The questionnaire is a structured technique for collecting primary data in a marketing survey. It is a series of written or verbal questions for which the respondent provides answers. A well-designed questionnaire motivates the respondent to provide complete and accurate information.
The survey questionnaire should not be viewed as a stand-alone tool. Along with the questionnaire there is field work, rewards for the respondents, and communication aids, all of which are important components of the questionnaire process.
Although there are no set rules for developing a flawless questionnaire, the collective experience of numerous researchers offer a broad set of guidelines for minimising the likelihood and the severity of data validity problems in designing questionnaires. However, some Do’s and Don’t’s need to be followed, viz.
- Do not ask question unless a truthful answer will be useful to decision making
- Should ask one question in different way to obtain a right answer.
The following steps to be helpful during questionnaire design.
DETERMINE THE INFORMATION REQUIRED
Since the questionnaire is the link between the information needs and the data to be collected, the researcher must have a detailed listing of the information needs as well as a clear identification of the respondent group. This step is normally the result of exploratory research and of the hypotheses development phases. The different forms of market response described in the preceding chapter will help the analyst to identify the concepts to be measured.
DETERMINE THE TYPE OF QUESTIONNAIRE TO BE USED
Data collection can be made by personal interview, mail or telephone. The choice among these alternatives is largely determined by the type of information to be obtained. It is necessary to decide on the type of questionnaire at this point since the content and wording of the questions (simple and clear meaning), the length of the questionnaire and the sequence of questions will all be influenced by this decision. A decision to use conjoint analysis. Thus, at this stage the market analysis must specify precisely how the primary data needed will be collected and also the type of analysis to be made with the data.
DETERMINE THE CONTENT OF INDIVIDUAL QUESTIONS
Once the information needed is known and the data collection method decided, the researcher is ready to begin formulating the questions. Several points should be reviewed systematically once the content of questions is determined. For example:
- Is the question necessary? Avoid including interesting questions, which are not directly related to the information needed.
- Are several questions needed instead of one? Some questions may have two or more elements and if these are left in one question, interpretation becomes impossible. This is typically the case for the ‘why’ question.
- Does the respondent have the information requested? Three sub-questions can be examined in order to determine this: Is the point raised within the respondent’s experience? Can the respondent remember the information? Will the respondent have to do a lot of work to get the information?
- Will respondents give the information? Even though they know the answer, respondents will sometimes not answer questions, because: They are unable to phrase their answer They do not want to answer.
DETERMINE THE TYPE OF QUESTION TO USE
In forming the actual questions, the researcher has the choice between three major types of questions.
- An open-ended question requires the respondents to provide their own answers to the question.
- A multiple-choice question requires the respondent to choose an answer from a list provided with the question. The respondent may be asked to choose one or more of the alternatives presented.
- A dichotomous question is an extreme form of the multiple-choice question which allows the respondent only two responses, such as yes/no, agree/disagree and so on.
In a multiple-choice question, when the proposed answers are ranked, the objective is not simply to identify a category (as in a nominal scale), but rather to ‘measure’ a level of agreement, a degree of importance or a level of preference. Two types of scale can be used:
- An ordinal scale where the numbers possess the property of rank order
- An interval scale, which has all the properties of an ordinal scale and, in addition, the differences between scale values can be meaningfully interpreted.
The distinction is important because the permissible mathematical operations are different with each type of scale. In practice, responses to questions on importance or preference are frequently assumed to form an interval scale. Different scales are currently used. The most common are:
- The Likert scale, with descriptor labels attached to each category
- The semantic differential scale, which uses bipolar adjectives
- The constant sum scale, where the respondent is instructed to allocate a given sum among two or more attributes on the basis of their importance to that respondent.
DECIDE THE WORDING OF QUESTIONS
The problem at this stage is to phrase the questions in a way that:
- The respondent can easily understand
- Does not give the respondent a clue as to how he or she should answer e.g. leading question.
Consider the following points in order to make sure questions are both comprehensible and unbiased.
- Define the issue clearly. Each question should be checked on six points – who, where, when, what, why and how – to be sure that the issue is clear.
- Decide whether to be subjective or objective. A subjective question puts the question in terms of the individual while objective phrasing tends to refer to what people in general think. Subjective questions tend to give more reliable results.
- Use simple words. Words used in questionnaires should have only one meaning, and that everyone should know meaning. There are many examples of misunderstanding of what seem to be everyday words. In particular, the technical jargon of marketing (brand image, positioning, etc.) should be avoided. The pre-test of the questionnaire is very useful in overcoming this difficulty.
- Avoid ambiguous questions. Ambiguous questions mean different things to different people. Indefinite words include ‘often’, ‘occasionally’, ‘frequently’, ‘many’, ‘good’, ‘fair’, ‘poor’ and so on; these may have many different meanings. For example, ‘frequent reading’ of a monthly magazine may be six or seven issues a year for one person and twice a year for another.
- Avoid leading or one-sided questions. A leading question is one that may steer respondents towards a certain answer. One-sided questions present only one aspect of an issue. A question should be constructed in as neutral a way as possible, by avoiding the name of a brand or a company or by presenting all the sides of an issue.
- Avoid double-barrelled questions. A double-barrelled question is one that calls for two responses and thereby creates confusion for the respondents. Such questions should be divided into two separate parts.
- Use split-ballot wherever possible. There is no ‘correct’ wording for a question. When there are two wordings from which to choose, but no basis on which to select one in preference to the other, one wording can be adopted on half of the questionnaires and the other on the other half.
DECIDE THE SEQUENCE OF QUESTIONS
There are generally three major sections in a questionnaire:
- The basic information sought
- The socio-demographic information useful in obtaining the profile of the respondent
- The identification sections to be used by the interviewer.
The general rule is to put the sections in that order: the main body of the questionnaire should go in first position and the socio-demographic questions at the end (unless they serve as filter questions to qualify respondents for the survey).
The following points should also be considered in order to determine the sequence of questions to maximum effect:
- Use simple and interesting opening questions. If the opening questions are interesting, simple to comprehend and easy to answer, the respondent’s co-operation will be gained.
- Use the funnel approach. The funnel approach involves beginning with a very general question on a topic and gradually leading up to a narrowly focused question on the same topic.
- Arrange questions in logical order. Sudden changes in subject confuse the respondent and cause indecision.
- Place difficult or sensitive questions near the end. Sensitive questions should be relegated towards the end of the questionnaire, once the respondent has become involved in the study.
A mail/an e-mail questionnaire raises specific sequence problems since it must sell itself. It is particularly important that the opening questions capture the respondent’s interest. Questions should then proceed in logical order. However, in a mail/an e-mail questionnaire, it is not possible to take advantage of sequence position in the same way as in personal interviews since it is the respondent who will decide the order of response. The layout and physical attractiveness is particularly important for a self-administered questionnaire.
PRE-TEST THE QUESTIONNAIRE
Before a questionnaire is ready for the field it needs to be pre-tested under field conditions. Pretesting involves administering the questionnaire to a limited number of potential respondents selected on a convenience basis, but not too divergent from the target population. It is not necessary, however, to have a statistical sample for pre-testing. The pre-testing process allows the researcher to determine whether the respondents have any difficulty in understanding the questionnaire and whether there are ambiguous or biased questions. Tabulating the results of the pre-test is also very useful to ensure that all the required information will be obtained.
Questionnaire collection majorly falls into two categories
- Postal questionnaire
- On-line questionnaire
- Interviewer administered
- Face to face questionnaire
- Telephone questionnaire