Accessibility has become one of the predominant ways of understanding the relationship between transportation and land use in urban areas. Traditional measures of accessibility understand it unimodally or comparatively, without consideration of the dynamics of a multimodal transportation system. Multimodal, or mode share weighted accessibility (MWA) measures, take into account observed mode shares of the underlying geographic units and apply them to the accessibility to employment provided by that mode share. The individual MWA values are then added to give a singular MWA value. In this research MWA models are created for over 20 Canadian census metropolitan areas. They’re presented at regional and census tract levels, where the latter are then used in regression models to understand correlations that exist between MWA and socioeconomic and demographic factors. Inferential statistics are used to estimate differences in means of the socioeconomic and demographic variables of the top and bottom quintiles of MWA in every region. Many of the socioeconomic factors were found to be significantly corelated with MWA, with higher MWA values being associated with higher median household incomes, lower proportions of renters, and typically lower population density and lower proportions of visible minorities and immigrants. This is the first study to use multimodal accessibility models to understand the relationships between accessibility and socioeconomic factors across large- and medium-sized metropolitan regions in Canada.