In late 2013, the Commissioner of the RCMP initiated an RCMP-led study of reported incidents of missing and murdered Aboriginal women across all police jurisdictions in Canada.
This report summarizes that effort and will guide Canadian Police operational decision-making on a solid foundation. It will mean more targeted crime prevention, better community engagement and enhanced accountability for criminal investigations. It will also assist operational planning from the detachment to national level. In sum, it reveals the following:
This report concludes that the total number of murdered and missing Aboriginal females exceeds previous public estimates. This total significantly contributes to the RCMP's understanding of this challenge, but it represents only a first step.
It is the RCMP's intent to work with the originating agencies responsible for the data herein to release as much of it as possible to stakeholders. Already, the data on missing Aboriginal women has been shared with the National Centre for Missing Persons and Unidentified Remains (NCMPUR), which will be liaising with policing partners to publish additional cases on the Canada's Missing website. Ultimately, the goal is to make information more widely available after appropriate vetting. While this matter is without question a policing concern, it is also a much broader societal challenge.
The collation of this data was completed by the RCMP and the assessments and conclusions herein are those of the RCMP alone. The report would not have been possible without the support and contribution of the Canadian Centre for Justice Statistics at Statistics Canada.
As with any effort of such magnitude, this report needs to be caveated with a certain amount of error and imprecision. This is for a number of reasons: the period of time over which data was collected was extensive; collection by investigators means data is susceptible to human error and interpretation; inconsistency of collection of variables over the review period and across multiple data sources; and, finally, definitional challenges.
The numbers that follow are the best available data to which the RCMP had access to at the time the information was collected. They will change as police understanding of cases evolve, but as it stands, this is the most comprehensive data that has ever been assembled by the Canadian policing community on missing and murdered Aboriginal women.
This report is organized around four topics: the numbers of murdered and missing Aboriginal females; homicide perpetrator characteristics; what we understand about the outstanding cases; and, victim circumstances.
Additionally, the report briefly outlines some immediate steps to be taken by the RCMP to build on present efforts. These are reflective of this report as a snapshot of an evolving understanding and not definitive conclusions.
One particular challenge in conducting this research was the use of the term "Aboriginal" because of its different meanings across multiple organizations and systems. Details on this issue, project methodology and data sources are available in Appendix A.
It would be premature to focus on research outcomes without first addressing the context of the research. Violence against women is a significant societal issue. According to the World Health Organization, it affects one-third of women around the globe and represents a health problem of "epidemic proportions".Footnote 1
This project focused on a subset of that challenge, particularly related to the disappearance and murder of females of Aboriginal origin in Canada. It treads a path similar to recent studies by the Native Women's Association of Canada (NWAC) and Dr. Maryanne Pearce.Footnote 2 This study seeks to contribute to the ongoing dialogue on this important matter.
What this project did differently was to supplement publicly available data with a comprehensive extract of information from law enforcement holdings from across all police jurisdictions in Canada. This fills a significant gap.
The result should serve as a foundation upon which to target crime prevention and crime reduction efforts by the police. It may also provide additional information for communities and all levels of government.
The totals of murdered and missing Aboriginal females outlined in this report differ from existing publicly available research for a variety of reasons including differences in scope, collection methodology, data mismatches, and/or purging of records from closed files.
The RCMP has almost completed cross-referencing the data it collected from police records with NWAC and Dr. Pearce's research. Reconciliation to date has been valuable in establishing these findings and improving police record data quality.
The totals in this report also differ from previous statements by the RCMP. These have always focused on datasets narrower in scope or more limited in time frames surveyed. For example, in September 2013, the RCMP reported to the UN Committee for Elimination of Discrimination Against Women (CEDAW) that it had 36 ongoing investigations of missing Aboriginal women where foul play was suspected and 327 Aboriginal female homicides in its jurisdiction. These numbers were smaller because they focused solely on RCMP jurisdictions and spanned a relatively short period of time.
When looking at the issue of murdered and missing Aboriginal females, it is important to remain cognizant of the broader reality of violence affecting Aboriginal women in Canada. They are at a higher risk of being victims of violence than non-Aboriginal females. According to the 2009 General Social Survey (GSS) on Victimization, nearly 67,000 Aboriginal females reported being a victim of violence in the previous 12 months.Footnote 3 The rate of victimization among Aboriginal females was close to three times higher than that of non-Aboriginal females.
The broader reality is particularly relevant to ongoing work to identify communities of concern. Preliminary analysis suggests there is not enough data for meaningful community-by-community analysis. Combined with data on communities with high rates of violence against women year-to-year, a more useful picture materializes. Many of these communities match locations in the data collected for this project.
The 1980 to 2012 time frame was chosen to align as much as possible with that of the Native Women's Association of Canada (NWAC) Sisters in Spirit initiative to allow for comparisons and corroboration with recent public work on murdered and missing Aboriginal women.
Police-reported incidents of Aboriginal female homicides and unresolved missing Aboriginal female investigations in this review total 1,181. This number includes 1,017 Aboriginal female homicide victims between 1980 and 2012, and 164 Aboriginal women currently considered missing. Of these, there are 225 unsolved cases of either missing or murdered aboriginal women.
According to recent data from the 2011 National Household Survey, 1.4 million people identified as Aboriginal in 2011, representing 4.3% of the Canadian population.Footnote 4 The proportion of Aboriginal females in Canada's female population is similar. In 2011, there were 718,500 Aboriginal females in Canada, representing 4.3% of the overall female population that year.Footnote 5
The largest percentages of persons identifying as Aboriginal as a total of the population were found in the territories (Nunavut at 86.3%, Northwest Territories at 51.9% and the Yukon at 23.1%), followed by Manitoba (16.7%) and Saskatchewan (15.6%). Persons self-identifying as Aboriginal represented less than 8% of the total population in the remaining provinces.Footnote 6
Thousands of people are missing in Canada on any given day. The whereabouts of many are established quickly.Footnote 7 As a result, there is a certain challenge in arriving at an accurate count of what is a moving picture. The total of missing Aboriginal females was based primarily on a report of all women listed as missing for more than 30 days across all police jurisdictions on the Canadian Police Information Centre (CPIC) system as of November 4, 2013.Footnote 8
According to this extract and subsequent quality assurance review and follow-up, there were 164 missing Aboriginal females as of November 4, 2013. They make up approximately 11.3% of the total number of missing females (1,455 total). The most recent was reported in September 2013.
It is possible that the total number of missing Aboriginal females in this data set is different than the actual number due to a variety of factors including a missing female not being identified as Aboriginal during the investigation and/or a disappearance not being reported to police. For example, a number of cases that surfaced in NWAC and Dr. Pearce's research may not have surfaced in this project.
The reasons that these 164 individuals are missing vary. An individual may be a victim of a violent crime, may have gone missing due to an accident, or may have disappeared voluntarily for personal reasons. The best understanding of this is entered as the "probable cause" in CPIC by the police of jurisdiction at the time the person is reported missing. This breaks down as follows: 37% (61 cases) unknown; 27% (44) foul play suspected; 27% (45) accident where a body was never recovered; 7% (12) lost or wandered off; and, 1% (2) runaway.Footnote 10
Recognizing the clearly non-suspicious nature of many of these disappearances and the reality that the longer an individual is missing for unknown reasons, the more reason there is to be apprehensive about their fate, the research has aggregated these probable causes further into suspicious/unknown circumstances (64%, 105 individuals) and non-suspicious, (36%, 59 individuals).
It might seem odd to describe any missing person's disappearance as non-suspicious. Unfortunately, however, this is frequently the case. A number of people disappear as a result of mishaps, e.g. a boating accident or recreational swimming drowning (43 of the 59 non-suspicious missing Aboriginal females are categorized as presumed drowning). While there is reason to believe they are very likely deceased, there is insufficient information to officially categorize them as such.
Five elements of the data related to murdered Aboriginal females are discussed below: over-representation as a proportion of victims; trends as a proportion of victims; location of death; cause of death; and, related criminality.
Between 1980 and 2012, there were 20,313 homicides across Canada, which averaged approximately 615 per year. Females represented 32% of homicide victims (6,551 victims) across all police jurisdictions between 1980 and 2012.Footnote 11 Every province and territory was implicated.Footnote 12
There were 1,017 Aboriginal female victims of homicide during this period, which represents roughly 16% of all female homicides — far greater than their representation in Canada's female population as described above.
Eastern Canada exhibited lower proportions of Aboriginal victims than Western Canada and the North, keeping in mind overall population demographics for the provinces and territories will significantly impact proportional percentages.
Over-representation of Aboriginal female homicide victims appears to hold for most provinces and territories (See Figure 2 for a breakdown). The finding of over-representation is consistent with other research conducted in Canada on homicides of Aboriginal peoples.Footnote 13
|Aboriginal||Non-Aboriginal||Unknown||Aboriginal victim proportion|
The number of Aboriginal female victims of homicide has remained relatively constant while the number of non-Aboriginal female victims has been declining (Figure 3).
Figure 3 — Female homicides per year, 1980-2012
Description of graph in tabular format follows.
Aboriginal women accounted for 8% of female victims in 1984 as compared to 23% in 2012 (Figure 4).
Figure 4 – Aboriginal female homicides as proportion
of total female homicides per year, 1980-2012
Description of graph in tabular format follows.
The growing proportion of Aboriginal female homicides is a direct reflection of a decrease in non-Aboriginal female homicides.
The Aboriginal female homicide rate per 100,000 population dropped from 7.60 to 4.45 between 1996 and 2011, the years data on this statistic was reviewed (Figure 5).
Figure 5 – Female homicide victimization rate
Description of graph in tabular format follows.
|Aboriginal Females||Non-Aboriginal Females|
Source: RCMP Revised Homicide Survey Dataset, Statistics Canada Table 051-0001. Aboriginal female population based on Census data for 1996, 2001, and 2006, and the Statistics Canada National Household Survey 2011
There were differences in the cause of death between Aboriginal and non-Aboriginal female victims. Additional research to understand these differences in detail is required.
Among Aboriginal female victims, approximately one-third (32%) died as a result of a physical beating. This was the most frequent cause of death among Aboriginal female victims and was reported almost twice as often as it was for non-Aboriginal female victims (17%). Aboriginal and non-Aboriginal female victims of homicide were similarly likely to be victims of stabbing (31% Aboriginal versus 27% Non-Aboriginal). Non-Aboriginal female victims were more likely to have died as the result of a shooting (26% versus 16%) and were also more likely to have been strangled, suffocated or drowned (22% versus 13%).
Figure 6 – Cause of death, female homicide victims, 1980-2012
Description of graph in tabular format follows.
|Strangulation, suffocation, drowning||13%||22%|
|Smoke, inhalation, burns||3%||3%|
Data on where the victim was killed was only available for the period from between 1991 and 2012. During this period, there were similarities between Aboriginal and non-Aboriginal female homicides. Approximately three quarters of victims were killed in a residence.Footnote 14 Vehicles, public transport and open areas were the next most frequent location of the homicide.
|Vehicle, public transport, street or parking lot||10%||9%|
One element captured was whether the offender had the original intent to kill the victim.Footnote 15 An associated or related offence led to the homicide in approximately one third of all female homicides.Footnote 16 Aboriginal female homicides were slightly more likely (39%) to involve an associated or related offence than non-Aboriginal female homicides (31%).
The two most frequent associated offence types were "other assault" and "sexual assault" and both were more likely to occur in Aboriginal female homicides.Footnote 17 "Other assault" was the precursor in 18% of Aboriginal female homicides (compared to 8% of non-Aboriginal female homicides); "sexual assault" was the precursor to 10% of Aboriginal female homicides (compared to 6% of non-Aboriginal female homicides).
It should be noted that only a small proportion of Aboriginal female homicides (approximately 2%) were described by investigators as linked to the drug trade or gang or organized crime activity.
Female homicide victims generally know the person who kills them – more than 90% had a previous relationship with them.Footnote 21 This is true for Aboriginal and non-Aboriginal female victims.
Figure 8 – Offender-to-victim relationship, female homicides, 1980-2012
Description of graph in tabular format follows.
Aboriginal female victims were most often murdered by an acquaintance (30% compared to 19%).Footnote 22 Breaking this down further, Aboriginal females were more likely to be murdered by a casual acquaintance (17% compared to 9%) or by someone with whom they had a criminal relationship (7% compared to 3%).
Spousal relationships (married, divorced, common-law and separated individuals) were also prominent, though Aboriginal female victims were less often killed by a current or former spouse (29% compared to 41%).
In cases of homicides where the offender and the victim are in a familial relationship, investigators are asked to note on the Homicide Survey whether they were aware of any previous history of violence between the two. More often, they cited a known history of previous family violence (which may or may not have been reported to police) between Aboriginal female victims and their offenders than their non-Aboriginal female counterparts (62% compared to 43%).
Offenders accused in both Aboriginal and non-Aboriginal female homicides were mostly male (89%) as opposed to female (11%). Their average age was 35 years old, with the youngest being a youth and the oldest being over ninety years old. Those accused of killing Aboriginal females, however, were slightly younger (30 compared to 36 years old).
Offenders accused of killing Aboriginal females were less frequently employed than those accused of killing non-Aboriginal females (41% compared to 26%). They were more often on some form of social assistance or disability insurance (24% compared to 10%). Those accused of killing Aboriginal females were less likely to be identified as making a living through illegal activities (5% compared to 8%).Footnote 23
Offenders accused of killing Aboriginal females were more likely to have a criminal record (71% compared to 45%). The most common serious conviction among offenders accused in both Aboriginal and non-Aboriginal homicides was related to a violent offence, yet those accused of killing Aboriginal females were more likely to have this type of offence on their criminal records (53% compared to 27%).
An offender consumed alcohol and/or drugs at the time of the incident in more than a third of female homicides (both Aboriginal and non-Aboriginal). Those accused of killing Aboriginal females were more likely to have consumed an intoxicating substance (71% compared to 31%).
Overall investigators indicated that 19% of offenders were likely to have, or be suspected of having, a mental or developmental disorder.Footnote 25 Perpetrators of Aboriginal female homicides, were less likely to have, or be suspected of having, a developmental disorder (10% compared to 20%).
Motive refers to the offender's purpose in killing the victim.Footnote 26
Between 2005 and 2012, the apparent motive was unknown in 14% of female homicides (possibly because a chargeable suspect has yet to be identified).
The most frequent motive in Aboriginal female homicides was "argument or quarrel" representing 40% of all incidents (compared to 23% for non-Aboriginal females). "Frustration, anger or despair" was the second most frequent motive identified in Aboriginal female homicides at 20% (compared to 30% for non-Aboriginal females).Footnote 27
Figure 9 – Offender motive, 2005-2012
Description of graph in tabular format follows.
|Argument or quarrel||40%||23%|
|Frustration, anger or despair||20%||30%|
|No apparent motive||7%||10%|
|Financial gain, settling accounts||6%||8%|
The data collected indicates that police solve almost 9 of every 10 female homicides, regardless of victim origin (88% for Aboriginal female homicides, 89% for non-Aboriginal female homicides). Other factors, such as victim involvement in certain occupations, may reduce the chance their murder will be solved.
The term "solve rate" or "solved" is used synonymously with clearance rate for the purposes of this report.Footnote 28
Description of graph in tabular format follows.
|Missing (unknown or
foul play suspected)
Figure 10 - Female Homicide Solve Rates, 1980-2012Footnote 29
Description of graph in tabular format follows.
Figure 11 – Unsolved Aboriginal Female Homicides and Unresolved Missing Aboriginal Females by province, all jurisdictions
Description of graph in tabular format follows.
Police-reported data indicates that solve rates are comparable between incidents involving Aboriginal and non-Aboriginal female victims of homicide. Overall, 5,770 of 6,551 incidents from 1980 to 2012 were solved.
In the data collected solve rates varied across the country, from as low as 80% (Nova Scotia) to as high as 100% (New Brunswick) for Aboriginal females, and as low as 84% (British Columbia) to as high as 100% (PEI, the Yukon and the Northwest Territories) for non-Aboriginal females. Solve rates for the provinces and territories are illustrated in Figure 10. These fluctuate significantly when the numbers involved are small, for example, in Atlantic Canada.
While solve rates remain similar between Aboriginal and non-Aboriginal female homicides, certain homicides appear to be solved less frequently than homicides overall. For example, homicides involving women who were reported to be employed as prostitutes were solved at a significantly lower rate than homicides overall; for Aboriginal victims in the sex trade, the solve rate was 60%, whereas for non-Aboriginals it was 65%. Similarly, homicides involving female victims employed in other illegal activitiesFootnote 30 were cleared at a lower rate than the overall; Aboriginal victims had their cases solved 65% of the time and non-Aboriginal victims 60%.
On average, homicides involving Aboriginal female victims are solved in roughly the same amount of time as ones involving non-Aboriginal female victims.Footnote 31 Where the data was provided, the average clearance time for Aboriginal female homicides was 224 days whereas the average for non-Aboriginal female homicides is 205 days.Footnote 32 The overall average time to solve was 212 days.
Criminal intelligence analysis of collected data will be necessary to ensure there are no previously unknown suspicious clusters of occurrences. While it is preliminary to draw any conclusions from the data, it appears that multi-agency task forces on missing or murdered women have been established in several of the highest volume areas of unsolved murdered and missing Aboriginal female incidents.
The total number of unsolved missing and murdered Aboriginal women occurrences as of the time of data collection was 225. This includes 105 cases of missing Aboriginal women categorized as "unknown" or "foul play suspected" and 120 unsolved Aboriginal female homicides. The RCMP is the police force of primary jurisdiction for about half of these cases (121), including 53 of the missing and 68 of the murdered.
Unresolved cases of missing and unsolved cases of murdered Aboriginal females stretch across the period studied.
Figure 12 – Unsolved Aboriginal Female Homicides (1980-2012) and unresolved missing Aboriginal females (according to CPIC as of Nov 4, 2013) by year of incident/disappearance, all jurisdictions
Description of graph in tabular format follows.
N.B.: Valid only as of data captured, may have been overtaken by events since.
Source: RCMP Revised Homicide Survey Dataset
Any discussion of victim characteristics is vulnerable to the accusation that blame is being assigned to the victim. There is no such intent here. There are certain factors that will make an individual more susceptible to violent victimization. The presence of these vulnerability factors in the cases of murdered Aboriginal women as opposed to the cases of murdered non-Aboriginal women may help provide some descriptive statistics to inform future social interventions or operational crime prevention planning. It should also emphasize the need to engage not just police tools, but broader response options (social services, health, education, etc.).
There is insufficient data relating to currently missing Aboriginal women from which to draw reliable conclusions with respect to risk factors contributing to their disappearance.
The data suggests differences between Aboriginal female homicide victims and non-Aboriginal female homicide victims in terms of employment status. Aboriginal victims were less likely to be employed than non-Aboriginal victims (16% versus 40%).Footnote 33 Aboriginal victims were more likely to support themselves through illegal means (18% versus 8%), be unemployed but still part of the labour force (12% versus 8%), and be on some form of social assistance or disability insurance (23% versus 9%). This is in line with data from the 2006 census about the Aboriginal population in Canada.
Another vulnerability factor that is more prevalent in the cases of murdered Aboriginal females is the consumption of drugs, alcohol or other intoxicants by the victim prior to the incident.Footnote 34 This information is generally obtained from the toxicology results on the Coroner's Report or from witness accounts of the event where available. From the data available between 1997 and 2012, Aboriginal females were more likely than non-Aboriginal females to have consumed some form of alcohol and/or drugs or other intoxicating substance prior to the incident (63% versus 20%).
The lack of a large difference in the prevalence of the involvement in the sex trade among victims is noteworthy. From 1991 to 2012, the survey identified 255 female homicide victims as being involved in the sex trade.Footnote 35 The percentage of Aboriginal female homicide victims involved was slightly higher than that of non-Aboriginal female homicide victims — 12% versus 5% respectively — which are both relatively small components of the available population.
As a result, it would be inappropriate to suggest any significant difference in the prevalence of sex trade workers among Aboriginal female homicide victims as compared to non-Aboriginal female homicide victims.
While the question was first introduced into the survey back in 1991, options for intoxicating substances — other than alcohol or drugs — were introduced in 1997. These other options were added to capture inhalants like glue, gasoline or other solvents. This information is generally obtained from the toxicology results on the Coroner's Report and from witness accounts of the event where available.
The completion of this report marks the beginning of the next phase of the RCMP's effort as a component of a more global response to the challenge of murdered and missing Aboriginal women. This effort will be sustained nationwide along four key lines:
RCMP Divisions have been provided with data from this project and have been directed to review all outstanding cases within their areas of responsibility to ensure all investigative avenues have been explored; and, ensure units responsible for missing and murdered cases are resourced sufficiently. Progress will be monitored at RCMP National Headquarters.
Of the unresolved cases of missing and murdered Aboriginal females, approximately 46% fall within the jurisdiction of other provincial or municipal police forces. The RCMP will provide partners with data relevant to their jurisdictions to assist in their efforts on these outstanding cases.
Additionally, the RCMP has dedicated resources to develop a National Missing Persons Strategy. The strategy will guide the RCMP's approach to missing persons cases and will increase the quality of investigations in the following ways:
The RCMP will use the data gathered to enhance its efforts at preventing Aboriginal females from going missing or being murdered.
Based in part on the data collected, the RCMP will identify communities with the highest risk of violence against women. Within these communities, intervention, diversion and family violence prevention initiatives will be focused towards at-risk individuals to maximize support and referrals to appropriate community treatment programs, such as the community-led, police-assisted program, Aboriginal Shield.
The RCMP will also work with other government departments and agencies to introduce and initiate crime prevention programs within these communities. For example, the RCMP will collaborate with Public Safety Canada and other federal partners to help these communities identify issues and mobilize resources through the Community Safety Plan process.
The RCMP will also track the progress of prevention and intervention initiatives through detachment performance plans to ensure appropriate accountability of local commanders.
To further discussions about policing issues involving Aboriginal women, the RCMP will maintain its dedicated NWAC liaison — an RCMP employee who works collaboratively with NWAC to develop public awareness tools and resources focused on the prevention of violence against Aboriginal women.
As a partner in the federal Family Violence Initiative, the RCMP's National Crime Prevention Services distributes funding to detachments to work with non-profit community organizations and all levels of government. This funding is used to support initiatives that address the issue of family violence as part of the Family Violence Investment Project. The RCMP will re-focus the funding for this project to target high-risk communities.
The RCMP will continue to work with stakeholders and partners to encourage the lawful and appropriate public release of data from this exercise.
Also, the RCMP will provide information collected during the project to Statistics Canada so that any new information can be included as part of its official holdings. Once this is completed, the information will be available for future publications or analysis via regular Statistics Canada disclosure mechanisms.
The list of outstanding murdered and missing Aboriginal females compiled for this project represents the most comprehensive list of police-reported murdered and missing data to date. It is important that it remains current because it provides a single common operating baseline from which agencies can assess and deploy response strategies.
To that end, the data on missing women has been shared with the National Centre for Missing Persons and Unidentified Remains (NCMPUR), which will work to publish these cases on its Canada's Missing website. This will greatly expand the number of cases that are in NCMPUR's holdings and that are publicly available on its website.
To continue to ensure there is solid data available for operational decision-making and to ensure RCMP members record the most relevant information possible for Statistics Canada, the RCMP will roll out changes to how it collects data on homicides and missing persons. As a result of this project, the RCMP will ensure that Aboriginal origin is captured as part of Homicide Surveys.
It is standard practice that all missing persons reported to police are entered into the Canadian Police Information Centre (CPIC) database, though policy and procedure may differ from one police service to another about what gets reported and when.
The figure of missing Aboriginal women in this report is derived from a file review of cases involving women in Canada who, on November 4, 2013, had been categorized on the CPIC database as missing for a period exceeding 30 days.
This limits the size of the data set described in this report. The missing women included herein are those who remained missing on November 4, 2013, and they are all deemed to be of Aboriginal origin.
Limiting the search was the first methodological step of the team assigned to this project. The second was to limit the subsequent file review to missing women whose CPIC entry categorized their "ethnicity" as "non-white" female (558) or "blank" (152) — a total of 710 records. (Aboriginal origin is a recent addition to the ethnicity field in CPIC.)
Of these cases, file reviews determined that 419 were non-Aboriginal females and 127 did not meet the parameters of the project (88 because they had been found despite remaining listed as missing on CPIC, 20 because they were confirmed to be deceased but remained listed as missing on CPIC, eight because they had yet to reach the 30 day mark, one because he was confirmed male but was listed as female on CPIC, and 10 because the file review could not confirm an ethnicity).
The remaining cases were assessed to be unresolved cases of missing Aboriginal women as of November 4, 2013.
The following were not included in this project: women missing for a period less than 30 days as of November 4, 2013; women who, since November 4, 2013, have become missing for more than 30 days; women who previously had been missing for a period exceeding 30 days but whose disappearance had been resolved; Aboriginal women missing for a period exceeding 30 days whose "ethnicity" field erroneously identifies them as "white;" and, women who were reported missing to a police service but never uploaded to CPIC. As a result, not all Aboriginal females who have — at one time — been reported to police as missing are included in this study.
A number of the missing Aboriginal females in the data set have a reported Date Of Birth that would now put their age past the average life expectancy of women in Canada. As they were listed on CPIC as missing on November 4, 2013, they remain part of data set collected as missing.
CPIC provides police officers with nine "probable cause" options to categorize why a person has gone missing: (1) abduction by a stranger; (2) accident; (3) wandered off/lost; (4) parental abduction with a custody order; (5) parental abduction — no custody order; (6) runaway; (7) unknown; (8) other; and, (9) presumed dead. This research did not identify any cases of "parental abduction" (either with or without a custody order) or "other." Researchers consolidated all categories of suspected violence as "foul play suspected."
As a result, the definitions of the probable cause categories used in this report are as follows:
The figure of 1,017 murdered Aboriginal women used in this report was compiled based on the Canadian Centre for Justice Statistics (CCJS) Homicide Survey data from 1980 to 2012 and follow-up review with police forces across Canada.
Every non-RCMP police service in Canada agreed to allow Statistics Canada to provide its data to the RCMP for the purposes of this project.
Information on the Homicide Survey and the methodology used to gather the data from police services can be found on the Statistics Canada website.Footnote 36 Homicide data reported to Statistics Canada only includes incidents that have been substantiated by investigators as an offence of culpable homicide and are a part of the official record — suspected homicides or deaths deemed suspicious are not a part of the analysis in this report.
The following were not included in this project: homicides that were never reported to Statistics Canada via the Homicide Survey; those homicides reported to Statistics Canada via the Homicide Survey where the victim was assessed by the investigator as not being of Aboriginal origin.
The project dataset includes both system-derived data (i.e. offline queries of CPIC) and project-specific research (i.e. file reviews). As a result, the final dataset must be understood to be an RCMP data holding, albeit one that is subject to significant third-party originator restrictions on further sharing.
The dataset was frozen in early April 2014. Developments on murdered cases described herein since that date are not included.
Collection of information about Aboriginal origin or identity is challenging for a number of reasons.
First, because the primary purpose for collecting this type of information in a police record is specifically to help identify an individual — whether as a potential victim, witness, person of interest or suspect — this can lend itself to a "perception-based" assessment by police.
In other words, it can correspond to how a police officer defines how an individual looks in terms of complexion and/or ancestry. Asking a police officer to judge a person's race based on his or her perception is difficult and can yield incomplete and inaccurate results. What a person looks like does not always reflect how s/he would self-identify.
Second, the use of the term "Aboriginal" as a descriptor has different definitions in the different data sources that make up this research project. For example, CPIC captures Aboriginal as an "ethnicity" whereas Statistics Canada's official position is that "Aboriginal" is not an ethnicity but rather an origin. Where possible, the above report attempts to use Statistics Canada compliant language.
Third, differences in police practice between agencies make it hard to create a data set that is comparable across jurisdictions. For example, in collecting data on homicides, some agencies use official Aboriginal "status" as the means to determine identity, others use officer discretion (as discussed above), and others rely on self-identification by individuals or their associations (family, friend etc.).
Fourth, historical police service (including the RCMP) adherence to jurisdictional and organizational policies has undermined the consistent collection and sharing of information on Aboriginal identity. This has meant a high number of Homicide Survey reports where the identity of the victim (and/or the accused) remained "unknown."
The designation of identity or origin in specific cases was made by the investigative agency in line with their policies and procedures and was not verified/adjusted by the RCMP without their input.
The RCMP approached Statistics Canada in November 2013 to obtain Homicide Survey information on female homicides between 1980 and 2012.
For Statistics Canada to share records with the RCMP in accordance with the Statistics Act, certain steps and conditions were necessary. Namely, all non-RCMP police services had to agree, in writing, for Statistics Canada to share its Homicide Survey data for the purposes of this project.
In addition, the RCMP was asked to sign an Undertaking of Confidentiality that outlines conditions to which the RCMP must abide. In December 2013, the Chief Statistician of Canada signed an order to disclose the records for all police services who signed the agreement.
This project did not definitively resolve the issue of data collection of the Aboriginal origin variable for the Homicide Survey, but it represents a first step towards this objective.
The file-by-file review brought the overall unknown/unavailable factor for the "Aboriginal origin" variable from 20% down to 1.5% nationally and established an origin descriptor for close to 1,200 victims whose origin descriptor had previously been unknown (over 250 as Aboriginal, the rest as non-Aboriginal).
In addition, as a result of the follow-ups, the reliability of the "solve rate" variable was enhanced due to the inclusion of cases that had been solved but had not been reported to Statistics Canada in the form of revised Homicide Surveys. As a result, approximately 40 homicide incidents were updated from the status of unsolved to solved.
Footnote 1WHO, Department of Reproductive Health and Research, London School of Hygiene and Tropical Medicine, South African Medical Research Council, (2013) Global and regional estimates of violence against women: Prevalence and health effects of intimate partner violence and non-partner sexual violence pg. 3.
Footnote 2See Native Women's Association of Canada. What Their Stories Tell Us: Research Findings from the Sisters in Spirit Initiative (2010). Internet; and Maryanne Pearce, An Awkward Silence: Missing and Murdered Vulnerable Women and the Canadian Justice System. (Diss.: University of Ottawa, 2013).
Footnote 3Shannon Brennan, Violent Victimization of Aboriginal Women in the Canadian Provinces, 2009 (PDF). Statistics Canada Catalogue no 85 002-x.
Footnote 4Statistics Canada (2011), "Aboriginal Peoples in Canada: First Nations Peoples, Metis, and Inuit" (PDF) National Household Survey.
Footnote 5Aboriginal female population based on census data and the 2011 Statistics Canada National Household Survey.
Footnote 6Statistics Canada (2013). "Aboriginal Peoples in Canada: First Nations People, Métis and Inuit". National Household Survey, 2011, Statistics Canada catalogue no. 99-011-X201100, 8-9.
Footnote 7The National Centre for Missing Persons and Unidentified Remains, Canada's Missing (2012). According to Canada's Missing, there were 61,096 missing person reports from CPIC in 2013, 85% of missing adults are removed from CPIC within a week of being reported missing because their case is resolved.
Footnote 8This extract captured all official police recorded data on CPIC of missing persons not bound by any start date but with an end date of October 5, 2013 (i.e. 30 days prior to November 4, 2013).
Footnote 9The figures in this chart are based on the analysis of an extract of CPIC for all Aboriginal females women listed as missing for more than 30 days as of November 4, 2013.
Footnote 10See appendix A for definitions.
Footnote 11The rate of victimization decreased 35% from 2.41 per 100,000 in 1980 to 1.56 per 100,000 in 2012. Source: Statistics Canada - CANSIM Table 253-001, Homicide Survey, number and rate per 100,000 population, Canada, province and territories. Retrieved April 15, 2014. Currently, the United States, Norway, Finland and Belgium have slightly higher homicide rates than Canada and rates in countries such as France, Australia, the United Kingdom, Ireland, Italy, the Netherlands, Sweden, Austria, Germany, Denmark, Switzerland and Japan are lower. See Jillian Boyce and Adam Carter, "Homicide in Canada, 2012" Canadian Centre for Justice Statistics, Statistics Canada, Statistics Canada Catalogue no. 85-002-x, p. 8.
Footnote 12These victims were associated to 6,233 case files (some incidents involved multiple victims). 5,991 offenders were identified as perpetrators of these cases (accused persons or charged/chargeable suspects).
Footnote 13For example, according to a Statistics Canada "between 1997 and 2000 the average homicide rate for Aboriginal people was 8.8 per 100,000 population, almost seven times higher than for that of non-Aboriginal people (1.3 per 100,000 population)." See Jodi-Anne Brzozowski, et. al. "Victimization and offending among the Aboriginal population in Canada". Canadian Centre for Justice Statistics, Statistics Canada Juristat, Statistics Canada Catalogue 85-002-XIE, vol. 26, no. 3, p. 1.
Footnote 14The term "residence" includes single homes, houses, townhouses, other residential dwelling units and apartment buildings.
Footnote 15This is captured by the associated or related offence being reported in the Homicide Survey administered by Statistics Canada. The question of associated or related offence is intended to capture whether an offence led to the homicide. To be captured, there must be a direct and immediate connection between the associated/related offence and the homicide itself, where the lesser offence gave rise to the homicide. For example, where a robbery occurs and ends in a homicide when the victim resists. This question was introduced to the Homicide Survey in 1991, and thus the analysis herein about associated/related offences represents the last 22 years of the reporting period only.
Footnote 16In 55% of all female homicides, no associated or related offence was established. The presence of an associated or related offence was unknown in an additional 13% of all female homicides.
Footnote 17"Other assault" offences includes common assault, assault with a weapon or causing bodily harm, aggravated assault, assault against police, peace, or public officers, criminal negligence causing bodily harm, and other assaults.
Footnote 18For the purposes of this report, the term offender represents those identified as the "charged" or "suspect chargeable" person on the Homicide Survey. In other words, it represents an individual against whom a charge of homicide has been laid by police, or recommended by police to the Crown. Alternatively, it may also be an identified suspect, against whom there was enough evidence to charge, but for reasons such as the suspect committed suicide/died, the incident was cleared otherwise.
Footnote 19The comparative percentages describing offender-to-victim relationships in this section reflect the victim's origin, and not the offender's.
Footnote 20The offender-to-victim relationship category in the Homicide Survey was significantly expanded in 1997, therefore these results are based on records from 1997 to 2012.
Footnote 21For the purpose of this report, offender-to-victim relationships were categorized as the following: spousal, other family, other intimate relationship, acquaintance and stranger.
Footnote 22The acquaintance category can be broken down further to include close friends, neighbours, authority figures, business relationships, criminal relationships and casual acquaintances. (i.e. a person known to the victim that does not fit in the other acquaintance categories).
Footnote 23"Employment status" only began to include "illegal activities" in 2005. As a result, analysis of this variable is limited to 2005-2012.
Footnote 24This may be based on information gathered during the investigation and may not be the result of an official medical diagnosis.
Footnote 25Can include mental disorders such as schizophrenia, depression, neurotic illness, psychotic and sociopathic tendencies, as well as developmental disorders such as fetal alcohol syndrome, mental disability and neurological injuries/illness.
Footnote 26Motive data prior to 2005 was problematic, likely coinciding with major revisions to the Homicide Survey and other administrative changes. For this reason, discussion of apparent motive was limited in this report to 2005 to 2012.
Footnote 27"Argument or quarrel" tends to be used to refer to an incident with an element of suddenness to the exchange, whereas "frustration, anger or despair" involves emotions that have built up for the offender over a longer period of time.
Footnote 28Clearance refers to whether or not a homicide incident was cleared: (1) either by the laying, or recommending of a charge to the Crown; or (2) where at least one suspect has been identified and against whom there is sufficient evidence to lay a charge, but where the incident is cleared otherwise (e.g. the suicide or death of the chargeable suspect is the most common reason for clearing otherwise in incidents of homicide).
Footnote 29Part of the Northwest Territories became the Territory of Nunavut in 1999. The lack of female victims identified as Aboriginal in Prince Edward Island, and the lack of non-Aboriginal female victims in Nunavut explains the missing bars.
Footnote 30This category involves a very small subset of the victim population for both Aboriginal and non-Aboriginal females. The employment status of the victims and offender is a data element within the Homicide Survey that was first introduced in 1991. However, it underwent some major revisions in format and in 2005 was expanded to include the option of selecting "illegal activities". For this reason, the analysis for this variable is limited to the eight year time frame (2005-2012).
Footnote 31See above for definition of "solved."
Footnote 32Of the 6,233 incidents (involving 6,551 female victims), approximately a third had no clearance date recorded.
Footnote 33 Employment status of the victim is collected within the Homicide Survey and was first introduced in 1991. However, it underwent some major revisions in format and in 2005 was expanded to include the option of selecting "illegal activities." For this reason, the analysis for this variable is limited to the eight year time frame (2005-2012).
Footnote 34The Homicide Survey collects information as to the consumption of alcohol and/or drugs prior to the incident by the victim.
Footnote 35Data about involvement in the sex trade was only captured as part of the Homicide Survey in 1991.
Footnote 36Statistics Canada - Homicide Survey