 |
The 130th Annual Meeting of APHA |
|
The 130th Annual Meeting of APHA |
Barbara Lepidus Carlson, MA, Mathematica Policy Research, Inc., P.O. Box 2393, Princeton, NJ 08543-2393, (609) 275-2374, bcarlson@mathematica-mpr.com and Frank J. Potter, PhD, Mathematica Policy Research, Inc, P.O. Box 2393, Princeton, NJ 08543.
The Community Tracking Study (CTS) is a multi-year multifaceted study of changes over time in the U. S. health care system conducted for the Center for Studying Health System Change, a nonpartisan policy research organization committed to providing objective research on the nation’s changing health system; the Center is funded by the Robert Wood Johnson Foundation The CTS has three survey components: a household survey, a physician survey, and a follow-back survey of health plans. Data from all three surveys are made available to researchers after each round of data collection via public use files and restricted use files. These surveys include a clustered sample with 60 sites and high unequal selection rates (and certainty selections) for 80% of the sites. For each survey, we provide analysis weights for different types of estimates and variance estimation parameters to account for the multiple selection stages, the certainty selections, and adjustment for the finite population. While a number of software packages are designed to estimate variances for data from complex sample designs, the complexity of the design limits the available software. In particular, SUDAAN allows for the finite population correction adjustment through joint inclusion probabilities. In this paper, we will present the various types of estimates that can be made, and the corresponding weights and variance estimation parameters to be used. We will also show the difference between variance estimates made using the joint inclusion probabilities and variance estimates assuming equal probability without-replacement and with-replacement sampling.
Learning Objectives:
Keywords: Survey, Methodology
Related Web page: www.hschange.com
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: Center for Studying Health System Change
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.