miatzSign in

Stratifying a population before the audit sample is drawn

Splits a large transaction population into sub-populations that share characteristics — value band, approver, location, posting route — so the auditor can size and draw a sample within each, and separately surfaces items whose attributes argue for specific testing outside the sample altogether.

Effort
Days of work
Skill level
Some technical skill
Organisation size
Mid-market
Value
Time saved, Quality

Tools named for this

  • A clustering or stratification routine over the ledger extract
  • A notebook or spreadsheet the engagement file can retain and re-run
  • The client's own general-ledger export

What to check before you ship it in India

  • SA 530 paragraph 8 requires the auditor to select items so that every sampling unit in the population has a chance of selection. A model that returns only the items it scores riskiest is not audit sampling at all — it is selection of specific items, and the file has to say so and conclude accordingly.
  • A stratification the engagement cannot re-run next year leaves a working paper whose basis nobody can reproduce, which is worse than a cruder split that anyone on the team can rebuild.

Sources

Every claim on this page traces to one of these, on the date it was read.