Abstract
Jenny Thompson (1994) "Mode Effects
Analysis of Major Labor Force Estimates," CPS Overlap
Analysis Team Technical Report 3.
The official monthly civilian labor force estimates from
January 1994 onward are based on data from a comprehensively
redesigned Current Population Survey (CPS). The redesign
incorporates changes in the basic questionnaire and
collection methodology. Bregger and Dippo (1993) discuss the
motivation for this redesign.
To gauge the effect of the CPS redesign on published
estimates, a Parallel Survey (PS) was conducted using the new
questionnaire and collection procedures from July 1992 to
December 1993. Annual average estimates from the PS were used
to examine the effect of the CPS redesign on major labor
force estimates. Polivka (1994) presents a comparison of the
labor force estimates from the PS and CPS during 1993.
Kostanich and Cahoon (1994) further consider the possible
influence of the differences in the two surveys' designs on
these comparisons.
A secondary consideration was an investigation into the
possible effect of selected factors associated with the new
questionnaire or collection mode on major labor force
estimates. Special studies were embedded within the CPS and
the PS during the same time period to provide data for
testing hypotheses about the effects of these methodological
differences on labor force estimates. The resultant
hypothesis tests attempted to link the differences in annual
average major labor force estimates between CPS and the PS to
the presence of specified collection mode differences.
This report centers on four possible effects: two centralized
telephone interviewing effects and two new questionnaire
effects. Three of the four hypothesis tests used subnational
estimates of CPS and PS data and were not necessarily
representative of the national differences. The other
hypothesis test used national estimates computed from
approximately one fourth of the full sample (for both
surveys).
The Background section of this report provides an overview of
the two surveys' designs, a description of the estimates and
variance estimates, descriptions of hypothesis testing
methodology, and detailed descriptions of the tested
hypotheses. The Results section presents the hypothesis tests
for major labor force characteristics. The Summary section
provides general conclusions.
Last Modified Date: July 19, 2008
|