A question like the following often appears on questionnaires sent to samples of college students:

Age : 1) below 18 2) 18-19 3) 20-21 4) over 21

If there is a specific need to generalize results to older or younger students, the question is valid. Also, such a question might be included to check on the representativeness of the sample. However, questions like this are often included in an apparently compulsive effort to characterize the sample exhaustive. A clear-cut need for every question should be established. This is especially important with respect to questions characterizing the responders, because there may be a tendency to add these almost without thought after establishment of the more fundamental questions. The fact that such additions may lengthen the questionnaire needlessly and appear to pry almost frivolously into personal matters is often overlooked. Some questionnaires ask for more personal data than opinions on their basic topics.

In many cases, personal data are available from sources other than the responders themselves. For example, computer files used to produce mailing labels often have other information about the subjects that can be merged with their responses if these are not anonymous. In such cases, asking the responders to repeat this information is not only burdensome but may introduce error, especially when reporting the truth has a negative connotation. (Students often report inflated grade-point averages on questionnaires.)