Sample Size and Saturation: A Three-phase Method for Ethnographic Research with Multiple Qualitative Data Sources

Abstract

Despite recent methodological advances in saturation, guidelines for its estimation in more complex research designs—such as ethnographic studies—have been lacking. We present an accessible, step-by-step approach to empirical assessment of data saturation, tested on a moderately sized ethnographic study with 109 combined direct observations and interviews. The three-phase method includes a priori sample size prediction, provisional saturation estimation during data collection, and post hoc confirmation. Post hoc analysis indicated we reached key thresholds of data saturation before ending fieldwork. We achieved 80% saturation at our 69th data collection event (63% of the full sample) and 90% saturation at our 91st data collection event (83% of the full sample). The method is designed to account for on-the-ground realities confronted in contemporary ethnographic research, including time constraints during fieldwork, and can accommodate studies with medium to large scopes, broad foci, heterogeneous populations, and analysis of multiple qualitative data sources.