Wealth measurement in the survey of consumer finances: Methodology and directions for future research

Article type: Research Article

Authors: Kennickell, Arthur B.

Affiliations: Board of Governors of the Federal Reserve System (Ret.), Washington, DC 20551, USA. Tel.: +1 202 452 2247; E-mail: Arthur.Kennickell@gmail.com

Abstract: This paper summarizes, as of 2000, the history of the Survey of Consumer Finances (SCF) and the supporting structure for measuring household wealth with that survey. Surveys are the large scientific measurement devices of the social sciences. Each set of observations rests heavily on theories - statistical theories about sampling, missing data, inference, and other such formal issues, cognitive theories about how people perceive the data collection process and respond to it, behavioral theories that deal with the different incentives that operate on all parties in the data collection, and even social theories that address the ways that we interpret the process and its outcomes. Each aspect of these theories has a role in characterizing the ``results'' of the cumulation of processes that make up a survey. For the SCF, the most pressing issues are persuading respondents to participate and, given participation, to provide accurate data. Many measurement problems are amplified by the fact that the distribution of wealth is highly skewed.

Keywords: Wealth measurement, survey of consumer finances, history

DOI: 10.3233/SJI-160274

Journal: Statistical Journal of the IAOS, vol. 33, no. 1, pp. 23-39, 2017