Multiple Frame Sample Design and Estimation for the 2008 National Sample Survey of Registered Nurses - Theory, Implementation, and Assessment

• Activity Number: 420
• Type: Contributed
• Date/Time: Tuesday, August 3, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Ralph DiGaetano*+ and Jim Green and Jay Clark and Marshall Fritz
• Companies: Westat and Westat and Westat and U.S. Department of Health and Human Services
• Keywords: Multiple frame surveys ; probability matching ; composite estimation ; bias ; variance ; design effects

Abstract

The 2008 NSSRN is the latest in a series of national surveys of RNs carried out roughly every four years since the late 1970s. For the 2008 NSSRN a new sample design and weighting process were used. RNs were sampled independently from state listings of currently licensed RNs. Many RNs were licensed in several states, resulting in multiple chances of sample selection. Probabilistic matching and questionnaire responses were used to identify RNs appearing on multiple listings. Compositing factors based on the number of strata containing an RN were applied to nonresponse adjusted weights assigned to the sampled records of responding RNs. The nonresponse adjustment process was enhanced compared to prior surveys. This paper discusses the theory behind the estimation approach used, the processes of implementing the approach, and the corresponding impact on survey precision and accuracy.