Chapter 11: Introduction to Interviewing Techniques

Prepared for the Interdisciplinary and Global Studies Division,
Worcester Polytechnic Institute
by James K. Doyle
Associate Professor
Department of Social Science and Policy Studies

In Depth Qualitative Interviews

A. Selecting Participants

Since qualitative interviews are so time-consuming to conduct and analyze, only a limited number of people can be interviewed. Thus, participants should be carefully selected for their special expertise or experience, and it is important to identify and seek the participation of the people who will be the most informative and helpful rather than interviewing only those people who are the easiest to access. It is also important to keep in mind that the criteria for selecting potential interviewees may change as the study proceeds. For example, at the start of an interview study it is generally important to interview individuals who can present "the big picture" so that all of the important sub topics can be identified. At later stages of the process, it may be more appropriate to interview people who have more detailed or specialized knowledge rather than generalists. One of the best ways to find such people is to ask for recommendations from those early interviewees who have been particularly helpful.

The obvious drawback to this strategy of selecting a limited number of interviewees who have special characteristics is that it is not possible to ensure that the final list of interviewees is representative of the larger category of people from which they were drawn. It is therefore a good strategy to deliberately seek out the widest possible range of opinions and experience (a technique called "maximum variation sampling") so that your study cannot be criticized for only interviewing people whose thoughts conform to the researcher's expectations or hypotheses.

B. Deciding How Many Interviews to Conduct

Since qualitative interviews do not attempt to gain high enough numbers of participants to allow hypotheses to be tested for statistical significance, the appropriate number of interviews to conduct is determined by the subjective judgment of the researcher. What this means is, you stop interviewing when, in your opinion, you have obtained a complete understanding of your chosen topic. If in your last few interviews the participants simply repeated information you had already heard from others, it's probably time to stop.

C. Preparing to Interview

Since qualitative interviews are intended to be flexible and adaptive, they cannot be planned out in specific detail in advance. However, it is possible and, in fact, necessary to prepare in advance a list of important topics so that you can refer to them during the interview; otherwise, you may forget to cover them. It is also necessary to make informed decisions about how to ask or phrase questions related to these topics and about what level of detail you would like to achieve in the answers to them before you start interviewing.

Careful thought should also be given to how to organize the interviews. Although qualitative interviews are flexible, they should not be unorganized or random. Effective qualitative interviews are typically divided into three stages. In the first stage, the goal is to establish the background of the interviewee. You should try to get them to talk about themselves, describe the experience they have that is relevant to the topic at hand, and explain the history of how they came to be where they are today. In the second stage, the focus shifts to the details of their present experience that are relevant to your topic. You might ask them, for example, to "reconstruct a typical day" or "talk about their relationships with the people they work with." Finally, having reflected on their background and experience, they will be ready in the third stage to report on the meaning their experience has for them. In this third stage typical questions are "what have you concluded from all this?" or "what sense does this make to you?" This basic tripartite structure should be followed for each interview to ensure that participants render judgments and opinions only after careful reflection.

You should also have done your homework on the topics of the interview, which usually requires a thorough review of relevant literature. Qualitative interviewing requires you to have a good idea in advance of what information you need and to think fast in order to adjust your questions depending on what the interviewee says - these goals can only be achieved if you have gained a command of the relevant topic area and associated terminology.

Finally, novice interviewers are well-advised to conduct a "practice" interview or two before the study begins in earnest. Qualitative interviewing requires the interviewer to do several things effectively all at once, including listening intently, monitoring the progress of the interview, picking up nonverbal cues, and remembering what has already been said. It is a difficult task that can only be mastered through practice and experience.

D. The Interview Process

Qualitative interviews place several demands on the interviewer. For example, the interviewer must give the interviewee their rapt attention while taking notes, following an agenda, monitoring the time, paying attention to nonverbal cues, and making decisions about whether to probe for more detail or move on to another question, all at the same time. Good qualitative interviewing is a difficult skill that is best acquired through practice and experience, rather than reading about it. However, as you begin this process it might be helpful to pay attention to some of the collected wisdom of social science researchers. The following are some of the most important "pointers" for novice interviewers culled from the literature on qualitative interview techniques:

Qualitative interviewing requires concentrated, rapt attention, so that you can follow up on interesting comments, detect when interviewees are giving answers meant for public consumption rather than their true opinions, and monitor the progress of the interview. It is therefore important that the interviewer not dominate the interview by talking too much but instead focus on listening.
Yes/no or other questions that can be answered briefly are not appropriate for qualitative interviews. Instead, you should ask open-ended questions, that is, questions that require a detailed narrative answer. For example, you might ask interviewees to "walk you through a typical day" or "tell you a story about a particular event that illustrates what they've been talking about" or "trace a story back to its very beginning."
3. It is important to give interviewees appropriate clues as to how detailed you would like their answers to be. If you tell them how many topics you plan to cover in the allotted time frame, listen carefully, and ask for further detail when appropriate at the beginning of the interview, they will probably catch on.
The questions in a qualitative interview should logically follow from previous questions. Rarely is one question on a topic enough - you must be prepared to follow up whenever you're not sure about what the person is saying by asking for clarification, details, and examples.
It is important to avoid leading questions, that is, questions that because of the way they are worded or phrased indicate to the interviewee what sort of answer the researcher expects to hear.
As a general rule, the researcher should interrupt the interviewee as little as possible. Things they mention that cry out to be followed up on can be noted and brought back up at a later point. However, if the interviewee strays drastically off course, it is appropriate to guide them back to the topic at hand.
Qualitative interviewers have to learn to tolerate silence. It is important not to get impatient, but to give participants a chance to think about what they want to say. It is a natural human tendency to fill in pauses in conversation, and if the researcher can avoid doing so the interviewee will often fill the silence with more information.
The interviewer should try to strike a balance between being formal versus casual. If the interviewer behaves too formally, the participants may not open up to them.. However, if the interviewer behaves too casually, the participants may not take the interview seriously and may stray off topic.
Finally, it is important to leave the interview on a positive note and to get the interviewee to agree to be contacted again in the future if necessary. You might need, for example, to contact them at a later date to clarify something they said or even to ask new questions that were raised by subsequent interviews.

E. Reducing and Analyzing the Data

The goal of a qualitative survey is to develop explanations and theories that are carefully grounded in the evidence, that is, the interview data. These data must therefore be recorded and preserved in some form to allow the interviewer (and perhaps other members of the research team) to review and analyze them long after the interview is over. For a variety of reasons, tape recording and transcribing the interviews verbatim is far and away the best strategy for preserving interview data:

It is just too difficult to do all the things you need to do during a qualitative interview and take detailed notes as well.
It is important to keep data collection and analysis separate. Conclusions should not be drawn until all of the data are in, and taking notes by hand during the interview requires the researcher to make judgments about what's important and what's not during the interview.
Taking notes efficiently requires the researcher to edit and rewrite the interviewee's comments, which may result in their meaning being inadvertently altered.
Like any other research effort, it is desirable to keep the original data complete and intact so that both you and other interested researchers can refer to it if necessary.

If for some reason tape recording an interview is not possible, an alternative strategy is to have two researchers participate in the interview: one who focuses exclusively on conducting the interview and a second who takes over the task of taking detailed notes.

Recording and transcribing interviews results in an unwieldy amount of verbal data. It should therefore come as no surprise that one of the main purposes of qualitative interview analysis is to reduce the amount of data to a more manageable level. The goal is to identify and extract the most important, meaningful, and interesting parts of the interview text. This should begin as a process of "discovering" what's in the material, rather than starting out with definite hypotheses in mind. In qualitative studies this process of data reduction and analysis is subjective, and to convince readers of your report that your analysis is appropriate and defensible you will have to (a) demonstrate that it is based on a careful reading of the interview texts; (b) make it clear that you explored any data that were inconsistent or contradictory across subjects; and (c) report in detail on how you carried out your analysis so that readers can judge its appropriateness for themselves.

A good way to get started in qualitative data analysis is to read carefully through each interview and underline or circle those items that you think will prove to be the most important or meaningful. You should then develop a "profile" of each interviewee, that is, a summary of the interviewee's background, experience, and opinions. Whenever possible, interviewee's opinions should appear in these profiles in their own words, to avoid the possibility that rewriting them may change their substance or emphasis.

Once each individual interview has been summarized, attention generally turns to making comparisons across individuals. This requires the researcher to return to the interview transcripts and review them looking for similarities and differences and patterns and thematic connections in the data. Each coherent interview segment should be given a code number to indicate the concept, category, theme, or argument it relates to. These segments can then be copied and separated from the transcripts (with either real or electronic scissors) and put back together thematically to serve as the raw material for an analysis of the general findings of the study and their implications for policy or decision making. Data reduction can be achieved by noting redundancies in the data and discarding all but the most interesting and compelling statements concerning a particular issue or theme. In some cases it may be desirable to note quantitative aspects of the interview data, for example, the number of participants that express a particular concern or who shared a similar experience.

It is a good idea, when practical, to give interviewees the opportunity to review their transcribed interviews and the profiles and other sections of the project report that are based on them as soon as they are available. This will give the participants a chance to catch any errors in the material and to express their opinion as to whether the report accurately captures what they were trying to say. If they express confidence that their transcripts and profiles are accurate, this will in turn increase the confidence that readers of your report will have in your work.

Maintained by webmaster@wpi.edu
Last modified: November 09, 2006 12:01:37