|
Activity Number:
|
14
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 11, 2002 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods*
|
| Abstract - #300612 |
|
Title:
|
An Experiment Embedded in a Health Survey with Respondent-Generated Intervals
|
|
Author(s):
|
Diane Miller*+ and S. Press
|
|
Affiliation(s):
|
University of California, Riverside and University of California, Riverside
|
|
Address:
|
245 W. Big Springs Rd. Apt. J, Riverside, California, 92507, USA
|
|
Keywords:
|
bias ; accuracy ; survey ; recall ; response rates
|
|
Abstract:
|
This paper presents results evaluating methods for gathering information from surveys that may reduce bias and increase response rates. This research is an extension of work by Press (2000), who developed the procedure of asking respondents to provide two values (upper and lower bounds), which enclose what they believe to be the true value for some factual information. These bounds generate Respondent-Generated Intervals (RGIs). In the experiment, five different versions of question constructs for a health survey were mailed to women in two age strata (total N = 3,000). The true value, available in computer-stored patient records, is compared to the answer provided. The study compares estimators based on the RGI versus those used in standard practice to ascertain whether RGI estimates perform better. Additionally, respondent preference for providing an RGI over a point estimate is determined. Question preference is evaluated in light of demographic and salience variables.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2002 program |
