|
Activity Number:
|
373
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #306778 |
|
Title:
|
The Impact of Medical Expenditure Predictors in MEPS Nonresponse Adjustments
|
|
Author(s):
|
Lap-Ming Wun*+ and Trena Ezzati-Rice and Steven B. Cohen and William Yu
|
|
Companies:
|
Agency for Healthcare Research and Quality and Agency for Healthcare Research and Quality and Agency for Healthcare Research and Quality and Agency for Healthcare Research and Quality
|
|
Address:
|
540 Gaither Road, Rockville, MD, 20850,
|
|
Keywords:
|
nonresponse ; logistic regression ; weight adjustment ; response propensity
|
|
Abstract:
|
The MEPS is a large and complex sample survey sponsored by the Agency for Healthcare Research and Quality. The sample is drawn from respondents to the previous year's National Health Interview Survey. The MEPS, like most sample surveys, experiences unit nonresponse. Weighting adjustment is a general method to compensate for nonresponse. Currently, in the MEPS, a weighting class nonresponse adjustment based on relevant covariates modeled through the software CHAID is used. An alternative method of using response propensities calculated from logistic regression models has been investigated and reported previously. This paper reports on the investigation of the impact of including a high medical expenditure predictor as a covariate in the logistic model for nonresponse adjustment. The impact is evaluated by applying the adjusted weights to selected analytical variables.
|
