Chapter 4 Questionnaire testing
Open access

Questionnaire testing is essential for minimizing error from respondents misunderstanding a question or being unable to provide an accurate response. In addition to asking peers to evaluate questions, a questionnaire must undergo cognitive testing with ten or more individuals drawn from the study population and, if funds are available, testing in a small pilot survey. Testing ensures that the data collected are of high quality, achieving content validity (the questions measure what the researcher wants to measure), inter-rater validity (questions are interpreted in the same way by different respondents) and reliability (respondents give accurate responses that result in very few false positives and false negatives). The chapter explains, in detail, how to conduct cognitive testing and the value of collecting paradata in an online pilot survey. The advantages and disadvantages of web probing, a relatively recent alternative to cognitive testing interviews, are also discussed.