MAIN POINTS

The foundation of all questionnaires is the question. The major considerations involved in formulating questions are content, structure, format, and sequence.

Most questions can be classified into two general categories. Factual questions are designed to elicit objective information from respondents regarding their background, their environment, their habits, and so forth. The most common type of factual question is the background question. Opinion questions explore respondents' opinions (specific expressions of underlying attitudes) and attitudes (general orientations). Survey questions about opinions and attitudes present more problems in construction than questions about facts, partly because the former are more sensitive to changes in wording, emphasis, and sequence.

Three types of question structures can be distinguished. Closed ended questions offer respondents a set of answers from which they are asked to choose the one that most closely represents their views. These questions are easy to ask and quick to be answered; they require no writing by either respondent or interviewer, and their analysis is straightforward. Their major drawback is that they may introduce bias, either by forcing respondents to choose from given alternatives or by making respondents select alternatives that might not have otherwise occurred to them.

Open ended questions are not followed by any kind of specified choice, and respondents' answers are recorded in full; they are flexible, have possibilities of depth, enable interviewers to clear up misunderstandings, and encourage rapport. However, they are difficult to answer and still more difficult to analyze.

Contingency questions reflect a special case of closed ended questions; they apply only to a subgroup of respondents. Relevance of such questions to a particular subgroup is determined by a preceding filter question.

There are different techniques for structuring the response categories of closed ended questions. One of the most common of these formats is the rating scale, used whenever respondents are asked to make a judgment in terms of sets of ordered categories. Such scales measure the intensity of feelings toward something.

The matrix question is a method for organizing a large set of rating questions that have the same response categories.

Ranking is used in questionnaires whenever researchers want to obtain information regarding the degree of importance or the priorities that people assign to a set of attitudes or objects.

Two general patterns of question sequence have been found to be most appropriate for motivating respondents to cooperate. In the funnel sequence, each successive question is related to the previous question, and the questions get progressively narrower in scope. In the inverted funnel sequence, narrower questions are followed by broader ones.

Questions must be worded so that they are comprehended by respondents. Questions must be concise for open-ended questions and provide context to ensure the most desirable answers. Each question should be worded so that the respondent understands its meaning and so that the question has the same meaning to each respondent.

A response set is the tendency to answer all questions in a specific direction regardless of the question's content. This can impact the results of the survey if the answers to the questions all point towards a particular topic, especially if the subject is controversial.

Leading questions are those that are phrased in such a manner that it appears to respondents that the researcher expects a particular answer. Respondents also tend to agree with statements that support accepted norms or are perceived as socially desirable, and therefore might skew answers away from more truthful, but non-normative answers.

Threatening questions are those that respondents may find embarrassing and therefore difficult to answer. It has been determined that the reporting of certain behaviors decreases as questions increase in their degree of threat.

Double barreled questions include two or more questions in one, and this poses a problem if respondents feel differently about the issues involved.

An introduction must succeed in overcoming any resistance or prejudice that respondents may have against the survey. This document should: 1) identify the sponsoring organization and/or the persons conducting the study; 2) explain the purpose of the study; 3) tell why it is important that respondents answer the questionnaire; and 4) assure respondents that the information provided will be held in strict confidence. Cover letters for mail questionnaires and online surveys must be more detailed than introductory statements for personal and telephone interviews.

Instructions should be included at the beginning of questionnaires and should accompany any questions that are not self explanatory. Instructions for a personal or telephone interview can be delivered at the beginning of the interview, and any concerns can be answered during the interview. As mail and online questionnaires do not have any person-to-person interactions, instructions must be as clear, concise, and all-encompassing as possible, particularly for open-ended questions.