Appendix II: Objectives, Scope, and Methodology
someone other than the cardholder was on the sales invoice, packing slip, bill of lading, or any other shipping or receiving document indicating receipt.
For statistical sample and data-mining transactions containing accountable or highly pilferable property, we performed an inventory to determine whether executive agencies maintained accountability over the physical property items obtained with government purchase cards. Because each agency had its own threshold for accountable property, we were not able to test accountable property against each agency’s threshold for this governmentwide audit. Consequently, we defined accountable property as any property item exceeding a $350 threshold and containing a serial number. We defined highly pilferable items as items that can be easily converted to personal use, such as cameras, laptops, cell phones, and iPods. We selected highly pilferable property at any price if it was easily converted to personal use.
The purchase card data provided by the banks did not always contain adequate details to enable us to isolate property transactions for statistical testing. Because we were not able to take a statistical sample of these transactions, we were not able to project failure rates for accountable and pilferable property. Consequently, our tests of property accountability were performed on a nonrepresentative selection of property that we identified when a transaction selected for statistical sampling or data mining contained accountable and pilferable property. For these property items, we identified serial numbers from supporting documentation provided by the agency and, in some cases, by contacting the vendors themselves. To minimize travel costs associated with conducting a physical inventory governmentwide, we requested that each agency provide photographs of the property items, which we compared against the serial numbers originally provided. When we were unable to obtain serial numbers from supporting documentation or from the vendors, we gave the agency the benefit of the doubt and accepted the serial numbers shown in agency-provided photographs as long as the product(s) and quantity matched. In some isolated instances, we performed the physical inventory ourselves.
To identify examples of fraudulent, improper, and abusive purchase card activity, we data mined purchase card transactions from July 1, 2005, through September 30, 2006. This period contained an additional 3 months of data subsequent to the period included in our statistical samples. For data-mining purposes, we also included transactions from federal agencies
GAO-08-333 Governmentwide Purchase Cards