State Sampling Allocation Strategies for the 2016 Redesigned National Health Interview Survey (NHIS)

• Activity Number: 445
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Van Parsons* and Erin Dienes
• Companies: National Center for Health Statistics and National Center for Health Statistics
• Keywords: multiple objectives ; oversampling

Abstract

The NHIS is a multi-purpose face-to-face health survey conducted annually by the National Center for Health Statistics (NCHS). Historically, the NHIS has been designed to produce accurate national-level statistics on an annual basis. The traditional design has been based upon a one-time probability sample of coarsely defined geographical units containing population dwellings to be sampled over a 10 year NHIS design cycle. The current NHIS objectives, however, are broader in scope than in the past. Now, the 10 year survey objectives also include the ability to achieve accurate estimates for state and targeted minority populations, but on an "as needed" basis. Budgetary and operational constraints have presented challenges in achieving an optimal design. As a compromise, the final redesign is structured with a nationally-focused design as its core, but the design contains large reserve samples that can be used to achieve state or minority objectives. A recent focus on state-level heath care has placed a priority on the allocation of sample to achieve state objectives. In this paper we present the state allocation strategies to be implemented in 2016.