Author ORCID Identifier
Matthew Ingram: https://orcid.org/0000-0001-6904-2178
Mixed-methods designs, especially those in which case selection is regression-based, have become popular across the social sciences. In this paper, we highlight why tools from spatial analysis—which have largely been overlooked in the mixed-methods literature—can be used for case selection and be particularly fruitful for theory development. We discuss two tools for integrating quantitative and qualitative analysis: (1) spatial autocorrelation in the outcome of interest; and (2) spatial autocorrelation in the residuals of a regression model. The case selection strategies presented here enable scholars to systematically use geography to learn more about their data and select cases that help identify scope conditions, evaluate the appropriate unit or level of analysis, examine causal mechanisms, and uncover previously omitted variables.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Ingram, Matthew C. and Harbers, Imke, "Spatial Tools for Case Selection: Using LISA Statistics to Design Mixed-Methods Research" (2019). Political Science Faculty Scholarship. 1.
This the publisher's PDF. The version of record can be found here: Ingram, M. C. and Harbers, I. (2020) “Spatial Tools for Case Selection: Using LISA Statistics to Design Mixed-Methods Research,” Political Science Research and Methods. Cambridge University Press, 8(4), pp. 747–763. doi: https://doi.org/10.1017/psrm.2019.3