Several aspects of interview research heretofore receiving little attention are discussed. A brief description of the different types of interview formats and levels of analysis is presented. Following a discussion of the problem of analyzing protocol data, some suggestions are offered about analysis procedures that derive from constructionist assumtions. A model is offered of the interview which describes its role in hypothesis formulation and hypothesis testing. Views on how the interview can be used in combination with other research methods to investigate problem solving are discussed. Finally, how interview research is currently being reported is examined, and recommendations concerning the types of information necessary for inclusion in such reports are offered. Suggestions are aimed at encouraging the researcher to remain skeptical of interpetations of protocol data, and to report the results of interview research fitting specific criticism from the research community. ERIC RIE #ED-204 400.
The CAUSE Research Group is supported in part by a member initiative grant from the American Statistical Association’s Section on Statistics and Data Science Education