Big Data versus a Survey

Stephan Whitaker

No 1440, Working Papers (Old Series) from Federal Reserve Bank of Cleveland

Abstract: Economists are shifting attention and resources from work on survey data to work on ?big data.? This analysis is an empirical exploration of the trade-offs this transition requires. Parallel models are estimated using the Federal Reserve Bank of New York Consumer Credit Panel/Equifax and the Survey of Consumer Finances. After adjustments to account for different variable definitions and sampled populations, it is possible to arrive at similar models of total household debt. However, the estimates are sensitive to the adjustments. Little similarity is observed in parallel models of nonmortgage debt. While surveys intentionally collect theoretically related variables, it may be necessary to merge external data into commercial big data. In this example, some education and income measures are successfully integrated with the big data, but other external aggregates fail to adequately substitute for survey responses. Big data offers sample sizes, frequencies, and details that surveys cannot match. However, this example illustrates why caution is appropriate when attempting to substitute big data for a carefully executed survey.

Keywords: Big Data; Survey Data; Household Debt (search for similar items in EconPapers)
JEL-codes: C55 C81 D12 (search for similar items in EconPapers)
Pages: 41 pages
Date: 2015-01-07
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://doi.org/10.26509/frbc-wp-201440 Persistent link
https://www.clevelandfed.org/-/media/project/cleve ... sus-a-survey-pdf.pdf Full text (application/pdf)

Related works:
Journal Article: Big Data versus a survey (2018)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:fip:fedcwp:1440

Ordering information: This working paper can be ordered from

DOI: 10.26509/frbc-wp-201440

Access Statistics for this paper

More papers in Working Papers (Old Series) from Federal Reserve Bank of Cleveland Contact information at EDIRC.
Bibliographic data for series maintained by 4D Library ().