Research Design and Sampling

Conclusion

The key to good research design is not to copy someone else’s idea of research, but to anticipate the kinds of data that are likely to help you resolve a particular research question, or evaluate a particular hypothesis, and then use or develop methods to acquire those kinds of data. Often this involves what some archaeologists call a “multi-stage” research design (Redman, 1973) in which researchers periodically reassess their progress, refine hypotheses or define new kinds of data with which to test them, and introduce new hypotheses that unexpected results may suggest. Other archaeologists characterize their research as cyclical, dialectical, or hermeneutic. While these approaches differ in detail and theoretical underpinnings, they share a cyclical or back-and-forth revision of hypotheses in light of new data and search for new data from the perspective of revised hypotheses (e.g., Gardin, 1980; Hodder, 1992; Johnsen and Olsen, 1992). The idea is that research is an iterative process that, to some researchers’ minds, gradually “homes in” on a meaningful understanding or a satisfying hypothesis, although perhaps never quite reaching it. For interpretive archaeologists, it homes in on one possible understanding, while other archaeologists with other interests or theoretical perspectives will approach quite different meanings or understandings, as long as they are not incompatible with the data (Hodder, 1992). Whether the data are from statistical samples or carefully selected observations, or are quantitative or not, depends entirely on the nature of the problem at a given stage in the research.