Rotation Group Bias in the Consumer Expenditure Survey

Abstract

Do respondents underreport purchases in later waves to avoid follow up questions? CE respondents take part in four waves of the survey, reporting hosuehold purchases. Each interview lasts about 90 minutes. Respondents may learn that each purchase they mention triggers several follow up questions about the price of the item. This presentation tests for rotation group bias in the CE.

Date
Event
Office of Survey Methods Research Seminar
Location
Bureau of Labor Statistics, Washington DC

Much of my research in the last ten years has explored Motivated Misreporting: the shortcuts respondents take to reduce the length or burden of a survey.

The research for this talk is part of my ASA / NSF / BLS fellowship.