Attribute sampling
Is the control working?For outcomes that are either a pass or a fail, such as checking whether invoices were authorised.
A practical audit sample size calculator
Audit sampling.
A clearer plan.
Calculate an audit sample size for attribute testing, variables mean precision or monetary unit sampling. Review the assumptions and download your sampling plan.
Free to use. No account. Calculations stay in your browser.
THE SAMPLING NOTE Illustrative example
A plan you can explain.
59
items to testAttribute sampling
Population5,000 items
Confidence95%
Expected deviation0%
Tolerable deviation5%
Sample proportion 1.18%
Start with the assumptions. Keep the reasoning.
01 / THE CALCULATOR
Find your sample size.
Set the parameters for your engagement.
See the result and the assumptions behind it.
Your sample plan
Example calculationAttribute sampling · ISA 530
—items to test
Your calculation will appear here.
See the calculation assumptions
A number is a starting point.
Use the result alongside your firm’s methodology and professional judgement. The standard reference does not certify compliance. Read the calculation basis.
02 / CHOOSE YOUR APPROACH
The method follows the question.
Three ways to approach your population.
Choose the one that fits the test.
Variables: mean precision
What is the numerical difference?A normal-based precision estimate for a numerical population mean, using supplied variability and an absolute margin in the same units.
Monetary unit sampling
Where might value be overstated?For positive recorded values and overstatement testing. Sizes represent monetary selections; a record can be selected more than once.
03 / KEEP THE CONTEXT
Clear assumptions.
Better conversations.
Good sampling starts with a defined population and an appropriate objective. The calculation is one part of the judgement you document.
Check the version of each standard applicable to your engagement.A FEW USEFUL ANSWERS
Before you begin.
The practical details, without the small print.
Do I need an account or a subscription?
No. The calculator is free to use and runs directly in your browser.
What happens to my calculation inputs?
The parameters are processed in your browser. Your audit inputs and results are not collected, and you do not upload client records. Analytics is enabled by default and counts page views as daily totals to help improve the site, without visitor profiles or tracking cookies. You can turn it off at any time in Analytics settings. Read the cookies and privacy details. Downloading a result creates a text file on your device.
Can I keep the result in my working papers?
Yes. Download or copy the result to keep a record of the method, inputs, factors and calculated sample size. Review it against your firm’s documentation requirements.
Is the calculated sample size a mandatory minimum?
No. It is the output of the stated statistical planning method. Neither standard prescribes a universal 25- or 60-item minimum. Your audit objective, the population and professional judgement still matter.
Why can large populations have the same sample size?
For attribute sampling, confidence and the expected and tolerable deviation rates drive the estimate. With 95% confidence, no expected deviations and a 5% tolerable deviation rate, the large-population binomial estimate is 59 items (a Poisson approximation gives 60). Large populations can therefore give the same result. Smaller populations use the without-replacement hypergeometric calculation. Full examination is recommended when the plan requires every available item.
Does this select the actual items I should test?
No. This tool calculates a sample size. Selecting items from your population and evaluating the testing results are separate steps.
YOUR NEXT SAMPLING DECISION