Interviews can take many forms and be more or less structured, depending on the researcher’s needs. Here an interviewer stops a passerby.
Janine Wiedel Photolibrary / Alamy Stock Photo
The use of interviews has a long history in sociology. One very early example is W. E. B. Du Bois’s ([1899] 1996) study of the “Philadelphia Negro.” A watershed in the history of interviewing in sociology was reached during World War II, when large-scale interview studies of members of the American military were conducted. Some of the data from those studies were reported in a landmark study, The American Soldier (Stouffer et al. 1949). More recently, Robert Wuthnow (2018) and his research assistants conducted over 1,000 in-depth interviews with individuals residing in rural communities to learn about how their norms, values, and local experiences are changing (see Trending box, Chapter 3).
Types of Interviews
The questions asked in an interview may be preselected and prestructured so that respondents must choose from sets of preselected answers such as agree and disagree. Or an interview may be more spontaneous, unstructured, and completely open-ended. The latter form is used by those who do observational research. An unstructured interview offers no preset answers; respondents are free to say anything they want.
Prestructured interviews are attractive when the researcher wants to avoid any unanticipated reactions or responses from those being studied. In a prestructured interview, the interviewer attempts to
Behave in the same way in each interview
Ask the same questions, using the same exact words, and in the same sequence
Ask closed-ended questions that the participant must answer by choosing from a set of preselected responses
Offer the same explanations when they are requested by respondents
Not show any kind of reaction to the answers, no matter what they might be
Interviews conducted in this way often yield information that, like data obtained from questionnaires, can be coded numerically and then analyzed statistically.
There are problems associated with prestructured interviews. First, interviewers often find it difficult to live up to the guidelines for such interviews:
They are frequently unable to avoid reacting to answers (especially unexpected or outrageous ones).
They may use different intonation from one interview to another.
They may change the wording, and even the order, of the questions asked (which can affect respondents’ answers).
Another problem is that respondents may not respond accurately or truthfully. For example, they may want to conceal things or give answers they believe the interviewer wants to hear. Finally, and most importantly, closed-ended questions limit the responses, possibly cutting off useful unanticipated information that might be provided in a more free-flowing interview.
The last problem is solved by the use of open-ended or unstructured interviews. The interviewer begins with only a general idea of the topics to be covered and the direction to be taken in the interview. The answers in unstructured interviews offer a good understanding of the respondents and what the issues under study mean to them. Such understandings and meanings are generally not obtained through structured interviews. However, unstructured interviews create problems of their own. For example, they may yield so much diverse information that it is hard to offer a coherent summary and interpretation of the results.
The Interview Process
Conducting interviews, especially those that are prestructured, usually involves several steps:
1 The interviewer must gain access to the setting being studied. This is relatively easy in some cases, such as when interviewing one’s friends in the student union or at a local bar. However, access would likely be much more difficult if one wanted to interview one’s friends in a sorority house or on the job. People might be less eager to talk to a researcher—to any outsider—in such settings.
2 The interviewer must often seek to locate a key informant (Brown, Bankston, and Forsyth 2013; Rieger 2007). This is a person who has intimate knowledge of the group being studied and is willing to talk openly to the researcher about the group. A key informant can help the researcher gain access to the larger group of respondents and verify information being provided by them.
3 The interviewer must seek to understand the language and culture of the people being interviewed. In some cases this is very easy. For example, it is not a great problem for an academic interviewer to understand the language and culture of college students. However, it is more difficult if the academician interviews people with their own, very different language and culture. Examples might include interviews with members of motorcycle gangs or prostitutes. In these kinds of cases, it is all too easy for the researcher to misunderstand or to impose incorrect meanings on the words of respondents.
4 The researcher must gain the trust of the respondents and develop a rapport with them. Establishing trust and rapport can be easy or difficult, depending on the characteristics of the researcher and the respondents. Well-educated and relatively powerful male researchers may intimidate less privileged female respondents. Older researchers may have trouble interviewing traditional-age college students. Depending on the field site, a researcher’s point of view and (perceived) similarities with the respondents may increase rapport.
Ask Yourself
Have you ever conducted or participated in an interview, perhaps for a job or as part of a study? How closely did it adhere to the guidelines mentioned here?
Survey Research
Survey research involves the collection of information from a population, or more usually a representative portion of a population, through the use of interviews and, most important, questionnaires. While some sociologists do their own surveys, most rely on data derived from surveys done by others, such as the U.S. government (the U.S. census, for example) and the National Opinion Research Center, which conducts various opinion polls.
Most survey research relies on questionnaires which are self-administered, written sets of questions. While the questions can be presented to respondents face-to-face, they are more often delivered to them by mail, asked over the telephone, or presented in a web-based format. Questionnaires are now increasingly being filled out on personal computers and over the phone (Snyder 2007).
Types of Surveys
There are two broad types of surveys. A descriptive survey is designed to gather accurate information about, for example, members of a certain group, people in a given geographic area, or people in a particular organization. The best-known descriptive surveys are those conducted by organizations such as