Structured interviews

A topic in research methodology

Interviews are sometimes classed as structured, unstructured or semi-structured.

The amount of structure appropriate depends upon the research questions being asked, and the methodology selected for the study.

"…if one was testing the hypothesis that 'girls are just as likely to experience bullying as boys', then one would wish to make sure that the same question was asked in the same way, so that the respondents' judgements were not influenced by different phrasing of the question, or differently cued by the question sequence. (Asking the question 'can bullying be verbal, or must it always be physical?' just before asking 'have you ever been the victim of bullying?' could well lead to a different response to the second question.) … In this situation…, a fully structured interview schedule would be appropriate"

(Taber, 2013: 275-276)

""structured interviews often produce quantitative data" (DiCicco-Bloom & Crabtree, 2006: 314)

Structured interviews are used when interviewing is used as a data collection technique within research studies using survey methodology.

Why interview in surveys (rather than just issue a questionnaire)?

Usually in surveys it would be more effective (and use much less resource) to use a paper or on-line questionnaire rather than use interviewers to read out the questions and note down responses. So, generally, interviews are not a sensible choice here. A major facility of interviews is the flexibility of the interviewer in interacting differently with different interviewees according to their responses.

"There would be little point in using the interview at all, if it simply resolved itself into a fixed list of stock questions put by the interviewer. For this would abandon a distinct merit of the interview in comparison with the questionnaire: the give-and-take which helps the interviewee decode and report the meanings which a situation held for him. It would mean the loss of that collaboration which encourages the interviewee to continue his self-exploration of an experience until some measure of clarity is obtained."

Merton, Fiske & Kendall, 1990

Not only is that flexibility not useful in surveys seeking quantitative results, it actually undermines them.

However there may be situations where interviews are a sensible choice in surveying rather than questionnaires:

Ability to clarify questions (but meant to be standard questions)

If the participant does not understand the question, or some of the vocabulary, the interviewer may be allowed to explain it.

In situ coding – can check response fits allowed category

The interviewer often has to assign the response to one of a number of predetermined response categories.

Response rate

Response rates to questionnaires are often low which can undermine confidence in the representativeness of those responding.

People may be more likely to answer questions verbally, than to complete a paper form or go on line – it may seem less effort, or they may prefer human interaction, or the presence of a human asking may encourage them to respond.

Control of sample make-up

If the sample has to be representative in terms of, e.g. age, gender, etc, then the interviewer can be selective in asking questions to those in categories with too few respondents. (Does this raise possibilities for biasing the sample?)

Allows participation

And, of course, there may be some groups who cannot access written questionnaires – such as very young children.

Sources cited:

• DiCicco-Bloom, B., & Crabtree, B. F. (2006). The qualitative research interview. Medical Education, 40, 314-321.
• Merton, R. K., Fiske, M., & Kendall, P. L. (1990). The Focused Interview. A manual of problems and procedures (2nd ed.). New York: The Free Press.
• Taber, K. S. (2013). Classroom-based Research and Evidence-based Practice: An introduction (2nd ed.). London: Sage.