Abstract:
|
The Current Population Survey (CPS) is a household survey sponsored by the Bureau of Labor Statistics and collected by the United States Census Bureau. CPS has a short panel/longitudinal component that acquires data through both CAPI and CATI modes that samples 72,000 households per month. Households are in the survey a total of eight months, with a rotating panel in which each household is in sample for four months, then out for eight months, then again for another four months. This research explores if the survey can reduce effort and maintain the same level of data quality through case prioritization of CAPI field cases and automated selection for CATI phone cases. Effectively assigning which cases should be prioritized and deprioritized for CAPI field cases and which cases should be selected for CATI for phone cases. For CAPI, we introduce the idea of using R-Indicators in order to deprioritize over-represented low impact cases. For CATI, we introduce a mechanism for automatic case selection that explores if the survey can reduce the number of CATI that are reassigned to the field.
|