This paper reports the results of research and analysis undertaken by Census Bureau staff. It has undergone a Census Bureau review more limited in scope than that given to official Census Bureau publications. This report is released to inform interested parties of ongoing research and to encourage discussion of work in progress.
In their final report, the NRC Panel on Poverty Measurement and Family Assistance recommended numerous changes to the method by which the US Census Bureau measures poverty. The Panel sought to make recommendations that could implemented. One of the Panel’s proposals was to subtract from the family’s resources the amount of medical out of pocket (MOOP) spending. Given that neither the Current Population Survey (CPS) nor the preferred data set, the Survey of Income and Program Participation (SIPP), collect information on the family’s medical spending, a natural question is how well can one impute this needed data from other sources to either the CPS or SIPP?
The purpose of this paper is to examine the current imputation strategy, discuss its potential shortcomings, and report upon efforts to re-estimate the MOOP model on data from the Consumer Expenditure Survey (CEX).