The Main Predictive and Intersecting Effects of Neighbourhood Income, Ethnicity, and Gender on Arrest in Canada

Background and Purpose: Indigenous Peoples, especially Indigenous females, are overrepresented across Canada’s criminal legal system, including contacts with police. Evidence suggests the extensive structural violence, including colonialism, Indigenous Peoples have faced contributes to their being disproportionately impacted by racial and social profiling by police (Ontario Human Rights Commission [OHRC], 2019). Although racial profiling occurs based on race, social profiling is based on other indicators related to social status, such as poverty (OHRC, 2019). Both racial and social profiling lead to over-policing, and arguably over-arresting, and thus, potentially overrepresentation across the criminal legal system. Demonstrating evidence of the intersection of gender-based and racial/ethnic profiling, recently, Alberton et al. (2023) found that Indigenous females living in Indigenous enclaves across Canada were at the highest risk for being arrested. However, no Canadian studies to date have examined the effects of ethnicity, gender, and neighborhood income on arrest.

Using the 34^th iteration of Canada’s General Social Survey and aggregated Canadian Census data, the following hypotheses were tested: (1) People who have experienced extensive structural violence will be more likely to have been arrested than those who had no such experiences. (2) People living in low-income neighbourhoods will be more likely to have been arrested than those living in the least impoverished neighbourhoods. (3) Indigenous Peoples living in low-income neighbourhoods will be more likely to be arrested, and this Indigenous disadvantage will be greater than that observed in other places. And (4) Indigenous females living in low-income neighborhoods will be most likely to have been arrested.

Methods: Bivariate analyses tested the magnitude and significance of between group differences and logistic regression models were built to test the main predictive, and two- and three-way interaction effects of individual- and structural-level factors on arrest in the past 12 months.

Results: The second strongest predictor of being arrested in the full model was extensive experiences of structural violence (OR = 2.83). The main predictive effect of neighbourhood income was not significant. However, the two-way (neighborhood income by ethnicity), and the three-way (neighbourhood income by ethnicity by gender) interaction effects were practically and statistically significant. Indigenous Peoples living in the low-income neighbourhoods were more than three times as likely (OR = 3.22) as Indigenous Peoples living in higher income neighborhoods to have been arrested. This neighbourhood income effect was not significant for White people. This risk was most profound for Indigenous females living in low-income neighborhoods (OR = 4.25). Although just approaching statistical significance (p = .08), a heightened risk for Indigenous men living in low-income neighbourhoods was also evident (OR = 2.90).

Conclusions and Implications: The current study provides evidence that profiling by police may be occurring at the intersection of ethnicity, gender, and social location; with the risk of arrest being especially profound for Indigenous females living in low-income neighborhoods. Addressing the overrepresentation of Indigenous Peoples across Canada’s criminal legal system will require collaborative and intersectional strategies to be implemented across all systems and sectors.