Family Violence and Homelessness: A Review of the Literature – Literature Review

Literature Review

Scope of the Literature

Research on the relationship between family violence and homelessness is still at an early stage of development. Integration of the two topics has been largely limited to references to an association, with virtually no empirical studies having explicitly investigated the links.

Studies of homelessness identify family violence as a highly prevalent feature in the histories of homeless people, and some suggest it is a precipitant cause of homelessness. Research on family violence identifies shelters as meeting a critical need of those who are escaping abusive home environments. It points to the lack of affordable housing as a factor that keeps victims from leaving or contributes to their decision to return to abusive spouses.

Research has also documented the changing demographic profile in the homeless population since the 1960s. While single men remain the largest group of visibly homeless people, there are increasingly more women, families, youth and children – labelled as the "new" homeless, in comparison to the "old" homeless who were generally described as socially disaffiliated men, many of them suffering from alcoholism (Rossi 1989).

As the problem of visible homelessness has increased, more researchers have sought to investigate the social forces behind it. Macro-level analysis of structural and policy changes has established that a lack of sufficient affordable housing is the predominant underlying cause of homelessness. Micro-level analysis, on the other hand, has focussed on investigating the characteristics and experiences of homeless people. Explanations of why "new" groups have joined the ranks of the homeless have evolved to acknowledge high rates of family violence and have begun to probe the contributing role of family breakdown, trauma and adverse childhood experiences.

Most studies tend to be retrospective or describe the current situation and experiences of people who are already homeless. There are very few longitudinal studies that have tracked people to examine change over time or the effects of various factors. Those studies that have tracked people have tended to focus on assessing whether "graduates" of targeted programs maintain a stable housing status after program completion, usually within a period of no longer than two years.

For example, an evaluation of about 500 transitional housing programs in the United States (about one-tenth of which were designed for abused women) found that most clients showed post-program improvements in housing stability, though the success rate was lowest for abused women (Novac et al. 2004).

While studies of criminal victimization of homeless people have measured the extent of harmful or violent incidents directed at them, they generally do not provide enough information about the offender to distinguish whether the relationship was familial (Wenzel et al. 2001). For example, in a Toronto health survey of more than 450 homeless individuals (Ambrosio et al. 1992), women were asked about sexual violence in the previous year. Almost half (46%) had been physically assaulted and one-fifth (21%) had been raped. Nearly half (43%) of the women and 14% of the men had experienced sexual harassment or assault. The researchers did not attempt to distinguish violence and abuse committed by intimates from that committed by strangers. Indeed, the lack of privacy and spatial control that is often an inherent part of homelessness makes it difficult to maintain boundaries between people, whether they are intimates or non-intimates.

Finally, there is very little literature that addresses the service needs of homeless victims of family violence or the implications of those needs for service sectors.

Associations Between Family Violence and Homelessness

Many studies in Canada and elsewhere have found histories of family violence to be quite common among homeless persons – especially youth and women – and more common than among non-homeless populations (Neal 2004; Novac et al. 2002b; Farrell et al. 2000; Gravel 2000; Gaetz et al. 1999; Poirer et al. 1999; Novac et al. 1999; Hagan and McCarthy 1997; Régie Régionale de la Santé et des Services sociaux de Montréal-Centre 1998: Mental Health Policy Research Group 1998: Peters and Murphy 1994: Breton and Bunston 1992; Radford et al. 1989: Janus et al. 1987; Kufeldt and Nimmo 1987). Based on a review of studies on homeless families, Rosenheck et al. (1999, 11) reached the following conclusion:

"Interpersonal violence may well be the subtext of family homelessness. Abuse and assault seem to be salient features of homeless mothers' childhood and adult experiences. Women suffer its devastating medical and emotional consequences for the rest of their lives."

In reporting their extensive review of U.S. research, Shinn and Baumohl (1999) concluded that studies have consistently found, in the histories of both individuals and families who are homeless, high rates of physical and sexual abuse in childhood, frequent foster care and other out-of-home placements, and a variety of other family disruptions. With regard to family homelessness or studies of homeless mothers, most studies have found higher rates of family violence among homeless families than among other poor families.

Burt et al. (1999) conducted a large U.S. study with a probability sample of about 4,200 clients of services to the homeless. They determined that childhood experiences of family violence were more common among homeless people than among those who had "exited" homelessness (29% and 23% respectively) and much more common among both of these groups compared to those who had never been homeless but did use food bank and drop-in programs (8%).

Family violence is also a common antecedent for institutional experiences that are highly prevalent among homeless persons. Among homeless youth and adults, there are high rates of involvement with the child welfare system or out-of-home placement during childhood. Among homeless single adults, especially women, there are high rates of psychiatric hospitalization. Finally, among homeless single men, there are high rates of incarceration.

Children who have been abused typically display behavioural problems in their youth and adulthood that are also associated with homelessness (Gilmartin 1994). These include developmental delays (Yawney 1996); poor school adjustment and disruptive classroom behaviour (de Paul and Arruabarrena 1995); school-age pregnancy (Smith 1996), truancy and running away (Manion and Wilson 1995; Kurtz et al. 1993); delinquency and prostitution (Manion and Wilson 1995); early use and adult use of illicit drugs and alcohol and addictions (Malinosky-Rummell and Hansen 1993; Chandy et al. 1996; Downs et al. 1996); and suicide and suicide attempts (Fergusson and Lynskey 1997). Among the long-term effects of childhood abuse, Kendall-Tackett (2002) identified impaired relationship skills, social isolation and low levels of social support, re-victimization, self-medication (e.g., substance abuse, eating disorders, and smoking), as well as foster care and other forms of family separation, all of which are also associated with the risk of homelessness.

Adverse Childhood Experiences

Questions about childhood experiences of maltreatment have been included in many studies of homeless populations, but they are not generally detailed enough to specify whether the perpetrator was a family member, parental figure, or intimate partner.

Among homeless adults

As mentioned above, many studies have found high rates of childhood abuse suffered by homeless people. For example, Holley and Arboleda-Florez (1997) found that, among 250 shelter users in Calgary, 37% had been physically abused during childhood.

There are clear gender differences in rates of childhood abuse among the homeless. In a sample of 230 homeless individuals in an Ottawa study, 42% of men and 76% of women had been physically abused during their childhood (Farrell et al. 2000). Similarly, in a Toronto study with a representative sampleFootnote 4 of 300 single adult shelter users, 51% of the women and 38% of the men had been physically abused during childhood (compared with 21% of women and 31% of men in the general population). About half (49%) of the women and one-sixth (16%) of the men participating in the Toronto survey experienced sexual abuse during childhood (compared with 13% of women and 4% of men in the general population) (Mental Health Policy Research Group 1998).

So, among homeless adults, women have experienced higher rates of abuse, and especially sexual abuse, than men. This pattern is evident among homeless youth as well (see below).

Herman et al. (1997) investigated whether adverse childhood experiences were risk factors for adult homelessness. Based on data from a probability sample of homeless people and a comparison group of never homeless people, they determined that lack of care from a parent during childhood "sharply increased the likelihood of subsequent homelessness, as did physical abuse" (Ibid., 249). Sexual abuse was not significantly associated with homelessness, but the combination of lack of parental care and either type of abuse dramatically increased the odds of homelessness, compared with those who had neither of these experiences. They concluded that adverse childhood experiences were powerful risk factors for adult homelessness.

Among homeless youth

With respect to homeless youth, some researchers have probed more extensively to investigate the nature, dynamics and implications of childhood abuse. Very high levels of family disruption, conflict and childhood abuse are a consistent finding in studies of street-involved and homeless youth in Canada (Novac et al. 2002b; Poirier et al. 1999; Hagan and McCarthy 1997; Caputo et al. 1997; Peters and Murphy 1994; McCall 1990; Kufeldt and Nimmo 1987; Lamontagne et al. 1987).

A small, qualitative study of homeless youth in downtown Vancouver explored the early stages of the process of becoming street involved.

"Common themes that emerge from the youths' stories about their previous lives include chaos and conflict at home, not fitting in at school, constant movement between households and between communities, and an absence of connections with supportive adults" (McCreary 2002, 33).

Estrangement from family is extremely high among street-involved teenagers. Almost all (96%) of the teens in one study were assessed as having low family connectedness, compared to 13% in a sample of high school students (McCreary 2001).

A study of Toronto street youth revealed that, on average, they had left home at the age of 13 and experienced a high rate of family disruption, e.g., unemployment, single parenthood, reconstituted families, and out-of-home placement (Hagan and McCarthy 1997). Their families were also more likely to have been violent — some type of force was used by a member of their families in 87% of cases and, for 60% of the youth, violence by their parents was sufficient to cause bruising or bleeding.

Perhaps the earliest study on homeless youth in Canada was a 1984 survey of 149 residents, aged 16 to 21, at a Toronto youth shelter (Janus et al. 1987). About three-quarters of the youth had been physically beaten during childhood. Almost three-quarters of the young women had been sexually abused, either within their families or after leaving. The young women who had been sexually abused were more likely to have been involved in delinquent activities (participated in physical violence, been arrested, been placed in jail or juvenile correctional institution).

The results of a subsequent survey at the same shelter showed that 81% of the residents had been physically abused. Onset of sexual and physical abuse began at an earlier age and was more frequent for female residents (Welsh et al. 1995). Female residents had initially left home at an earlier age than male residents and were much more likely to have been sexually abused by their fathers or stepfathers (36% vs. 3%). Moreover, female residents were more likely than male residents to have experienced violent sexual assault (16% vs. 9%) and rape (38% vs. 13%) and to have been sexually abused more frequently and more severely (Ibid.). Among female residents, a caregiver or caretaker had been the usual perpetrator (84%), while male residents cited caretakers and others equally.

Gaetz et al. (1999) surveyed 360 homeless youth in Toronto and found that 40% of the female youth and 19% of the male youth cited sexual abuse as a key factor leading to their life on the streets. Physical abuse was cited as a factor by 59% of the females and 39% of the males.

Among a sample of more than 500 high risk and street involvedFootnote 5 teenagers in several British Columbia communities, 87% of the females and 65% of the males had been abusedFootnote 6 – physically or sexually or both – compared to 26% of females and 14% of males in a large sample of high school students (McCreary 2001). The rate of physical and sexual abuse among non-heterosexual youth was even higher (89% versus 64% for physical abuse and 65% versus 29% for sexual abuse). Only one-third to one-half of the street-involved youth in the six B.C. communities surveyed by the McCreary Centre identified themselves as heterosexual, compared to 85% in a sample of high school students (McCreary 2001).

Negative impacts associated with childhood sexual abuse were severe. A higher proportion of those who had been sexually abused had also been in care, experienced sexual intercourse before age 13, traded sexual favours, developed an addiction problem, and attempted suicide in the previous year. About half (46%) of the girls and 18% of the boys had been forced or coerced to have sex, compared to 3% in a sample of high school students (Ibid.).

A survey of 15 youth sheltersFootnote 7 was conducted as part of an evaluation of Canada Mortgage and Housing Corporation's Shelter Enhancement Program, which provides funds for repairs and enhancements to existing shelters. Youth shelter staff estimated that 79% of their clients were homeless as a result of family violence. Most of the abuse was perpetrated by family members in the home, but in some cases, particularly those of female youth, it included dating violence. Staff attributed their clients' substance abuse and addiction problems to long-term abusive relationships (SPR Associates 2002b).

Family violence is clearly the predominant reason for homelessness among children and youth. Poverty is not as strongly associated with homelessness among youth as it is among adults. Especially high rates of physical and sexual abuse of girls and young women may partly account for the gender ratio variation by age. Among visibly homeless adults, about one-quarter to one-third are female; this ratio increases as age decreases. Among younger teens who are homeless, half are female (Novac et al. 2002b).

Out-of-Home Placement, Family Violence and Homelessness

Canadian studies of homeless youth (e.g., Clarke and Cooper 2000; Leslie and Hare 2000; Gaetz et al. 1999; McCarthy 1995) have found high rates of current or previous involvement with the child welfare system (40-49%). Among street-involved teenagers in British Columbia, 37% had been in some type of government care, including group homes or foster care. Also, 59% had been in the youth justice system, for a combined total of 69% in institutional care (McCreary 2001).

Family abuse and neglect are the predominant reasons for which children are taken into public care, but most youth-in-care do not end up homeless (Novac et al. 2002b). While it is clear that public care will increase the odds that some youth will become homeless, the literature suggests that, among youth-in-care, those most at risk of later homelessness are those who have suffered the more extreme forms of family violence.

A 2002 Montreal study (Robert et al. 2002) compared two groups of more than 100 teenagers, aged 12 to 17, who had been taken into public care during a 12-month period in 1998/99 – teenagers who had previously been homeless for at least two months and those who had never been homeless. The homeless teens had experienced more conflict with their families and were more likely to have been mistreated by their parents and experienced a higher level of family violence (suffering physical and verbal abuse and excessive punishment), had more child welfare placements, and felt a greater sense of alienation from their families. It appears that the degree of family violence, which is probably highly correlated with more child welfare placements and alienation from family, affects the likelihood of subsequent homelessness for youth.

Child protection efforts can be implicated in the housing circumstances of families. In a substantial number of child welfare cases in Toronto, inadequate housing was identified as a factor in the decision to place a child in temporary care. In some cases, a child's return home was delayed because of a housing-related problemFootnote 8 (Cohen-Schlanger et al. 1995). Among families receiving social assistance or a housing subsidy, changes in household size and composition can threaten their benefits, potentially leaving them unable to adequately house a returning child (Novac et al. 2002b).

Family Violence as a Precipitant of Homelessness

A recent Canadian study cited family violence as one of the main causes of homelessness among families. For more than 40% of the 59 homeless families interviewed in 10 major cities across Canada, family violence was among the factors specified as having caused them to leave their homes. Family violence often occurred in combination with other events that led to homelessness, including marriage breakdown and eviction (Social Planning and Research Council of B.C. 2003).

Longitudinal analysis of Toronto shelter use data for the period from 1988 to 1996Footnote 9 suggests that family violence is increasingly declared a reason for homelessness by shelter users. The number of admissions to Toronto shelters linked to family breakdown or violence almost doubled (from 14% to 26%) during that eight-year period (Springer et al. 1998).Footnote 10 More specifically, the number of shelter users who gave "spousal abuse" or "parental abuse" as the reason for their service use doubled (from 7.3% to 14%) between 1988 and 1996 (Springer et al. 1998, 32).

According to the 2003 Statistics Canada statistical profile on family violence (Canadian Centre for Justice Statistics 2003), in a snapshot survey taken on April 15, 2002, almost three-quarters (73%) of the women and more than four-fifths (84%) of the children admitted to family violence shelters across the country were fleeing abusive situations. (The remaining women may have experienced family violence at an earlier time.)

The 2002 Calgary homelessness study determined that "running from an abusive relationship"Footnote 11 was the reason given for their homelessnessFootnote 12 by 29% of the absolutely homeless women, 5% of the absolutely homeless men, 37% of the relatively homeless women, and 5% of the relatively homeless men (Gardiner and Cairns 2002).

Family Violence and Repeat Homelessness

Most shelter users enter the system once and do not return (Springer et al. 1998). Those who are repeatedly homeless are of special concern. Is family violence a factor in chronic homelessness? The research findings vary by type of abuse and family status.

Gardiner and Cairns (2002) developed a profile of the repeatedly homeless, whom they characterized as "less resilient" and having histories of neglect by parents or other forms of family violence.

A Toronto study of 300 homeless single adults found that rates of childhood abuse were equally high among both the repeatedly homeless and those who were homeless for the first time (Goering et al. 2002). Neither were there statistically significant differences between the groups in rates of psychiatric and substance abuse disorders and previous hospitalization. The two groups did differ with respect to out-of-home placements and early experiences of homelessness, however. Of the repeatedly homeless, 37% had out-of-home childhood placements and 49% had been homeless for at least one week before the age of 18 (compared to 22% and 33% respectively among those who were homeless for the first time). The researchers concluded that chronic or repeated homelessness is associated with the lack of a secure and stable place to live during childhood. It is possible, however, that this association is confounded with other factors. For example, it may be that both out-of-home placement and homelessness at an early age are associated with more severe experiences of familial conflict and abuse. To clarify this matter, there is need for more detailed data on childhood abuse histories, including data indicating whether the perpetrator was a family member.

A two-year U.S. study found that rates of childhood sexual abuse were higher among the repeatedly homeless than among first-time homeless mothers (Bassuk and Perloff 2001). As well, first-time homeless mothers who experienced partner violence after being re-housed were more than three times as likely to experience a second homeless episode, even controlling for the effects of childhood sexual abuse. As well, repeatedly homeless mothers were twice as likely to have been sexually abused as children, and they were more likely to have experienced random anger from both parents and to have histories of running away. (These same factors are predictive of post-traumatic stress disorder during adulthood.) It is also noteworthy that social support resources before the first episode of homelessness neither mediated nor moderated the relationship between childhood sexual abuse and repeated episodes of homelessness.

The contrast in the findings of these two studies may be related to the differences in the membership of their samples. The Toronto study was conducted with homeless single people, while the U.S. study focussed on homeless mothers. The presence of dependent children, among other factors, may account for the different results. Moreover, in Toronto, Goering et al. (2002) studied the effects of childhood abuse, while in the U.S., Bassuk and Perloff (2001) examined the effects of partner abuse as well as sexual abuse during childhood.

To explore the links between family violence and homelessness, an Australian study consulted 161 stakeholders (including staff from women's accommodation services, public housing, police, and court administration) and 52 women who had experienced family violence. The results highlight how extensively social and personal disruption and financial disadvantage can affect some women and their children. Prolonged legal settlements, long periods of post-separation violence and stalking, and lack of national residency status can each contribute to extended homelessness (Chung et al. 2001).

Zappardino and DeBare (1992) identified additional factors that may contribute to the greater likelihood of repeat homelessness among victims. They include:

  • abrupt, unanticipated homelessness (unlike that experienced by someone who knows that an eviction is imminent);
  • crisis compounded by trauma;
  • use of denial as a coping mechanism that often impairs problem-solving ability; and
  • child custody disputes.

A longitudinal analysis of shelter use in New York City revealed that homeless mothers who reported family violence had greatly increased odds of repeated shelter use. There was evidence that certain family dynamics (having young children in the household, family instability, violence) were all associated with an increased risk of experiencing additional episodes of shelter use. Family violence and having young children also decreased the likelihood of making a successful shelter exit (Metraux and Culhane 1999).

Using a rigorous research design, Shinn et al. (1998) studied predictors of homelessness and housing stability among first-time homeless families in New York. Positive social ties were not predictive, but early disruptive experiences and adult family violence were clear risk factors for homelessness. Their main finding was that – regardless of poverty, behavioural disorders, impoverished social networks, and loss of affordable housing – the receipt of subsidized housing was the primary predictor of housing stability up to five years later.


It is important that our understanding of the nexus of family violence and homelessness be situated within an appreciation of intersectionality, i.e., an understanding of the particular and unique situations and experiences of groups according to age, gender, sexual identity, ethno-cultural background and disability. With the exception of a focus on gender differences and the particular situation of youth, this level of specificity is rarely addressed in the research literature, though some distinctions have been noted or are beginning to be investigated.

Aboriginal People

Aboriginal people experience a higher rate of both family violence and homelessness than non-Aboriginal people. According to the results of the General Social Survey (1999), Aboriginal women are three times more likely to have experienced spousal abuse than non-Aboriginal women, and they are twice as likely to have experienced it than are Aboriginal men (Besserer et al. 2001). Aboriginal people are also over-represented among the homeless in studies conducted in several Canadian cities, including Calgary, Edmonton, Winnipeg, Vancouver and Toronto.

Immigrant and Racial Minority Membership

Members of some immigrant and/or ethnocultural groups may experience social identities and multiple oppressions that affect their views of, and responses to, family violence. Their cultural backgrounds and immigration status may also affect their use of, and treatment by, service agencies. A recent study of spousal abuse court cases in Montreal found that ethno-racial minority women who were recent immigrants were over-represented, and in almost one-quarter of the cases the victim did not speak either English or French (Oxman-Martinez et al. 2002). Some immigrant women who move out of the marital home and set up independent households to avoid family violence risk being ostracized and socially isolated by their communities.

Recent refugees appear to be more vulnerable to both family violence and homelessness. Racial minority households, especially black households, whether immigrant or not, face discrimination in housing and labour markets that increases the stress they experience and makes it harder for victims to establish and maintain independent households (Novac et al. 1996; Novac et al. 2002a).

Rural and Remote Communities

Because emergency shelters are generally available only in cities, homelessness has been seen as an urban experience. According to Daly (1996), because it is hidden, it is incorrectly assumed that rural homelessness is non-existent in Canada. People in rural, remote and northern communities have few housing alternatives, short of doubling up with friends or relatives or moving to a larger community. But homeless people are, of necessity, quite mobile, and a distinction drawn between an "urban" and "rural" homeless population would be misconceived (Cloke et al. 1999).

Victims of family violence in rural Canada are no different from those in urban environments (Scott 1995), and there is no reason to believe that family violence occurs less frequently in rural environments (Hornosty and Doherty 2002). It is undeniable, however, that in rural settings services are more likely to be absent or more difficult to access. Martz and Sarauer (2000) referred to an urban bias in specialized services responding to family violence, caused by decades of cost-related decisions to centralize services in urban centres. Victims also move to cities because the urban environment can offer a greater degree of anonymity; maintaining confidentiality in small communities is very difficult.

In a report on their study of victims of spousal abuse in rural Saskatchewan, Martz and Sarauer (2000) noted that some women who wanted to leave an abuser could not afford, or were unwilling to suffer, significant economic losses in household and farm property. Limited transportation was identified as another barrier to leaving. Few victims were aware of any legal remedies that could remove the abuser and gain them exclusive possession of their home, such as may be possible by obtaining an Emergency Intervention Order under the Saskatchewan Victims of Domestic Violence Act, 1995. Most of the women who did leave their homes went to shelters in Saskatoon and Moose Jaw. The nearest shelter was an apartment with no staff or form of support, and the one woman who did use this option during the time span of the study felt isolated and scared.

Similar dynamics apply to the situation of military wives, especially those living on military bases far from urban services. Frequent moves can place such a woman in communities in which the prevalent language is unknown to her (Family Violence and the Military Community Research Team 2000).

Older Persons

A relatively small proportion of family violence literature addresses abuse and homelessness experienced by older persons. While abuse of older persons has become a recognized phenomenon, professionals working to prevent it say it typically occurs in a closed setting where the abused person is a "prisoner of their milieu." They report very few cases in which an older person being abused will choose to leave. In situations in which an older person is in an abusive family environment but is not considered competent, he or she may be removed and placed in a protected setting, such as a seniors' home.

In Toronto, from 1988 to 1996, only 15% of those who used shelters were aged 45 or older, and very few (1.8%) were 65 or older (Springer et al. 1998). A count of people using homeless services in Ottawa in December 2001 revealed a similarly small proportion of seniors (2%) (Aubry 2002).

Within small Inuit communities, situations of overcrowding, when pervaded with simmering resentments among adult children who are forced to remain in their parents' homes, have been linked to physical and financial abuse of seniors (Pauktuutit Inuit Women's Association 1994).

Hightower et al. (2001) investigated the situation of older women who experienced family violence. In many cases, there was a long history of wife assault; some women had repeatedly left and returned to the family home. The majority of abusers were husbands, and their abuse was less likely to be financial than that of sons, daughters, and even grandchildren. The impacts of leaving are not the same as those experienced by younger women; in the course of a long marriage a woman tends to accumulate many assets and belongings, including treasured possessions from the past and mementos that become increasingly precious in later years. For a woman who has invested a lifetime of care, leaving a home that holds such personal effects can be shattering and tantamount to a loss of self.

Severe Mental Illness

The recent historical trend of deinstitutionalization and withdrawal of services for persons with intellectual disabilities and mental health problems constitutes another significant factor in homelessness, especially among long-term homeless single women. It appears that the relationship among these factors is dynamic, i.e., women with psychiatric disabilities who live in poverty are at very high risk of being abused and being homeless. Being abused and being homeless also constitute traumas that exacerbate, if not cause, mental health problems.

Goodman et al. (1995) found that the lifetime risk for violent victimization among episodically homeless women with serious mental illness is so high that rape and physical assault are normative experiences. Only three of their 99 respondents reported no experience of physical or sexual abuse in either childhood or adulthood, and for most the abuse was severe. The risk for recurrent abuse was also extraordinarily high. A third of the women reported physical or sexual assault during an episode of homelessness.

Davies-Netzley et al. (1996) investigated the rates of childhood abuse among homeless women with severe mental illness. They found that more than 75% of the women had been exposed to either physical or sexual abuse between the ages of 6 and 18. Women with histories of childhood abuse were much more likely to become homeless during childhood, and those who had been both physically and sexually abused during childhood were more than 15 times more likely than non-abused women to experience homelessness prior to the age of 18. The combination of childhood abuse and the lack of outside social and financial support can put young women at high risk for homelessness and also be a precursor to mental health problems.

Dynamics of Family Violence and Homelessness

Family violence is disproportionately a phenomenon of youth. As victims and offenders, the rates are highest for those between the ages of 18 and 30. Males are more likely than females to be offenders, and females are more likely than males to be victims of physically injurious assaults. In societies around the world, the greater the inequality between men and women and the greater the degree of social disorganization, the higher the rate of assaults against wives. While family violence occurs in all social and economic groups, the risk of child abuse, wife abuse and elder abuse is greater among those who are poor or unemployed or holding low-prestige jobs. As well, those with fewer resources and more stress are more likely to use violence (Gelles 1993).

Homelessness and housing unaffordability are also predominantly problems of the young (Novac et al. 2002b), the result of poverty and inadequate social resources. Those conditions cause a high level of stress and increase the likelihood of being victimized.

As the above review of research findings indicates, the relationship between family violence and housing status is complex. Clearly, housing problems, such as crowding and unaffordable costs, are linked to household stress and can contribute to abusive behaviour within the family (Pauktuutit Inuit Women's Association 1994). It is therefore not surprising that women living in urban public housing experience a greater rate of violence from intimates than do other women (DeKeseredy et al. 1999). The role of housing in preventing or addressing family violence is multi-faceted and fundamental (Weisz et al. 1996). Unequal power — economic, physical and social — within society and within family or household relations plays a role in determining whether family conflict and violence occur, who abuses whom, and what access to alternative housing will be available.

Various policies and institutions are implicated in these dynamics. Bufkin and Bray (1998) argued that a weak response to family violence on the part of the criminal justice system is a crucial intervening variable related to homelessness. Cumberland (2001) has suggested that those made homeless because of family violence should be seen as distinct, in that their homelessness is the result of a crime committed against them, and that governments should be concerned about this as a distinct justice imperative.

Gardiner and Cairns (2002) identified several policy orientations and institutional practices that can ameliorate outcomes of homelessness. They include child protection and child care services (access to good quality child care, early intervention in cases of abuse and neglect, and improved effectiveness of the foster care system) and spousal violence legislation (to facilitate removal of the abusive partner from the home, to ensure consistent law enforcement). Health and housing agencies also shape paths to homelessness for victims of family violence and affect how victims achieve safety, recovery and housing stability.

In their review of Canadian research on woman abuse, DeKeseredy and Hinch (1991) noted that subsidized housing and short-term emergency shelter are required to prevent continued female victimization by male partners and to promote women's safety, autonomy and self-reliance.

Because women's access to housing is largely dependent on their position in families (Novac 1995), when marital relationships break down, the economic consequences are considerably different for them than for men. After divorce, the poverty rate among women increases almost threefold. Their household income drops by more than 40%, while men's increases slightly (Finnie 1993). Single women and single mothers account for almost half of households with affordability problems (CMHC 2000). Discrimination in labour and housing markets has not been eliminated, especially for Aboriginal and other racial minority women. Such socio-economic inequalities affect the power dynamics within family relationships. Dependent members are under pressure to stay because of their inability to afford alternate housing, and this can fuel more conflict.

Since the early 1970s, the battered women's movement in Canada and other countries has stressed that the lack of affordable housing is a barrier that prevents abused women and their children from "moving on" after shelter stays (OAITH 1989, 1990, 1998; Walker 1990; Schechter 1982). The economic inability of women to set up independent households plays a clear role in their decisions to stay with or return to abusive partners.

"... Homelessness is not resolved for women by having a roof over her head unless this roof is accompanied by a sense of safety and security" (Neal 2004, 1).

Breton and Bunston (1992) found a dramatic reduction in the rates of physical and sexual assault experienced by homeless women in comparison to those in their prior living situations.

Temporary homelessness and supportive, safe shelters can provide a much-needed temporary respite for women and children facing harsh, ongoing realities of poverty, partner abuse, victimization and trauma (Styron et al. 2000).

Clearly, family violence is a trigger or proximate reason for homelessness for a significant proportion of women, youth and children who use shelters (a very high proportion of those using family violence and youth shelters, and a smaller but possibly increasing proportion of those using homeless shelters).

The dynamics of family violence can contribute to poverty as well as financial and social vulnerability. Men who abuse their wives or partners commonly also restrict their social relationships with friends and other family members and control their ability to work outside the home; this may contribute to the abused woman having relatively limited social and economic resources available to help her deal with an event of homelessness.

"When women have no option to reside somewhere else, men are often able to strengthen their control over women's lives, which may increase the opportunity for some men to be abusive. By providing women with housing options, whether or not they use them, their control over their lives is increased and the message that spousal assault will not be tolerated is reinforced" (Weisz et al. 1996, 14).

Implications for Service Provision

Most victims of family violence do not seek out services for help. Findings from a Statistics Canada survey show that most victims seek informal support rather than formal assistance. The majority (80%) of male victims and almost half (48%) of female victims who experienced spousal violence during the five years prior to the study did not use any type of social service agency for assistance. Only 11% of female victims and none of the male victims used a shelter (Canadian Centre for Justice Statistics 2003). They tend to stay with friends or relatives and thereby become members of the hidden homeless.

There is very little literature that explicitly addresses the implications for service provision with relation to family violence and homelessness or the kinds of service that are most appropriate and effective.

Patterns of Shelter Use

The common tendency for women to move into the homes of relatives or friends immediately after a separation may mark the beginning of a spiral into homelessness (McCarthy and Simpson 1991). A woman may resort to using a shelter only after she has exhausted the good will of family and friends or been evicted after struggling to pay rent that she cannot afford. In other words, family violence may trigger a series of moves, yet not be the immediate reason for using a shelter. A pattern of temporary stays with friends and family ("couch-surfing") is also common among homeless youth (Novac et al. 2002b). Shelter use and an intermittent pattern of public and hidden homelessness is likely to follow.

Most people admitted to shelters in Toronto from 1988 to 1996 as a result of spousal abuse stayed from five days to six months. A small number of them (9%) spent only one or two days; another small proportion (11%) spent one year or more in the system. Over time, fewer of them were able to obtain subsidized housing on leaving a shelter (from 32% in 1988 to 14% by 1996). Very few (2.4%) said they were returning to their partner (Springer et al. 1998).

The episodically or repetitively homeless are defined as those who entered the shelter system more than five times in one year. One-third of this group were found to be women fleeing abusive partners (Toronto 2001, 7).

A survey of those leaving family violence shelters across Canada showed that 25% moved to "alternate" housing and 19%, to "other" housing; 12% went to stay with friends or relatives; a similar percentage returned to their spouse; and 8% returned to their home without their spouse. The destination of the remaining 25% was unknown (Canadian Centre for Justice Statistics 2003). This suggests that about one-third of victims who flee violent homes remain homeless or unstably housed (i.e., they "double up" with another household or family, or staff are unaware of where they go) and about one-tenth return to being "homeless at home."

Not surprisingly, shelter use was found to be strongly correlated with the severity of the violence. Twice as many of the shelter users had suffered an injury – 80%, compared to 45% of those who did not use shelters – and 63% had at some time been injured severely enough to seek medical attention (compared to 19% of all abused women). As well, shelter users were more than twice as likely as other abused women to have feared for their lives (85% versus 39%).

In the case of youth,Footnote 13 those who returned to their parents tended to leave the shelter system quickly. However, in cases in which youth were abused by their parents, the most common time pattern was a shelter stay of one month to two years (Springer et al. 1998).

Shelter Service Issues

In Toronto, women and children who are victims of family violence are increasingly using homeless shelters. In 2001 the City of Toronto estimated that as many abused women and their children stayed in homeless shelters as in family violence shelters (Toronto 2001). In that same year in Vancouver, 26% of the approximately 2,100 people using emergency shelters during an 11-month period reported that family abuse or breakdown was the predominant reason for their homelessness (Greater Vancouver Regional Steering Committee 2002).

This trend may be occurring in part because some shelter staff co-operate to accept each others' overflow, and in part because some victims of family violence may prefer to use homeless shelters. There are several imaginable reasons for this possible preference: they might not actually consider themselves victims of family violence; they might fear the associated stigma; they might seek shelters with fewer restrictive rules and policies; if they are parents, they might fear being reported to child welfare authorities or they might want to avoid separation from older male adolescent children who are generally excluded from family violence shelters.

On the other hand, many victims of family violence may require benefits and services that are offered only at family violence shelters – most notably the greater security from abusive family members that such shelters provide, and the personal counselling they need to deal with their traumatic experiences.

Homeless shelters are not as likely to be able to provide the same level of safety for victims of family violence. (There are documented instances of abusive family members and dating partners pursuing young women who were staying in youth shelters (Novac et al. 2002b).) This is critical, as the likelihood of partner abuse and escalated violence increases for some women who have left their partner. Based on an analysis of more than 1,200 cases of women killed by their male partners between 1974 and 1994, Gartner and Crawford (1998) found several associated risk markers, the primary one being recent separation. This risk must be seriously considered in safety planning and service provision. It is not mitigated by the fact that, for many women, becoming homeless is safer than remaining in an abusive relationship.

Similarly, without appropriate counselling and other services to assist women and children to deal with trauma, there is a tendency toward re-victimization, especially among women who have been sexually abused in childhood (Wyatt et al. 1993).

Additional characteristics of homeless shelters that may discourage their use by victims of family violence include:

  • presence of male residents,
  • absence of staff who have been trained regarding family violence issues,
  • absence of services for children,
  • absence of any confidentiality safeguards,
  • absence of any safety planning for abused women,
  • inadequate building security, and
  • rules that require residents to spend the day outside the shelter (Novac et al. 2002b).

Browne (1993a) has suggested that, because a high proportion of homeless women have suffered physical and/or sexual abuse repeatedly in their lives, shelters serving women must be able to provide ongoing support in a safe environment away from their abusers. As well, staff must be educated about the dynamics of family violence and homelessness.

Homeless youth tend to prefer to use shelters and other services that are designed for youth rather than adults. The legacy of family violence among homeless youth includes high levels of alienation, mistrust and self-imposed isolation. All of these conditions can contribute to the development of psycho-social impediments, post-traumatic stress disorder and depression. They also deter homeless youth from seeking help from adults. For these reasons, youth-specific services are required. Currently, however, they are available only in major cities (Novac et al. 2002b).

As well, most youth shelters are gender-mixed, which raises additional issues of safety. Young homeless heterosexual women are subject to high levels of sexual exploitation and dating violence while staying in gender-mixed youth shelters (Ibid.).

Recovery and Mental Health

In her research on trauma, Herman emphasized that recovery from family violence victimization is a slow process, "a gradual shift from unpredictable danger to reliable safety, from dissociated trauma to acknowledged memory, and from stigmatized isolation to restored social connection" (Herman 1992, 155). Both safe shelter and supportive relationships are critical to recovery.

Some researchers have urged service providers to be aware of the effects of family violence among their clientele. For instance, Bassuk and Perloff (2001, 137) observe that "providers and policy makers must be aware of the pervasiveness of childhood sexual abuse and recent partner violence and its relationship to repeated shelter use." But researchers rarely suggest how professional practice should be altered.

In practice, staff and residents may not quite agree on the service needs of family violence victims, including those who suffer from symptoms of post-traumatic stress disorder. Gorde et al. (2004, 702) found that staff in shelters and transitional housing projects placed more emphasis on residents' mental health and therapeutic needs than did the residents themselves. The residents "demonstrated defensive avoidance, which can hinder engagement in treatment." Residents were focussed on the priority to find a safe and secure place in which to live and manage their own lives. This suggests that, even for severely traumatized victims, the provision of permanent affordable housing is a prerequisite to the recovery process.

When victims are ready to begin healing, the lack of sufficient mental health services covered by public insurance is a major barrier to recovery for both homeless youth and adults.

Post-Traumatic Stress Disorder (PTSD)

As mentioned above, residents of abused women's shelters have experienced more severe levels of physical, sexual and psychological abuse, and are more likely than other abused women to have suffered a severe injury and feared for their lives (Trainor 1999). High lifetime rates of both childhood abuse and assault by intimate partners (Browne 1993b) are associated with both high levels of subsequent PTSD (North and Smith 1992) and long-term homelessness (Goodman et al. 1995). Post-traumatic stress disorder is one of the most common manifestations of traumatic victimization and is likely to coexist with depression and substance abuse (Bassuk et al. 1998; Davis 1998; and Browne 1993b).

Compared to poor women who are housed, homeless victims of family violence who suffer PTSD have more chronic health problems and more problematic relationships with health care providers (Bassuk et al. 2001). Experiences of sexual violence may affect women's capacity for trust-building and subsequent adherence to preventive and ongoing health care plans (Goodman et al. 1997).

Services for Older Adults

Some researchers believe that senior victims of family violence are not generally well served by family violence shelters for several reasons: the shelters are not structurally designed to accommodate the needs of older people; the noise and activity levels are too stressful for seniors; there is a lack of peer support; and the shelters are not designed to accommodate the needs of a couple who may be escaping abuse together (Bergeron 2000 cited in Canadian Centre for Justice Statistics 2003, 26).

Services for Homeless Children

Most children who accompany their mothers to family violence shelters are between one and four years of age. Shelter staff report that poor self-esteem is the most common problem among these children, followed by behavioural problems, poor social skills, poor school achievement (CMHC 2001), as well as mental health problems, chronic health issues, substance abuse, malnutrition and related disorders (Social Planning and Research Council of B.C. 2003). Because of limited resources, only some shelters are able to provide any follow-up services to families (CMHC 2001).

Services for Homeless Youth

U.S. researchers have explored the psycho-social consequences of maltreatment suffered by homeless youth. Those who were victimized in the home by their parents are more at risk of further victimization while homeless than those who were not (Whitbeck and Simons 1990). Adolescents who had been both physically and sexually abused within their families exhibit the most severe psychological problems and are at greatest risk for re-victimization (Ryan et al. 2000). Whitbeck and Hoyt (1999) found that homeless young women who had been sexually abused by an adult caretaker were twice as likely as young men to be re-victimized.

Adolescent runaways forced to flee sexual abuse and physical brutality in their families have special emotional needs that set them apart from youth who are escaping overly strict parents or leaving home for other reasons. They have more severe separation problems, unresolved issues with their parents, and difficulties in their post-runaway relationships (Powers et al. 1990).

According to the "risk amplification" developmental model proposed by Whitbeck and Hoyt (1999), the negative effects of early psychological harm from abusive families are amplified while adolescents are on their own. With increasing emotional separation from parents, adolescents become more reliant on peers who provide information and support and help socialize them regarding street survival skills. Deviant social networks and high-risk behaviours increase the likelihood of serious re-victimization, leading to assaults and exploitation. Re-victimization and coercive relationships reinforce what they learned in their families. This process is very hard to reverse. Attempts to force submission only confirm their aggressive/ coercive world view. Inevitable encounters with the legal system do the same.

Kurtz et al. (1991) argued that temporary protection, crisis intervention and counselling are not sufficient for youth who have been abused nor for the "doubly homeless" who have run from public care. For them, more in-depth assessment, treatment and placement services are required and they should be delivered through multi-agency co-ordination and case management of services that are "flexible and forgiving." Services should provide assistance with the provision of housing arrangements that differ from the foster or group homes from which many have run away.

Programs that focus on training and employment readiness are not appropriate for all homeless youth. Those who have been traumatized by family violence may not be able to seek or maintain employment until recovery is well under way. Novac et al. (2002b) identified the need for a shelter, specifically for young women, that would help them cope with past abuse and deal with their poor self-esteem and issues relating to sexuality and relationships.

Need for Transitional, Supportive and Permanent Housing

On a selected day in 2002, about one-quarter of family violence shelters across Canada (115 out of 482) turned away people in need of shelter, usually because the shelter was full (Canadian Centre for Justice Statistics 2003). A survey of 391 abused women's shelters found that one-quarter identified the need for more shelters. One-third also identified the need for more second-stage or transitional housing units (SPR Associates 2002a).

Many transitional housing programs have been developed to assist homeless individuals and families. Although program models vary, the concept usually involves temporary provision of housing for a period of three months to three years, in combination with an array of support services. An evaluation of more than 500 transitional housing programs in the United States found that those families who completed the programs achieved a high rate of housing stability (90%). However, the success rate was substantially lower among abused women (41% to 61%), and their employment status generally did not improve (Novac et al. 2004).

Whether or not transitional housing programs are an appropriate response to homelessness, especially for families with dependent children, their effectiveness is predicated on the availability of affordable move-on housing (Ibid.). In its 2003 report on homelessness, the City of Toronto identified a need for more supportive housing units (permanent, subsidized housing with support services, with no time limit imposed on the duration of stay) for several sub-groups of homeless people, including women fleeing abusive relationships (Toronto 2003).

A large study conducted in nine U.S. cities established that provision of permanent subsidized housing was more effective, at least for homeless families, than transitional housing programs. A majority (88%) of the 781 chronically homeless families remained housed 18 months after being provided with permanent housing and short-term support services, despite the fact that a very high proportion of the mothers had experienced childhood risk factors, were poorly educated, had health problems, had experienced family violence, and were alcohol and drug dependent (Rog et al. 1995).

Another U.S. program that was designed to prevent family violence or reduce re-victimization provided a range of services to increase women's power to escape violent relationships. Of all the services offered — education, legal support, childcare, independent housing, job training, employment assistance, independent living skills, assistance in negotiating with the welfare system, etc. — the provision of housing was critical to reducing the likelihood of further violence. Of the women who obtained independent permanent housing, 98% reported no further victimization. Of those women who returned to their former residence and abuser, 86.3% reported further victimization. While employment also protected women, housing was the factor most strongly associated with women's long-term safety and was deemed more effective than criminal justice remedies (Webscale and Johnson 1998 cited in Pascall et al. 2001).

Child Welfare

Raychaba (1993) criticized the public care system in Canada for creating too much transiency in placement locations and, more importantly, in relationships. He argued that a stable, trusting, long-term relationship with caregivers is what works best for youth in care and what is least available. In Ontario, Crown wards average one change of placement or social worker a year.

Two-thirds of former Crown wards residing in a large Toronto youth shelter said they had not been prepared for independent living. They were less likely than other homeless youth to have a supportive network of family or friends. This included young women who again came to be involved with the child welfare system later as parents (Leslie and Hare 2000). Leslie and Hare (2000) recommended a review of the Ontario child welfare system's mandatory "aging-out" policy and suggested that contractual service arrangements should be available for former Crown wards up to 21 years of age.

As well, the foster care system often fails to provide children with therapy to help them deal with their familial experiences of abuse and neglect. In a portion of cases, children have experienced further abuse within their foster families or group home residences (Roman and Wolfe 1997; Downing-Orr 1996; and Zlotnick et al. 1998).

Special Priority Housing Policies

Special priority housing policies for victims of family violence are in effect in various jurisdictions across Canada. They are intended to quickly re-house family violence victims who become homeless by giving them priority allocations of subsidized housing. Unfortunately, there have been no assessments of their effectiveness in meeting their primary objective or in preventing homelessness.

In Ontario, where such a policy was adopted in 1986, some local housing agencies have reportedly escalated their demands for verification and details of abuse to establish eligibility and have ignored claims that related only to non-physical abuse (OAITH 1998). Service providers in Australia and Britain have reported similar difficulties with such policies (Dillon 2001; Malos and Hague 1997). Davis (2003) attributed these shortcomings to several factors, chief among them the lack of consultation with housing staff during the policy's development and inadequate training of staff on the dynamics of wife assault.

In 2002 the Ontario Coroner's office held an inquest into the death of Gillian Hadley. She was murdered by her husband after he had been convicted of abusing her and been barred from their home while she was waiting to be re-housed. The inquest identified failings of the Special Priority Policy – in particular, the requirement that the application for housing "should be submitted within three months of separating from her abuser." Given the common pattern of abusive spouses continuing to stalk, threaten and attack their former partners for several years after separation, and the fact that the greatest risk to women is during the first year after separation, the three-month time limit constitutes an unnecessary barrier for women who are being abused or are under threat of abuse.

Among its many recommendations to the inquest jury, the Ontario Association of Interval and Transition Houses requested removal of the three-month restriction, an audit of housing authorities and agencies to ensure consistent, standardized application of the policy, and provision of a transparent complaint process for subsidized housing applicants.

Family Violence and the Prevention of Homelessness

Despite the high rates of family violence in the backgrounds of people who become homeless, Shinn and Baumohl (1999) questioned what a program to forestall potential violence would look like. They argued that, as a form of homelessness prevention, addressing the basic housing issue would be more effective than any efforts to prevent family violence. They acknowledged that risk factors for homelessness could be reduced by universal strategies to prevent family violence (by changing norms of acceptable behaviour, punishing perpetrators, and providing support and education to parents) and to reduce the need for, and increase the quality of, foster care. Yet the vast majority of abused and placed children do not become homeless, and benefits for the prevention of homelessness would be very gradual.

Identifying families at risk of violence and determining the appropriate interventions pose significant challenges. Although programs to prevent family violence are in themselves useful, these researchers did not support such programs as an approach to prevent homelessness. In situations in which households are already experiencing violence, they advised against any attempts to maintain the household. Instead, they concluded that emergency shelters and permanent, subsidized housing are the only effective prevention measures. They reiterated the findings of a large New York study (Shinn et al. 1998) which noted that homeless families who received subsidized housing were able to achieve long-term residential stability, regardless of their histories of childhood abuse, out-of-home placements, childhood poverty, and adult family violence.

Re-Housing Victims

There are some gender differences in the paths to homelessness. Homeless women more often have histories of family violence and high rates of mental illness, while homeless men more often have histories of unemployment, incarceration and substance abuse. Brown and Capponi (1993) counselled caution regarding the potential risks in new housing projects that mixed formerly homeless women and men with little or no attention to their different histories.

Many such gender-mixed projects were developed in Toronto during the late 1980s and early 1990s. Most consisted of small, self-contained units; some were shared apartments. An investigation of women's experiences in such projects found that although they were satisfied with their new dwellings in many ways, they faced an extremely high level of harassment and abuse from male residents. Sexual harassment incidents were generally not reported to managers, but when women did make complaints of such harassment, they were very disappointed with the outcomes. Housing developers, managers and other staff need training to anticipate the risks, prevent sexual harassment and offer meaningful remedies (Novac et al. 1996). Just as employers have had to learn to deal with problems of sexual harassment in the workplace, housing managers will have to do so with regard to their environments.

Research Gaps

More than a decade ago, Fischer (1992, 237) noted that researchers were slow to acknowledge the role of family violence in patterns of homelessness.

"The realization of the central role of victimization in creating and sustaining homelessness has come relatively late to investigators, arising mostly from incidental studies of early findings. Next to surface in early studies examining pathways to homelessness was the suspicion that homelessness might result directly from certain forms of victimization — principally domestic violence casting women from their homes .... Perhaps the least suspected effect of victimization was suggested by early, almost serendipitous findings of what appeared to be excessively high rates of various indicators of early childhood dysfunction, especially of foster care placement and physical and sexual abuse."

Research associating family violence with homelessness has advanced since then but is still in its early stages. A great many questions remain unanswered. In their assessment of the implications of the links between family violence and homelessness, Braun and Black (2003) pointed to research gaps in two areas:

  • the overlooked, complex realities of women experiencing family violence and homelessness, and
  • service delivery best practices.

Researchers have suggested investigations that would address gaps in both areas, as the following list indicates.

Complex Realities

  • Bassuk and Perloff (2001, 316) have called for additional research to clarify the impact of violent victimization on repetitive homelessness, specifically "to understand the extent to which childhood molestation is a marker of chronic homelessness, in contrast to recent partner violence, which may be a marker of episodic homelessness."
  • There is some evidence of a cyclical pattern among young homeless women who have histories of traumatic childhoods of family conflict and violence, involvement with the child welfare system, and early pregnancy (Novac et al. 2002b). There may also be an inter-generational element. However, no empirical studies in Canada or elsewhere have investigated these dynamics.
  • Several analysts have pointed out that we know virtually nothing about intermediate causes of homelessness, nor protective mechanisms that may mitigate against it. "Network analysis [could tell] us about the ways in which families, friends, informal networks, neighbourhoods, peer groups and sub-cultural groups operate and how they may be preventing or exacerbating housing exclusion" (Avramov 1998, 20). Are there informal protective mechanisms that may prevent homelessness among victims of family violence and, if so, can they be bolstered with public supports or provided in more formal ways with public resources?
  • It appears that a history of family violence increases the likelihood of repetitive homelessness, but we have little more than conjecture as to why this occurs. Longitudinal research with homeless victims of family violence could identify the factors that contribute to repetitive homelessness and possibly suggest effective interventions.
  • Research to study the factors contributing to the over-representation of Aboriginal women among the homeless is especially warranted in this context.
  • The link between poverty and homelessness continues to be deserving of focused research that will provide reliable statistics. Related to this broad issue is its relationship with the availability of public housing and sufficient affordable housing.
  • Bufkin and Bray (1998) asked whether homeless women experience the same problems with their abusers as do housed women, e.g., stalking, harassment, and physical abuse incidents.

Services and Prevention

  • We lack information on abused women's use of and experiences with homeless shelters, as opposed to family violence shelters. For example, Krishnan and Hilbert (1998) recommended collaboration between homeless and family violence shelters to better deal with the needs of women who use homeless shelters. Research should determine the extent to which this occurs and whether it is a good idea.
  • Given that the child welfare system collects no information on what becomes of their clients on discharge (e.g., whether they become homeless on leaving care), research is required to establish the dynamics of the path from public care to homelessness, the extent of the problem, and the factors that distinguish those who become homeless from those who do not, with particular attention to the role of family violence.
  • As established above, compared to other abused women, women staying in family violence shelters have experienced more extensive or severe levels of physical, sexual, and psychological abuse and are more likely to have suffered a severe injury and feared for their lives (Trainor 1999). This may also be the case for their children. If we knew that their risk of future homelessness is greater than that of victims of family violence who use homeless shelters, we might conclude that targeting more resources to family violence shelters would be a wise minimal investment.
  • We know almost nothing about the long-term housing and support service needs of family violence victims.
  • There has been no research on the effectiveness of Canadian legal remedies to remove perpetrators of family violence from the family home or otherwise protect victims from financial loss. Research conducted in the UK showed that judges were reluctant to order abusive husbands out of their homes. A study of 656 women in family violence shelters established that only 8% relied on protective legislation to return to their homes with the abuser excluded. Moreover, one year later, only 4% were still living there (Binney et al. cited in Pascall 1986, 154-6). In Australia, feminist housing workers have lobbied for reform of landlord/tenant legislation to allow women to break their lease when they are leaving a property because of violence, and thereby be exempt from liability for rent payments after they have left (Dillon 2001).

Page details

Date modified: