When the Auditor says they plan to conduct “sampling” to determine the extent of review or level of acceptance, pay attention! Don’t wait until sampling is completed.
(From IRS Audit Techniques Guide)
In certain cases, use of sampling methodologies may be a desirable examination tool. Sampling is quite often used in research credit cases for greater audit efficiency. In fact, sampling should be considered in a research credit case whenever excessive amounts of time or resources are anticipated in examining all of the taxpayer’s expenses or projects.
When a sampling technique is contemplated, the examiner, along with a Computer Audit Specialist Program (CAS), should meet with the taxpayer and design a sampling plan that will result in an acceptable number of items to examine. The formulation of questions, questionnaires, or surveys to be used should be customized to each taxpayer’s specific circumstances. Specific sampling procedures should be thoroughly documented. While a judgment (non-statistical) sample will often require less examination work than a statistical sample, a judgment sample generally requires taxpayer assent as to how results the judgment sample results will be applied, while a valid statistical sample does not. Agreement between taxpayer and examiner should be negotiated in the early stage of the examination, and should generally be set forth in a closing agreement (Form 906), binding the parties to apply the results of the agreed-upon methodology.
Check out this link: