Record sampling (attributes sampling)

Record sampling is a statistical sampling method for estimating the rate of deviation from a prescribed control in an account or class of transactions.

If your analysis of the sampled data will yield a yes/no or pass/fail result for each record, then you should use record sampling.

Tip

For a hands-on introduction to the end-to-end process of record sampling in Analytics, see Record sampling tutorial.

How it works

Record sampling allows you to select and analyze a small subset of a population, and based on the result estimate the rate of error or deviation from an internal control for the entire population.

You can then compare the estimated rate to the rate that you judge is acceptable, and make a determination regarding the control.

Record sampling supports making this sort of statement:

  • There is a 95% probability that the deviation rate from the prescribed control does not exceed 4.35%, which is below the tolerable deviation rate of 5.0%. Therefore the prescribed control is operating effectively.

Overview of the record sampling process

Caution

Do not skip calculating a valid sample size.

If you go straight to drawing a sample of records, and guess at a sample size, there is a high likelihood that the projection of your analysis results will be invalid, and your final conclusion flawed.

The record sampling process involves the following general steps:

  1. Calculate the required sample size

  2. Choose a sample selection method:

  3. Draw the sample of records
  4. Perform your intended audit procedures on the sampled data.
  5. Evaluate whether the observed rate of control deviation in the sampled data represents an acceptable or unacceptable deviation rate in the entire population.

How record sampling selects records

Record sampling uses the following process for selecting sample records from an Analytics table:

  • When you specify record sampling, the sampling unit is an individual record with a unique record number. You do not specify a particular field.
  • Using one of the sample selection methods, Analytics selects samples from among the record numbers. The selected records are included in the sampling output table.

Example

In a table with 100 records, Analytics could select the following record numbers:

  • 9
  • 13
  • 40
  • 52
  • 78
  • 91
  • 99

Unbiased sample selection

Record sampling is unbiased and it is not based on the amounts contained in a record. Each record has an equal chance of being selected for inclusion in the sample. A record containing a $1000 amount, a record containing a $250 amount, and a record containing a $1 amount all have the same chance of being selected.

In other words, the probability that any given record will be selected has no relation to the size of the amount it contains.

Considerations

Record sampling is appropriate for use with controls testing that results in a yes/no, or pass/fail, result. In controls testing, you are more concerned with the rate of errors in the total population than with the cumulative monetary amount of the errors.

Because record sampling does not consider the amounts contained by records, there is a significant chance that large monetary transactions may be excluded from the sample.