|
Activity Number:
|
567
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 6, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Government Statistics
|
| Abstract - #305373 |
|
Title:
|
Assessing the Effects of Imputation on Income Estimates in the U.S. Consumer Expenditure Survey: Comparison with the Current Population Survey
|
|
Author(s):
|
William D. Passero*+
|
|
Companies:
|
Bureau of Labor Statistics
|
|
Address:
|
2 Massachusetts Avenue, NE, Washington, DC, 20212,
|
|
Keywords:
|
Income imputation ; Data comparisons ; Data quality
|
|
Abstract:
|
The Consumer Expenditure Survey (CE) provides data on the buying habits of American consumers. The data are widely used in economic research as well as for weights in the Consumer Price Index. CE collects extensive demographic and socioeconomic data, including income data. CE staff assesses the quality of the survey's data by comparing CE data to similar data from other sources. Comparisons of income estimates first became possible with 2004 data, the year for which CE began to impute missing income values. This paper analyzes changes in the ratio of post-imputation CE income estimates to income estimates from the 2004-2007 Current Population Survey. The paper also gauges the impact of imputation on CE income estimates, comparing income reports prior to imputation in 2002 and 2003 with post-imputation data from 2004 to 2007.
|
