Section 1: Overview of family violence
By Maire Sinha
- Defining family violence in Canada
- Measuring family violence in Canada
- Overview of family violence in Canada
- Detailed data tables
Defining family violence in Canada
Defining family violence is integral to accurately profiling the issue. While there is no universally accepted definition of family violence, two elements must be considered in any definition: the forms of violence to be included and the types of family relationships. Within the Family Violence Initiative, family violence has been conceptualized as "a range of abusive behaviours that occur within relationships based on kinship, intimacy, dependency or trust" (Family Violence Initiative Performance Report, 2008). This definition is far-reaching and can encompass physical, sexual, verbal, emotional, and financial victimization, or neglect. Within this publication, analysis of violence within the family is primarily based on statistical data that are consistent with Criminal Code definitions, unless otherwise stated.
Determining the family relationships for inclusion in a definition of family violence is also a question of scope. Past publications have addressed the different dimensions of family violence for three primary victim groups: spouses (legally married, separated, divorced, and common law partners), children and youth under 18 years of age, and seniors aged 65 years and older. Essentially, family relationships have been defined by the accused person's relationship to the victim through blood, marriage, co-habitation (in the case of common-law partners), foster care, or adoption.
Recently, there has been some consideration within the research community both nationally and internationally and by federal, provincial and territorial governments towards including all types of intimate partner relationships, including dating partners, in a definition of family violence (see, for example, Justice Canada 2009, PEI Premier's Action Committee on Family Violence Prevention Administration Committee 2010). Violence against dating partners has been argued to fall within the definition of family violence, due to its many similarities with spousal violence. For example, previous research has found that the characteristics of police-reported dating violence generally mirror those of spousal violence (Hotton Mahony 2010). In addition, from a prevention perspective, understanding the characteristics of violence in dating relationships is important to the development of effective programming. Research has found that individuals' experiences in early dating relationships can have an impact on future patterns of violence for both victims and abusers (Wolfe 2006; Close 2005).
As a result, for the first time, this publication will explore the impact of including dating violence within a definition of family violence. This exploration will involve an examination of the overall prevalence of family violence with and without the inclusion of dating violence. For most sections, the analysis of risk factors and offence characteristics will be based on a definition of family violence that excludes victims of dating violence. The one exception is the section on intimate partner violence. For this section, differences between spousal and dating partner violence will be delineated, including both the analysis of the prevalence of each form of intimate partner violence and the examination of detailed victim, accused and incident characteristics.
Measuring family violence in Canada
As with previous editions of Family Violence in Canada, two main sources of information are used to measure and analyze family violence: police-reported information from the Uniform Crime Reporting (UCR) Survey and the Homicide Survey, and self-reported victimization data from the General Social Survey on Victimization. These data sources yield complementary yet different types of information on violence within families.
As mandatory annual censuses of all police services in Canada, the UCR Survey and the Homicide Survey provide trend data, as well as national, provincial/territorial, and census metropolitan area (CMA) level information on Criminal Code offences that are reported to and substantiated by police. Both surveys collect data on the characteristics of victims, accused and incidents. Consequently, risk factors such as age, sex and marital status can be examined for all victims of family violence as well as indicators of offence severity.
While all Criminal Code offences are captured by the UCR Survey, the survey does not record victim information for non-violent crime, such as the relationship of the accused to the victim. As a result, it is not possible to identify non-violent incidents that targeted family members. For instance, it is not possible to examine financial abuse against family members such as theft, forgery and fraud, that are criminal in nature but do not have a violent component.
The UCR Survey is limited to only those incidents that come to the attention of police, which may be a greater issue for incidents of family violence, as these violent acts have historically had lower levels of reporting to police (Bala 2008). The General Social Survey (GSS) on Victimization complements police reported data in that it helps to explore the large amount of crime in Canada that goes unreported to police.Note 1 It is also able to provide information on non-violent forms of abuse (e.g., emotional and financial abuse), consequences of victimization, levels of reporting to police, children's exposure to spousal violence, and social service utilization. Since the GSS is a voluntary household survey, it relies on the willingness of Canadians to participate. While some Canadians refuse to participate, other groups may be unable to participate because of cognitive impairments, compromised mental health, restricted access to a telephone (for example, individuals without a landline telephone and victims of family violence whose activities are severely restricted) or inability to communicate in English or French.Note 2
Some populations that are potentially vulnerable to family violence, including those who are dependent on others for their basic needs, are not included within the GSS on Victimization. In particular, the GSS solely asks persons aged 15 years and older living in private dwellings about their experiences of victimization. As a result, children under 15 years of age and individuals living in institutions, such as long-term care facilities, are not eligible to participate in the survey. Only official sources of information, such as reports from police, child welfare agenciesNote 3 and medical and social service surveillance systems are able to yield information on direct violence against children under the age of 15 and persons living in institutions.Note 4 That said, even with these reports from authorities, the prevalence of abuse against children and institutionalized seniors is difficult to measure because it often relies on other individuals to detect and report the abuse.Note 5
In addition to these police-reported and victimization surveys, other data sources are presented in the current report. Contextual information is also included throughout the report to better explain the nature of victimization, as it is understood that family violence does not exist in isolation, but occurs alongside a range of individual and social factors (Diem and Pizarro 2010; Thomas and Bennett 2009).
Overview of family violence in Canada
Until 30 years ago, violence committed against family members was largely seen as a private matter and remained hidden (Montalvo-Liendo 2009; Richie 2006; Dawson 2001). Since then, there have been substantial changes in the criminal justice system response and social intervention to family violence. This has been accompanied by a shift in the public's understanding and awareness of the issue. Violent acts committed against family members are now recognized as serious violent crimes (Bala 2008; Schneider 2007).
While the Criminal Code does not contain separate violent offences based on the relationship of the victim and offender, perpetrators of violent acts against family members can be charged with the appropriate criminal offence, such as homicide, assault, sexual assault, or criminal harassment. Further, the Criminal Code considers the abuse of a spouse or child or any position of trust or authority to be an aggravating factor at sentencing. Other Criminal Code provisions can assist victims of family violence, including the availability of protection orders and the enhancement of testimonial aids for vulnerable victims (Justice Canada n.d.). Beyond the Criminal Code, some provincesNote 6 and all three territories have also adopted civil legislation specific to family violence. These pieces of legislation serve to provide additional supports and protection to victims of family violence.
Procedurally, police, courts, and corrections have also recognized the unique needs of victims and offenders of family violence. Criminal justice initiatives have included changes to policing protocols (such as pro-charging policies), domestic violence investigation units within police services including programs for dating partners, specialized training programs for police and Crown counsels, dedicated domestic violence courts, and family violence treatment interventions within correctional systems (Correctional Services of Canada n.d.; Public Health Agency of Canada 2008).
As previously mentioned, understanding the unique nature of family violence is the theme of this year's publication. The analysis will examine the following research questions for family violence as a whole and for each victim group (intimate partners, children and youth, and seniors):
- How are the socio-demographic risk factors for family violence, such as age and sex, different from other forms of violence?
- Does the severity of violence perpetrated by family members differ from violence committed by non-family members?
- Are perpetrators accused of family violence more likely than other persons accused of violence to be charged by police?
- Are regional variations in the prevalence of family violence similar to those in non-family violence?
One in four victims of violent crime was victimized by a spouse or other family member
Based on the traditional definition of family violenceNote 7 which excludes dating violence,Note 8 there were almost 99,000 victims of family violence in 2010, accounting for one-quarter (25%) of all police-reported victims of violent crime (Table 1.1). Almost an equal proportion of these family violence victims were spouses (49%) or other types of family members, such as children, parents, siblings or extended family members (51%).
To understand the relative prevalence of family violence, rates of family violence can be compared to other forms of violence. In 2010, there were 294 victims of family violence for every 100,000 Canadians (Chart 1.1). This police-reported rate of family violence was similar to the rate of stranger violence (307 per 100,000), but was nearly half the rate involving acquaintances or friends (574 per 100,000). Included in the latter category are dating violence victims, including those in a current and former dating relationship with the accused. In all, dating violence victims accounted for 28% of victims of acquaintance and friend-related violence.
Victims of police-reported violent crime, by type of accused-victim relationship, 2010
1. The traditional definition of family violence excludes dating violence. According to this definition, family relationships are defined by the accused person's relationship to the victim through blood, marriage, co-habitation (in the case of common-law partners), foster care, or adoption.
2. The expanded definition of family violence includes dating violence. According to this definition, family relationships are defined by the accused relationship to the victim as an intimate partner (including spouses and dating partners) or through blood, foster care, or adoption.
Note: Rates are calculated on the basis of 100,000 population. Populations based upon July 1st estimates from Statistics Canada, Demography Division.
Source: Statistics Canada, Canadian Centre for Justice Statistics, Incident-based Uniform Crime Reporting Survey.
If the definition of family violence is expanded to include dating relationships, the representation of family violence victims as a proportion of all victims of violence would increase from 25% to 39% of all victims. As well, family violence would then have the highest rate of violent crime among all major relationship categories. At 457 victims per 100,000 population, the rate would be 11% higher than the rate for acquaintances or friends (411) and 49% higher than the rate for strangers (307) (Chart 1.1).
However, regardless of the definition of family violence used, when violence culminated in the killing of the victim, friends or acquaintances were most often responsible (Table 1.2). In particular, between 2000 and 2010, family members excluding dating partners accounted for 35% of all solved homicides, lower than the proportion (49%) committed by friends and acquaintances. The inclusion of dating homicides in the definition of family homicides does not impact the overall patterns, as family homicides would still represent a smaller proportion of solved homicides than the share of acquaintance or friend-related homicides (39% versus 45%). The one exception to this finding was homicides involving children and youth victims under the age of 18. The vast majority of these homicides over the past 10 years were perpetrated by family members (59% excluding dating partner homicides and 62% using a definition including dating partner homicides).
Rates of family homicide continues to decrease
Tracking changes in rates of family violence can help inform the development and evaluation of programs and policies designed to reduce this form of violence. As previously mentioned, issues of underreporting to police may be particularly evident among incidents family violence (Bala 2008). For this reason, trends in homicides are often used as a barometer of family violence (Nivette 2011, Gannon et al. 2005). Mirroring trends in homicide overall, rates of family homicide excluding dating homicide have been generally decreasing over the past thirty years, with a rate in 2010 that was 41% lower than in 1980 (Chart 1.2).
Victims of family and non-family homicide, Canada, 1980 to 2010
Note: Family homicide refers to homicide committed by spouses, parents, children, siblings, and extended family. Non-family homicide refers to homicide committed by friends, casual acquaintances, dating partners, business associates, criminal associates, authority figures, and strangers. Excludes homicides where the age and/or sex of the victim was unknown. Rates are calculated on the basis of 1,000,000 population. Populations based upon July 1st estimates from Statistics Canada, Demography Division.
Source: Statistics Canada, Canadian Centre for Justice Statistics, Homicide Survey.
Family violence victims primarily girls and women
Unless otherwise noted, the following sections examine the characteristics of family violence according to the definition of family violence exclusive of dating violence. An examination of risk factors for family violence (excluding dating violence) reveals that unlike other forms of violence, victims of police-reported family violence are predominantly female. In 2010, 7 in 10 (70%) victims of police-reported family violence were girls or women.
Looking at rates, the risk of becoming a victim of police-reported family violence was more than twice as high for girls and women as it was for boys and men (407 per 100,000 versus 180 per 100,000). This heightened risk of family violence among girls and women was true regardless of age, but was most pronounced among those aged 25 to 34 years (Chart 1.3). Females in this age group were over three times more likely than their male counterparts to become a victim of family violence (rate of 709 per 100,000 versus 216 per 100,000 population). The main factor behind females' increased risk of family violence is related to their higher representation as victims of spousal violence. Women aged 15 years and older accounted for 81% of all spousal violence victims.
Victims of police-reported family violence, by sex and age group of the victim, Canada, 2010
Note: Family violence includes violence committed by spouses, parents, children, siblings, and extended family. Spouses include those aged 15 to 89 years, while other family members include those aged 0 to 89 years. Excludes incidents where the age and/or sex of the victim was unknown. Rates are calculated on the basis of 100,000 population. Populations based upon July 1st estimates from Statistics Canada, Demography Division.
Source: Statistics Canada, Canadian Centre for Justice Statistics, Incident-based Uniform Crime Reporting Survey.
While the rate of police-reported non-family violence was generally lower among females than males, this was not consistently the case across all age groups. In particular, women aged 15 to 24 years were more likely than similarly aged men to be victims of non-family violence (2,259 versus 2,176 per 100,000 population) (Chart 1.4). This can be largely explained by young women's increased vulnerability to dating violence. Further, differences between age-based rates for females and males were smaller for non-family violence, compared to differences in rates of family violence.
Victims of police-reported non-family violence, by sex and age group of the victim, Canada, 2010
Note: Non-family violence refers to violence committed by dating partners, friends, casual acquaintances, business associates, criminal associates, authority figures, and strangers. Includes victims aged 0 to 89 years. Excludes incidents where the age and/or sex of the victim was unknown. Rates are calculated on the basis of 100,000 population. Populations based upon July 1st estimates from Statistics Canada, Demography Division.
Source: Statistics Canada, Canadian Centre for Justice Statistics, Incident-based Uniform Crime Reporting Survey.
Physical assaults more common among incidents of family violence than non-family violence
Police-reported data suggest that although there are some similarities in the types of violent offences committed against family and non-family victims, there are two notable exceptions. First, family violence is more likely to involve physical assaults. In 2010, 70% of all victims of family violence were victims of physical assault (Table 1.3). The same was true for 58% of non-family violence victims.
Second, family violence less frequently involved robbery offences. This violent crime accounted for less than 1% of family violence incidents, compared to 9% of non-family violence incidents.
Victims of family violence more likely than victims of non-family violence to sustain injury
As a reflection of the higher volume of offences that more often result in injury to victims, notably physical assault, a greater proportion of victims of family violence (46%) sustained injury compared to other victims of violent crime (41%).
However, in examining specific types of offences, there was little difference in the incidences of injury between victims of family and non-family violence. As would be expected, the prevalence of injury tends to be higher for those offences that characteristically involve the use of physical force or weapons, regardless of the relationship between the accused to the victim. For example, police-reported common assaults resulted in injuries to 58% of family violence victims, about the same proportion as non-family violence victims (59%). Also, an equal proportion of sexual assault victims of family violence and non-family violence suffered injuries (21% each). Offences that most often involved threatening behaviour rather than the use of physical force or weapons were least likely to result in injuries to the victims. For example, an equal proportion of family violence victims and victims of non-family violence suffered injuries as a result of criminal harassment (1% each).
Charges laid by police more often when incidents involve family members
As previously mentioned, police response to the issue of family violence has evolved over the years, particularly with the introduction of pro-charging policies in the 1980s (Garner and Maxwell 2009). In general, those accused of victimizing their spouses or other family members were more likely than other perpetrators to have charges laid or recommended by police, with the exception of 'other assaults', which includes such offences as unlawfully causing bodily harm and discharging firearm with intent. In 2010, 56% of accused family members were charged (or had charges recommended), compared to 43% of other accused.Note 9 Also, a higher proportion of accused family members than other perpetrators were cleared by other means (29% versus 25%). Incidents may be cleared by other means for a variety of reasons, including the complainant declined to lay charges, use of departmental discretion, and reasons beyond the control of the department.
There were notable gender differences in charging patterns among family violence incidents. For example, when the victim was female, police were more likely to lay a charge against the accused family member (60% versus 46% incidents against male victims). While this was also true for non-family violence incidents, the gendered pattern in charging was partly driven by dating violence incidents. In dating violence incidents, the accused was more likely to be charged if the victim was female (69% versus 57% of incidents involving male victims).
Ontario records the lowest rate of family violence
In all provinces and territories, rates of police-reported family violence were lower than the combined violent crime rate against friends, dating partners, acquaintances, and strangers without exception. In 2010, the rate of family violence in Ontario was the lowest in the country and except for Prince Edward Island and Nova Scotia, rates in all other provinces and territories were above the Canadian average (Table 1.4). By comparison, rates of non-family violence were lowest in Quebec, followed by Ontario, Prince Edward Island, Newfoundland and Labrador, and Alberta. For both family and non-family violence, rates were highest in the territories, Saskatchewan and Manitoba.
An examination of family homicide reveals regional variations similar to overall rates of family violence. Rates of family homicide were lowest in Prince Edward Island and Nova Scotia (Table 1.5).
Contrasting the similarity in regional patterns between family homicide and family violence overall, regional patterns for non-family homicide differ somewhat from overall rates of non-family violence. Prince Edward Island had the lowest rate of non-family homicides, followed by Newfoundland and Labrador and New Brunswick. As with family violence overall, the territories, along with Manitoba and Saskatchewan had the highest rates of family-related homicide.
Rates of family violence lower among CMAs than non-CMAs
For the first time, rates of family violence can be examined at the census metropolitan area (CMA) level.Note 10 In 2010, CMAs generally had lower levels of family violence than non-CMAs, which include small cities, towns and rural areas. That said, there were significant variations in rates of family violence across the CMAs. For instance, the rate recorded in Saint John was over four times higher than the lowest rate recorded in Ottawa (Table 1.6).
For the most part, the CMAs with the highest rates of family violence also had rates of non-family violence above the national average. The only departure from this pattern was the Quebec CMAs of Gatineau, Montréal and Saguenay. These cities all recorded rates of family violence above the Canadian average but had rates of non-family violence below average.
For the past thirteen years, Statistics Canada has released an annual report on family violence in Canada. This year marks the first time the publication has undertaken an exploration into the statistical impact of expanding a definition of family violence to include dating partners. This exploration revealed differences in the overall prevalence of family violence depending on whether dating violence was excluded or included from a definition of family violence. When dating violence was excluded, the rate of family violence was similar to the rate for stranger violence but lower than the rate for friends and acquaintances. In contrast, the rate of family violence including dating violence was higher than all other major relationship categories.
This year's focus on a comparative analysis of family violence and non-family violence reveals some important differences between the two types of police-reported violence. In 2010, victims of police-reported family violence (excluding victims of dating violence) were disproportionately female, contrasting the findings for non-family violence. Further, family violence is sometimes more severe than other forms of violence. Despite the similarity in incidences of injury between victims of family and non-family violence for particular types of offences, the higher volume of physical assaults translates into an overall higher frequency of injury among victims of family violence. In general, victims of family violence were also more likely have physical force used against them. Those accused of family violence were also more often than other perpetrators to have charges laid or recommended by police.
Detailed data tables
Bala, N. 2008. "An historical perspective on family violence and child abuse: Comment on Moloney et al., Allegations of Family Violence, 12 June 2007." Journal of Family Studies.Vol. 14, no. 2/3. p. 271-278.
Close, S. M. 2005. "Dating violence prevention in middle school and high school youth." Journal of Child and Adolescent Psychiatric Nursing. Vol.18. p. 2-9.
Correctional Service of Canada. n.d. Correctional Programs: National Family Violence Prevention Programs. (accessed February 21, 2012).
Dawson, M. 2001. "Examination of declining intimate partner homicide rates." A Literature Review. Ottawa. Department of Justice Canada.
Diem, C. and J. Pizarro. 2010. "Social structure and family homicides." Journal of Family Violence.Vol. 25, no. 5. p. 521-532.
Gannon, M., K. Mihorean, K. Beattie, A. Taylor-Butts and R. Kong, 2005. Criminal Justice Indicators. Statistics Canada Catalogue no. 85-227-X.
Garner, J.H. and C.D. Maxwell. 2009. "Prosecution and conviction rates for intimate partner violence." Criminal Justice Review. Vol. 34, no. 1. p. 44-79.
Hotton Mahony, T. 2010. "Police-reported dating violence in Canada, 2008." Juristat. Statistics Canada Catalogue no. 85-002-X.
Justice Canada. 2009. Family Violence: Department of Justice Canada Overview Paper. (accessed December 12, 2011).
Justice Canada. n.d. Family Violence Initiative: Laws. (accessed December 12, 2011).
Montalvo-Liendo, N. 2009. "Cross-cultural factors in disclosure of intimate partner violence: An integrated review." Journal of Advanced Nursing.Vol. 65, no. 1. p. 20-34.
Nivette, A. 2011. "Cross-national predictors of crime: A meta-analysis." Homicide Studies. Sage Publications. Vol. 15, no. 2. p. 103-131.
Prince Edward Island (PEI) Premier's Action Committee on Family Violence Prevention Administration Committee. 2010. Federal, Provincial/Territorial and Other Organizations' Definitions of Family Violence in Use in Canada. Unpublished.
Public Health Agency of Canada. 2008. Canada's Treatment Programs for Men Who Abuse Their Partners. Catalogue no. HP20-7/2008.
Richie B.E. 2006. "Foreword." Domestic Violence at the Margins: Readings on Race, Class, Gender, and Culture. Sokoloff N.J. and Pratt C., (eds.). Rutgers University Press. New Brunswick, N.J. p. 15–18.
Schneider, E. M. 2007. "Domestic violence law reform in the twenty-first century: Looking back and looking forward." Family Law Quarterly.Vol. 42, no. 3.
Thomas, M. D. and L. Bennett. 2009. "The co-occurrence of substance abuse and domestic violence: A comparison of dual-problem men in substance abuse treatment and in a court-ordered batterer program." Journal of Social Work Practice in the Addictions. Vol. 9, no. 3. p. 299-317.
Wolfe, D.A. 2006. "Preventing violence in relationships: Psychological science addressing complex social issues." Canadian Psychology. Vol. 7, no. 1. p. 44-50.
- Results from the General Social Survey (GSS) on Victimization are based on findings for the provinces. Data from the Northwest Territories, Yukon and Nunavut were collected using a different methodology and are therefore excluded.
- In 2009, the response rate for the GSS on Victimization was 61.6%. Types of non-response included respondents who refused to participate, could not be reached, or could not speak English or French. For further information, see the Data sources section.
- The Canadian Incidence Study on Child Neglect and Abuse collects data on reports of neglect and abuse that come to the attention of provincial/territorial child welfare authorities.
- The Uniform Crime Reporting (UCR) Survey and the Homicide Survey are not able to indicate if the violence against the senior took place in a long-term care facility.
- In some cases, child welfare workers and health and social service workers may be in regular contact with children and seniors because of previous concerns of abuse or because of existing health or cognitive conditions in the case of seniors.
- The provinces with civil legislation are Newfoundland and Labrador, Prince Edward Island, Nova Scotia, Manitoba, Saskatchewan, and Alberta.
- Includes those related by blood, marriage, co-habitation, foster care, or adoption. Violent crime includes violations causing death, attempted murder, sexual assaults, assaults, robbery, criminal harassment, uttering threats and other violations involving violence or the threat of violence.
- Dating relationships include current or former boyfriends and girlfriends, as well as 'other intimate relationships'. 'Other intimate relationships' are defined in the Incident-based UCR2 Survey as "a person with whom the victim had a sexual relationship or a mutual sexual attraction but to which none of the other relationship options apply".
- Excludes Montreal police service due to the unavailability of clearance data.
- A census metropolitan area (CMA) consists of one or more neighbouring municipalities situated around a major urban core. A CMA must have a total population of at 100,000 of which 50,000 or more live in the urban core. To be included in the CMA, other adjacent municipalities must have a high degree of integration with the central urban area, as measured by commuting flows derived from census data. A CMA typically comprises more than one police service.
- Date modified: