FORUM on Corrections Research

Federal offender family violence: Estimates from a national file review study

Federal offenders appear to possess many of the characteristics routinely used to describe men who are violent toward their families.

A recent literature review⁽²⁾ identified certain demographic (such as male, younger than 30 and unemployed), psychological (such as angry, aggressive and immature), attitudinal (such as externalizes blame and has rigid definitions of male and female roles) and behavioural (such as substance abuse and threatens suicide or homicide) factors as characteristic of abusers. However, the criminal behaviour literature⁽³⁾ identifies many of these same characteristics as well-established correlates of general criminal behaviour.

Further, federal offenders' histories of generally violent behaviour suggest that this group is likely to be at high risk of perpetrating violence against family members.

This article summarizes a recent Correctional Service of Canada file review study that provides strong empirical support for the argument that offenders tend to be violent toward family members. The study identified a large proportion of randomly sampled federal inmates whose official record contained evidence of their committing some form of family violence. Methodology This file review study was conducted as part of the Correctional Service of Canada's family violence initiative and was the first national study aimed at identifying the incidence of family violence among federal offenders.

File reviewers recorded all evidence (including criminal charges and unofficial reports) of violence against family members contained in the offender files studied. Evidence as to childhood victimization of the offenders was also collected.

The study sample was composed of men admitted to Service facilities between June and November of 1992. A total of 935 files were randomly sampled and examined by file reviewers. This represents roughly a third of all admissions during this period (there were 2,806 total admissions). The sample was stratified by region to increase the number of files randomly selected from smaller regions. Demographic and offence characteristics The basic characteristics of the study sample were very similar to those of all offenders admitted to federal custody during this period. The average age (at the time of admission) of the sample was 31.8, most of the offenders had not completed high school (71.1%), 40.5% of the offenders had previously received a federal sentence, 41.3% were serving a sentence of less-than three years and 3.9% were serving a life sentence.

Further, the major admitting offence (the offence for which the offender is currently serving the longest sentence) of 39.4% of the offenders was a violent non-sexual offence (such as murder, manslaughter or assault). Approximately 20.7% of the offenders had been convicted of a property offence, 15.4% of a sex offence, 11.9% of a drug offence and 12.7% of other offences.

At the time of admission, 44.6% of the offenders were married or living in a common-law relationship. Just 16.3% said they had never been married (legally or common law) and 18% had three or more previous marital unions. Approximately 59.1% of the sample had children or stepchildren. Family violence File information indicates that about one third (33.7%) of the sample had been violent, at some time, toward one or more members of their family. This estimate includes sexual, physical and psychological abuse. About 1 in 10 (10.6%) files contained indications that the offender had sexually assaulted a family member, 3 in 10 (26.9%) pointed to physical assaults, and about 5 of every 100 files mentioned psychological abuse.


Figure 1

Further, more than half of the files (56.3%) containing evidence of family violence indicated that more than one family member had been victimized and just under half of the "violent" files (47.8%) suggested that the victim required medical assistance.

Further analysis, however, reveals an alarming statistic. Official charges were laid against 80.2% of the offenders whose files contained evidence of family violence. This extremely high charge rate suggests that our estimate of family violence incidence (approximately 33%) falls far short of the actual number of offenders who have been violent toward a family member(s).

Why? Research indicates that only a small proportion of family violence leads to criminal charges. For example, a recent Statistics Canada survey revealed that just 26% of wife assaults were reported to police and only 28% of reported incidents led to official charges.⁽⁴⁾ Therefore, it is probable that this file review study greatly underestimates the incidence of family violence among the federal offender population.

Overall, there was a higher proportion of family violence perpetrators identified in the Service's Prairie and Atlantic regions and these differences were statistically significant (see Figure 1). There was also a higher proportion of family violence perpetrators among Aboriginal offenders. Indications of family violence were found in 51.9% of Aboriginal offender files, compared with just 31.2% of the files of non-Aboriginal offenders. Spousal violence The file review data clearly indicate that women were the most frequent victims of the family violence perpetrated by federal offenders. Approximately 91.6% of these violent incidents involved female victims, while just 24.2% involved male victims (some incidents had more than one victim).

Overall, there was evidence of family violence against a female partner in about one quarter of the files. However, almost one third of offenders who had (at some time) been married or involved in a common-law relationship committed some type of abuse against a female partner (see Table 1).

Table 1

Abuse Directed Against Female Partners
Abuse type	All offenders (935)	Offenders involved (at same time) in a marital union (721)	Offenders involved in dating only (153)
Sexual abuse	2.6%	3.1%	1.2%
Physical abuse	22.1%	26.7%	7.4%
Sexual and/or physical abuse	22.9%	27.6%	7.4%
Psychological abuse	4.2%	5.3%	0.5%
Any family violence	24.1%	29.0%	7.9%
Note: Not all files contained enough information to determine whether the offender had ever been involved in a relationship with a female partner.

The most frequent type of abuse reported against a female partner was, by far, physical abuse. References to psychological abuse were much less frequent - only about 4% of the files contained information suggesting that the offender had been psychologically abusive toward a female partner. The file review method is, therefore, probably an unreliable source of information as to the extent of psychological abuse. This type of abuse occurs relatively frequently in the general population.⁽⁵⁾

The large majority (72.7%) of female partner abuse situations resulted in official charges. In fact, one of every five offenders (21%) who had had a female partner (at some time) had been charged for spousal abuse. Further, resulting injury requiring medical attention was reported in about half of the cases of spousal abuse (46.4%), and about half of the offenders identified as spousal abusers (45.5%) had abused more than one partner. Violence against children Child abuse was relatively less frequent than violence against female partners. The file data indicated that about 13.3% of the men who had children or stepchildren had been abusive toward them. Sexual abuse accounted for 83% of all identified instances of child abuse. Similar to spousal abuse, the majority of these abuse cases resulted in official charges (87.3%). Childhood victimization About half of the offender files (50.2%) suggested that the offender had been abused by one or more family members as a child. This figure combines instances of physical, sexual or psychological abuse, neglect and witnessing the abuse of other family members (see Table 2).

Table 2

Evidence of Childhood Victimization of Offenders (935 Offenders)
Abuse Type	Offenders abused
Sexual abuse	12.0%
Physical abuse	34.6%
Sexual and/or physical abuse	39.6%
Psychological abuse	8.7%
Neglect	6.8%
Witnessed abused of another family member	23.8%

About 1 in 10 (12%) offenders had been. victimized sexually by a member of their family, while more than a third of the files (34.6%) referred to some form of childhood physical abuse. Witnessing abuse was also very common.

The files also provide considerable detail about the nature of the abuse. In most physical or sexual abuse cases (74.7%), the victimization began before the age of five and some form of the abuse continued through ages 12 to 16.

Figure 2

Fathers were most often the abusers (75.7%), although perpetration by mothers (42.4%) and other family members (20.2%) was not uncommon. Further, 5.6% of the cases involved abuse by an institutional authority figure outside the family.

Of the offenders who witnessed abuse, the most typical file entries related to the abuse of other family members by their father (84.1% of those who had witnessed abuse). The victim of the abuse was most often a mother or other adult female partner (67.5%). However, 63.5% of these offenders witnessed the abuse of other children in their family. Overall, physical abuse was the type of abuse most frequently witnessed (88.9%).

This victimization data seems to confirm the popular belief that offenders tend to have troubled childhoods characterized by repeated witnessing and experiencing of abuse.

Consistent with current victimization research,⁽⁶⁾ childhood victimization was also correlated with the later perpetration of family violence. Offender files containing evidence of childhood family violence (witnessed or experienced) contained evidence of a higher rate of adult perpetration of family violence than files containing no evidence of childhood abuse.

More specifically, offenders victimized as children were almost twice as likely (1.8 times) as those not victimized to become abusers themselves. Further, this relationship between victimization and perpetration was consistent for both violence against female partners and against children (see Figure 2). A problem that must be addressed The results of the file review study were remarkably consistent with a similar British Columbia file review study.⁽⁷⁾ Offender files suggest that at least one third of the federal admission population has been involved in some form of violence against family members.

Further, there is ample evidence that this figure underestimates the actual family violence rate for this population.

It is clear, therefore, that federal offenders are at high risk of victimizing members of their families. Such risk must be taken into account in the supervision of offenders on conditional release in the community and in the management of institutional visits with family members. Most important, however, this study provides compelling support for a continued focus on family violence programming for federal offenders.

---

⁽¹⁾Correctional Research and Development, Correctional Service of Canada, Second Floor, 340 Laurier Avenue West, Ottawa, Ontario K1A 0P9. A more detailed report of the results of this study is available.
⁽²⁾B. Appleford, Family Violence Review: Prevention and Treatment of Abusive Behaviour (Ottawa: Correctional Service of Canada, 1989).
⁽³⁾D. A. Andrews, "Recidivism Is Predictable and Can Be Influenced: Using Risk Assessments to Reduce Recidivism, Forum on Corrections Research, 1, 2 (1989): 11-18.
⁽⁴⁾"The violence Against Women Survey," The Daily (Ottawa: Statistics Canada, November 18, 1993), Catalogue No. 11-001E.
⁽⁵⁾L. MacLeod, Battered But Not Beaten...Preventing Wife Battering in Canada (Ottawa: Canadian Advisory Council on the Status of Women, 1987).
⁽⁶⁾C. S. Widom, "The Cycle of violence," Science, 244 (1989): 160-166. See also M. T. Tolman and L. W. Bennett, "A Review of Quantitative Research on Men Who Batter," Journal of Interpersonal Violence, 5 (1990): 87-118.
⁽⁷⁾D. G. Dutton and S.D. Hart, "Risk Markers for Family violence in a Federally Incarcerated Population," Forum on Corrections Research, 5, 2 (1993): 26-28. See also D. G. Dutton and S.D. Hart, "Risk Markers for Family violence in a Federally Incarcerated Population," International Journal of Law and Psychiatry, 15 (1992): 101-112.