Performance Improvement 2011-2012. What Were The Effects Of Imputation On The Current Population Survey Income And Poverty Series?


The Current Population Survey has been the source of the official U.S. poverty estimates since their inception. Since then, many changes have occurred in society and in the willingness of survey respondents to report the cash income used to construct these poverty figures. Improvements have also occurred in the data collection instruments. Poverty rates were examined by type of imputation from 1981 until 2007 focusing particularly on how poverty series for reporters and for those with item and whole imputes have trended over this period. Differences for blacks and whites were determined.

Poverty rates for persons with positive income trended downward. Over most of the period, poverty rates for persons reporting all of their income and for those with at least some of their income imputed paralleled each other, however, poverty rates for imputers were 2.5% to 3% lower than for reporters from year to year. Persons with no imputes have the highest poverty rates and those with item imputes the lowest poverty rates. Blacks have significantly higher poverty rates than whites for all types of imputation.

Report Title: Effects of Imputation on CPS Income and Poverty Series: 1981-2007
Agency Sponsor: OASPE-OSDP, Office of Science and Data Policy
Federal Contact: Joan Turek, 202-690-5965
Performer: Joan Turek, ASPE/OSDP
Record ID: 9245 (Report issued November 30, 2009)

View full report


"PerformanceImprovement2011-2012.pdf" (pdf, 701.44Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®