A3-M8-Sampling, Part 2

An auditor most likely would stratify a population in meaningful groups if the population has highly variable recorded amounts. The auditor may be able to reduce the required sample size by separating items subject to sampling into relatively homogenous groups on the basis of some characteristic related to the specific audit objective.

In selecting an appropriate sample for a substantive test, the auditor most likely would stratify the population if the auditor plans to give greater representation to large recorded amounts. Stratification is often used when the population has highly variable recorded amounts and involves grouping the population into smaller groups, such as grouping large recorded items as a group.

Ratio estimation is most effective if there is a correlation between book values and audit amounts.

A stratified sample generally is more efficient than an unstratified sample since the population is classified in a manner that emphasizes the higher dollar value items. The result is an estimate having a desired level of precision with a smaller sample size.

Stratification involves the grouping of transactions sharing some characteristic (such as recorded amounts). The goal of stratification is to ensure selection of items for which potential misstatements may individually equal or exceed tolerable misstatement. Thus, the auditor should stratify the sample such that the unusually large transactions are selected.

Sample size has a direct relationship with the expected misstatement, standard deviation (population variability), and assessed level of risk. Sample size has an inverse relationship with the tolerable misstatement and the acceptable level of risk.

After identification of misstatements in the sample, the next step is to project the detected error to the entire population.

After performing an audit data analytic procedure, the auditor should group the data for items that have common characteristics. For the groups that contain possible misstatements that are not clearly inconsequential (in aggregate), the auditor should perform further analysis of the groups to determine whether they may represent actual misstatements.

In PPS sampling, the auditor controls the risk of incorrect acceptance by specifying that risk level for the sampling plan. The inputs for PPS are tolerable misstatement, risk of incorrect acceptance (reliability), and the recorded amount of the population being sampled.

PPS sampling is a method designed to estimate overstatement errors. Zero balances, negative balances, and understated balances require special design considerations.