	Number	Percent	Number		Percent	Number
Atlantic	7.3	3886	4.4	3	0.07	8.4	1003
Québec	18.6	9876	29.4	20	0.2	26.9	3229
Ontario	31.2	16577	36.8	25	0.15	28.8	3455
Prairie	9.9	5279	17.6	12	0.23	22.2	2666
Pacific	33.0	17,504*	11.8	8	0.05	13.6	1635
Total	100	53,122	100	68		99.9	11,988

* 7366 of these TAs were from one institution and were taken by 201 different offenders.

Table 7 presents data for the number of TAs by the security of the institution where the TA originated from. The majority (52%) of TAs are taken from minimum security institutions although they hold only 12.5% of inmates. Only 9% of TAs are taken from maximum security institutions, but 30% of the offender population are held in these institutions. Ninety-five percent of TAs from maximum security institutions have an escort, 66% have a security escort, and of those with a security escort 95% are for medical reasons. Some inmates in maximum security institutions do not require such a high level of security but may request to be in one of these institutions for work reasons, or to be closer to home.

Table 7: Number of TAs and Failures by Security of Institution¹

Security	All TAs		Failed TAs		Percentage of Failures	Offender Population (March 31, 1991)2
	Percent	Number	Percent	Number		Percent	Number
Maximum	9.0	4799	13.2	9	0.19	25.6	3409
Medium	23.4	12,408	47.1	32	0.26	46.1	6141
Minimum	67.6	35915	39.7	27	0.08	14.2	1889
Total	100	53,122	100	68			11,439

¹ Offenders may serve their sentence in an institution with a higher security level than they require so as to be closer to home or for work opportunities.

² 550 offenders are serving their sentences in provincial institutions.

The distribution of TAs by major admitting offence is presented in Table 8 along with the failure rates for each offence type. Relative to their percentage, 13.1% of the offender population (see Table 2 offenders receiving TAs by major admitting offence) murderers receive a larger percentage of TAs than other offenders, but in terms of the number of offenders receiving TAs (10.7%) murderers are under represented. TAs are the only way for murderers to experience community contact until shortly before their parole eligibility date which is 25 years for first degree murder, and is set by the court for other murder convictions. In addition, the data indicate that this group has one of the lowest failure rates. The less serious crime of break and enter has the highest percentage of failures for TAs, at 0.45%. Those convicted of theft, robbery and similar offences are less likely to receive TAs (19.9%) than would be expected given their representation in the offender population (26.2%), but this group also has one of the highest failure offender population (26.2%), but this group also has one of the highest failure rates (0.26% of TAs). These data suggest that the risk of offenders failing while on a TA cannot be predicted by the seriousness of the offences for which they were convicted. One reason for this is likely that, of those offenders convicted of the most serious crimes, only those who are least risk to the community have the opportunity for a TA.

Table 8: Number of TAs and Failures by Major Admitting Offence

Security	All TAs		Failed TAs		Percentage of Failures
	Percent	Number	Percent	Number
Murder	17.4	9202	2.9	2	0.02
Sexual Offences	12.4	6558	0	0	0
Assault, injury	15.2	8044	5.9	4	0.04
Theft, robbery	19.9	10528	39.7	27	0.26
Break & enter	8.8	64648	30.9	21	0.45
Attempts, conspiracy, fraud	7.4	3911	7.4	5	0.13
Drug offences	11.4	6046	5.9	4	0.07
Other	7.4	3893	7.4	5	0.13
Total	99.9	52830	100.1	68

Slightly more than half of all TAs (as shown in Table 9), which require a security escort are part of group TAs. This arrangement makes it possible to reduce risk for the community while making effective use of correctional officer resources. In addition, a large percentage of group TAs (64%) are supervised by resocialization escorts. The percentage of failures for group TAs is extremely low (0.01%). Approximately 30% of single TAs have a resocialization escort, and another 31% are unsupervised UTAs. Of all TAs only 18% have a single security officer escorting a single offender.

Table 9: Number of TAs by Escort and Type (Group or Single)

Type of TA	Type of Escort
	Security		Resocialization		UTA no escort
	Percent	Number	Percent	Number	Percent	Number
Group	35.8	10,238	64.1	18,339	0.1	24
Single	38.1	9,337	30.5	7,472	31.4	7,712
Total		19,575		25,811		7,736

Table 10 presents the number of TAs by the purpose of the TA and the type of escort. Approximately 50% of TAs requiring a security escort are for medical reasons, and the other 50% are for family and community contact. Where escorts are required, resocialization escorts are used frequently (more than 50% of the time), and the majority of these provide for family and community contact. Eighty-one percent of TAs for community and family contact have an escort to reduce the risk to the community, and of these 57% are resocialization escorts. UTAs which represent the highest risk TAs provide only 18.8% of the family and community contact TAs.

Table 10: Number of TAs by Purpose of TA and Escort Type

Purpose of TA	Type of Escort
	Security		Resocialization		UTA no escort
	Percent	Number	Percent	Number	Percent	Number
Family & Community Contact	23.7	9,219	57.5	22,384	18.8	7,325
Administrative	8.4	125	86.5	1,280	5.0	74
Compassionate	30.7	160	53.0	276	16.3	85
Medical	82.6	10,071	15.3	1871	2.1	253
Total		19,575		25,811		7,736

The data in Table 11 show that the use of the different levels of security for TAs varies across the regions. Resocialization escorts are most commonly used in the Atlantic and Ontario regions, while Quebec and the Prairie regions use UTAs more frequently than the other regions. For most regions resocialization escorts are the most common, but in the Prairie region all three types of TA escorts are used about equally often.

Table 11: Number of TAs by Region and Escort Type

Region	Type of Escort
	Security		Resocialization		UTA no escort
	Percent	Number	Percent	Number	Percent	Number
Atlantic	20.3	789	71.0	2759	8.7	339
Québec	27.3	2696	49.7	4908	23.0	2272
Ontario	17.6	2916	70.9	11,752	11.5	1909
Prairie	36.0	1903	36.4	1922	27.5	1454
Pacific	64.4	11,271	25.5	4470	10.1	1763
Total		19,575		25,811		7,736

Group TAs

Group TAs are an efficient way to provide offenders with access to external resources because one escort can supervise many offenders. Table 12 presents data on the size of groups which are used for group TAs. Approximately 52% of group TAs have between 2 and 5 people in them, and an additional 26% have between 6 and 10 people. Groups greater than 10 account for 15% of the groups which are used for TAs. The 4744 group TAs in 1990/91 allowed for 28601 absences from federal institutions, that is, more than half of the TAs were as part of a group.

Table 12: Size of Groups Used for Group TAs

Size of Group	Percent	Number of Groups
1 Person1	5.9	282
2 People	18.7	887
3 People	13.1	622
4 People	11.1	527
5 People	9.6	454
6 to 10 People	26.3	1249
11 to 20 People	13.8	656
21 to 30 People	1.4	67

¹One person groups occur because of cancellations for those who should be part of group TA.

Group TAs are used most frequently for family and community contact (82% of group TAs), and less frequently for medical reasons. More than two thirds (72%) of the group TAs for family and community contact have a resocialization escort rather than a security escort. These data further illustrate the efficiency of using group TAs to provide offenders with the community contact while maintaining a low level of risk for the community.

Table 13: Number of Group TAs by Purpose and Escort Type*

Purpose of TA	Type of Escort
	Security		Resocialization		UTA no escort
	Percent	Number	Percent	Number	Percent	Number
Family & Community Contact	27.1	1053	72.7	2827	0.2	9
Administrative	6.2	12	93.8	180	0	0
Compassionate	20.0	1	60.03	3	20.0	1
Medical	83.4	549	16.4	108	0.2	1
Total		1615		3118		11

*Note: These data represent the number of Group TAs, and not the number of people who participated in group TAs.

Group TAs are used in all regions with Ontario and the Pacific regions making greatest use of them as shown in Table 14. In relative terms, the Prairie region and Quebec region make less use of group TAs than the other regions. Resocialization escorts are most frequently used with group TAs, but security escorts are commonly used in the Pacific region.

Table 14: Number of Group TAs by Region and Escort Type

Region	Type of Escort
	Security		Resocialization		UTA no escort		Total
	Percent	Number	Percent	Number	Percent	Number	Percent	Number
Atlantic	3.9	16	96.1	393	0	0	8.60	409
Québec	19.1	106	80.4	447	0.5	3	3	556
Ontario	9.1	137	90.8	1369	0.1	2	2	1508
Prairie	25.3	80	73.7	233	1.0	3	3	316
Pacific	65.3	1276	35.6	676	0.2	3	3	1955
Total		1615		3118		11	100	4744

Single ETAs

Table 15 presents the type of escorts used for single ETAs. Roughly half of single ETAs require security escorts, and the data in Table 16¹ indicate that most of these TAs are for medical reasons. Single ETAs are very successful with 99.96 of those requiring a security escort ending successfully and 99.93 of those using a socialization escort ending successfully. After medical reasons for single ETAs the most common reason is for community and social contact and 88% of these use a resocialization escort. Detailed data on purpose of single ETAs are presented in Table 16.

¹ Differences in the number of cases in Tables 15 and 16 are the result of missing data for some variables, and the exclusion of TAs for prerelease from estimates of TA success.

Table 15: Type of Escort for Single ETA's by Success

Successful	Security Escort		Resocialization Escort
	Percent	Number	Percent	Number
Yes	55.0	8873	45.2	7266
No	50.0	5	50.0	5
% Successful	99.96		99.93
Total		8878		7271

Table 16: Type of Escort for Single ETA's by Purpose of ETA

Purpose	Security Escort		Resocialization Escort
	Percent	Number	Percent	Number
Family & Community	11.6	697	88.4	5315
Administrative	14.7	84	85.3	486
Compassionate	36.8	157	63.2	269
Medical	85.7	8399	14.3	1402
Total		9337		7472

The Atlantic, Quebec and Ontario regions have similar distributions of security and resocialization escorts, but the Prairie and Pacific regions are more likely to use security escorts as shown in Table 17.

Table 17: Type of Escort for Single ETA's by Region

Region	Security Escort		Resocialization Escort
	Percent	Number	Percent	Number
Atlantic	54.6	728	45.4	604
Québec	54.4	2255	45.6	1888
Ontario	47.4	2525	52.6	2797
Prairie	63.8	1714	36.2	971
Pacific	63.6	2115	36.4	1212
Total		9337		7472

Failed TAs

The percentage of failed TAs were presented along with the other data and need not be repeated here. Some detailed analyses on failures were performed to determine if those who failed were given another opportunity for a TA, and if they subsequently had problems on TAs. One restriction to these data is that they apply only to the fiscal year used for the analysis, that is, 1990/91. Therefore, when numbers of TAs before and after a failure are presented, they represent only a snapshot from one year.

As presented in the previous data, there were 68 TAs that were classified as failures. Two offender had two failures, so there were 66 offenders who had a failure. Prior to their failures, the failure group had received 361 TAs including 114 UTAs. They had, therefore, probably established themselves as reasonably acceptable risks. After recording a failure, 21 offenders received 133 additional TAs. Of these 29% (38) were for medical reasons, and 114 were with either a security or resocialization escort. There were also 19 successful UTAs taken. Forty-five of those who unsuccessfully completed a TA were not granted additional opportunities during the year. However, a number of these (8) were still UAL at the end of fiscal year, and others may have been held by police for extended periods of time making them ineligible for TAs.

Summary

The data on TAs indicate that a large number of TAs are taken by offenders in the federal correctional system, The success rate for TAs is very high, has been high for periods covered in this report, and have been improving during the past few years. Group TAs and TAs with resocialization escorts are used frequently and these allow for the most efficient use of correctional officers for the supervision of TAs. The majority of single TAs requiring a security escort are for medical reasons, and it is unlikely that the number of these could be reduced.

Analysis of Those Receiving Temporary Absences

The previous section presented data on the number of TAs. However, there is also a need to look at the people who took TAs. During 1990/91 approximately 7200 offenders took one or more TAs, and it is this group which serves as the basis for the analyses which follow. Although all analyses could be conducted using the TA data, these analyses would be biased when considering data which relate to individuals, rather than TAs. The bias is introduced because of multiple TAs for a single offender. If one offender within a subgroup receives a large number of TAs that offenders personal data will bias any analyses which are conducted. For offender variables, the most appropriate data to use are those which count each offender only once.

In addition to providing tables comparing various variables on the basis of whether or not a TA was taken this section presents some data on the percentage of sentence served before a TA was received. For these analyses it was decided that TAs which were least discretionary need not be considered. If someone requires a medical TA they will receive it even if it requires a very high level security escort. If a compassionate TA, or a administrative TA is granted it is because of a need. On the other hand, TAs for community and social contact are more discretionary and may be viewed as a privilege earned. Therefore, in the analyses of percentage of sentence served we look at UTAs as a group and we look at ETAs for community and family contact. It is predicted that higher risk offenders will need to wait a longer period of time before receiving their first UTA or their first family and community ETA.

It should be remembered throughout these analyses that one very large group of offenders is excluded from all of the data. These are the offenders who did not receive a TA. Although the data may indicate a relationship between the variables this relationship may only exist for those who were granted a TA, not for all offenders. The process leading to the granting of TAs, and particularly the more discretionary TAs, ensures that those offenders who would pose the highest risk to society are not released. The sample of offenders who receive a TA is a select sample of the offender population.

5 Year Trends

Figure 9 presents data on the number of offenders receiving TAs by their major admitting offence. The number of offenders receiving TAs has remained relatively constant, during the past five years, for some offences (murder, sexual offences, drug offences and other offences), but has declined for others (break & enter; theft & robbery; assaults and injury to people; attempts, conspiracy & fraud). It is likely that the screening process for TA releases has been fairly consistent for those offences showing little change over the five year period, but that it has been made more stringent for those offence groups which have experienced a decline.

Figure 9: Types of Major Offences for those on Temporary Absences

Figure 10 presents the number and percentage of offenders in each of the offence categories averaged over 5 years, and this can be compared to the number of offenders in each offence group for the entire population of offenders (March 31, 1991) in Figure 11. The data in Figure 11 would suggest that those offenders convicted of the most serious offences are less likely to receive TAs than those convicted of less serious offences.

The racial distribution of those receiving TAs during the past 5 years is presented in Figure 12. The data indicate that the relative number of native and other racial groups receiving TAs has remained relatively constant, while the percentage of caucasians receiving TAs has declined. The actual number of offenders per year in each racial group (averaged over five years) is presented in Figure 13 as a pie chart. For those offenders receiving TAs approximately 9% were Native peoples and 5% were from other racial groups with the balance being listed as caucasians. These numbers are compared to the percentage of each racial group in the offender population in Figure 14. Figure 14 indicates that for the noncaucasian groups their numbers are underrepresented in the group receiving TAs. More detailed analyses presented later suggest that the offence profile of these groups indicate more serious crimes were committed and therefore they are less likely to receive TAs.

The vast majority of offenders in federal prisons are males. Approximately 2% are female and about the same percentage received TAs over the 5 year period. The data, as shown in Figure 15, suggest that slightly fewer females may have received TAs than their relative proportion of the population of offenders. This difference is not found in the detailed 1990/91 data.

1990/91 Data

More detailed data on offenders who received TAs in 1990/91 are presented in the following tables. Along with the data on the distribution of offenders for each variable the number of offenders who failed on a TA are also presented. The TA failure rates were presented in the previous section and will not be presented again in this section. In addition, these tables include data on the distribution of offenders in the population as of March 31, 1991.

Figure 10: Major Admitting Offence for those Released on Temporary Absences (average of past 5 years)

Figure 11: Distribution of Major Admitting Offence for those Receiving Temporary Absences and Institution Population

Figure 12: Distribution by Race of those Released on Temporary Absences

Figure 13: Race of those Released on Temporary Absences (Average of past 5 years)

Figure 14: Race Distribution for those Receiving Temporary Absences and Institution Population

Figure 15: Gender Distribution for those Receiving Temporary Absences and Institution Population

Table 18, which presents data for the major admitting offence, indicates that most offence groups receive TAs consistent with their representation in the population, except murderers (less likely to receive TAs) and those serving time for drug offences (more likely to receive TAs). The majority (84%) of TAs for those convicted of murder and sexual assault are escorted, with 31% requiring a security escort, and 80% of these are for medical reasons.

Table 18: Major Admitting Offence for Offenders Receiving TAs

Offence	Offenders Receiving TA's		Failed TA's		Offender Population (March 31, 1991)
	Percent	Number	Percent	Number	Percent	Number
Murder	10.7	758	2.9	2	13.1	1573
Sexual Offences	13.2	934	0.0	0	13.4	1603
Assault, Injury	13.9	985	5.9	4	14.9	1788
Theft, Robbery	26.0	1844	39.7	27	26.2	3136
Break & Enter	14.4	1025	30.9	21	14.6	1747
Attempt, Conspiracy	6.5	460	7.4	5	4.9	588
Drug Offences	10.0	707	5.9	4	8.1	972
Other	5.5	387	7.4	5	4.9	582
Total	100.2	7100	100.1	68	100.0	11,989

As discussed earlier the percentage of sentence served to first UTA and to first family and community ETA were calculated. These data for major admitting offence are presented in Table 19. For most offence groups slightly less than 30% of the sentence is served before a family and community ETA, but for those convicted of attempts, conspiracy and fraud, drug offences, and other offences approximately 20% of time is served before being permitted an ETA of this type. For UTAs this percentage increases to about 35% and 25% for the two groups of offences.

Table 19: Percentage of Sentence Served Before First UTA or Family and Community ETA by Major Admitting Offence*

Offence	UTA			Community and Family ETA
	Mean	Number	Standard Deviation	Mean	Number	Standard Deviation
Sexual Offences	36.5	187	15.1	27.2	372	16.9
Assault, Injury	31.4	226	13.5	28.4	399	18.3
Theft, Robbery	33.6	490	17.0	29.9	698	18.8
Break & Enter	35.9	259	20.8	28.0	418	20.4
Attempt, Conspiracy, fraud	25.6	170	14.0	21.1	254	17.2
Drug Offences	24.3	279	11.6	16.9	437	11.3
Other	28.2	115	13.5	20.3	189	16.2
	F(1,1719) = 22.3, p < .001			F (1,2760) = 35.5, p < .001

* The murder group was dropped because of the very small sample

A series of analyses were conducted to compare offenders receiving TAs in terms of the seriousness of their offence and their criminal history. These analyses, presented in Tables 20 to 24, provide data on the federal penitentiary term the inmate is serving (each term represents a mandatory supervision failure or the commission of another offence with a federal sentence), the type of sentence (standard, indeterminate, indefinite or life), whether the offender has been classified as Category 1 by the National Parole Board (requiring special review before release), or is a scheduled offender under bill C67 (allowing the offender to be detained when the mandatory supervision date is reached).

Table 20 presents the data for Term of incarceration in a federal institution. In general the percentage of each group receiving a TA is similar to their representation in the offender population, but those who show repeated offences are slightly less likely to receive TAs. The data also suggest that multiple term offenders may be at greater risk of failing a TA than other offenders. Those least likely to fail while on a TA are those serving a first term.

Table 20: Term of Incarceration for Offenders Receiving TAs

Term	Offenders Receiving TAs		Failed TAs		Offender Population (March 31, 1991)
	Percent	Number	Percent	Number	Percent	Number
1	61.0	4393	45.6	31	58.9	7071
2	17.5	1264	17.6	12	17.2	2063
3 or more	21.5	1548	36.8	25	24.1	2865
Total	100.0	7205	100	68	100.2	11,989

Table 21 presents the data for the different types of sentences. The majority of TAs are received by those serving regular sentences and these are the people most likely to have a failed TA. The more serious offenders, who are serving life or indeterminate sentences, still receive TAs, but the failure rate is extremely low.

Table 21: Sentence Type for Offenders Receiving TAs

Sentence	Offenders Receiving TAs		Failed TAs		Offender Population (March 31, 1991)
	Percent	Number	Percent	Number	Percent	Number
Regular	87.1	6,303	97.1	66	83.4	9,999
Lt. Gov. Warrant	0.1	6	0.0	0	0.1	8
Habitual Offender	0.0	3	0.0	0	0.0	2
Dangerous Sex Offender	0.3	19	0.0	0	0.3	31
Life	11.9	862	2.9	2	15.3	1,834
Dangerous Offender	0.5	37	0.0	0	0.8	99
Life (Commuted from death)	0.1	5	0.0	0	0.1	16
Total	87.5	7,235	97.1	68	83.7	11,989

Tables 22 and 23 present data for Category 1 and C67 offenders, those who have committed the most serious crimes. The data indicate that Category 1 offenders are less likely to receive TAs than those who are not Category 1 offenders. In addition, relative to their TA exposure these offenders are less likely to have a TA failure. Those whose offences are listed in bill C67 receive TAs equal to their proportion in the offender population.

Table 22: Category 1 Designation for Offenders Receiving TAs

Category	Offenders Receiving TAs		Failed TAs		Offender Population (March 31, 1991)
	Percent	Number	Percent	Number	Percent	Number
Not Category 1	43.5	3136	58.8	40	38.2	4585
Category 1	56.5	4069	41.2	28	61.8	7404
Total	100.0	7205	100.0	68	100.0	11,989

Table 23: C67 Offence Designation Offenders Receiving TAs

C67 Offence	Offenders Receiving TAs		Failed TAs		Offender Population (March 31, 1991)
	Percent	Number	Percent	Number	Percent	Number
Not C67	52.6	3793	60.3	41	53.0	6358
C67	47.4	3412	39.7	27	47.0	5631
Total	100.0	7205	100.0	68	100.2	11,989

The percentage of time served before receiving a UTA or a family and community ETA was analyzed for C67 and Category 1 offenders and the results are presented in Table 24. The data indicate that offenders in these classifications must wait significantly longer for nonessential TAs. For ETAs the difference is about 7% of the sentence and for UTAs it is 4% and 5% for category 1 and C67, respectively.

Table 24: Percentage of Sentence Served Before First UTA and First Family and Community ETA

		UTA			Community and Family ETA
		Mean	Number	Standard Deviation	Mean	Number	Standard Deviation
C67	Yes	34.0	819	15.3	29.2	1334	17.4
	No	28.9	911	16.9	16.9	1439	18.0
		F (1,1728) = 28.6, p <.001			F (1,2271) = 104.3, p <.001
Category 1	Yes	33.6	791	15.9	29.2	1285	17.6
	No	29.4	939	15.4	22.3	1488	17.8

Number of people granted TAs and failure rates by region are presented in Table 25. Some regions are less likely to grant TAs (differences in proportion of offenders and proportion of offenders who receive TAs). TA failures seem to be less likely in the Atlantic and Pacific regions, but more likely in the Ontario region. The percentage of sentence served is presented in Table 26 by region. These data suggest that those in the Prairie and Pacific regions serve a greater percentage of their sentence before receiving a UTA, or a family and community ETA.

Table 25: Regional Distribution for Offenders Receiving TAs

Region	All TAs		Failed TAs		Offender Population (March 31, 1991)
	Percent	Number	Percent	Number	Percent	Number
Atlantic	9.8	697	4.4	3	8.4	1003
Québec	30.0	2127	30.9	21	26.9	3229
Ontario	28.8	2001	35.5	24	28.8	3455
Prairie	15.8	1122	17.6	12	22.2	2666
Pacific	16.2	1153	11.8	8	13.6	1635
Total	100.0	7100	100	68	99.9	11,988

Table 26: Percentage of Sentence Served Before First UTA and First Family and Community ETA by Region

Region	UTA			Community and Family ETA
	Mean	Number	Standard Deviation	Mean	Number	Standard Deviation
Atlantic	33.1	173	15.3	22.2	410	17.8
Québec	29.0	561	15.5	24.7	792	16.0
Ontario	30.5	558	16.7	24.6	819	19.4
Prairie	34.5	158	14.8	29.2	340	18.0
Pacific	34.4	280	17.8	29.0	412	18.5
	F (4,1725) = 7.7, p < .001			F (4,2768) = 11.8, p<.001

Table 27 presents the TA data by gender. It would appear from these data that males and females are equally likely to receive TAs, and that the failure rates for the two groups are about the same.

Table 27: Gender of Offenders Receiving TAs

Gender	All TAs		Failed TAs		Offender Population (March 31, 1991)
	Percent	Number	Percent	Number	Percent	Number
Male	97.3	6907	97.1	66	97.7	11,717
Female	2.7	193	2.9	2	2.3	272
Total	100.0	7100	100	68	99.9	11,988

The data on race are presented in Table 28. Relative to their representation in the offender population Native peoples are slightly less likely to be granted TAs. Additional data on Native peoples are presented in the section on special analyses.

Table 28: Race

Race	All TAs		Failed TAs		Offender Population (March 31, 1991)
	Percent	Number	Percent	Number	Percent	Number
Caucasian	85.1	6035	85.3	58	81.3	9718
Native	9.3	656	11.8	8	11.8	1411
Other (stated)	5.6	398	2.9	2	6.9	822
Total	100.0	7089	100	68	100	11,951

Tables 29 and 30 present data on the marital status and religious orientation of offenders receiving TAs. The data indicate that those involved in a marriage or common law relationship are slightly more likely to receive a TA, but the failure rate for these offenders is also slightly higher. Single offenders are slightly less likely to receive TAs. The distribution of offenders receiving TAs by their stated religious affiliation is consistent with the religious distribution in the offender population.

Table 29: Marital Status of Offenders Receiving TAs

Marital Status	All TAs		Failed TAs		Offender Population (March 31, 1991)
	Percent	Number	Percent	Number	Percent	Number
Single	45.0	3187	48.5	33	49.5	5893
Married	14.2	1006	16.2	11	12.4	1473
Common Law	27.7	1963	29.4	20	26.3	3137
Widowed	1.3	92	0.0	0	1.1	133
Separated	4.3	307	1.5	1	4.0	479
Divorced	7.4	524	4.4	3	6.8	806
Total	100.0	7079	100	68	100	11,921

Table 30: Religion

Religion	All TAs		Failed TAs		Offender Population (March 31, 1991)
	Percent	Number	Percent	Number	Percent	Number
Protestant	26.1	1716	23.8	15	25.4	2782
Roman Catholic	59.6	3917	65.1	41	58.0	6365
Other (listed)	1.8	117	3.2	2	1.8	202
Other (not listed)	9.9	651	4.8	3	11.7	1283
Atheist	2.7	176	3.2	2	3.1	335
Total	100.0	6577	100.0	63	100.0	10,967

Summary

These analyses suggest that those offenders convicted on the most serious offences are less likely to receive TAs and they serve a larger percentage of the their sentence before receiving an ETA for family and community contact or a UTA. Multiple term offenders and those classified as Category 1 by the NPB are also less likely to receive TAs. The data on TA failures indicate that failures for those offenders convicted of murder and other serious crimes is lower than the already very low failure rate for other offenders. This suggests that caution is exercised in determining when an offender can safely be exposed to the community. The data also suggest that there are regional differences in the percentage of people who get TAs and in the percentage of sentence that must be served before being permitted a UTA or family and community ETA. These regional differences may be due to differences in the offender profile in each region.

Special Analyses

Three special analyses were conducted on race, gender, and sex offenders. The analyses for race were conducted to determine why Native peoples were less likely to receive TAs than non-Natives. There was also concern about differences between the way males and females were treated with respect to TAs. Finally, the opportunity existed to take a detailed look a sex offenders and to determine if different types of sex offenders were treated differently.

Race

Tables 31 through 41 present analyses comparing offenders who had described themselves as either Caucasian, Native peoples, or from another racial group. Native peoples represent 11.8% of the offender population and given that all else is the same we would expect that they would receive 11.8% of the TAs. In fact they receive approximately 8% of the TAs. The lower rate of TAs for Native peoples may be the result of a more serious offence profile, rather an indication of bias within the Correctional Service.

Table 31 presents the number of group and single TAs for each racial group. The data indicate that Native peoples receive 8.4% of the group TAs, and only 7.7% of the single TAs.

Table 31: TAs by Race and Type of TA

Race	Type of TA
	Group		Single
	Percent	Number	Percent	Number
Caucasian	83.3	23619	85.9	20862
Native	8.4	2384	7.7	1868
Other	8.3	2354	6.5	1569
Total	100	28,357	100.1	24,299

Data on the purpose for which TAs were granted are presented in Table 32. These data indicate that Native peoples receive larger percentages of compassionate and administrative TAs than their representation in the offender population, but receive lower numbers of TAs for family and community contact and for medical reasons. Members of other racial groups receive TAs in proportion to their numbers in the offender population (6.9%), except in the case of compassionate TAs, where they are much less likely to receive a TA.

Table 32: Number of TAs by Race and Purpose

Purpose	Race
	Caucasian		Native		Other
	Percent	Number	Percent	Number	Percent	Number
Compassionate	80.9	420	16.8	87	2.3	12
Medical	86.4	10,420	7.7	928	5.9	707
Administrative	79.6	1168	14.2	209	6.1	90
Family & Community Contact	84.1	32,473	7.8	3028	8.1	3114
Total		44,481		4252		3923

* Note: Native people are 11.8% of the offender population and other racial groups are 6.9%.

Table 33 presents the racial distribution of the number of TAs by region and Table 34 presents the racial distribution by region in the offender population (March 31, 1991). The distribution of Native peoples in the offender population is similar to that for TAs in all regions except the Prairies. Native peoples represent 37% of the offender population in the Prairie region, but receive only 27% of the TAs. However, 70% of all Native peoples offenders are in the Prairie region. If this group has a more serious criminal profile then the observed difference in the Prairie region may be explained. There are differences in the distribution of TAs and representation in the offender population for Other racial groups, but no clear pattern is present.

Table 33: Number of TAs by Race and TA Region

Region	Race
	Caucasian		Native		Other
	Percent	Number	Percent	Number	Percent	Number
Atlantic	92.1	3351	3.7	134	4.3	155
Québec	98.0	9879	0.3	31	1.7	173
Ontario	87.3	14,110	3.8	615	8.9	1440
Prairie	68.9	3746	27.3	1485	3.9	210
Pacific	77.3	13395	11.5	1987	11.2	1945
Total		44,481		4252		3923

Table 34: Distribution of Racial Groups by Region (On register population less federal parole of March 31, 1991)

Region	Race
	Caucasian		Native		Other
	Percent	Number	Percent	Number	Percent	Number
Atlantic	91.5	918	3.9	39	4.6	46
Québec	94.9	3049	1.0	31	4.1	133
Ontario	84.4	2904	3.7	128	11.9	409
Prairie	58.1	1547	37.2	989	4.8	126
Pacific	79.7	1300	13.7	223	6.6	108
Total		9718		1410		822

A larger percentage of Native peoples and Other racial groups are female than is found for caucasians. Details of the gender distributions are found in Table 35.

Table 35: Number of Offenders Receiving TAs by Race and Gender

Gender	Race
	Caucasian		Native		Other
	Percent	Number	Percent	Number	Percent	Number
Male	98.0	5913	95.4	626	89.7	357
Female	2.0	122	4.6	30	10.3	41
Total	100	6035	100	656	100	398

One method of measuring the seriousness of criminal behaviour is to look at the number of times an offender has been in the federal penitentiary system. Data on the number of federal terms by race are presented in Table 36. Sixteen percent of Native peoples who had a TA were serving their 4th or higher term in a federal prison, while only 12% of Caucasians were serving 4th or higher terms.

Table 36: Offenders Receiving TAs by Number of Terms in Federal Institutions and by Race

Number of Terms	Race
	Caucasian		Native		Other
	Percent	Number	Percent	Number	Percent	Number
1	59.5	3591	57.9	380	79.2	315
2	18.5	1118	15.1	99	10.8	43
3	10.4	625	11.4	75	6.5	26
4 or more	11.6	701	15.5	102	3.5	14
Total	100	6035	100	656	100	398

Table 37 presents the racial distribution of major admitting offences. Amongst Native peoples 27% were serving sentences for crimes of assault or ones that cause injury, and 18% were serving sentences for sexual offences. By comparison, 12% of Caucasians were serving sentences for assault, and 13% for sexual offences. There was very little difference in the percentage of offenders in each racial group who were serving time for murder. Native peoples were much less likely than others to be serving time for drug offences, whereas, those from other racial groups more likely to be serving time for these offences.

Table 37: Number of Offenders receiving TAs by Race and Major Admitting Offence

Offence		Race
		Caucasian		Native		Other
		Percent	Number	Percent	Number	Percent	Number
C67	Yes	46.8	2,822	64.3	422	41.2	164
	No	53.2	3,213	35.7	234	58.8	234
	Total	100	6.035	100	656	100	398
Category 1	Yes	56.0	3380	72.6	476	52.5	209
	No	44.0	2655	27.4	180	47.5	189
	Total	100	6035	100	656	100	398

Native peoples were more likely to have been convicted of an offence scheduled under Bill C67, or listed as Category 1 by the National Parole Board. Sixty-four percent of Native peoples committed C67 offences and 73% committed Category 1 offences. This compares to 47% of Caucasians convicted of C67 offences and 56% convicted of Category 1 offences.

Table 38: Number of Offenders receiving TAs by Race and Designation as Category 1 and C67

Offence		Race
		Caucasian		Native		Other
		Percent	Number	Percent	Number	Percent	Number
C67	Yes	46.8	2,822	64.3	422	41.2	164
	No	53.2	3,213	35.7	234	58.8	234
	Total	100	6.035	100	656	100	398
Category 1	Yes	56.0	3380	72.6	476	52.5	209
	No	44.0	2655	27.4	180	47.5	189
	Total	100	6035	100	656	100	398

Data presented in Table 39 indicate that Native peoples serve a larger percentage of their sentence before receiving a UTA or family and community ETA. Members of other racial groups serve the same or slightly less time before receiving their first UTA or ETA.

Table 39: Percentage of Sentence Served before First UTA and First Family and Community ETA

		Mean	Number	Standard Deviation
UTA	Caucasian	31.0	1540	16.4
	Native	37.8	95	16.8
	Other	29.5	95	13.3
ETA / Family and Community	Caucasian	25.2	2372	18.1
	Native	30.9	229	19.3
	Other	22.6	167	14.0

Table 40 presents the distribution of offenders who received TAs by race and region. These data need to be compared with those in Table 34 to see racial differences for those receiving in TAs. As was found for the data on the number of TAs Native peoples are less likely to receive TAs in the Prairie region than would be expected based on the representation in the offender population.

Table 40: Number of Offenders Receiving TAs by Race and Region

Region	Race
	Caucasian		Native		Other
	Percent	Number	Percent	Number	Percent	Number
Atlantic	91.5	638	4.6	32	3.9	27
Québec	96.5	2049	0.7	15	2.8	60
Ontario	85.9	1715	4.2	84	9.9	197
Prairie	62.7	703	33.7	378	3.6	40
Pacific	80.8	930	12.8	147	6.4	74
Total		6035		656		398

Summary

These analyses have shown that in terms of both the number of TAs granted, and the number of offenders who receive TAs, Native peoples are under-represented. However, the analyses also show that Native peoples are more likely to have been convicted of more serious offences involving violence and are more likely to have served multiple federal prison terms. They may therefore, represent a greater risk to the community, and as with other offenders in these categories, are less likely to be granted TAs. In addition, it was shown that in the case of compassionate TAs, Native peoples are more likely to be granted an absence than other racial groups.

Gender

Tables 41 to 47 present an analysis of differences between males and females with respect to TAs. Data in Table 41 indicate that females are more likely than males to have a single, rather than a group TA. In 1990/91 females represented 2% of the offender population and received 3% of the TAs.

Table 41: Number of TAs which were Group or Single Gender

Group or Single	Male		Female
	Percent	Number	Percent	Number
Group	54.7	28,046	27.2	427
Single	45.3	23,217	72.8	1140
Total	100	51,263	100	1567

Table 42, which presents the purposes for which TAs were granted, indicates that females more frequently have TAs for medical reasons, and are less likely than males to have TAs for family and community contact. Females are more likely to have resocialization escorts than males, and are also less likely to have a UTA than males (see Table 43).

Table 42: Number of TAs by Gender and Purpose

Purpose	Male		Female
	Percent	Number	Percent	Number
Medical	22.6	11610	30.2	473
Compassionate	1.0	505	0.9	14
Administrative	2.6	1359	6.9	108
Family and Community Contact	73.7	37789	62.0	972
Total	99.9	52163	100.0	1567

Table 43: Number of TAs by Gender and Type of Escort

Type of Escort	Male		Female
	Percent	Number	Percent	Number
Security	37.2	19,087	24.8	389
Resocialization	47.9	24,544	70.2	1100
Unescorted	14.9	7632	5.0	78
Total	99.9	52163	100.0	1567

Females are more likely to be serving sentences for crimes of assault and causing injury, and for drug offences than are males. The data in Table 44 also indicate that males are more likely to have been convicted of sexual offences, break and enter, and theft and robbery.

Table 44: Number of Offenders Receiving TAs by Gender and Major Admitting Offence

Major Admitting Offence	Male		Female
	Percent	Number	Percent	Number
Murder	11.6	799	15.5	30
Assault, injury	13.0	898	23.8	46
Sexual offences	13.4	924	1.6	3
Theft, robbery	27.0	1864	21.2	41
Break & enter	14.1	975	2.6	5
Attempts, conspiracy, fraud	5.8	404	8.8	17
Drug offenders	9.2	633	20.7	40
Other	5.9	410	5.7	11
Total	100.0	6907	99.9	193

The data in Table 45 indicate that females are less likely than males to be serving multiple federal terms of incarceration. Although males and females are about equally represented in terms of C67 offences females offenders are more likely to be serving time for a Category 1 offences as shown in Table 46.

Table 45: Number of Offenders Receiving TAs by Gender and Number of Terms in a Federal Prison

Number of Terms	Male		Female
	Percent	Number	Percent	Number
1	60.1	4148	76.7	148
2	17.8	1234	13.5	26
3	10.4	718	4.2	8
4 or more	11.4	807	5.7	11
Total	100.0	6907	100.1	193

Table 46: Number of Offenders Receiving TAs by Gender and Designation as Category 1 or C67

		Male		Female
		Percent	Number	Percent	Number
C67	Yes	48.1	3323	46.1	89
	No	51.9	3584	53.9	104
	Total	100	6907	100	193
Category 1	Yes	57.2	3950	61.7	119
	No	42.8	2957	38.3	74
	Total	100	6907	100	193

There is very little difference between males and females in the percentage of time served before first UTA (25.6% vs 22.3%) and first community and family ETA (31.3% vs 29.5%).

Table 47: Percentage of Sentence Served Before First UTA and First Family and Community ETA by Gender

		Mean	Number	Standard Deviation
UTA	Female	29.5	32	14.1
	Male	31.3	1698	16.4
	F (1,1728) = .4, ns
ETA	Female	22.3	100	16.4
	Male	25.6	2673	18.1
	F (1,2771) = 3.19, ns

Summary

The comparison of male and female offenders suggests that females are not treated differently from males in terms of the number of TAs received. However, there is evidence that females receive less family and community contact TAs than males. The offence data suggest that females are more likely to be serving time for crimes involving violence and this may explain the relatively fewer family and community TAs. Another possible reason for fewer family and community TAs may be the centralization of prison services for women in Kingston.

Sex offenders

On March 31, 1991 a census was conducted of all sex offenders in federal prisons and under supervision. Included in the census were offenders who were serving time for a sexual offence, or who had been convicted of a sexual offence in the past.

The census questionnaire contained a description of the sexual offences which had been committed, including a description which reflected the level of violence and severity of the offence. The census identified 3089 sex offenders, 2183 of whom were in prisons. Of these, 464 offenders had received a TA in 1990/91 so they were in our TA file, and we had determined the date of their first UTA or their first Family and community ETA.

The 464 offenders were included in our analysis group. Since the analysis sample was only 25% of the original sex offender population (from the census) comparisons were made between the groups. The detailed results of these comparisons are not included here, but they indicated that we had in fact obtained a reasonable sample of sex offender cases with one important distinction. The combined file contained slightly higher percentages of the less serious sex offence characteristics and a slightly lower percentage of the more serious sex offence characteristics. The difference is not surprising given that those convicted of more serious offences are less likely to appear in the temporary absence data. That is, those offenders convicted of more serious offences are less likely to have received an unescorted temporary absence or an escorted temporary absence for family and community contact.

Table 48: Percentage of Sentence Served Before First UTA and First Family and Community ETA by level of Physical Injury to Victim

Level of Injury	UTA			Family and Community ETA
	Mean	Number	Standard Deviation	Mean	Number	Standard Deviation
No injury	34.1	39	12.2	22.9	72	13.7
Slight injury (no weapon)	43.5	31	21.6	30.7	59	22.6
Slight Injury (weapon)	35.6	7	15.8	29.0	19	17.9
Treated and released	33.5	13	11.5	24.6	30	17.7
Hospitalized	36.1	9	7.4	32.6	19	12.1

Table 48 presents the percentage of sentence served by level of injury inflicted on the victim during a sexual offence. Percentage of sentence served to first family and community ETA was lowest for those offenders who had inflicted no injury on their victim. It was highest for those who had hospitalized the victim as a result of the attack. The data for percentage of time to first UTA shows no consistent pattern. The difference in percentage time served between no injury and hospitalizing the victim is minimal, and the largest percentage served is for those who inflicted only slight injury without the presence of a weapon.

Percentage of time served to first family and community ETA by the level of force used during a sexual offence shows a consistent increase as the level of force rises. Data in Table 49 indicate that for those offenders using no force they received their family and community ETA after serving 24% of their sentence while those who assaulted their victim had to serve 31% of their sentence.

Table 49: Percentage of Sentence Served Before First UTA and First Family and Community ETA by Level of Force Used

Level of Force	UTA			Family and Community ETA
	Mean	Number	Standard Deviation	Mean	Number	Standard Deviation
No coercion	34.3	32	11.7	23.6	55	15.3
Threatened (no weapon)	36.9	29	13.4	26.4	53	14.8
Threatened (weapon)	39.0	11	12.3	28.0	25	14.8
Minor assault	35.2	26	14.4	30.3	52	20.9
Brutal assault	40.5	6	7.0	30.7	13	10.7

The pattern for UTAs is similar, but not as consistent. Those who used no coercion while committing the offence received their first UTA after 34% of their sentence was served while those who used extreme force had to serve 40.5% of their sentence. Note that eligibility for a UTA is normally after one sixth of the sentence, and parole eligibility is after one third of the sentence.

Table 50 demonstrates how the presence of alcohol and drugs during the commission of a sexual offence affects the amount of time that offenders serve before receiving a UTA or a family and community ETA. For those who had no alcohol or drug involvement with their offence they served 23% of the sentence before a family and community ETA, but where alcohol and drugs were elements in the offence, offenders served 30% or more of their sentence before receiving an ETA. The differential is similar for UTAs with the presence of alcohol during the current or a past offence resulting in larger percentage of the sentence being served before a UTA. It is also interesting to note that alcohol and drug involvement in previous offence had a greater influence on percentage of sentence served. It may be that such involvement is indicative of an addiction problem that contributes to the return to a federal institution.

Table 50: Percentage of Sentence Served Before First UTA and First Family and Community ETA by Alcohol and Drug Involvement in Offence

Alcohol & Drugs		UTA			Family and Community ETA
		Mean	Number	Standard Deviation	Mean	Number	Standard Deviation
Current Offence	Yes	38.9	65	12.0	29.6	130	19.5
	No	34.1	50	17.6	23.4	80	14.8
Previous Offence		35.2	26	14.4	30.3	52	20.9
Yes		42.7	22	24.2	33.3	62	22.4
No		35.1	77	12.7	23.4	108	14.5

Seeking treatment for sexual problems reduces the amount of time served before first UTA and first family and community ETA as shown in Table 51. For family and community ETAs simply starting treatment is associated with reduced time to ETA, but stopping the treatment after is has begun it associated with much higher percentage of time served.

Table 51: Percentage of Sentence Served Before First UTA and First Family and Community ETA by Treatment within the Correctional Service

Treatment	UTA			Family and Community ETA
	Mean	Number	Standard Deviation	Mean	Number	Standard Deviation
Yes, completed	33.6	33	12.4	22.5	48	16.5
Yes, in progress	36.5	19	11.6	25.7	38	15.5
Yes, but Stopped	33.2	5	13.2	34.4	8	19.7
No	38.3	58	18.1	28.2	125	18.6

Summary

The data for sexual offenders seems to indicate that those who have committed more serious offences must wait longer for the opportunity to receive TAs that are discretionary. Given that the percentage of time served is influenced by the length of the original sentence, and that more serious assault are likely to result in longer sentences, it is probable that those offenders committing the most serious sexual assaults are spending considerably more time in institutions before receiving the opportunity for a temporary absence. As noted earlier, it must also be remembered that the offenders who have committed the most serious offences do not even appear in these data, so the time served presented here is a conservative estimate.

Figure 9: Types of Major Offences for those on Temporary Absences