87 – Developments in Survey Design and Implementation
Allocation of Sample for the 2010 Redesign of the Consumer Expenditure Survey
Stephen Ash
U.S. Census Bureau
David Swanson
Bureau of Labor Statistics
This paper discusses the allocation of sample of the Consumer Expenditure Survey (CE) for the 2010 redesign. CE is a national survey of U.S. households that measures the expenditures made by households on different categories of items. The survey actually consists of two surveys. Small expenditures are collected in the CE Diary Survey and larger expenditures are collected in the CE Quarterly Interview Survey. Both CE surveys use a two-stage sample design, where the first-stage is a sample of counties or groups of counties and the second-stage is a sample of households within the counties or groups of counties. In this paper we discuss methods for optimally allocating the survey’s overall sample to the two surveys where optimality is with respect to minimizing the survey’s overall variance. Then, we compare several alternative methods for optimally allocating the sample to the strata of the first-stage sample design.
