|
Activity Number:
|
240
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2003 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #301819 |
|
Title:
|
Rating Medicare: Does Mode Matter?
|
|
Author(s):
|
Celia Eicheldinger*+ and Vincent G. Iannacchione and Lisa Carpenter
|
|
Companies:
|
RTI International and RTI International and RTI International
|
|
Address:
|
5613 Olde South Rd., Raleigh, NC, 27606-9250,
|
|
Keywords:
|
survey mode effects ; CAHPS ; Medicare FeeSFor-sevice ; telephone survey
|
|
Abstract:
|
Results of the 2001 Consumer Assessment of Health Plans Study (CAHPS) survey of Medicare Fee-For-Service (MFFS) beneficiaries revealed that telephone respondents to the follow-up portion of the survey rated Medicare 12 percentage points higher than those who completed a self-administered questionnaire. These apparent mode effects remained significant in regression models even after the inclusion of numerous demographic and health utilization measures. The evaluation, however, was adversely affected because assignment to the telephone follow-up depended on the survey team's ability to find a telephone number for each beneficiary. To reduce possible confounding for the 2002 CAHPS MFFS survey, a generalized random-block design (GRBD) experiment was embedded within the group of beneficiaries having a telephone number. Two blocks were established; one block resembled the group of beneficiaries with a telephone and the other without. Within each of the blocks, two random samples were chosen, one sample received the phone survey and the other a mail survey. Our analysis compares the survey results of the GRBD.
|
