Studying Causal Mechanisms Using In-Depth Case Studies (QMMR D)
Tasha Fairfield
Half Day, 9:00 AM – 1:00 PM
Los Angeles Convention Center, 406B
The study of causal mechanisms (aka causal processes) is ubiquitous in the social sciences. The promise of process-focused research using in-depth case studies is that we can gain a better understanding of how things work and under what conditions using actual cases instead of controlled comparisons across cases using experimentally manipulating treatments to gain knowledge about mean causal effects. However, the potential gains of process-focused research have not been fully reaped in the social sciences because of the tendency to reduce causal processes to simple one-liners that do not unpack what is actually going on in a case (e.g. that grievances are linked to democratization through social mobilization). By not unpacking process theoretically, we are unable to evidence how they work empirically because empirical material is only processual evidence when we can identify the theorized part of a process that it is evidence of.
Inspired by the mechanistic turns in fields such as medicine, policy evaluation and policy studies (e.g. Clarke et al, 2014; Cartwright and Hardie, 2012; Cartwright, 2021; Capano, et al, 2019), the first session of the course discusses what ’good’ processual explanations can look like in the social sciences. The course introduces a conceptual language of actors, activities and linkages that enables us to move beyond one-liner theories to theorize the inner workings of causal processes, while at the same time not getting lost in the gory details.
The second session presents the developing standards in the natural and social sciences for what constitutes ‘good’ mechanistic/processual evidence, and how we can evaluate it. The final session discusses practical applications, including what and how we can ‘generalize’ from processual case studies, and how process-focused research can be used as an adjunct method to improve social science experiments in designing the experiment and interpreting the data.