Tips for formulating interview guides for semi-structured qualitative interviews (Part 2)

Developing interview guides goes hand-in-hand with reflecting on what you hope to learn. For any research study, you might start with two or research questions, and develop interview questions that will elicit information that will help you to examine the research questions posed. Here are a few more issues to think about as you develop an interview guide.

The relationship of research questions and interview questions. Remember that research questions are usually NOT the same as interview questions. Of all the theoretical perspectives one might use in a study, phenomenological research questions seem to most closely resemble an interview question. For example, to use Kathleen deMarrais’s example (deMarrais, 2004), “What are women teachers’ experiences of anger in school settings?” is the research question; the initial interview question is: “Think of a time when you experienced anger in a school setting, and tell me about that.”

Attributes of a population. Questions to do with attributes of a population (e.g., age, income, educational attainment, no. of years of work experience etc.) might be better examined via use of a “cover sheet” that the participant completes themselves after an interview has been conducted, as these sorts of questions do not typically generate conversation.

When standardized surveys might be a better fit. If you want to seek factual information, you may be better served by using a standardized survey. Qualitative interviews are useful for eliciting people’s descriptions of perceptions, events, experiences, opinions, beliefs, understandings and so forth. This is best done by asking questions in a way that elicits stories. That means that you need to:

• Avoid asking “closed” questions that may be answered yes/no.
• Work on developing interview questions that elicit “stories.” E.g., use “Tell me about….” “Tell me how you define….”
• Ask one question at a time. (i.e., avoid multiple-part questions).
• Avoid asking questions that have a prior assumption embedded in them (i.e. a potential answer or what you hope to find).

“Questions about factors that impact X”: You can includes these sorts of questions in a semi-structured interview. Before you do though, think about the evidence you are getting. Answers provided by interviewees will be their perception/s of “factors that impact” – that is, what people attribute to be the reason for X. It has long been known that people do not necessarily know everything about themselves, or perhaps believe things that are not actually “true.” Thus if you want to examine research questions about “what are the factors that [actually] impact X”, consider what other forms of evidence you might need to gain an understanding of those.

Knowledge questions. These can sometimes be threatening, since people think that they are being examined or evaluated in some way. On the other hand, you might want to know what their knowledge is about a topic. If that is the case, you could introduce this topic further; E.g.

• Tell me what you have noticed about X.
• What stands out to you with respect to Y?

Open-ended questions. Think about the formulation of open-ended questions, since sometimes these are too broad, and participants will need further guidance. It is useful to experiment with different wordings prior to an interview (try these out with your friends and family!). For example, “Tell me about your education” may be too broad of a question for a participant to answer. Thus, consider what you want to learn, and ask an open-ended question that is focused enough that participants know what you want to learn. For example, if you want to learn about a person’s educational history, you could start with: “Tell me about your high school years. What stands out?”

Fewer questions. If you have a long list of questions, consider drafting fewer questions, with potential probes that get at details. When asking follow up questions, the easiest and most effective way to do this is by using the participant’s words, e.g.,

• You mentioned X, describe an example of how that works.
• You talked about Y, what usually happens then?

If people have already answered a question, you might not have to ask the follow up questions you have on your interview guide. That is something you will have to judge on a case-by-case basis in the interview. You cannot draft follow up questions in advance, since you don’t know what people will say. However, I always remind myself of the sentence starters above, since they focus on what the participants say, rather than my own interpretation of what they have said.

Kathy Roulston

Reference

deMarrais, K. (2004). Qualitative interview studies: Learning through experience. In K. DeMarrais & S. D. Lapan (Eds.), Foundations for research: Methods of inquiry in education and the social sciences (pp. 51-68). Mahwah, NJ: Lawrence Erlbaum Associates.