Vragenlijsten met en zonder vragen: een experiment

Samenvatting

Sociologists are often concerned with the statistical treatment of no response or "don't know" answers to questionnaire items. Part of the problem may lie in the overall design of the questionnaire, or in wording that confuses the issue for the R. With these points in mind, a statistical reprocessing is attempted of data collected by J. van Heerden & J. Hoogstraten ("Response Preference as a Function of Instructions in an Unstructured Questionnaire," Perceptual & Motor Skills, 1980, 50, 227-230). It is shown that variables such as type & sequence of questions, degree of homogeneity of the sample, & clarity of instructions are key factors that must be considered in experimental design. 3 Tables, 9 References. M. Meeks.