Dissertation chapters or working papers are available upon request prior to publication.
Social and health researchers are increasingly interested in the ways that macro-societal structures can affect human health and shape population distributions of health and well-being. Among health researchers, this interest in social structures has led to debates over whether the current ‘causal inference’ paradigm, which draws from the clinical trial setting and empiricist philosophy, is able to provide a fruitful methodology for the study of health inequalities. My dissertation research was motivated by the need for new frameworks that can promote rigorous and insightful research on social structures and health, particularly for quantitative research in health geography.
To that end, the dissertation brings realist theories of social science into conversation with plausible reasoning (PR), a tradition of work that aims to develop a logic of probable inference. Whereas realist theories of science offer a general causal ontology that can make sense of social structures, questions of epistemology have been relatively neglected. I argue that realism requires epistemological principles that can be stated with relative independence of realism’s core ontological claims. A similar position was previously articulated by Harold Jeffreys, a twentieth-century theorist of probability and proponent of what he called (perhaps as early as 1939) `critical realism’. The objective of this research is to transpose insights from PR into principles that can help advance research methodology in the social sciences. I argue that PR can be deployed heuristically to support the development of non-experimental study designs, to assess and debate findings, and develop explanations.
The main case study of the dissertation applies principles of PR to structure an investigation into the evolution of the colorectal cancer (CRC) burden in the Dallas-Fort Worth and Houston metropolitan regions. The initial motivation for the study was a persistent and substantial black-white CRC inequality. A growing literature in social epidemiology and health geography now draws heavily from the ‘neighborhood effects’ paradigm to explain health inequalities. This case study employs principles from PR to formulate a series of research questions that stress-test such explanations with respect to CRC in urban Texas. The research involves three inter-connected components: description of the expansive and polarizing urbanization processes reshaping these regions and their neighborhoods, a multivariate disease mapping analysis, and a test of the conjecture that the vast majority of the black-white racial disparity is accounted for by the excess disease burden inside historically ghettoized African-American neighborhoods. The findings highlight important limitations of the neighborhood effects paradigm for explanation in health geography—particularly with respect to urbanization processes and their consequences for disease burdens—and argues that PR can help support geographical research about macro-societal structures and health inequalities.
Connor Donegan. Plausible Reasoning and Heuristic Methodology in Human Geography: An Investigation of Colorectal Cancer Incidence and Inequalities in Urban Texas, 1999-2019. Doctoral dissertation, The University of Texas at Dallas. Dallas, Texas, 2023.