Journal of Official Statistics, Vol.6, No.3, 1990. pp. 275–293
Data Collection Organization Effects in the National Medical Expenditure Survey
Steven B. Cohen and D. E. B. Potter
Abstract:The Household Component of the National Medical Expenditure Survey (NMES) was designed to provide unbiased national and regional estimates of the health care utilization, expenditures, sources of payment, and health insurance coverage for the U.S. civilian noninstitutionalized population. The complex survey design of the NMES was complicated by combining two independently drawn national samples from Westat, Inc. and the National Opinion Research Center (NORC). It was assumed that since the designs of both national area samples were similar, they would allow for the independent derivation of unbiased national estimates of study parameters. However, even though the two survey organizations operate under a common set of survey conditions with comparable samples, the actual data generated may differ, independent of differences due to pure sampling error. In this paper, national parameter estimates of key study measures are produced separately for each survey organization to determine whether significant differences exist. The analysis includes a comparison of item nonresponse rates to determine whether differences exist in data quality across survey organizations. A comparison of survey design effects is included to determine whether the precision in survey estimates was affected by design differences across organizations.
Keywords:House effects; The National Medical Expenditure Survey.
Copyright © Statistics Sweden 1996-2018. Open AccessISSN 0282-423XCreated and Maintained by OKS Group