Questionnaires are used to collect data in MARKET RESEARCH. Whether used in MAIL SURVEYS and INTERNET SURVEYS or TELEPHONE SURVEYS and PERSONAL-INTERVIEW SURVEYS, questionnaire design is critical to successfully obtaining the information needed to achieve the research objectives. While there will be some variation depending on how the questionnaire is being administered, most questionnaire introductions
• state who is conducting the research
• introduce the surveyor
• indicate to respondents that the questionnaire is for a research study, not a sales effort
• describe the general topic of the research
• state approximately how long the interview will take
• assure individual respondents that their answers will be kept confidential
• state any honorarium or other incentive for participating
Generally market researchers try to use closeended questions rather than open-ended questions. Close-ended questions include dichotomous choice, yes or no type questions, and multiple-choice questions are easy for respondents to answer and easy to tabulate. Sometimes researchers use FOCUS GROUPS to help identify the likely responses to questions in advance of conducting a survey. Likert scale questions, ranging from “strongly agree” to “strongly disagree” or “poor” to “excellent,” are used to elicit more information about people’s feelings and opinions. The scale used and the number of choices in the scale can influence responses. Researchers have to decide in advance whether to include “neutral” or “no opinion” choices. For example, the first scale below forces respondents to agree or disagree, while the second allows respondents to not express an opinion.
• strongly agree / agree / disagree / strongly disagree
• strongly agree / agree / neither agree or disagree / disagree / strongly disagree
While researchers try to avoid using open-ended questions, the “other” choice in a multiple-choice question and the “Is there anything else you can tell us” question sometimes can be the most important source of information. Researchers are often so involved in their work that they cannot anticipate what consumers are thinking. Open-ended questions can be a valuable source of new ideas. Question design is a critical part of successful business research. Five types of errors are common in question design

• Double-barreled questions: How would you rate our
burgers and fries?
Excellent good fair poor

What if the respondent thought the burgers were excellent but the fries were poor? Most people would probably just circle “good,” and the researcher would never know there was a problem with the fries.

• Ambiguous wording: Do you eat at Fast Freddy’s regularly?
Yes No

What is “regularly”? For some people it might be once a week, for others once a day.

• Unanswerable question: When did you eat your first hamburger?

Most consumers cannot remember when. Questions asking people to recall events or experiences even just a week or two earlier are difficult to answer accurately.

• Missing alternatives: Where do you live? dormitory apartment own your own home

What if the person rents a home?

• Leading question: Why do you like Freddy’s fries?
hot spicy crisp salty

What if the respondent does not like Freddy’s fries?

Market researchers pretest questionnaires to avoid problems with question design. In addition, researchers create mock data results. Reviewing a set of hypothetical data allows researchers to evaluate whether the question will generate the information needed and whether the question is needed at all. Most first-draft questionnaires can usually be reduced in length by pretesting and doing mock data analysis. Market-research companies charge a design fee and a fee per question. Critical analysis during the questionnaire design stage can reduce the cost of a study.