This project will examine the hypothesis that provinces experiencing greater housing unaffordability will associate with higher rates of child vulnerability, after controlling for other provincial characteristics. Housing unaffordability will be conceptualized as the relationship between average provincial home prices relative to local earnings.
Outcome variable: The Early Development Instrument (EDI) is well-established as a population measure of child development for Kindergarten children. The EDI measures vulnerability according to five domains -- physical, social, emotional, language/cognitive, and general communication.
Sample: There have been 30 population-level provincial EDI data collections completed across the 10 provinces between 2004 and 2019. Our study will examine the relationship between home prices, local earnings and EDI vulnerability rates across all Canadian provinces, in different years, with a focus on BC, Ontario and Quebec.
Input variables: Beyond housing and earnings, we will control for a range of socioeconomic variables that are theoretically relevant to child development, including GDP, unemployment, female labour force participation, poverty, child care fees, the share of the population that is indigenous and subject to the harmful legacy of colonization, etc.
Analytic approach: We will use a diversity of statistical techniques, including a series of “best subset” linear regressions, to identify the subset of input measures which account for the largest amount of EDI vulnerability between provinces while maintaining a theoretically coherent and reproducible pan-Canadian model of inter-provincial differences and intra-provincial trends over time. The project’s hypothesis will be supported if indicators of housing unaffordability emerge as important inputs in the statistical model.