|
Activity Number:
|
127
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 4, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #301279 |
|
Title:
|
Using Continuous Variables as Modeling Covariates for Net Coverage Estimation
|
|
Author(s):
|
Vincent T. Mule*+ and Don Malec and Jerry Maples and Teresa Schellhamer
|
|
Companies:
|
U.S. Census Bureau and U.S. Census Bureau and U.S. Census Bureau and U.S. Census Bureau
|
|
Address:
|
15007 Nebraska La, Bowie, MD, 20716,
|
|
Keywords:
|
Dual System Estimation ; Logistic Regression
|
|
Abstract:
|
The Census Bureau uses dual system estimation as one method to evaluate the coverage of the decennial census. This estimation for previous censuses has used post-stratification to minimize the impact of correlation bias on the population estimates. For 2010, we are planning on using logistic regression modeling instead of post-stratification cells. Logistic regression gives us the option of using variables in the model as continuous variables instead of having to form groupings. This research present our initial results of the impact of using certain continuous variables in the model development and the resulting population estimate.
|
