e. Statistical Sampling Guidelines

Statistical Sampling Guidelines

Download the IRS Statistical Sampling Memorandum here: IRS Memo
 

Without solid contemporaneous project accounting data, taxpayers who claim the R&D Credit often attempt to argue a statistical sampling of their project information to minimize the burden of recreating project documentation.  However, under Tier I, the IRS has focused on a flaw in this strategy known as “biased judgment sampling“.

In light of this strategy, on November 3, 2009, the IRS published a memorandum to establish guidelines in evaluating sampling estimates by taxpayers.  The memorandum does not supersede any formal regulations (e.g., Rev. Proc. 2007-35, 2007-1 C.B. 1349, and Rev. Proc. 2004-29, 2004-1 C.B. 918) that address the specific application of statistical principles, and does not relieve you of your responsibility to maintain any documentation required by section 6001 of the IRC.

If you attempt to use a statistical sample in an audit, the memorandum instructs the auditor to determine:

1. Have you appropriately used a probability sample?  Probability samples are generally considered appropriate if there is a compelling reason for their use and you cannot reasonably obtain more accurate information.  However, probability samples are generally not considered appropriate if more accurate evidence is available, or if your use of sampling does not conform to Generally Accepted Accounting Principles (GAAP).
2. Does your final answer represent a valid estimate?  In general, an estimate from your sample is considered valid if all of the following conditions are met:
   • You have maintained all of the proper documentation to support the statistical application (as defined in Attachment A of the memorandum).
   • The estimate is based on a statistical sample using either a simple random sampling method or stratified random sampling method.  The total sample size must be at least 100 units, and each stratum for which a population estimate is made should contain at least 30 sample units.  The confidence limits will be determined using the Hypergeometric, Poisson, or Binomial distribution.
   • The estimate is computed at the least advantageous 95% one-sided confidence limit. The “least advantageous” confidence limit is either the upper or lower limit that results in the least benefit to the taxpayer.

These guidelines address only the statistical requirements that must be met for a probability sample to meet preliminary acceptance.  Any fatal error in statistical methodology which renders the probability sample invalid will preclude the use of any statistical estimate based on the sample.

Here’s a better idea:  Document all of your research activities as you go, and you’ll never have to consider these sampling guidelines when defending your R&D Credit.  Not sure how to do it cost effectively?  Take Titan Armor’s free trial or call us for more information.