This dissertation proposes a heuristic theoretical framework for understanding dynamics that impact environmental health including social/built environmental settings, individual residents' behavioral patterns, location activity spaces (LAS), environmental quality, exposure, and health outcomes. I examined the relationships between factors included in the framework based on individuals' LASs, and represent a hypothetical geographic boundary in which an individual is expected to spend his/her time in daily life. In addition to the individual level exposure, I characterized built environmental quality for subsidized housing neighborhoods in Los Angeles and Orange Counties, which have not been the focus of previous affordable housing studies. In Chapter 2 and Chapter 4, I empirically demonstrated the framework for residents in neighborhoods near the Expo Right Rail Transit line and the Boyle Heights community in Los Angeles. With OLS regression analysis, I found that bigger LAS were associated with lower walkability, more non-residential land use, higher transit stop density, shorter length of residency, working out of home, and higher income. I examined the relationship between the probability of a census block group (BG) having at least one subsidized unit and associated BG built environmental qualities. Based on logistic regression models, I found that subsidized housing units tended to be located in BGs with better transit access, lower walkability, more mixed-use, and lower air pollution concentrations.