|
Activity Number:
|
71
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #303766 |
|
Title:
|
Investigation of Variance Properties of Noise-Infused Estimates for the Survey of Business Owners (SBO)
|
|
Author(s):
|
Irene Brown*+ and Marilyn Balogh and Anthony Caruso and Beth Schlein and Katherine J. Thompson
|
|
Companies:
|
U.S. Census Bureau and U.S. Census Bureau and U.S. Census Bureau and U.S. Census Bureau and U.S. Census Bureau
|
|
Address:
|
8524 Willow Bend Ct APT 14, Springfield, VA, 22152,
|
|
Keywords:
|
variance estimation ; disclosure avoidance ; noise infusion ; random group variance estimator
|
|
Abstract:
|
Beginning with the 2007 data release, the Survey of Business Owners (SBO) will employ random noise instead of cell suppression to perform disclosure avoidance processing. This paper reports the results of a simulation study conducted to assess the impact of noise infusion on the statistical properties of the SBO variance estimates. We present two alternative methods of estimating the additional variance component due to noise infusion, while using the SBO random group variance estimator to estimate the sampling variance component. We examine the coverage and bias properties of the alternative variance estimators for level estimates and percentage change estimates over repeated samples. Our study showed that the effects on the variance estimates over repeated samples due to the addition of noise was negligible for the SBO estimates, due to the survey's large sampling variances.
|
