Structured Interview

A structured interview is an interview in which the questions the interviewers are to ask and, in many cases, the answer categories the respondents are to use have been fully developed and put in an interview schedule before the interview begins. Structured interviews, sometimes also referred to as standardized interviews, are almost universally used in social surveys designed to produce statistical descriptions of populations.

Structured interviews can be contrasted with SEMISTRUCTURED INTERVIEWS and UNSTRUCTURED INTERVIEWS. When an interviewer is conducting a semistructured interview, he or she has a list of topics or issues to be covered, but the specific wording of the questions is left to the interviewer. In addition, the interviewer is free to ask about additional topics or issues that were not anticipated in the interview outline. In an unstructured interview, interviewers are free during the interview to decide what topics to ask about, as well as how to word the specific questions.

The use of structured interviews is essential if the goal of the interviews is to produce statistical data. The answers to questions cannot be tabulated meaningfully unless all respondents answered the same question.

Example: Do you approve or disapprove of the way the President of the United States is doing his job?

A question similar to this is used routinely to tabulate the approval rating of presidents. Samples of people are asked this question, and the percentage of people saying they “approve” is calculated. The percent approving is tabulated over time to get a comparative sense of how people are feeling about the president, compared to other times in his administration and compared to other presidents.

Suppose, however, that some people were asked not about the president but about whether they approved or disapproved of the way the administration was conducting its affairs. Further suppose that some people were asked not whether they approved or disapproved, but whether they supported or opposed what the president was doing. These changes in wording would mean that people were generally being asked the same question, but research has shown that small changes in the wording of questions or answer categories can have a definite effect on the answers that are obtained. In short, if there was this kind of variation in the wording of questions that people answered, the tabulation of their answers would not be interpretable.

The Evolution of Structured Interviews

In the early days of survey research, semistructured interviews were the norm. Interviewers were sent out with a list of topics, but they were free to devise their own questions to find out what people had to [Page 1095]say. Three considerations have led to the evolution of survey methods so that structured interviews are almost universally used:

1.
Studies showed that small changes in the wording of questions or in the alternatives offered had discernible effects on the answers people gave.
2.
When interviewers were free to decide on the wording of questions, not surprisingly they differed in the ways in which they worded questions to achieve the same objective. As a result, the answers obtained to surveys were often related to which interviewer did the interviewing.
3.
When respondents were allowed to answer in their own words, it often was difficult to compare what they had to say. For example, if one respondent described her health as “not bad,” while another described her health as “fairly good,” could we reliably say that the first person's health was better, worse, or about the same as that of the second person's? Researchers have found that they can tabulate answers more reliably, and compare the answers of different respondents, if they have respondents choose from the same set of response alternatives.

Thus, in order to ensure that the differences in answers people give stem from differences in what they have to say, rather than differences in the questions they were asked or the way they chose to phrase their answers, researchers now routinely use structured interviews, in which both the wording of the questions and the response alternatives are designed ahead of time and are the same for all respondents.

...