Testing whether numeric data contained on an Excel spreadsheet conforms with Benford's law, is a simple audit test that can be performed in just minutes. The audit objective is to reasonably preclude that the data was "made up" or that it contains any unusual patterns or number distributions. This exercise is based upon data contained in the 26th annual report to Congress. To demonstrate the audit procedure, follow these steps:
1. Download the Excel workbook and select the first sheet, which is named Table 3.1.
2. Select the range $A$4:$D$60, and copy it to the clipboard (Ctrl-C or Edit|Copy).
3. Open EZ-R XL and click "Paste". If you don't have EZ-R XL, you can download a full working copy at no cost.
4. Click the name of the column to be analyzed, in this case it is called "#". (You will see an "X" marked in the column named select).
5. From the menu on the left, select "benford".
6. Click the "Run" bottom at the bottom of the dialog.
7. You should now see the Benford chart displayed, along with the measure of fit, the Kolmogorov-Smirnov "d-statistic". In this instance, the d-statistic is .0550, which indicates a fairly good fit. You can also see from a looking at the chart that there is good correlation (except for a minor difference regarding digit "8").
8. Document the results in the work-papers, including the data source, d-statistic and perhaps a print out of the chart.
Below is the end result that you should have obtained...
Note: the graphic image created is saved to your desktop, and named "benford.png".
You can leave (anonymous) comments, if you wish, at our audit blog.
There is also a very similar procedure available which does not require the installation of software. This procedure can be run directly from the Internet. Details and a help file are available.
Should you need further detail results and statistics on this audit procedure, it is recommended that you use EZ-R XL for Windows, which works in a similar fashion.
There is also a web analytics version which can be run directly from the Internet using Excel or other data source (in tab separated value format). View Web Analytics details. All software is provided at no cost.