Abstract
While police are gatekeepers to the criminal justice system for intimate partner violence (IPV), women are often divided about their helpfulness. The current secondary, mixed-methods analysis examined 575 Canadian women's impressions regarding police intervention (50.3% Indigenous, 43.4% White, and 6.4% visible minority). In more than half of the cases (53.3%), the police arrested/removed their partner/ex-partner, 53% said the police “just talked to them,” and 8% were also arrested. While 40% were satisfied with the police response, 21.5% thought police did not take them seriously, and 10.9% were explicit about the police not treating them well. Practice and policy implications are provided.
Women whose partners or ex-partners use intimate partner violence (IPV) against them often experience significant physical injuries, which is the major reason they report incidents to the police (Barrett et al., 2020; Ogrodnik, 2006). Sexual violence (Tarzia, 2021; Tutty & Nixon, 2022) is also often a factor in IPV but is much less often reported to the police compared to physical violence (Bielen et al., 2022; Messing et al., 2014; Wright et al., 2022). In addition, many women are subject to coercive control, defined as a pattern of controlling emotional abuse (Stark & Hester, 2019), which, although a significant problem for women, is unlikely to be a factor in police charges (Aspinall et al., 2024; Hilton et al., 2023), especially in jurisdictions that have not criminalized coercive control (e.g., Canada).
The Canadian criminal justice system, including police, Crown prosecutors (district attorneys), defense lawyers, judges, and probation/prison officers deal with an enormous caseload of IPV incidents every year (Statistics Canada, 2024; Tutty & Koshan, 2013). Police officers are the front-line implementers of justice policy; their actions constitute the gateway into the criminal justice system (Russell & Light, 2006). The police response to IPV has been of long-standing concern to those who work closely with abused women, as it often seems insufficient (i.e., Saxton et al., 2021, 2022). The question of whether the police respond differently to Canadian women of diverse racial/ethnic backgrounds who report IPV, the focus of the current research, remains understudied (Couture-Carron et al., 2022; Tam et al., 2016).
This paper begins with a summary of research on IPV on racial/ethnic groups in Canada, research on the police response to IPV incidents, including relevant policies and studies of the racial/ethnic backgrounds of women who call the police. A secondary data analysis of women's perceptions of police intervention in their IPV cases in a sample of 575 Canadian women with diverse racial/ethnic backgrounds follows.
Theoretical Perspective: Intersectionality
Intersectionality is “a theoretical or analytical approach that simultaneously considers multiple categories of identity, difference, and inequality (gender, race, class, sexual orientation, disability, as well as others)” (Else-Quest & Shibley Hyde, 2016, p. 155). Originally developed to refute the idea that all women are at equal risk of IPV, intersectionality is commonly used in IPV research (Baird et al., 2021; Bent-Goodley, 2007), including IPV and disability research (Sasseville et al., 2022; Tutty & Nixon, 2024) and mothering and IPV (Broughton et al., 2022). An intersectional perspective is appropriate in the current study of a large group of Canadian women of diverse racial/ethnic origins and their views of the police response to their reported IPV incidents.
IPV in Diverse Canadian Women
Not all Canadian women face the same risks of IPV or have the same IPV experiences. Indigenous people (including First Nations, Métis, and Inuit peoples) represent about 4.6% of Canada's population (Statistics Canada, 2013a). According to Brownridge (2008), Indigenous women are at risk of being abused by intimate partners three times higher than non-Indigenous Canadian women (21% compared to 7%). As noted by Heidinger (2021), Indigenous women often endure the most serious physical and sexual IPV. Structural issues in Indigenous communities contribute to a greater risk, including high rates of poverty, lack of services, loss of traditional lifestyles, and loss of parental role models because of residential schools and colonization (Brownridge, 2008). This should be viewed in the context of colonization (Baskin, 2020; Ogden & Tutty, 2023) and the systematic oppression and dehumanization of Indigenous men and women and children through institutions such as residential schools (Olsen Harper, 2006).
In Canada, individuals who are non-White are referred to as “visible minorities” and make up a noteworthy portion of the population of Canadian women. In 2011, the percentage of visible minority women and girls was 19.3% of the total female population, according to Statistics Canada (2013b), with South Asian, Chinese, and Black the most common visible minority groups, representing a continuous, upward trend since the 1980s. In a study comparing visible-minority Canadian women to White Canadian women, Hyman et al. (2009) found no differences in the self-reported rates of physical and or sexual violence in the last 5 years.
The Police Response to IPV in Canada and the United States
Until the 1970s, the police response to incidents of IPV was limited, as it was generally considered to be a private matter. Since the mid-1980s, all Canadian jurisdictions support a pro-charging directive that the police lay charges in all IPV cases when warranted, effectively shifting the onus of laying charges from the victim to the state (Bader et al., 2019; Fraehlich & Ursel, 2014). These pro-arrest policies were initially developed to provide better protection for IPV victims by ensuring a uniform police response to cases and providing general deterrence.
However, researchers and community professionals became increasingly concerned about the number of dual arrests that have been documented (Leisenring, 2011; Poon et al., 2014). A dual arrest (or cross-charge) is when both individuals involved in an IPV incident (i.e., the male and female partners) are arrested. Some see this as a negative side effect of the pro-arrest laws (Finn, 2004; Powers & Bleeker, 2023). As an example of the numbers of dual charges, in a study of 72 Canadian women IPV survivors in Alberta and Saskatchewan (Tutty et al., 2008), the police charged the male partner in almost three-quarters of the cases (72.2%); a smaller proportion was dual charges (11.1%); and in 2.8% of cases the woman alone was charged.
In both the United States and Canada, dual arrests have increased in some jurisdictions since the advent of pro-arrest policies (Durfee, 2012; McCormack & Hirschel, 2021; Poon et al., 2014), although who this affects varies. In McCormack and Hirschel’s (2021) US study, White women were more likely to be subject to dual arrests than Black women, and in Durfee’s (2012) US study, women's arrest rates increased much more than men's.
Primary aggressor policies are an attempt to address problems with dual arrests (Fraehlich & Ursel, 2014; Grace, 2019; Hirschel & Deveau, 2017) by training criminal justice professionals, mainly police officers and prosecutors, to identify whom to charge in incidents where it may be difficult to determine who is the victim and who is the perpetrator. Ambiguity can arise, for example, when both parties display injuries, and it is not clear whether these resulted from self-defense. It can also be unclear when both parties are making accusations of assault, but only one party displays injuries. Typically, police officers examine the level of violence perpetrated and the degree of coercive control (Henning et al., 2006), although this does not necessarily remove the subjectivity involved with this decision. As well, the effectiveness of primary aggressor policies has been questioned. In a recent US study (Hirschel & Deveau, 2017), the policy reduced the likelihood of dual arrests, but when instructed to arrest the primary aggressor, the officers often decided to arrest neither party. Further, a US 10-year review (2000–2009) of primary aggressor policies from 5,481 jurisdictions (Hirschel et al., 2021) reported that, while the laws appeared to result in officer reluctance to make any arrest, there was no statistically significant reduction in dual arrests. Notably, there was a 41% decrease in dual arrests with respect to Black offenders and victims. Grace's recent Canadian qualitative study (2019) with 18 women solely charged by police with IPV assaults in Ottawa, Canada, raises questions about whether the police are implementing the primary aggressor policies they were directed to use.
According to Dawson and Hotton (2014), there is no official Canadian national charging policy on IPV, and data pertaining to the adoption of pro-arrest versus primary aggressor policies is generally lacking. Regarding the sites of the current research, Fraehlich and Ursel (2014) note that primary aggressor policies were adopted in the city of Winnipeg, Manitoba early in the 2000s, with a concomitant decrease in the number of both dual arrests and the arrests of women as the sole IPV offender. The province of Alberta first mentioned primary aggressor policies in 2008 (Alberta Justice, 2008), beyond the current data collection. Adding to the complexity, different police agencies are used in different regions. In Canada's three prairie provinces, Alberta, Saskatchewan, and Manitoba, for example, the Royal Canadian Mounted Police (RCMP) service primarily rural settings and local municipal police officers service most cities.
Who Contacts the Police?
According to the General Social Survey (GSS) 2004 report that tabulated data for a 10-year period (1994–2004), which is close to the data collection period for the current study (Ogrodnik, 2006), only 28% of victims of IPV turned to the police for help, more often in response to IPV from ex-spouses (43%) than current spouses (36%). The majority of IPV offenders (81%) were reported to police only once during the 10 years; repeat offenders (with 2–4 IPV police-reported incidents) accounted for 18% and chronic abusers (with 5 or more reported incidents) accounted for the remaining 1% of offenders.
Examining the police response to 383 Canadian women (of 920 who reported IPV in the past 5 years in the 1999 GSS), Barrett et al. (2011) reported that 77.8% had contacted the police themselves and had experienced extreme IPV, including fear for their lives and experiencing multiple violent incidents, which is congruent with other studies (i.e., Dichter & Gelles, 2012; Leisenring, 2012). In contrast, some women choose not to contact the police for reasons such as privacy concerns, fear of reprisal, fear of the perpetrator, desire to protect the perpetrator, and concerns about a poor police response (Felson et al., 2002; Saxton et al., 2021; Wolf et al., 2003).
Satisfaction With the Police Response
Among the relatively few studies to assess women's responses to, and satisfaction with, police interventions for IPV, Peirone et al. (2021), using the Canadian 2009 GSS, reported that 65% of female victims were satisfied with the actions of the police. Previously, Russell and Light's (2006) Canadian study, using interviews with 63 women and 28 police officers, concluded that three dimensions affected victim satisfaction: integrated team versus isolated unit (i.e., whether or not police officers were embedded within a community response to IPV or not); police perceptions of a deserving versus undeserving victim (i.e., judgements of the extent to which victims required services, attention, and direction); and proactive versus pro-forma police responses (i.e., the extent to which police actively intervened, including conducting thorough interventions, arresting assailants, and offering information and referral).
Three US studies add to this body of knowledge. Johnson (2007) interviewed 50 women (about half of whom were African American) about their perceived safety after a police response. In about 75% of the cases, the victim called the police; 56.4% of the victims reported that the officers were very helpful, and 53.8% reported that they had seemed interested in her welfare. About one-third of the women regretted calling the police; in about one-fifth of these cases, the victim wished that the offender had been arrested, but he was not, which is consistent with Leisenring (2012). In Leisenring's study, over half of the 40 women were dissatisfied with the police response, mostly because two-thirds had experienced an unwanted arrest—either their partner's or their own and 35% of the women were arrested, mostly as the sole arrestee. Gezinski’s (2022) qualitative study with 43 American IPV survivors also identified problems with police responses. Particularly in rural areas, the women referred to inadequate responses, such as minimization and victim-blaming. They also noted difficulties in identifying the primary aggressor, which resulted in the women's criminalization and described how some IPV perpetrators received preferential treatment from the police officers.
Women of Diverse Backgrounds and the Police
Canadian Indigenous women victimized by partners report higher rates of IPV (Heidinger, 2021) and have long expressed concerns about the police response to them (Ursel, 2006). Among women who had experienced IPV and responded to Canada's 1999 GSS, immigrant and racialized women were somewhat less positive about the police response than other women (Johnson, 2007).
In a review of 28 qualitative studies of the perceptions of IPV survivors from diverse populations respecting the police responses, the following themes were identified: (a) revictimization by the police, (b) police negligence, (c) discrimination, (d) cultural differences, and (e) positive experiences (Belisle et al., 2024). Revictimization was reported most often and was defined as the police gaslighting, threatening, or physically or verbally assaulting victims. Police negligence refers to police officers not upholding their duty by not acting against the perpetrator of the abuse (e.g., not upholding protection orders). Common among studies of racial/sexual minority women, discrimination refers to using derogatory slurs or mistreating survivors because of their identity. Except for racially/ethnically minoritized women, some positive experiences with the police were noted in studies of all groups, for example, the police connected survivors with resources and followed their arrest wishes.
Two Canadian studies of how racialized women perceive the police response in IPV situations were included in the Belisle et al. (2024) review and so merit additional attention. Couture-Carron et al. (2022) interviewed 90 immigrant, racialized women, finding few demographic differences between those who did and did not contact the police about IPV; the women who called reported more serious IPV than those who did not call. A majority (80%) reported positive experiences with the police, indicating that they were treated respectfully and that the police complied with their wishes. The second study, Tam et al. (2016), involved interviews with 14 racialized women from two Canadian prairie provinces (as in the current study) and identified the following factors as influencing them to seek assistance from the police: safety, severity of the violence, and their children's well-being. Difficulties with the justice response included apathetic justice personnel, discrimination, and cultural insensitivity. An additional, earlier Canadian study (Singh, 2010) concluded that immigrant women have special reasons for not contacting the police, as removing their partners may be detrimental if the women depend on their partners for their immigration status, financial assistance, and linguistic support.
Rationale for the Current Study
Relatively few studies have assessed women's perceptions of police interventions for IPV, the exceptions being the US studies already cited. Most of the cited Canadian studies focused on women identifying as racially minoritized but excluded Indigenous women, an obvious gap. The current study addresses these gaps by examining how women of different racial/ethnic backgrounds perceive their interactions with the police in a region of Canada with the highest rates of IPV, the prairie provinces of Alberta, Saskatchewan, and Manitoba (Johnson, 2006; Ursel, 2006). Studies from this region also report higher rates of police-laid major assault charges involving bodily harm or weapon use (Ogrodnik, 2006; Ursel, 2006); dates congruent are with the current study data collection. Unfortunately, though, the police-reported data seldom identify the racial/ethnic backgrounds of perpetrators or victims, so we know relatively little about Indigenous and visible minority women and the police response. The current study provides the opportunity to do so.
Method
A secondary descriptive mixed-method data analysis was undertaken with a sample of 575 women from the “Healing Journey.” The original study included 665 Canadian women abused by intimate partners in the prairie provinces, focusing on mental health and general wellbeing (Tutty et al., 2021a), experiences of mothering (Ateah et al., 2019; Nixon et al., 2017; Tutty et al., 2025a), and their use of IPV-specific services (Tutty, 2024; Tutty et al., 2025b). The women were followed over 2.5 years in seven waves between 2005 and 2009 (Tutty et al., 2021b). Both academics and community agency personnel were part of a large research team.
The research protocols were approved by the Research Ethics Boards of the six associated universities (Universities of Calgary, Manitoba, Regina, Brandon, Lethbridge, and Winnipeg). Ethical considerations included informed consent, confidentiality, and permission to withdraw. The women were recruited from community agencies that address IPV and met the following criteria for inclusion in the study: (1) a minimum 18 years of age; (2) the most recent incident of IPV no sooner than 3 months and no longer than 5 years prior; (3) commitment to stay in the study for the full 4 years; and (4) no significant mental health issues. The women answered the study questions in a location of their choosing: The community agency from which they were recruited, their home or the local research institution, whichever they considered safer.
Demographics of the Research Participants
As can be seen in Table 1, in response to the yes/no question about whether they had ever contacted the police (another 90 did not answer the question), the 575 women were an average age of 36.7 years (SD = 11.0) while their partners were slightly older (39.1 years; SD = 11.4). Slightly more than half of the women self-identified as Indigenous (50.3%), 43.4% identified as White, and 6.4% identified as visible minority. The identified racial/ethnic backgrounds of the abusive partners were 46% White, 46.3% Indigenous, and 7.7% visible minority.
Women's Demographics by Ever Called the Police (N = 575).
p < .05.
Most had children (90.3%), with 84% of the mothers having at least some children younger than 18 years. Most (81.7%) no longer lived with their abusive partners. About half (36%) had some form of post-secondary education, while 42.3% had not completed high school. About one quarter worked full time (22.4%), 16.7% worked part-time/casual, and 60.9% were not currently employed.
Quantitative Component
The nature of the IPV was assessed using the 30-item Composite Abuse Scale (CAS) (Hegarty et al., 1999). This screening measure rates items for their frequency in the past 12 months on a 6-point scale from never to daily, with a possible total score of 150. The four subscales are Severe Combined Abuse (eight items, range of scores = 0–40, suggested cutoff of 1), Physical Abuse (seven items, range of scores = 0–35, cutoff of 1), Emotional Abuse (11 items, range of scores = 0–55, cutoff of 3), and Harassment (four items, range of scores = 0–20, cutoff of 2). The suggested clinical cut-off for the total CAS score is 7 to minimize false positives. The scale has demonstrated convergent and discriminant validity (Hegarty et al., 1999). Cronbach's alpha for the CAS in the original Healing Journey study is .93.
In addition, the women were asked forced-choice questions about any police involvement, including, “Did you ever call the police?” “If yes, how many times?” “Have other people called the police?” and a series of items about how the police responded, including arresting the partner (see Table 2 for the questions).
Police Response by Woman's Ethnic/Racial Background (N = 575).
Bolded writing indicates significant differences.
p < .05; p < .01.
Research Procedures: Qualitative Component
The qualitative component entailed two open-ended questions: How did the police respond? And what did you think of the police response? Descriptive qualitative analysis (Neergaard et al., 2009; Sandelowski, 2000) was used to code the comments of 497 women who offered their views of the police response. Qualitative secondary analysis re-uses pre-existing qualitative data from previous research (Heaton, 2008), an important option since qualitative studies often produce a wealth of data not used in the primary analysis. Secondary analysis is especially appropriate with vulnerable populations to prevent repeated questioning that might be traumatic (Long-Sutehall et al., 2011; Sullivan-Bolyai et al., 2005). Neergaard et al. (2009) suggested that descriptive qualitative analysis is particularly appropriate for mixed methods research as it uses the practicality of the research question as the guiding principle, rather than epistemological confines of qualitative traditions such as grounded theory or phenomenology.
The descriptive analysis, conducted by the first author, followed established inductive content analysis processes (the meaning emerges from the quotes) (Elo & Kyngäs, 2008; Sandelowski, 2000). First-level coding entails word-by-word scrutiny of the narratives to identify similarities and differences about data segments, called meaning units (Braun & Clarke, 2006; Coleman & Unrau, 1996). While reading and rereading data, the analyst mentally connects similar meaning units together as categories. Finding themes involves locating patterns that appear repeatedly in the categories. Second-level coding is more abstract and examines the themes and subthemes to identify similarities, differences, and gaps using the constant comparative method (Coleman & Unrau, 1996; Thorne, 2000).
Results
As is evident in Table 1, about one-fifth (112 or 19.5%) of the 575 women had never contacted the police, while 463 (80.5%) had called the police themselves. Those who called the police differed from those who did not on only one demographic characteristic; the non-callers had a higher proportion of adults who had both adult children and those under age 18. There was no difference in the severity or the type of IPV experienced by the two groups.
Table 2 shows the responses to the forced-choice questions about the police response, broken down by racial/ethnic identity. The women described their partners as having a substantial number of previous arrests: Only 15.6% had no previous arrest history, but 68.4% had previous arrests because of assaulting them. Overall, 53.3% reported that the police had arrested/removed their partner/ex-partner.
There were a few statistically significant differences between the racial/ethnic groups. A smaller proportion of Indigenous women (16%) had called the police “only once” in comparison to White women (28.2%). The police told a larger proportion of White women (18.6%) that the incident was not criminal compared to visible minority (3.6%) or Indigenous women (10.6%). A larger proportion of Indigenous women (61.1%) than either White (45.1%) or visible minority (46.4%) women reported that the police had arrested or removed their partners. A larger proportion of White women (11%) noted dual arrests than Indigenous (4.7%) or visible minority (7.1%) women.
Women's Views of the Police Response
In response to the open-ended question, “How did the police respond?” (see Table 3), 497 women commented about various issues, including the arrest history. Note that most of these comments were short one- to three-sentence responses. The coded responses were broken down by the racial/ethnic background of the women and of their partners (noted by White [W], Indigenous [I] or visible minority [V] firstly for the woman and, secondly, the partner given our interest in possible differences, although these cannot be compared statistically. By inspection, 12.5% mentioned that the police response was fast (the largest proportion was for Indigenous women with Indigenous partners, 16%). A comparable proportion (13.3%) characterized the police response as slow (from hours to days after the initial phone call, to not responding at all)—the largest proportion was for White women with Indigenous partners (22.4%).
Number of Qualitative Comments About Women's Views on “How Did the Police Respond to IPV Calls?”
As reported in Table 2, 53% of the women endorsed the item asking if the police just spoke with them. This was also mentioned by 12.9% of the women in their comments but was not necessarily perceived as a negative encounter, for example, “Explained how subsidized housing works and gave me good advice. They were professional; told me he was a textbook abuser (W-I)”; Other women were less content with this police response, for example, “Cop talked about how he remembers tough times when first married and to try to work it out (W-W)”; “They said it was ‘just another domestic’ (V-W)”; “Said it wasn’t their jurisdiction (W-W).”
About half of the women (53.5%) endorsed the item asking if the police had arrested their partner or removed him (see Table 2). However, in their open-ended comments (see Table 3), only 16.5% of the women mentioned this. Overall, their comments had a positive tone, for example, “They laid charges. They got him out of my house every time (I-I)”; “Police charged him because they had the evidence like wounds, torn shirt, broken nose (I-V)”; “They insisted on charging him when I wouldn’t so I’m glad they did it. I probably wouldn’t be here if they hadn’t arrested him (W-W).” However, the men were not necessarily kept in custody for long, for example, “They saved me from being strangled by my ex-husband. The next day, they phoned and told me that he would be released in one hour, so I had one hour to get out (I-I)”; “I thought the police should have kept my ex-husband in jail at least for one night but they released him within four hours (W-W)”; “I was dissatisfied with length of time my ex-partner stayed in jail (quick release) (I-W).”
In their open-ended comments (see Table 3), 11.5% of the women mentioned that the police did not lay charges. In some cases, their comments implied that they thought charges were warranted, for example, “A couple of times I was black-eyed, bleeding, and the police said it was not a criminal matter—just sober-up and go home (I-W)”; “Police said there was nothing they could do (W-W)”; “They didn’t want to do anything. They didn’t pick him up until four days later (W-I)”; “I called the day after an assault—was told that they couldn’t press charges and that I should have called the day of the assault. It was his word against mine—they were not helpful (V-W).”
As noted in Table 3, a small proportion of the women (3.9%) commented that the police told them to bring charges against the partner, for example, “They asked if I wanted to charge him (W-I)”; “Police were not compassionate because I was afraid to charge him (W-W)”; “Police said I had to charge him because it is ‘no tolerance’ issue and so they required me to charge him before they would help me get to the shelter (I-I).”
As indicated in Table 2, a small proportion of the women endorsed the item about a dual arrest (8.9%), and an even a smaller proportion (1.6%) mentioned dual arrests in their open-ended comments (see Table 3), all of whom were Indigenous with Indigenous partners, for example, “We were both arrested because partner accused me of threatening him with a knife. Charged me with assault (I-I)”; “When they arrested both of us, they were rude about it. I was covered in blood, and they didn’t let me clean myself (I-I)”; “The police saw scratches on my partner's face and arrested and charged both of us (I-I).”
The questions about arrest history documented in Table 2 did not include a question about the woman being solely charged. However, as can be seen in Table 3, 3.9% of the women commented about being solely charged in their open-ended comments, for example, “In 2002, the police arrested me, threw me around violently. They treated me with disrespect, my partner with respect. I had bumps, cuts, bruises, and he talked his way out of it! (I-I)”; “Police arrested me for mischief and took my kids away (I-I)”; “They charged me with public mischief. They don’t realize how afraid I am and how much courage it takes to take action (W-W).”
Satisfaction with the Police Response
Of the women who commented on their satisfaction with the police response, 40% indicated that it was good (see Table 4), for example, “The police treated me respectfully. They listened, didn’t judge me or make me feel like I was lying (I-I)”; “Police very good—phoned to check on me—let me know what was going on. Felt he was a danger and decided to confiscate weapons. Kept checking on me (I-W)”; “The RCMP has been helpful with abusive behaviour to watch for. They validated my fears and offered advice (W-W)”; “I found the police kind and helpful. They took my children to their grandmother's house and me to the hospital. They made me feel secure, told me to call anytime (W-I).”
Qualitative Comments About Women's Views of the Nature of the Police Response.
Remembering that some women were describing multiple calls to the police, 5% of the comments about the police response indicated mixed satisfaction (most often from Indigenous women with Indigenous partners, 8.3%), for example, “There were mixed responses. There were times when I felt heard and they focused on getting the facts straight. Other times, they just brushed me off (I-I)”; “Some good, some not, overall, pretty good. One wouldn’t enforce the restraining order—he didn’t take the time to read it—made assumptions about what it said (V-V)”; “Once they were rude, yelled at me because they thought I was lying (after many complaints to the police about abuse). Mostly good support, take me seriously (I-W).” Some of the mixed comments were about an inconsistent police response, for example, “At first, I had nothing bad to report: I called for help, and they came. Then, they were alerted that my ex-partner was a known criminal. They were only interested in getting me to divulge things to them, then would tell him who I ratted on. I came to regret that because he would come back and hurt me. Sometimes I thought I’d be killed (I-I)”; and “They have been to my home more than 20 times and have always taken him away. They were good to me sometimes and other times they weren’t. They knew my address: I was a frequent caller (W-W).”
About one-fifth of the women (21.5%) did not believe that the police took them seriously (the largest proportions were for White women with visible minority partners, 31.6%, and Indigenous women with White partners, 27.4%). Comments included: “Treated me with disrespect. I was told to stop bothering them (V-W)”; “The first time, 2003, he abused me, blackened my eye. They came and talked to me; took his name and that was it (I-I)”; “They weren’t doing their job; they were trying to get out of doing the paperwork. They didn’t arrest him. They said it wasn’t that bad and I should let my ex off this time (I-W)”; “They always said, ‘don’t worry, he’ll sleep it off.’ I know there are good officers, but I get all the idiots (W-W).”
Another 10.9% of the women were explicit about the police not treating them well (the largest proportions were for Indigenous women with Indigenous partners, 13.6%, and White women with Indigenous partners, 13.8%). Comments included: “It was bullshit the way they talked to me. They didn’t like me; didn’t want to help. The cops said he didn’t pull my hair. I had to show them my pulled-out hair (I-I)”; “Towards the end, they started arresting me. I was really angry. They said it was easier to put me in jail than him (W-W)”; “I found a loaded gun in my closet and when the police came and found guns, they trashed my house and charged me because it was in my house (W-I).”
A small proportion of women (3.8%) believed that the police took their partner's side, for example, “I feel they were protecting him because they knew him well. He used to fish with some of them and I think they didn’t want to see him get into trouble (W-W)”; “They spent more time listening to my ex-husband than to me (W-V)”; “I quit calling because they wouldn’t help me. Ex was very convincing—getting police to be sympathetic to him (V-W)”; “They didn’t believe me but believed the bullshit stories my husband was feeding them. They should have seen the red flags in his story a mile away (I-W).”
Finally, 2.2% of the women commented that the police had made racist comments or seemed to have racist attitudes (mostly Indigenous women with Indigenous partners), for example, “I found the police prejudicial; there is a lack of respect for me as an Aboriginal that I feel deeply (I-I)”; “I feel the police are racist. Have called me and family members ‘drunken Indians’ in the past (I-I)”; “They acted ignorant and treated me with disrespect because I am Aboriginal (I-I).”
Discussion
The results of the study are largely consistent with previous research. The proportion of abused women who reported calling the police themselves (80.6%) was comparable to 77% in another Canadian study (Barrett et al., 2011) and to 75% in Johnson’s (2007) US study. Interestingly, there was no difference in the severity of the IPV when comparing women who had police contact versus those who did not. This may be explained by the nature of the convenience sample in our study, which was recruited from IPV-related services. A larger proportion of women in our study, 53.3%, reported that the police had arrested/removed their partner/ex-partner compared to only 27.3% in Barrett et al. (2011), which may be due to differences in the wording of the questions. The number of dual arrests (8%) is congruent with two Canadian studies from the prairie provinces (Fraehlich & Ursel, 2014; Tutty et al., 2008) in which 7% were dually arrested, but somewhat higher than that reported in Poon et al. (2014), a third Canadian study in which 11% were dually charged. In our study, the women were not directly asked if they had been charged, but a small number mentioned in their open-ended comments that they had (3.8%), consistent with 2.8% in Tutty et al. (2008) and 9% in Fraehlich and Ursel (2014).
In our study, few differences were identified in the police response when considered by racial/ethnic background. Unlike Barrett et al. (2011), we found that women who had contacted the police were not significantly more likely to be racialized or Indigenous, to have lower household incomes, to possess a high school diploma or general equivalency diploma, or to live in rural regions. In contrast, a higher proportion of Indigenous women had called the police more often; the police told White women more often that the issue was not criminal; and more Indigenous women reported that the police had arrested or removed their partners. Finally, more White women were subject to dual arrest.
Although we were not necessarily expecting differences in the women's perspectives of the police response across, the reason for the relatively few issues that emerged is difficult to ascertain. It may be that the women who engaged in the study were more similar across racial/ethnic groups than we had anticipated. The questions about the police response were only a small aspect of the Healing Journey study, which focused on health/mental health, custody, as well as other IPV services such as violence against women shelters. Continuing to study potential racial/ethnic differences in the police response is important, as the diverse dynamics of IPV in these groups has been well-documented (i.e., Heidinger, 2021; Johnson, 2007; Singh, 2010; Ursel, 2006). Additional research on Canadian women's perceptions of the police response to IPV is recommended.
In commenting about their experiences, almost half of the women in our study were satisfied with the police response (40% good and 5% mixed, which was mostly associated with separate police-reported incidents), about 10% less than the 53% in Johnson’s (2007) US study. While the question is appropriate, satisfaction with the police response should not be treated the same as program evaluations, when women voluntarily sign up for generally lengthy interventions with clear goals and outcomes. Almost one-fifth of the women in our study had never contacted the police themselves; 41.6% had had others contact the police at least once. As is clear from other research (Barrett et al., 2011), women's wishes for police action are often immediate, such as simply wanting the partner to be removed; but may change over time such as not wanting to testify against the partner in criminal proceedings, so regretting having involved the police.
The themes that emerged from the analysis of the women's comments about the police response were similar to those identified by Belisle et al.'s (2024) review of 28 qualitative studies of diverse populations of survivors: revictimization by the police (treating women poorly, 10.9% in the current study); police negligence, that is, not laying charges, 11.5%; just talking to the woman, 12.9%; siding with the partner, 3.8%; discrimination, 2.2%; and positive experiences, 45% good/mixed).
Notably, almost 50% of the women reported that the police had asked them if they wanted charges laid (see Table 2), effectively placing the responsibility on the women, despite existing Canadian criminal justice system policy from the mid-1980s that the police must lay charges if there is sufficient evidence (Bader et al., 2019; Poon et al., 2014). There are several ways to interpret this. Police officers may not be aware of this policy, they may be disregarding the victim, or they are appropriately seeking support for such a consequential action as arresting the partner (which they may or may not heed as is evident in women's comments about this). If the arrest were to proceed to criminal court, the woman would be asked to provide evidence against her partner, and assessing her willingness to do so may represent the context behind the question. However, their discussions should reflect this rather than what seems a clear request for her to charge him.
Practice and Policy Implications
Clearly, the police have a difficult task when attending IPV calls, as the dynamics of the abuse are often complex and may not be evident in a single event. Risk assessment measures (Saxton et al., 2022) such as the ODARA (Ballucci et al., 2020) are one possible resource to aid police interpreting the severity of the IPV. To our knowledge, none of these were in use in the prairies when the study was conducted.
Comprehensive IPV training remains the core recommendation to improve the police response and enhance victim's satisfaction (Eigenberg et al., 2012; Russell & Light, 2006), a direction recognized as important by the police themselves (Gill et al., 2021). Primary aggressor training provides an additional venue to educate police members in the complex mechanisms with which IPV abusers control their partners. Having experts, such as shelter advocates, provide the training to police, a hope that was embedded in developing primary aggressor policies, would facilitate a deeper understanding of IPV.
In examining police policies related to responding to IPV, we could not determine whether any of the police services were using primary aggressor policies when the study was conducted (i.e., Alberta Justice, 2008; Fraehlich & Ursel, 2014). Nevertheless, comprehensive research on the effects of implementing primary aggressor policies has not demonstrated that it reduces dual arrests or increases arrests in IPV calls (Hirschel et al., 2021; Hirschel & Deveau, 2017). Cross (2024), a Canadian IPV advocate, is among those calling for the abolishment of mandatory charging/pro-arrest policies. However, a study in South Carolina, where mandatory arrest policies were rescinded (Mason & Petrie, 2021), identified continuing problems with the police criminalizing some populations (i.e., Black men). Further discussion and research on mandatory charges and primary aggressor policies are clearly needed.
For the women in our study, perhaps the greater concern was to be listened to with respect, being believed and being offered community resources, rather than an arrest outcome, elements that stood out in the comments of the women who were pleased with the police response. This mirrors other studies of women's concerns with the police, such as Russell and Light (2006) and Johnson (2007). Police officers are trained to ask questions in an investigative manner rather than with empathy but, recently, some have suggested a focus on the latter (Eliasson & DeHart, 2023; Jakobsen, 2021), which could enhance victims’ sense of being heard and respected.
Study Limitations and Strengths
When using secondary data sources, one is limited by the original study, which relied on a convenience sample of women from VAW shelters or counselling agencies. This, of course, limits the generalizability of our results. Further, while the data set is from 2004 to 2005, secondary analysis allowed us to capture important information not highlighted in previous research. It is not unusual to conduct secondary analyses in older data sets; for example, Hilton et al. (2023) examined data from 1997. Police responses, including adopting primary aggressor polices, have certainly changed since the data were collected, although questions about whether these decrease dual arrests in Canada have been posed (i.e., Grace, 2019). We could identify only a few Canadian studies on police interventions that considered the racial/ethnic backgrounds of either the woman or the male partners (i.e., Couture-Carron et al., 2022; Tam et al., 2016). The women in our study were not asked to specify when the police contact occurred, so incidents may have happened years before data collection commenced in 2004–2005.
An important contribution is that the current study includes a large sample of Indigenous women whose experiences of the police response to IPV have not been well researched. Some of the differences between the data in Tables 2 and 3 about the police response, such as whether the partner was arrested, reflect variations in the way the questions were asked, that is, forced choice versus open-ended. All women were first asked the questions documented in Table 2, while the data in Table 3 emerged from open-ended questions to which the women's responses did not necessarily refer to a specific IPV incident to which the police responded. In answering the two open-ended questions, “How did the police respond?” and “What did you think of the police response?” the women may not have repeated details that had already been recorded in the Table 2 questions. The comments may represent what the women most wished to be documented about their experiences, regardless of how many. As the questions about the police response were only one aspect of a complex research project, future research that focuses more directly on Canadian women's perceptions about the police response across racial/ethnic backgrounds should be conducted.
Conclusion
That less than half of the women (45%) in the current study were satisfied with the police response, which is less than the 56.4% in Johnson’s (2007) US study, suggests the need for improvement. The reasons for women's dissatisfaction are congruent with previous studies (i.e., Belisle et al., 2024; Dichter & Gelles, 2012; Johnson, 2007; Leisenring, 2012; Russell & Light, 2006) and include the police not taking the women seriously, treating them disrespectfully, siding with the partner, and, in a few cases, behaving in a racist manner. New research that focuses on how women perceive the police response to IPV and how the police respond to IPV incidents is recommended given the adoption of primary aggressor policies across Canada (Alberta Justice, 2008; Fraehlich & Ursel, 2014; Grace, 2019).
Most of the partners in the current study had arrest histories: 84.4%, with 68.4% of the charges specific to IPV. These are dangerous men and, as such, it is essential that women perceive the police to be a resource that will take their needs seriously and perform their duties according to the law. For many, as identified in the comments, it is quite simply “a matter of life or death,” confirming the need for the police to consider their perspectives.
Footnotes
Acknowledgments
The CURA team: Dr. E. Jane Ursel and Marlene Bertrand (Manitoba Department of Family Services and Housing, MB) are the Co-Principal Investigators; Dr. Kendra, L. Nixon; Dr. Christine Ateah; Dr. Janice Ristock; Dr. Lori Wilkinson; Colin Bonnycastle; Dr. Jocelyn Proulx (University of Manitoba); Dr. Johanna Leseho; Dr. Roberta Graham (Brandon University); Dr. Linda DeRiviere; Dr. Michelle Owen (University of Winnipeg); Anna Pazdzierski (Nova House, Selkirk, MB); Karen Peto (YWCA Brandon); Margaret Marin & Darlene Sutherland (Osborne House, Winnipeg); Dr. Mary R. Hampton; Dr. Bonnie Jeffery; Dr. Darlene Juschka; Dr. Wendee Kubik (University of Regina); Dr. Stephanie Martin (University of Saskatchewan); Carol Soles (Prince Albert Emergency Shelter for Women); Debra George (Family Services Regina); Dr. Karen Wood (Tamara's House, Saskatoon); Maria Hendrika (Provincial Association of Transition Houses Saskatchewan); Angela Wells (Family Support Centre, Saskatchewan); Dr. Leslie M. Tutty; Dr. H. L. Radtke; Dr. Wilfreda Thurston; Dr. Erin Gibbs Van Brunschot (University of Calgary); Dr. Caroline McDonald-Harker (University of Alberta); Dr. Ruth Grant Kalischuk (University of Lethbridge); Jan Reiner & Carolyn Goard (Alberta Council of Women's Shelters); Brenda Brochu (Peace River Regional Women's Shelter); Kristine Cassie (YWCA Lethbridge); Pat Garrett (WINGS of Providence, Edmonton).
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Social Sciences and Humanities Research Council (SSHRC) Community University Research Alliance (CURA); Alberta Centre for Child, Family, & Community Research; Alberta Heritage Fund for Medical Research; The Prairieaction Foundation; and TransCanada Pipeline.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
