Definition
A random sample requires a numbered or ordered population, a source of random numbers, and a documented mapping from the numbers to the items. Systematic selection, which takes every nth item from a random start, is accepted by AU-C 530 as a way to draw a sample when the population is not arranged in a pattern. Statistical sampling requires random selection, or systematic selection with a random start.
Why it matters in an audit or review
Randomness is what lets the auditor argue that the sample represents the population and evaluate results statistically. Selection that is influenced by convenience, by what the client pulled, or by the auditor's own preferences brings bias that sample size cannot repair.
Example
A population of 3,812 vouchers (illustrative) is numbered from the system listing, which the auditor reconciled to the ledger total. A random number generator is set to produce 60 integers between 1 and 3,812 with seed 2026 documented in the file. The auditor pulls those voucher numbers and any duplicates are redrawn.
Common mistakes
- Sampling from a listing the client produced that was not agreed to the general ledger.
- Not saving the seed or parameters, so the selection cannot be reproduced.
- Dropping unwieldy items and replacing them with easier ones.
FAQ
Is systematic selection random?
Only if the starting point is random and the population has no pattern that aligns with the interval.
Sources
- AICPA, AU-C Section 530, Audit Sampling (random selection, systematic selection and statistical sampling)
Definitions paraphrase AICPA and FASB guidance; read the standard itself before relying on a requirement. Example figures are illustrative. This page is guidance, not professional advice. Report an error