If the questionnaire is too long, subjects may refuse to complete it. This is an important factor in any population survey, as a high response rate is essential. If the response rate is not high, subjects who take part in the study cannot be said to be representative of the population from which they were selected. As a guideline, a single questionnaire should not take longer than about 20 minutes to complete to minimise the potential for a detrimental effect on response rate.

If the questionnaire is very long, subjects may become fatigued or bored, impairing concentration and the accuracy of responses. Additional problems with a long questionnaire are that if using a paper form that contains many pages, subjects may turn over two pages together and hence miss out a number of questions. Also, if questions are squashed together in an attempt to reduce the number of pages, subjects may not be able to read it or they may miss questions. As a guideline, for subjects with normal sight it is not advisable to use a point size of less than six. If the survey includes middle-aged or elderly subjects, a point size of less than 10 is not advisable.