70 – Robust Inference in Social Science with Wrong but Useful Models
The Supplemental Poverty Measure in the Survey of Income and Program Participation
Kathleen Short
U.S. Census Bureau
Katherine G. Giefer
U.S. Census Bureau
This year the Census Bureau released its second report on the Supplemental Poverty Measure (SPM). The report used the Current Population Survey Annual Social and Economic Supplement (CPS) for these estimates. This measure followed suggestions from an interagency working group headed by OMB that urged the Census Bureau to publish it along with the current official measure of poverty. The SPM generally follows recommendations from the National Academy of Sciences Panel on Poverty and Family Assistance (NAS) that released a report in 1995 recommending revising the current official poverty measure. In that report, the Panel recommended using the Survey of Income and Program Participation (SIPP) to measure poverty. All of the information the NAS Panel included in a new poverty measure is directly collected in the SIPP. This exercise provides insight into how well the CPS measures the SPM concept. Differences in sample design and data collection, however small, have a significant effect on measurement outcomes. The exercise will also provide information on how a redesigned SIPP will measure the SPM.