Evaluating The Questionnaire and Layout
Is the question necessary? Perhaps the most important criterion for this phase of questionnaire development is judging the need for a given question. Sometimes researchers and brand managers want to ask questions because "they were on the last survey", or because " it would be nice to know". Excessive demographics questions are very common. Education data, number of children in multiple age categories and excessive demographic on the spouse are simply not warranted by the nature of many studies. Every question must serve a purpose. Either it must be a screener, an interest generator, a required transition, or directly and explicitly related to the stated objectives of this particular data acquisition survey. Any question that fails to satisfy at least one of these criteria should be omitted.
Is the question too long? At this point, the researcher should role-play the questionnaire with volunteers acting as respondents. Although there is no magic number of iterations, the length of time it takes to complete the questionnaire should be averaged over a minimum of five trials. Any questionnaire to be administered in a mall or over the telephone that averages longer than 20 minutes should be cut. Sometimes mall interviews can run slightly longer if an incentive is provide to the respondent. Common incentives are movie tickets, pens, pencil sets and cash or vouchers. The use of incentives often actually lowers the survey costs because response rates increase and terminations during interviews decrease.
Will the questions provide the desired information to accomplish the research objectives? The researcher must make certain that a sufficient number and type of questions are contained within the data acquisition questionnaire to meet the decision-making needs of management. A suggested procedure is to review carefully the written objectives for the research project. Next, the researcher should go down the questionnaire and write each question number next to the objective that the particular question will accomplish. For example, question 1 applies to objective 3; question 2, to objective 2; and so forth. If a question cannot be tied to an objective, the researcher should determine whether the list of objectives is complete. If the list is sufficient, the question should be omitted. Also, if after going through the entire questionnaire there is an objective with no response listed beside it, appropriate questions should be added to the questionnaire to answer them.
The objective of a good data acquisition questionnaire layout is to make the tasks of the interviewer and respondent as clear, logical and simple as possible. Evaluating the questionnaire and layout is rather time consuming for the research team, but is an essential part to really finalising the finished questionnaire and making sure it will work effectively out in the field.