|
Small Area Income
& Poverty Estimates
Model-based Estimates for States, Counties, & School Districts |
|
Small Area Income and Poverty Estimates (SAIPE) Program
Model Changes for Release of 2005 Estimates
The SAIPE program produces model-based estimates of poverty and median household income for states and counties and poverty and population estimates for school districts. The following items provide a high-level description of major changes that have been incorporated into the estimation procedures for the release of the 2005 estimates.
States and Counties: Incorporate American Community Survey (ACS) data into the model-based estimation process for states and counties by replacing the current poverty data, measured by the Current Population Survey (CPS) Annual Social and Economic supplement (ASEC), with ACS measures.
-
The National Academy of Sciences recommended changing to the ACS for SAIPE estimates as soon as possible.1
- ACS now provides the official direct state and county estimates of poverty.
- ACS sample size exceeds CPS ASEC (3 million versus 100,000 addresses, respectively), and ACS samples all counties compared with approximately forty percent sampled in the CPS ASEC. Also, the much larger sample size of the ACS results in fewer counties with direct estimates of zero poor children - a limitation in the CPS ASEC based model.
- At the county level, one year of ACS direct survey data can be used, whereas three years of CPS ASEC data were needed.
- Smaller margin-of-error associated with ACS direct estimates, compared to CPS ASEC, produce county estimates with substantial variance reduction over the current method.
School Districts: A new shares method using Federal Tax Information (FTI) from the IRS to better reflect current sub-county poverty distribution for school districts.
- Decreases reliance on Census 2000 by using more current data at the school district level.
- The use of FTI allows sub-county distribution of poverty to vary during intercensal years.
External Review: In September 2007 an external review of the proposed ACS model change was conducted. Technical recommendations were provided, and the switch to using ACS data was supported.
For more detailed information please see
Estimation Procedure Changes for 2005 Estimates.
1National Academy of Science, National Research Council (2000). Small Area Income and Poverty Estimates: Priorities for 2000 and Beyond (pp:117).
National Academy of Science, National Research Council (2007). Using the American Community Survey: Benefits and Challenges (pp: 69-80; 215-216).
Source: U.S. Census Bureau, Data Integration Division, Small Area Estimates Branch
For assistance, please contact the Demographic Call Center Staff at 301-763-2422 or 1-866-758-1060 (toll free) or visit
ask.census.gov for further information.