Evaluating Quality in CE's Proof of Concept Test

• Activity Number: 286 - Tracking and Evaluating Quality on the Road to a Redesigned Consumer Expenditure Survey
• Type: Topic Contributed
• Date/Time: Tuesday, August 1, 2017 : 8:30 AM to 10:20 AM
• Sponsor: Government Statistics Section
• Author(s): Laura Erhard* and Brett McBride and Yezzi Lee and Safia Abdirizak
• Companies: Bureau of Labor Statistics and Bureau of Labor Statistics and U.S. Bureau of Labor Statistics and U.S. Bureau of Labor Statistics
• Keywords: data quality ; household surveys ; diaries ; financial records ; incentives

Abstract

The U.S. Bureau of Labor Statistics (BLS) Consumer Expenditure Surveys (CE) released a redesign plan in 2013 that includes two identical waves of a single-sample expenditure survey, each consisting of two in-person interviews, personal multi-mode diaries for eligible householders, incentives, and record use. In order to test the basic structure and components of the new design, a Proof of Concept (POC) test was fielded in 2015 that included one wave of the redesign plan. One objective of the test was to evaluate the quality of data collected using the redesign protocols compared to CE's current design, which includes no incentive use and independent samples, one collecting expenditures from weekly household paper diaries and the other collecting expenditures in a 3-month recall interview. This presentation will provide a high level overview of the POC design and discuss results of the POC test related to data quality. We will compare levels and counts of reported expenditures, counts of missing expenditures, and editing rates between the POC data and a matched CE production sample. This will be of interest to survey methodologists, data producers, and data users.