Respondent Burden: A Test of Some Common Assumptions

Abstract

With an experimental design, the correlates of respondent burden were measured in 500 households in the suburban portion of the Philadelphia SMSA. The research design provided for variation in the length of the instrument, the effort required to answer some of the questions, and the administration of a second interview approximately one year after the first. Respondent burden was measured using behavioral indicators and responses to a self-administered reaction form. Instrument length was the only experimental variable which yielded statistically significant (although generally small) differences in burden perception. Two attitudinal factors—belief in the usefulness of surveys and denial of the privacy-invading character of survey questions—were strongly associated with low burden perception.