Sample Design Changes For The NHANES 2011-2014 Annual Surveys

• Activity Number: 175
• Type: Contributed
• Date/Time: Monday, August 1, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Sylvia Dohrmann*+ and Hongsheng Hao and Leyla K. Mohadjer and Lin Li and Wendy Van de Kerckhove and Lester R. Curtin
• Companies: Westat Inc. and Westat Inc. and Westat Inc. and Westat Inc. and Westat Inc. and Centers for Disease Control and Prevention
• Keywords: Sample design ; stratification ; oversampling ; household surveys

Abstract

The National Health and Nutrition Examination Survey (NHANES) is an ongoing annual survey with PSUs selected for four years at a time. This process minimizes overlap in annual samples and allows for adjusting the design overtime to address changes in the national health concerns. A major design change for the 2011-2014 sample reflected the decision to oversample the Asian population to produce adequate sample sizes for given subdomains. In addition, the stratification scheme for sampling the PSUs was re-designed to ensure PSUs comprising annual and multi-year samples are distributed evenly in terms of health level, geography, urban-rural distribution, and population characteristics. The re-designed PSU stratification scheme was based on grouping the states by health-related measures. Cluster and factor analysis were used to form state groups using a number of state-level health indicators. Within each state group, NSR PSUs were further subdivided into substrata by geography, urban-rural distribution of population, race/ethnicity density, and poverty level. Various options of strata formation methods were performed and compared in the sample design to reach the design goal.