Online Program Home
My Program

Abstract Details

Activity Number: 475 - SPEED: Predictive Analytics with Social/Behavioral Science Applications: Spatial Modeling, Education Assessment, Population Behavior, and the Use of Multiple Data Sources
Type: Contributed
Date/Time: Wednesday, August 1, 2018 : 8:30 AM to 10:20 AM
Sponsor: Social Statistics Section
Abstract #330141
Title: Supplemental Nutrition Assistance Program (SNAP) Integrity Monitoring
Author(s): Zhicong Zhao*
Keywords: SNAP; fraud detection; scan statistics; anomaly detection; cluster detection

The Supplemental Nutrition Assistance Program (SNAP) has provided monetary resources to aid the nutritional needs of millions of low-income individuals and families for almost a century. As a result of the Farm Bill of 2008, the SNAP program adopted the Electronic Benefit Transfer (EBT) card to disburse SNAP benefits. Previously benefits were disbursed via paper coupons frequently referred to "Food Stamps". Most EBT transactions have been recorded and stored since 2009. This study is the result of a fiscal year 2015 United States Food and Drug Administration (USDA) SNAP integrity grant. Anomalous price point clusters were identified in the SNAP EBT transaction data consistent with documented cases of fraud and an interesting historical explanation was uncovered. However, identifying all anomalous price point clusters required the development of new statistical methodology. New methods involving scan statistics are presented alongside the results. Overall, program integrity appears very much intact, but significant anomalous activity is detected.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program