How to Ask The Questions in Questionnaires and Interviews

How to Ask The Questions in Questionnaires and Interviews

Written by Bruce W. Tuckman in Conducting Educational. New York: Harcourt Brace College Publishers. 1999. pp.238-239.
 

Certain forms of questions and certain response modes are commonly used in questionnaires and interviews. This section deals with question formats and the following section addresses response modes.

1. Direct Versus Indirect Questions
The difference between direct and indirect questions lies in how obviously the questions solicit specific information. A direct question, for instance, might ask someone whether or not she likes her job. An indirect question might ask what she thinks of her job or selected aspects of it, supporting the researcher’s attempt to build inferences from patterns of responses. By asking questions without obvious purposes, the indirect approach is the more likely of the two to engender frank and open responses. It may take a greater number of questions fo collect information relevant to a single point, though. 

2. Specific Versus Nonspecific Questions
A set of specific questions focuses on a particular object, person, or idea about which a researcher desires input regarding an attitude, belief, or concept; nonspecific questions probe more general areas.
For example, an interviewer can ask a factory worker (specifically) how he likes operating a lathe or (nonspecifically) how he likes operating machinery or working at manual tasks. An interviewer can ask a student (specifically) how much she likes a particular teacher versus (nonspecifically) how satisfied she feels with a particular class taught by the teacher. Specific questions,like direct ones, may cause  respondents fo become cautious or guarded and to give less-than-honest answers.Nonspecific questions may lead circuitously to the desired information while  provoking less alarm by the respondent.

3. Questions of Fact Versus Opinion
An interviewer may also choose between questions that ask respondents to provide facts and those that request opinions. A factual question might ask a respondent the type of car he or she owns or to specify marital status. An opinion question might ask about preference for Ford or Chevrolet models or reasons why (or why not) a respondent thinks that marriage contributes to a meaningful relationship between a man and a woman. Because the respondent may have a faulty memory or a conscious desire to create a particular impression, factual questions do not always elicit factual answers. Nor do opinion questions necessarily elicit honest opinions, because they are subject to distortions based on social desirability; that is, respondents may reply in ways that show themselves in the most socially acceptable light. With both fact and opinion questions, questionnaires and interviews may be structured and administered to minimize these sources of bias.

4. Questions Versus Statements
To gather input on many topics, an interviewer can either ask a respondent a direct question or provide a statement and ask for a response. To a question, a respondent provides an appropriate answer. For a statement, the respondent indicates whether he or she agrees or disagrees (or whether the statement is true or false). Applied in this manner, statements offer an alternative to questions as  way of obtaining information. In fact, attitude measurement instruments more commonly present statements than ask questions.
Consider an example:
l Do you think that the school day should be lengthened?
YES     NO
1. The school day should be shortened.
AGREE DISAGREE

These two formats are indistinguishable in their potentialfor eliciting honest responses. Usually, researchers choose between them on the basis of response mode, as discussed in the next section.

5. Predetermined Versus Response-Keyed Questions
Some questionnaires predetermine the number of questions to be answered; they require respondents to complete all items. Others are designed so that subsequent questions may or may not call for answers, depending upon responses to keyed questions.  For example, a keyed item may ask a respondent if he is a college graduate. If the response is no, the respondent isinstructed to skip the next question. The decision whetheror not to answer the question is keyed to the response tothe previous question. Consider another example of responsekeying. An interviewer asks a school superintendent if herdistrict is using a nationally known curriculum. Twopossible questions are keyed to the response. If the superintendent says that the district is using the curriculum, the next question asks about its effectiveness;if the superintendent says the district is not using the curriculum, the next question asks why.