|
Activity Number:
|
127
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Social Statistics Section
|
| Abstract - #305736 |
|
Title:
|
Using Income Imputation Techniques to Improve Evaluation of Outreach Efforts for Low-Income Medicare Beneficiaries
|
|
Author(s):
|
Frank Funderburk*+ and Christopher Koepke and Thomas Kickham
|
|
Companies:
|
Centers for Medicare and Medicaid Services and Centers for Medicare and Medicaid Services and Centers for Medicare and Medicaid Services
|
|
Address:
|
7500 Security Blvd, Baltimore, MD, 21244,
|
|
Keywords:
|
social marketing ; health care ; multiple imputation ; Medicare
|
|
Abstract:
|
Effective social marketing efforts require a good understanding of the target audience. Consumer surveys are often conducted for this purpose, but sometimes have missing data on key variables. Developing an appropriate analytic strategy for using such data to inform policy is central for success. We examined the results of a survey (n = 4,208) to evaluate outreach effectiveness of direct marketing strategies for low-income Medicare beneficiaries. We found that 28% of survey respondents refused to provide income information. Logistic regression revealed that those who refused to provide income information differed systematically from other respondents on demographic, attitudinal, and decision-making style measures. We discuss how we evaluated the extent of the potential biases and how multiple imputation procedures helped make the results more useful for policy makers.
|
