Does Data Quality Decrease Over the Course of an Expenditure Survey?

• Activity Number: 232
• Type: Topic Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): David Cantor*+ and Cong Ye and Brandon Kopp
• Companies: Westat and American Institute for Research and Bureau of Labor Statistics
• Keywords: Respondent Burden ; Fatigue

Abstract

Common wisdom is that the length of the interview has a negative effect on data quality. One manifestation of this is that respondents may exert less cognitive toward the end of an interview. Despite the logical nature of this assertion, there is very little empirical evidence of this phenomena. This study investigates this hypothesis by comparing answers to global questions that were randomly placed either at the beginning or at the end of a telephone survey. The survey was an adapted version of the Consumer Expenditure Quarterly interview survey (CEQ), which collects quarterly expenditures for a consumer unit (CU). The interview has a series of interleafed global questions on expenditures for the CU. This survey averaged approximately 50 minutes. The research was sponsored by the Bureau of Labor Statistics and carried during the 2012 Practicum at the University of Maryland's Joint Program for Survey Methodology. A total of 677 CU's were interviewed. The presentation will report on analyses comparing affirmative answers to global questions by section placement and whether any differences vary by characteristics of the respondent and by interviewer.