Using Meta-ethnography to Synthesize Research: A Worked Example of the Relations between Personality and Software Team Processes

Context: The increase in the number of qualitative and mixed-methods research published in software engineering has created an opportunity for further knowledge generation through the synthesis of studies with similar aims. This is particularly true in the research on human aspects because the phenomena of interest are often better understood using qualitative research. However, the use of qualitative synthesis methods is not widespread and worked examples of their consistent application in software engineering are needed. Objective: To explore the use of meta-ethnography in the synthesis of empirical studies in software engineering through an example using studies about the relations between personality and software team processes. Methods: We applied the seven phases of meta-ethnography on a set of articles selected from a previously developed systematic review, to assess the appropriateness of meta-ethnography in this domain with respect to ease of use, and usefulness and reliability of results. Results: Common concepts were identified through reading and interpreting the studies. Then, second order translations were built and used to synthesize a model of the relationships between personality and software team processes. Conclusions: Meta-ethnography is adequate in the synthesis of empirical studies even in the context of mixed-methods studies. However, we believe that the method should not be used to synthesize studies that are too disparate to avoid the development of gross generalizations, which tend to be fruitless and are contrary to the central tenets of interpretive research.