Methodology

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.

Data sources

General Social Survey on Victimization

In 2004, Statistics Canada conducted the fourth victimization cycle of the General Social Survey (GSS). The previous cycles had been conducted in 1988, 1993 and 1999. The survey is designed to produce estimates of the extent to which persons are the victims of eight types of offences (assault, sexual assault, robbery, theft of personal property, breaking and entering, motor vehicle theft, theft of household property and vandalism); to examine the risk factors associated with victimization; to examine the rates of reporting to the police; and to evaluate the fear of crime and public perceptions of crime and the criminal justice system.

The GSS target population includes all non-institutionalized persons aged 15 and older. In 2004, the GSS sample consisted of 24,000 households in the provinces. Households were selected using random digit dialling, which yielded a response rate of 75%. The use of telephones for sample selection and data collection means that the 2004 GSS sample in the provinces represents only the 96% of the population that has telephone service.

Data limitations

The data that are used in this profile are based on estimates drawn from a sample of the Canadian population, and they are therefore subject to sampling error. The difference between the estimate obtained from a sample and the estimate based on the total population is sampling error.

This profile uses the coefficient of variation (CV) to measure sampling error. Any estimate with a high CV (more than 33.3%) was not published because it was too unreliable.

When we compare estimates to detect significant differences, we test the hypothesis that the difference between two estimates is zero. We construct a 95% confidence interval around this difference, and if the interval contains zero, we conclude that the difference is not significant. However, if the confidence interval does not contain zero, we conclude that there is a significant difference between the two estimates.

Additionally, non-sampling errors may also have been introduced. Non-sampling errors may include a respondent's refusal to report, a respondent's inability to remember or report events accurately, or errors in the coding or processing of the data. Also, people who could not speak English or French well enough to take part in the survey were not included. For these reasons, the data on victimization should be used with caution.