Statistical models were designed to estimate national improper payments due to certification error on an annual basis using district-level data. This enables FNS to update its estimates of national improper payment rates for the NSLP and SBP in future years without having to conduct full rounds of primary data collection.

The second Access, Participation, Eligibility and Certification Study (APEC II) included a follow-on report that provided statistically-derived state-level estimates of school meals erroneous payments. However, while APEC II provided a rough indicator of relative risk for groups of states (e.g., higher than average, about average, lower than average), it was not a state-representative direct measure, and creating actual annual measures of such erroneous payments at the state level using APEC methodology is cost-prohibitive. This report explores alternative approaches to developing measurement-based state-specific estimates that are responsive to year-to-year changes in the actual underlying rate in each state. It also provides cost and burden estimates for the implementation of each of these methods.

This is the 11th in a series of annual reports that examines the administrative accuracy of eligibility determinations and benefit issuance for free or reduced-price meals in the National School Lunch Program.

This document summarizes the preliminary findings of two projects being completed by Mathematica Policy Research, Inc. on the subject of certification accuracy in the National School Lunch Program: Evaluation of the National School Lunch Program Application/Verification Pilot Projects: Findings on Deterrence, Barriers and Accuracy, and Case Study of Verification Outcomes in Large Metropolitan Areas.

From July to September 2002, FNS reviewed the free and reduced price eligibility determination process (i.e., application, verification, reapplication, meal ticket status) for each of 3,474 applications selected for verification in 14 large school food authorities in the 2001-02 school year. These SFAs enroll nearly one million children, among whom 45 percent were approved for free meals and 7 percent were approved for reduced price meals as of Oct. 31, 2001.