Measures of functional limitations: The effects of person-level vs. household-level questionnaire design

This paper presents results on functional limitations and the use of special aids from the Census Bureau's 1999 Questionnaire Design Experimental Research Survey (QDERS), which included a split-ballot test comparing person-level questions to household-level questions. We find some evidence that the use of a household-level design results in lower survey estimates than the person-level design. The household-level approach, however, produces somewhat more reliable data than the person-level approach and results in a shorter interview than the person-level design. Interviewers administer survey questions equally well in both treatments, and respondents have little difficulty understanding and answering survey questions in either treatment. Item non-response is trivial in both treatments.