Glossary›Assurance, audit evidence and working papers›Sampling (assurance)
Glossary term
Cluster A · A27
Tier 2 · differentiator
Sampling (assurance)
Definition
Sampling is testing a selection of items rather than every item, then drawing a conclusion about the whole population. Assurers use statistical or judgemental selection, weighting toward large or unusual items. If a sample yields errors, the assurer projects the error rate across the population and may expand testing.
On this page
In practice
Sample sizes in climate assurance are driven by assessed risk and the quality of the process. A well-controlled, reconciled emissions process attracts a smaller sample than a spreadsheet-based one, which is where the cost difference between a well-prepared and a poorly-prepared file actually shows up.
What the assurer does with it
The assurer defines the population, which for emissions is often the hardest step because the population is not held anywhere. They then select, test and evaluate. Projection is the part entities dispute: two errors in a sample of forty becomes an estimated error across the whole population, which can be many times larger than the errors found. The answer to a projected error is to test more of the population yourself, not to argue about the projection.
Sources
1
ASSA 5000 General Requirements for Sustainability Assurance Engagements
AUASB
2
Review status
Review required
Last reviewed
15 September 2026
Editorial pass, unsigned
Reviewer required
Registered company auditor
Next scheduled review
1 July 2027
Part of
Cluster A, Assurance, audit evidence and working papers
47 terms on what an assurance provider tests, what they accept as evidence, and what a preparer has to be able to produce.
Related terms
The test that decides how large a sample must be
The procedure most often applied to a sample
What a population that cannot be defined can become
Related questions
What does an assurance provider actually test in year one?
−
In year one they test the disclosures inside the ASSA 5010 first-year scope: Scope 1 and Scope 2 emissions, governance disclosures, and specified strategy paragraphs on climate risks and opportunities. The work runs in a predictable order, from understanding your reporting process and your boundary, through sampling source documents and reperforming calculations, to written representations at the end.
What happens if the assurance provider disagrees with our numbers?
+
Disagreement is normally resolved before it reaches the conclusion: the practitioner raises a query, you produce more evidence or adjust the number, and the file moves on. A modified conclusion only arrives if you decline to adjust something material, or if they cannot obtain the evidence they need. The second case is far more common in a first year and is entirely preventable through documentation.
Other terms in this cluster
Sampling (assurance)