Interviews and questionnaires serve the purpose of enabling people to report information about themselves--about their life condition, beliefs, or attitudes. In interviews, questions eliciting people's reports are asked orally. In questionnaires, the queries are printed and require written responses. Observations, in contrast to interviews and questionnaires, are people's reports about other individuals, whereas tests are reports of people's intellectual or physical performances.

The term life condition, in relation to interviews and questionnaires, refers to characteristics of individuals that identify their status in terms of gender, age, place of residence, vocation, income, education, religious affiliation, ethnic background, and the like. Such information is typically used to place respondents in categories that are easily compared, on the assumption that the categories may be associated with the study's target variables. In a project focusing on attitudes toward birth control methods, the investigator may estimate that females' and males' attitudes could differ significantly. Thus, it is important to know the gender of each respondent. But if there is no reason to suspect that a particular aspect of the respondents' status--gender, age, religious affiliation, or such--might be correlated with a target variable, there is no good reason to include information about that aspect in an interview or a questionnaire.

The word beliefs refers to respondents' knowledge and convictions about a topic--what occurred during a city council meeting, typical child-rearing practices of an ethnic group, children's rights, immigration laws, religious doctrine, or the average cost of attending college.

Attitudes are underlying tendencies for people to act in certain ways. The tendencies derive from the individuals' collection of values, which can be of various kinds, including those bearing on moral behavior, etiquette, fair play, human rights, financial responsibility, job efficiency, artistic taste, protecting the environment, and more. A researcher asks for opinions on the assumption that information about people's preferences can help explain and predict their behavior in decision-making situations.

Despite the similar aims of interviews and questionnaires, the two approaches are sufficiently different to warrant our discussing them separately. This section treats interviews; the next section treats questionnaires.

Alternative Interview Strategies

Interviews are often employed in case studies, ethnographic research, biographies, and surveys. Their use in historical studies and experiments is less frequent.

Researchers with little experience planning interviews are often prone to devise their interview questions in a rather haphazard fashion, when they would be better advised to design the questions to fit an intentional strategy. The forms and purposes of different strategies can be illustrated with examples of four types labeled loose, tight, converging, and response-guided.

Loose-question strategy

The aim of a loose or broad question approach is to reveal the variable ways respondents interpret a general question. Consider, for instance, a proposed study of religious beliefs that includes this query:

What diverse meanings do people attach to words and phrases that relate to religions; and what implications does such diversity hold for people's behavior toward others (with "others" possibly including both humans and supernatural beings or spirits)?

Because the purpose of the study is to expose the extent of variability among individuals' interpretations, the interviewer plans to pose questions in a very general form, offering interviewees unrestricted freedom to tell their interpretation of a particular word or phrase.

In pursuing a loose strategy, interviewers resist respondents' attempts to have questions rephrased in greater detail, since the intent of the approach is to expose the variety of interpretations to which different respondents subscribe.

Tight-question strategy

A tight-question approach is sometimes enriched by the interviewer asking respondents to support their decisions with reasons for selecting the answer they chose.

One of the appealing advantages of a tight-question approach is the ease with which the results of the study can be compiled. The researchees job of organizing the answers merely requires that the percentage of people selecting each option be reported. In contrast, organizing the answers to open-ended questions (including respondents' rationales) is often a complex, demanding task.

Converging-question strategy

A converging approach is intended to incorporate the advantages of both the loose and tight strategies. The interviewer first asks broad, open-ended questions to discover what seems uppermost in the respondent's mind in relation to the topic at hand. Then, following the respondent's answer, the interviewer asks one or more limited-choice questions. The label converging-question strategy refers to a funnel-like approach with broad queries followed by one or more sharply focused questions.

By starting with broad questions, the interviewer optimizes the likelihood of eliciting diverse opinions. If the process were conducted in reverse, with specific multiple-choice options (yes/no on the death penalty or on affirmative-action policies) posed first, followed by general open-ended questions, respondents' answers to the open-ended queries might be influenced by the options suggested in the multiple-choice phase.

Response-guided strategy

A response-guided approach consists of the interviewer beginning with a prepared question, then spontaneously creating follow-up queries relating to the interviewee's answer to the opening question. This technique enables the researcher to investigate in some depth the respondent's detailed comprehension of issues related to the initial question. Perhaps the best known version of such a strategy is the clinical method popularized by the Swiss child psychologist, Jean Piaget ( Inhelder & Piaget, 1964). Piaget defended his deviation from using a single, standard set of questions by explaining that all children do not interpret a given question in the same way. Thus, the experimenter probes the child's understanding and may then cast the problem in a different form to help ensure that the problem situation is the same for each child, even though the wording may not be identical each time. In effect, the child's initial answer guides the interviewer in devising additional questions to pose.

A typical interview of this type is illustrated in the following passage in which eight-year-old Per was asked about some flowers--primulas (primroses) and other varieties--that the interviewer placed before the child. The interviewer's purpose was to discover how Per classified objects into a general set (flowers) and into subsets within the general set (primulas, violets, tulips). At the point we enter the discussion, Per has already responded to the initial question that asked her to order the flowers into three levels of classes: yellow primulas, primulas, and flowers (adapted from Inhelder & Piaget, 1964, p. 107).

Interviewer: Can one put a primula in the box of flowers (without changing the label)?

Per: Yes, a primula is also a flower.

Interviewer: Can I put one of these flowers (a tulip) in the box of primulas?

Per: Yes, it's a flower like the primula.

Interviewer: Suppose I remove all the primulas, will there be any flowers left?

Per: Oh, yes, there will still be violets, tulips, and other flowers.

Interviewer: Well, suppose I pick all the flowers, will there be any primulas left?

Per: No, primulas are flowers. You're picking them, too.

Per: The same number. Primulas are flowers.

Interviewer. Count the primulas.

Per: Four.

Interviewer: And the flowers?

Per: Seven.

Interviewer: Are they the same number?

Per (astonished): The flowers are more.

It's apparent that the experimenter in this example not only was interested in gathering information about Per's reasoning processes, but also had a didactic aim in mind--that of advancing Per's command of logic by confronting her with inconsistencies resulting from her initial mode of classifying the flowers.

As the foregoing examples of strategies have suggested, it is important for researchers to design their interview techniques carefully to suit the particular aims of the research project at hand.

Advantages of Interviews

It should be apparent that many interview questions could be presented to respondents in questionnaire form rather than as part of a personal conversation. Distributing questionnaires to a group of participants enables a researcher to save the time that interviewing would require. In addition, a far larger number of people can participate in a questionnaire survey than would be possible through individual interviews. Nevertheless, substantial advantages that interviews provide make interviewing the preferred data-gathering technique for certain kinds of research.An investigator's taking the time and trouble to conduct personal interviews rather than simply passing out questionnaires to a classroom of students or sending forms through the mail suggests to respondents that the researcher particularly values their opinions. This display of sincere interest in respondents' views can enhance the diligence and care with which interviewees answer questions. Furthermore, the interview setting enables a researcher to clarify items that participants may find confusing. Interviews also make it easy for participants to amplify their answers or to digress from the central topic in ways that prove useful to the investigator. Furthermore, interviews can provide an indepth understanding of a respondent's motives, pattern of reasoning, and emotional reactions that is not possible with questionnaires.

Interview Resources