|
Activity Number:
|
338
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, August 14, 2002 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Government Statistics*
|
| Abstract - #300362 |
|
Title:
|
The American Community Survey and Small Area Estimation
|
|
Author(s):
|
Warren Brown*+
|
|
Affiliation(s):
|
Cornell University
|
|
Address:
|
292 Caldwell Hall, Ithaca, New York, 14853, U.S.A.
|
|
Keywords:
|
estimates ; ACS ; households ; neighborhoods
|
|
Abstract:
|
The American Community Survey (ACS) will sample 3% of households on an annual basis. The sample design will allow for estimates down to Census Tracts and Block Groups based on five-year rolling samples. The survey will estimate characteristics of housing units, households, and persons controlling to independent estimates of the total number of these. The Census Bureau currently uses a distributed housing unit method to prepare estimates of population for subcounty municipalities, using "persons per household" as the link between occupied housing units and the household population. We explore ways to improve upon the estimates of persons per household. We believe there are two keys. The first is to determine how best to allow the ACS results on occupancy rates and persons per household to be used in preparing the estimates of household population that in turn will be used as controls for the survey estimates without contaminating the results. The second is in classifying Block Groups by their stage in neighborhood household status, and "borrowing strength" from similar Block Groups to make reliable estimates from the 3% samples for large collections of Block Groups.
|
- 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 |
