Interviews are typically described as structured, semi-structured and unstructured (Merriam, 1988). Interviews may also be collaborative, or dialogical (McLeod, 1999).
Structured interviews. In this approach, the interview questions, and the order of asking them, is determined ahead of time. It is important that all questions are asked in the same way, in the same order, in each interview. In structured interviews, the interviewer takes the lead and controls the interview.
Structured interviews are commonly used when large numbers of people person is trained to conduct the interviews. Structured interviews are also useful for gathering demographic information.
Semi-structured interviews. These are guided by questions or issues that the researcher wants to explore, but the exact wording and order of questions is not necessarily determined ahead of time. In this way, researchers can make adjustments and respond to what comes out in the questions. In semi-structured interviews, the researcher sets the agenda but follows the lead of the person who is being interviewed.
Unstructured interviews. These are exploratory and are aimed at learning about a situation, sometimes in order to develop questions for further interviews. Unstructured interviews are often used along with observations in the early stages of research. They are similar to everyday conversations, except that they are for the purpose of doing research. You might have such informal interviews while working with students or program participants or observing their work, provided they have consented to participate in your research. For example, if you are trying out a new teaching strategy, you might use informal interviews or conversations to gauge how practitioners are using it (Mills, 2000; Olson, 2002).
Collaborative interviews. McLeod (1999) describes these "dialogical methods" as approaches in which the researcher and participant build meaning together. I was drawn to this idea, because in some interview situations I have entered into conversation or dialogue with the person I'm interviewing rather than sticking with my questions and listening. the interview will be "dialogical" rather than just letting it go that way. When I've planned semi-structured interviews with a number of people, and then gone off on tangents with individuals I'm interviewing, I've had some challenges pulling information together from across the interviews.