Content area
Full Text
TECHNOLOGY WOPRKSHOP
Learn the formulas, functions, and techniques that enable efficient Benford analysis of data sets.
In 2015, Atlanta's Six Flags White Water theme park opened a new 10-story slide called the Dive Bomber. The ride begins with a virtual free fall nearly straight down before it gradually levels out toward the bottom. Most people who look at the Dive Bomber see a thrill ride-one they may or may not want to experience. CPAs who gaze upon the Dive Bomber may see something else-an image they could possibly use to detect and prevent fraud.
What could a water slide possibly have to do with fighting fraud? The answer lies not in the nature of the slide, but in its shape. The Dive Bomber's curve closely matches the curve Frank Benford discovered in 1938 that depicts mathematical expectations related to naturally occurring (or nonfabricated) numbers (see the sidebar, "The History of Benford's Law"). Affixing the shape of this water slide in your mind now may help you detect fraudulent data later. Though the merits of the analogy may not be immediately obvious, this article explains Benford's Law, how it relates to the shape of a water slide, and how to make simple Excel-based calculations that can help spot and stop fraudulent activity.
WHEN TO USE BENFORD'S LAW TO SPOT FRAUD
Briefly explained, Benford's Law maintains that the numeral 1 will be the leading digit in a genuine data set of numbers 30.1% of the time; the numeral 2 will be the leading digit 17.6% of the time; and each subsequent numeral, 3 through 9, will be the leading digit with decreasing frequency. This expected occurrence of leading digits can be illustrated as shown in the chart "Benford's Law."
The resulting curve pictured in this green bar chart closely resembles a steep water slide and is sometimes referred to as the Benford curve. Today, armed with any version of Microsoft Excel, CPAs can count the leading digits contained in virtually any data set, chart the findings, and compare the results to Benford's curve to see if that data set obeys the expectations set forth by Benford's Law.
The Excel-based procedures described in this article for counting and charting a data set's leading digits are the same...