|DATE:||August 2, 2006|
|SUBJECT:||Calculating the Program Access Index with the American Community Survey|
All Regional Directors
The final regulation establishing Food Stamp High Performance Bonuses was published in the Federal Register on Feb. 7, 2005. The regulation established the Program Access Index as one of the criteria – among others – on which states could qualify for a share of the funds made available for bonus awards. The Program Access Index is a ratio of the number of people served by the Food Stamp Program in each state to the number of low-income people in that state. The regulation provided for use of the Current Population Survey as the basis for estimating the number of low-income people, but reserved the right to use new and better data should it become available.
The preamble to the final regulation indicated that the Food and Nutrition Service (FNS) was particularly interested in using the American Community Survey (ACS) in place of the Current Population Survey (CPS) because of its substantially larger sample. Based on a comparative analysis of the merits of the two surveys (attached), we have determined that the ACS provides a better basis for the calculation of the Program Access Index. In summary:
- The fully implemented ACS sample is substantially larger than the CPS sample (consisting of approximately 3 million households starting in 2005, compared to roughly 100,000 households in the CPS).
- The larger ACS sample produces much more precise state estimates, with substantially smaller confidence intervals. Because the CPS-based calculations are based on fairly small samples in some areas, the reliability of those estimates – particularly the estimated change from one year to the next – has been a matter of concern, especially for smaller states. Using an index based on the American Community Survey will generate substantially better and more stable estimates for all states.
- Response rates to the ACS are higher, in part because cooperation is required by law, and higher response rates should produce better quality data.
The Census Bureau has announced plans to release poverty data based on the full American Community Survey on Aug. 29, 2006. Given the strengths of the ACS relative to the CPS, FNS will calculate the 2005 Program Access Index using the new ACS data. FNS will also re-calculate the Program Access Index for 2004 using the ACS for the purpose of identifying states with the most improved performance between 2004 and 2005. High Performance Bonus Awards to states with the highest and most improved PAI will be announced in September 2006.
Arthur T. Foley
Program Development Division
The contents of this guidance document do not have the force and effect of law and are not meant to bind the public in any way. This document is intended only to provide clarity to the public regarding existing requirements under the law or agency policies.