Abstract

The Consumer Expenditure (CE) Diary Survey at the U.S. Bureau of Labor Statistics (BLS) is fielded using primarily in-person interviews to place diaries, including a choice between a paper or online diary, and to train respondents how to record the household's daily expenditures. Interviewers also check in with respondents during the diary-keeping period to answer questions and follow up at the end of the period to collect any missed expenditures. To explore the impact of a fully self-administered mode, in contrast to CE's current interviewer-aided protocols, BLS conducted a test in a probability-based online panel (Ipsos KnowledgePanel). In CE's Online Probability Self-Administered Diary (OPSAD) test, panel respondents were invited to complete a self-administered Household Characteristics Survey and subsequent two-week Spending Diary, relying solely on self-guided online instructional materials and without the aid of an interviewer. To evaluate the data quality implications of the OPSAD test design, results are compared to CE diary production data from the corresponding months (November 2021 to January 2022). Data quality measures examined include item nonresponse, expenditure counts and amounts, and adherence to diary-keeping instructions to itemize expenditures. Analyses will account for compositional differences in the underlying samples compared. The presentation will conclude with a discussion of future directions for research and development of improvements to the CE online diary instrument and protocols.