39 – Modeling and Modeling Error
Using Targeted Lists for Studies of Rare Populations: The Super Wealthy
Ned English
NORC at the University of Chicago
Steven Pedlow
NORC at the University of Chicago
Lee Fiorio
NORC at the University of Chicago
Catherine Haggerty
NORC at the University of Chicago
Benjamin Page
Northwestern University
Jay Seawright
Northwestern University
The Survey of Economically Successful Americans (SESA) is designed to understand the influence of exceptionally wealthy individuals on the American political process. Our pilot study had the goal of targeting the top 1/10 of 1% of households, estimated at $20-40 million in net-worth. One challenge was the absence of a sampling frame that efficiently captured such high-wealth individuals, and limitations in publicly-available sources such as the American Community Survey. We created a composite frame from market-research sources, including lists of business executives and "wealthy" individuals. Most sources carried limitations in data resolution i.e. top-coding, as well as inconsistent accuracy. Our current research uses external data sources to enhance our results with the goal of improving both the coverage and hit-rate of our methodology. Examples of newly available data sources include estimates of total liquid assets, home value, and stock-sales that were not available during our pilot phase design. We present models that outline the most efficient approach for conducting nationally-representative household surveys of very wealthy populations.