Editing strategies used by the US Bureau of Labor Statistics in Data Collection over the Internet

Article type: Research Article

Authors: Cohen, Stephen H.^{; 1}

Affiliations: Office of Survey Methods Research, Mathematical Statistics Research Center, U.S. Bureau of Labor Statistics, Washington, DC 20212, USA. Tel.: +1 202 691 7400; Fax: +1 202 691 7426; E-mail: [email protected]

Note: [1] This is a slightly revised version of a paper presented to the UNECE Work Session on Statistical Data Editing in Madrid, Spain, 20–22 October 2003. Views expressed in this paper are those of the author and do not necessarily represent the official views of the Bureau of Labor Statistics or the Department of Labor.

Abstract: The U.S. Bureau of Labor Statistics (BLS) is the principle Federal agency charged with disseminating statistics in the broad area of labour economics. In particular BLS publishes the principle Federal economic indicators on unemployment and inflation in the U.S. BLS attempts to provide data providers with as many options as possible for submitting their reports such as over the Internet. The BLS approach to Internet data collection is to provide a singular, manageable, secure architecture for all surveys. This paper will explore the editing strategy used by BLS on the Internet. The paper will summarise the types of edits used in post collection review and the strategy employed to incorporate edits into web-based instruments. Edit experience from two Internet survey instruments will be discussed including implementation strategies, edit failure experience, and comparisons with edit failures of data collected by standard protocols.

DOI: 10.3233/SJU-2003-203-406

Journal: Statistical Journal of the United Nations Economic Commission for Europe, vol. 20, no. 3-4, pp. 265-278, 2003