| Abstract: | Welfare policy is multidimensional because of the political compromises, competing goals, and federalist structure underpinning it. This complexity has hindered measurement and, therefore, the comparability of research on race and welfare policy. This paper describes a measurement strategy that is transparent, replicable, and attuned to matching the assumptions of statistical models to the policy process. We demonstrate that this strategy leads to more nuanced conclusions regarding the relationship between minority caseloads and the flexibility of state welfare policies. The strategy and recommendations are adaptable to research agendas that scholars bring to the comparative study of welfare in the U.S. states, countries, or other units—and to other complex policies enacted in federal systems. |
