Janice Lent, Stephen Miller, Martha Duff, and Patrick Cantwell (1998) "Comparing Current Population Survey Estimates Computed Using Different Composite Estimators," Proceedings of the Section on Survey Research Methods, American Statistical Association.
We examine the effect of a new composite estimation method on estimates from the Current Population Survey, the U.S. labor force survey. Before December 1997, the AK composite estimator was applied directly for each characteristic of interest. To ensure consistency, the same coefficients were used for all estimates, though they were selected to be optimal only for unemployment totals. The new "composite weighting" method provides increased operational simplicity for micro data users and allows optimization of compositing coefficients for different labor force categories. Using both additive and multiplicative models, we develop adjustment factors that allow comparison of labor force estimates computed using the old and new composite estimators. The models appear to fit the data well; we illustrate their use for some key CPS statistics.
Last Modified Date: July 19, 2008