Sample Allocation for the Redesigned National Compensation Survey

• Activity Number: 87
• Type: Contributed
• Date/Time: Sunday, July 29, 2012 : 4:00 PM to 5:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Tiandong Li*+ and Hyunshik James Lee and Gwyn Ferguson and Chester H Ponikowski and Bradley Rhein
• Companies: Westat and Westat and Bureau of Labor Statistics and Bureau of Labor Statistics and Bureau of Labor Statistics
• Keywords: Logistic regression ; viability and usability rates ; between- and within-PSU variances ; intra-class correlation

Abstract

The National Compensation Survey (NCS) conducted by the Bureau of Labor Statistics (BLS) is an establishment survey for which the sample selection methodology has been redesigned to reflect a change in the scope of the survey. The new sample design will be implemented in 2012. The main feature of the redesign is the change from a three-stage area based design to a new two-stage design. Under the new design, establishments are selected by the probability proportional to size (PPS) sampling method as the primary sampling unit (PSU) at the first-stage, and occupations again by the PPS method at the second-stage from the sampled establishments. For this new design, sample allocation needs to be done to meet the precision objectives of the survey. For this purpose, variance components are estimated, response rates are projected, and based on this information the national private industry samp