Appendix 1: Multivariate analyses

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.

Archived

This page has been archived on the Web.

[an error occurred while processing this directive]85f0033m[an error occurred while processing this directive] [an error occurred while processing this directive]

Many factors contribute to a person's risk of being the victim of a violent crime. Multivariate analysis can be used to find out what impact a variable has on risk of victimization by keeping other factors that also affect risk constant. In this case, since the dependent variable is dichotomous, i.e. it can only take two forms (being a victim or not being a victim), the logistic regression approach was used.

The risk factors were chosen based on what was observed by Gannon and Mihorean (2005). To those factors, we also added the region of residence since there are proportionally fewer persons with activity limitations in Quebec, while it is also in that province that the lowest victimization rates are observed.

The table below shows the Odds ratio for each risk factor compared to their reference group. When a variable's Odds ratio is equal to 1.0, that means that that variable has little or no effect on risk of victimization. When the ratio is greater than 1.0, that means that the presence of that characteristic increases the risk of victimization compared to the reference category and, inversely, when it is less than 1.0, that indicates that it reduces the risk of victimization.

For example, the Odds ratios from the first model indicate that persons with activity limitations have a 89% (1.89) higher risk than persons without limitations (reference group). Also, persons with household incomes over $100,000 have a 25% (0.75) lower risk of victimization than persons with household incomes under $15,000 (reference group).

Table 2 Logistic regression - odds ratio for all variables