Academic journal article The Journal of Government Financial Management

Use of Capture/recapture Estimation in Property Audits

Academic journal article The Journal of Government Financial Management

Use of Capture/recapture Estimation in Property Audits

Article excerpt


Bias-Is the chance that error will be introduced into the sample results.

Estimator-A methodology chosen which will allow the auditor to accurately estimate the number of items not being tracked.

Model Variance-takes into account the bias that may be inherent in any one estimator model.

Population-The total number of objects under review

Sampling Variance-takes into the bias that may be inherent in sample selection.

Standard Error-How far the sample projection might deviate from the true value that could be obtained by a 100 percent validation.

Property audits traditionally use a simple approach-select a list of properties and then go out and find them. Usually the auditors obtain a listing of properties by location and select a statistical sample for review. The results are either positive (the property is found) or negative (the property cannot be located). This approach, while useful, only addresses part of the issue-the property we think we have. But what about property that was acquired but never recorded? From a fraud detection standpoint this is the greatest vulnerability-someone using the organization's funds to acquire property, which is never recorded in the books. How do you validate that which is not recorded? Auditors normally do not have the time to go through all purchase documents and determine what was or was not subsequently recorded in the property records. Given that it is not feasible to check the population of acquired assets, how can the auditor be assured that all property that was purchased was recorded? Practitioners in the fields of biology and zoology have provided the solution to this problem.

What do biology and zoology have to do with property audits? On the surface not much, except for having developed a remarkable population estimation technique known as capture/recapture.

Step 1-The Reverse Sample

A reverse sample is the basis for the capture/recapture analysis. A reverse sample-also called a floor-to-record sample-is simply selecting property and tracing it back to a data base. The items selected for the reverse sample should be randomly chosen and meet the criteria for inclusion in the property data base. For example, a recent property audit at a U.S. Department of Energy (DOE) laboratory found that property was located in several states as well as overseas. This laboratory had established specific criteria (age and value) regarding what types of property would be tracked.5

If the organization has multiple locations it may not be possible to travel to all sites.Therefore, the auditors must be careful to ensure that representative sites are chosen. For example, it was not possible for auditors to travel to the all of the DOE laboratory's locations.Therefore, the auditors initially chose those locations that had the majority of the property and then selected certain other sites at random. This allowed the auditors to have greater assurance that the results of the sample were representative. Once at the locations, property items were chosen purely at random.

However, not all property items should be subject to inclusion in the sample. Only those items, which meet the tracking criteria, should be included in the reverse sample. For example, if the organization has determined that only property of a certain age and acquisition value should be tracked, then the reverse sample should only include items meeting those criteria. Some organizations have different classes of property, such as sensitive and nonsensitive property, with different tracking criteria. Care should be taken to ensure that the tracking criteria are followed exactly, thus ensuring that only material items are included in the reverse sample.

Once the items are chosen they must now be traced back to the property data base. To ensure the accuracy of audit results a three-way test should be performed. That is, if a property identification number is available, compare it with that in the data base. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.