|
Activity Number:
|
329
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #305861 |
|
Title:
|
Improved Evaluation of the Quantitative Survey Response Variance
|
|
Author(s):
|
Patrick Flanagan*+
|
|
Companies:
|
U.S. Census Bureau
|
|
Address:
|
1314 Bristol Ridge Place, Crownsville, MD, 21032,
|
|
Keywords:
|
surveys ; concordance correlation coefficient ; index of inconsistency ; response error ; response variance ; reinterview
|
|
Abstract:
|
When evaluating the survey response variance of quantitative survey data using a reinterview, the Index of Inconsistency has an advantage over the Pearson correlation coefficient and Lin's concordance correlation coefficient in that it estimates the more illustrative ratio of the simple response variance to the overall variance (though under restrictive conditions they all produce the same results). The typical formula for the Index of Inconsistency for quantitative data is biased if the equal means or equal variances assumptions are violated. I propose an improved Index of Inconsistency "testimator" to adjust the estimation for violations of those assumptions, particularly focusing on the equal variance assumption. The results of simulations illustrate the effect of violation of assumptions on four possible estimators, comparatively illustrating their properties.
|
- 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 2006 program |
