|
Activity Number:
|
153
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #309000 |
|
Title:
|
Weight-Trimming in the National Immunization Survey (NIS)
|
|
Author(s):
|
Sadeq Chowdhury*+ and Meena Khare and Kirk Wolter
|
|
Companies:
|
NORC at the University of Chicago and National Center for Health Statistics and NORC at the University of Chicago
|
|
Address:
|
1350 Connecticut Ave, NW, Washington, DC, 20036,
|
|
Keywords:
|
Extreme weights ; outliers ; trimming ; truncation
|
|
Abstract:
|
Excessively large sampling weights can unduly inflate variances of survey estimates. The NIS weighting procedure involves a series of nonresponse and undercoverage adjustments that introduce a considerable variation in sampling weights, even though the sample is selected with equal probability within each of 78 immunization action plan (IAP) areas. To control such undue variance in estimates, extreme weights are truncated and the remaining weights are adjusted to compensate for the truncation. We plan to present an analytical approach to assess the effects of weight trimming on the NIS estimates. The results will provide an indication whether there is scope for further optimization of the trimming procedure. Also, an alternative approach for detecting extreme weights and setting trimming cutoff values under the assumption of an exponential distribution of the weights will be discussed.
|
