How Do We Know an Ethnography Is Good?

Questions of validity, that is, how we know that research findings are trustworthy, are constants in conversations about research methodology. Like all researchers, ethnographers must convince readers and reviewers that their work is credible; that events in the field are described accurately, and in scientific terms; and that the research findings are valid. Ethnographers define validity in particular ways. According to Roger Sanjek (1990), ethnographic validity is determined “according to three canons: theoretical candor, the ethnographer's path, and fieldnote evidence” (p. 395). In other words, validity-rich ethnography is transparent.

The ethnographer shows how her choices about fieldwork—from the initial broad net to the more selective and systematic data collection that follows—were guided by emerging theory. She also clearly describes her research path, specifying the actual network of informants and contacts with whom she engages. Ideally, information about the ethnographer's path includes the ethnography's size and range; demographic data about the informants (gender, occupation, age); and the path from one informant to the next. Ethnographic validity is strengthened by knowing who and how many people participated in the research, and the range of perspectives they brought to the inquiry. Finally, it is incumbent on an ethnographer to show the relationship between the ethnographic report and the field notes, that is, actual excerpts of the data, on which it is based. However, as Sanjek (1990) notes, “My own admiration of fieldnote-rich ethnographies is obvious, but the canon of fieldnote evidence requires only that the relationship between fieldnotes and ethnography be explicit. Ethnographic validity is served by, but does not require, extensive fieldnote documentation” (p. 403). For ethnographic writers the challenge is to include selected or filtered field notes that make the results credible and the ethnographer's arguments plausible.

Triangulation, the process of using data from different sources (for example, historical documents, interviews, informal conversations, observations) to support a conclusion, is important to establishing validity. As Fetterman (1989) wrote, triangulation “is at the heart of ethnographic validity, testing one source of information against another to strip away alternative explanations and attempt to prove an hypothesis” (p. 94). Triangulation corroborates the researcher's findings across sources or techniques, seeks convergence of information on a common finding or concept, and is useful in proving hypotheses or constructing models.

External reliability, a term typically coupled with validity in discussions on the rigor of research, refers to the extent to which a study and its results can be replicated by another researcher. Yet, as Margaret LeCompte and Judith Goetz (1982) point out, the ethnographic process is “personalistic; no ethnographer works just like another” (p. 36). That said, issues of reliability can be challenging for ethnographers. LeCompte and Goetz maintain that ethnographers enhance the external reliability of their data when they describe their own status position in fieldwork relations, identify and justify their choices of informants, recognize how social situations and conditions shape informants' disclosure, explicitly identify the assumptions that underlie analytic constructs and premises, and clearly present methods of data collection and analysis. Transparency in regard to process is again key here.

Internal reliability refers to the degree to which another researcher would agree with the ethnographer's data analysis and conclusions. Ethnographers do several things to reduce threats to internal reliability. They use low-inference descriptors, or terms that are as concrete and specific as possible, including verbatim quotes and narratives, in their field notes. When possible they do member checks by having their informants review their own interview transcripts for accuracy. They confirm the accuracy of field notes with local informants, and they verify their results through the review of their peers. And when ethnographers use devices that record and preserve their data, they have a data trail that can be reviewed by other researchers (LeCompte & Goetz, 1982).