Shail Butani, Lawrence Cahoon, Robert E. Fay, and Donna Kostanich (1994) "Measurement Of Different Design Effects", Proceedings of the Section on Survey Research Methods, American Statistical Association, forthcoming.
The Current Population Survey (CPS), a monthly labor force survey that the Census Bureau conducts for the Bureau of Labor Statistics (BLS), uses a state based probability sample of about 60,000 households. In January 1994, CPS interviewers began collecting data in a completely computer-assisted environment, using a new questionnaire. Before introducing the new data collection method, the Census Bureau and BLS tested it on a separate, national based probability sample of 12,000 households. The results of this parallel survey, which ran from July 1992 through December 1993, indicated that the CPS annual average unemployment rate would have been 0.5 percentage point higher in 1993 had the new data collection method been used. In this paper, we discuss the measurement of differences between the design effects of the parallel survey in December 1993 and the CPS in January 1994 onward, in order to better understand any transition effects. We concentrate our analysis on the effects of 1) differences in the proportion of households interviewed from centralized facilities; 2) use of 1990 based population controls, adjusted for census undercount, instead of 1980 based controls formerly used; 3) differences in ratio adjustment methods; 4) use of composite estimation to reduce variance; and 5) seasonal adjustment.
Last Modified Date: July 19, 2008