Sharpening causal reasoning in applied ethnographic research

Abstract

We present practical guidelines to help applied anthropologists and others use ethnographic methods to improve causal inferences in their research and practice. We do so by implementing principles from an ongoing “causal revolution” occurring in the social and behavioral sciences. We offer advice about how to collect and analyze ethnographic evidence to better approximate the logic of causal inference in randomized experiments, a strategy central to what that causal revolution has developed for other kinds of observational research practices. We highlight how ethnography is well-suited to developing causal explanations. For example, ethnographers can use observations and informal interviews to illuminate associations among a study’s key factors, including the temporal sequences connecting them, while also identifying how other factors might “confound” (or render non-causal) interpretation of those relationships. Overall, we argue that ethnographers can improve the certainty of their research conclusions, and the effectiveness of interventions based on them, by using ideas and strategies we offer here.