Leora Courtney-Wolfman, Austrian Academy of Sciences
The study presents an interim hierarchical cluster analysis of 4869 of Canada’s 5162 census subdivisions (CSD), which will be used in a subsequent small-scale mortality analysis that pays special attention to Indigenous populations. Because of Canada’s large size and uneven spatial population distribution, existing subnational composites overshadow heterogeneity in age structure, mortality, and socioeconomic characteristics—particularly for groups with poorer health and mortality outcomes. Unlike similar composites, the model includes median age and household size as proxies for age structure. The initial results suggest four “Canadas”: Cities with high human capital, large shares of immigrants and visible minorities, and slightly younger age structures; remote, majority-Indigenous CSDs with high socioeconomic deprivation and a young age structure; two forms of rural, White CSDs. Of the rural CSDs, the first type reflects national population averages, while the second demonstrates socioeconomic deprivation and older populations. The findings also suggest that other models may spuriously group Indigenous, immigration, and visible minority status, while overlook more important spatial and socioeconomic distinctions. Likewise, the results show that age and household size can provide insights about disparities in highly developed countries like Canada, while reducing inaccurate groupings based on ethnicity and immigration status.
Keywords: Spatial dependence/heterogeneity, Census data, Population geography, Culture, ethnicity, race, religion and language
Presented in Session 197. Spatial Networks, Clusters and Accessibility