Journal Article

A Review of Measurement Error Assessment in a U.S. Household Consumer Expenditure Survey

Scott Fricker, Brandon Kopp, Lucilla Tan and Roger Tourangeau

in Journal of Survey Statistics and Methodology

Published on behalf of American Association for Public Opinion Research

Volume 3, issue 1, pages 67-88
Published in print March 2015 | ISSN: 2325-0984
Published online January 2015 | e-ISSN: 2325-0992 | DOI: http://dx.doi.org/10.1093/jssam/smu025
A Review of Measurement Error Assessment in a U.S. Household Consumer Expenditure Survey

Show Summary Details

Preview

This paper examines what is known about error in the U.S. Consumer Expenditure Survey (CE). It reviews likely sources of error in consumer expenditure surveys and describes the various approaches that have been used to evaluate estimates from the CE in the United States and from similar surveys around the world. In the United States, many efforts to assess the quality of the CE have compared CE estimates to estimates from other surveys or from the National Income and Product Accounts (NIPA). In addition, other studies have used data internal to the CE in an attempt to assess the level of error. We review studies that take each of these approaches and assess their strengths and weaknesses. Our review shows that, as with other national and international surveys on household spending, the direction and magnitude of measurement errors in the CE are not well understood. The general impression that the CE expenditure estimates are dominated by underreporting is based largely on comparisons of the CE with NIPA estimates. However, other results suggest that the magnitude and direction of errors in the CE depend on the expenditure category examined, characteristics of the respondent and the survey protocol, and the method used to assess these errors. New research clearly is needed to understand the sources of error in surveys collecting consumer expenditure data.

Keywords: Assessment methods; Household expenditures; Measurement error; Survey design

Journal Article.  7678 words.  Illustrated.

Subjects: Social Research and Statistics

Full text: subscription required

How to subscribe Recommend to my Librarian

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.