Big Data Meets Survey Science

12Investigating Alternative Data Sources to Reduce Respondent Burden in United States Census Bureau Retail Economic Data Products

Rebecca J. Hutchinson

US Census Bureau, Economic Indicators Division, Washington, DC, USA

12.1 Introduction

Retail store closures, mergers, and acquisitions among major retailers, innovative industry disruptors, and the evolution of online shopping dominate the daily business news. Official statistics that accurately and consistently measure retail sales have long been closely watched economic indicators, but in this dynamic retail environment, they are even more thoroughly monitored. At the same time, response rates are declining for many United States Census Bureau surveys, including the retail surveys. Respondents often cite the burden of completing multiple surveys monthly and/or annually as one reason for nonresponse (Haraldsen et al. 2013). Recognizing these challenges and the growing needs of its data users, the Census Bureau is committed to exploring the use of Big Data or alternative data sources to produce high‐quality data products while reducing respondent burden (U.S. Census Bureau 2018).

To reduce respondent burden, the Census Bureau has been exploring the use of third‐party data in its retail survey programs. If a retailer is already providing another party with data similar to what the Census Bureau collects on surveys and censuses, can that data consistently be used to replace what a retailer is asked to provide to Census Bureau ...