Survey descriptions

Warning View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

[Full article in HTML] [Full article in PDF]

Incident-based Uniform Crime Reporting Survey

The Incident-based Uniform Crime Reporting (UCR2) Survey collects detailed information on criminal incidents that have come to the attention of, and have been substantiated by Canadian police services. Information includes characteristics pertaining to incidents (weapon, location), victims (age, sex, accused-victim relationships) and accused persons (age, sex). In 2011, data from police services covered 99% of the population of Canada.

The UCR2 Trend Database (2009 to 2011) represents 99% of police services in Canada. Analysis of this three-year trend database is limited to only those offences that have complete victim records and where UCR offence classification has remained constant over the three-year period. For the purpose of this Juristat article, the offences included in the trend analysis include attempted murder, physical assault (levels 1, 2, and 3) and sexual assault (levels 1, 2, and 3).

Homicide Survey

The Homicide Survey collects detailed information on all homicides that have come to the attention of, and have been substantiated by, Canadian police services. Information includes characteristics pertaining to incidents (weapon, location), victims (age, sex, accused-victim relationship), and accused persons (age, sex). Coverage for the Homicide Survey has represented 100% of the population since recording began in 1961. The count for a particular year represents all homicides reported in that year, regardless of when the death actually occurred.

General Social Survey on Victimization

In 2009, Statistics Canada conducted the victimization cycle of the General Social Survey (GSS) for the fifth time. Previous cycles were conducted in 1988, 1993, 1999 and 2004. The objectives of the survey are to provide estimates of Canadians' personal experiences of eight offence types, examine risk factors associated with victimization, examine reporting rates to police, measure the nature and extent of spousal violence, measure fear of crime and examine public perceptions of crime and the criminal justice system.


The target population included all persons 15 years and older in the 10 Canadian provinces, excluding full-time residents of institutions. The survey was also conducted in the three Canadian territories using a different sampling design and its results were published in a separate report in 2012. Households were selected by a telephone sampling method called Random Digit Dialling (RDD). Households without telephones or with only cellular phone service were excluded. These two groups combined represented approximately 9% of the target population (Residential Telephone Service Survey, (RTSS), December 2008). Therefore, the coverage for 2009 was 91%.

Once a household was contacted, an individual 15 years or older was randomly selected to respond to the survey. The sample in 2009 was approximately 19,500 households, a smaller sample than in 2004 (24,000).

Data collection

Data collection took place from February to November 2009 inclusively. The sample was evenly distributed over the 10 months to represent seasonal variation in the information. A standard questionnaire was administered by telephone using computer-assisted telephone interviewing (CATI). A typical interview lasted 45 minutes. Prior to collection, all GSS questions went through qualitative and pilot testing.

Response rates

Of the 31,510 households that were selected for the GSS Cycle 23 sample, 19,422 usable responses were obtained. This represents a response rate of 61.6%. Types of non-response included respondents who refused to participate, could not be reached, or could not speak English or French. Respondents in the sample were weighted so that their responses represent the non-institutionalized Canadian population aged 15 years or over, in the ten provinces. Each person who responded to the 2009 GSS represented roughly 1,400 people in the Canadian population aged 15 years and over.

Data limitations

As with any household survey, there are some data limitations. The results are based on a sample and are therefore subject to sampling error. Somewhat different results might have been obtained if the entire population had been surveyed. This Juristat article uses the coefficient of variation (CV) as a measure of the sampling error. Any estimate that has a high CV (over 33.3%) has not been published because the estimate is too unreliable. In these cases, the symbol 'F' is used in place of an estimate in the figures and data tables. An estimate that has a CV between 16.6 and 33.3 should be used with caution and the symbol 'E' is referenced with the estimate. Where descriptive statistics and cross-tabular analysis were used, statistically significant differences were determined using 95% confidence intervals.

Using the 2009 GSS sample design and sample size, an estimate of a given proportion of the total population, expressed as a percentage is expected to be within 0.95 percentage points of the true proportion 19 times out of 20.

Methodology for the multivariate analysis

Risk factors for victimization seldom exist in isolation. For example, being young is often associated with participating in evening activities, both of which can be risk factors for violent victimization. To evaluate the independent effect of each factor to the risk of victimization, logistic regression modeling techniques were used. By doing so, it was possible to identify factors that independently predict women's risk of self-reported violent victimization, even after controlling for the potential effects of other factors. 

Separate logistic regression models for women and men predicting spousal as well as non-spousal victimization were tested. For both spousal and non-spousal models, a 12 month reference period was used, since many risk factors, such as age, can change over time.Note 1

Transition Home Survey

The Transition Home Survey (THS) was developed under the federal government's Family Violence Initiative in consultation with provincial and territorial governments and transition home associations. Since 1993, the objectives of the survey have been to collect information on the characteristics of all residential services for abused women and their children during the previous 12 months of operation. In addition, the THS collects information on selected characteristics for the women residing in these shelters on a specified 'snapshot' day.

Included in the scope of the THS are shelters that serve a population broader than abused women, such as shelters admitting men as well as women and shelters admitting people for reasons other than abuse. Facilities that exclusively serve males and shelters that do not provide residential services are not included in the THS. The scope of the THS is limited to those facilities that come to the attention of Statistics Canada through its consultations with provincial and territorial governments and transition home and other associations.

Of the 593 residential facilities providing services to abused women and their children, 518 returned their questionnaires for a response rate of 87%. For those respondents who did not provide their information through the questionnaire, and for those respondents who did not answer some of the key questions on their survey forms, an imputation procedure was used to estimate the missing data.

Victim Services Survey

The Victim Services Survey (VSS) is funded by the Department of Justice Canada's Policy Centre for Victim Issues. It was developed in consultation with federal, provincial and territorial ministries responsible for justice and victim services, as well as a number of victim service providers from across Canada. The objectives of the survey are to provide a profile of victim service providers, information on the types of services offered and an overview of the clients who use them through a snapshot of clients on a given day. In addition, the survey collects standardized information from criminal injuries compensation and other financial benefit programs regarding applications for compensation and awards to victims of crime.

The VSS is a mail-out/mail-back paper questionnaire and is intended to be a census of victim service providers that fall within its scope. For administrative reasons, some head office locations submitted one form including data for all service locations under their administration. Of the 917 victim service providers and criminal injuries compensation programs eligible to respond, 389 sent forms representing data for 787 victim service providers and six criminal injuries compensation programs and other financial benefit programs. This resulted in a response rate of 86%.

The majority of service providers deemed ineligible to respond had either closed or were otherwise classified as outside the scope of the survey. An examination of provincial and territorial response rates showed that Newfoundland and Labrador, Prince Edward Island and Manitoba had a response rate of 100%. This was followed by New Brunswick (95%), Saskatchewan (92%), Ontario (88%), Alberta (88%), Quebec (87%), British Columbia (82%), Yukon (80%), Nova Scotia (76%), Nunavut (50%) and Northwest Territories (33%). The response rate for individual survey questions on which data in this article are based ranged from 86% to 100%.

In 2009/2010, the VSS questionnaire was modified to more accurately reflect the situation of victim services in Canada. The frame for the survey was updated for the 2009/2010 cycle of the VSS. Comparison with previous cycles is not possible for all variables.


  1. Sexual orientation was not included in the models since only those aged 18 years and older were asked about their sexual orientation, while other factors were assessed for those aged 15 years and over. While it would be possible to limit analysis to those aged 18 years and over, this would reduce the sample size.
Date modified: