Decide on wording MR questions

A number of other important ideas, however, should be considered on question content and types of questions and much has been said on question wording. Unfortunately, these ideas are more rules of thumb that have been developed from experience as they are underlying concepts.

Define the Issue: Beginning newspaper reporters are admonished to include in their lead paragraphs the six points: who, where, when, what, why and how. This can also serve as a guide to the researcher preparing a questionnaire. Each question should be checked against these points to be sure that the issue is clear. Who, what, where and when are particularly important. The why and how may be applicable in some questions.

Take the question: What brand of cigarettes do you smoke? Who? In this case the ‘you seem clear, but in some cases this word may leave confusion as to whether it applies to the respondents, their families their companies, or some other plural application. What? Brand of cigarettes and smoke are stated. The word brand may be slightly confusing but otherwise the ‘what’ in this question is clear. When? This is not clear. Does the question mean usually, always last time ever or what? The question makes the fundamental error of assuming something. It assumes that all respondents smoke one band of cigarettes to the exclusion of all others. Of course this is not true. A better way to get the same information might then be to ask: What brand of cigarettes did you buy the last time you bought cigarettes? This makes the ‘when’ specific. In this question the where, how, and why are not applicable.

Should the question be subjective or objective? Many questions can be stated in either subjective or objective form. Do you think the Ford is a better car than Chevrolet? Is the Ford better car than the Chevrolet? The first phrasing is subjective; it puts the question in terms of the individual and is apt to elicit a response in terms of the individual’s feeling. The second phrasing tends to cause the respondent to think more in terms of what people in general think. One study of the effect of subjective-objective statements used questions of the following type: Did you see a demonstration of Foley Kitchen Utensils in the house wares department? Was there a demonstration of Foley Kitchen Utensils in the house wares department? The results of this study indicated that subjective rather than objective questions tended to give more reliable results. Researchers have no available rules to follow in deciding whether to make their questions subjective or objective. They must be aware, however of the fact that the choice will influence their results.

Positive or negative statement: In a survey to determine the attitudes of executives towards advertising, each questions was worded in two different ways – one positively and one negatively. In an interspersed way half the respondents received the positive wording and half the negative wording Several issues were presented in the question, and respondents were to indicate one of five alternative reactions to each statement; agree generally agree, partially agree, can’t say, disagree, partially disagree, generally disagree. Presumably one who agreed with a favorable statement of one issue would disagree with an unfavorable statement of the same issue.

This study used positive and negative statements alternatively to average out the effect of each wording. Another approach would have been to state both alternatives the positive and the negative in each question, as follows: Do you think advertising should be increased or decreased to hasten recovery in a recession?

Use Simple Words: Words used in questionnaires should be words with only one meaning, a meaning known by everyone. Unfortunately, it is not easy to find such words. Many ordinary words have different meanings listed in the dictionary and even other meanings among certain groups of the population or in certain sections of the country. What is ‘soda pop’ in some sections is ‘tonic’ in others and ‘root beer’ and ‘sarsaparilla’ are the same things.