Free account
How many animals do you need?
Work out animals per group from the effect you expect, or with the resource equation when there's no effect size to go on. You'll get the total with expected losses and a paragraph for your IACUC protocol.
Your numbers show here.
Enter the difference you expect and its standard deviation. Nothing you type leaves your browser.
The rule
Why your IACUC asks for the numbers
Under 9 CFR 2.31(e)(2) and PHS Policy IV.D.1.b, a protocol gives a rationale for the number of animals it uses. A power calculation, or the resource equation for exploratory work, gives that rationale in numbers.
- 9 CFR 2.31(e)(opens the official source) Checked 10 Nov 2026.A proposal to the IACUC must identify the species and the approximate number of animals (e)(1), and give a rationale for the appropriateness of the species and numbers of animals to be used (e)(2).
- PHS Policy IV.D.1(opens the official source) Checked 10 Nov 2026.Applications to the IACUC identify the species and approximate number of animals (IV.D.1.a) and give a rationale for the appropriateness of the species and numbers used (IV.D.1.b).
- ARRIVE guidelines 2.0(opens the official source) Checked 10 Nov 2026.Percie du Sert N, et al. The ARRIVE guidelines 2.0: Updated guidelines for reporting animal research. PLoS Biol. 2020;18(7):e3000410. Item 2 asks how the sample size was decided.
- Festing and Altman, 2002(opens the official source) Checked 10 Nov 2026.Festing MF, Altman DG. Guidelines for the design and statistical analysis of experiments using laboratory animals. ILAR J. 2002;43(4):244-58.
- Arifin and Zahiruddin, 2017(opens the official source) Checked 10 Nov 2026.Arifin WN, Zahiruddin WM. Sample Size Calculation in Animal Studies Using Resource Equation Approach. Malays J Med Sci. 2017;24(5):101-105.
The math
How the numbers are worked out
Power calculation
For two equal groups compared with a two-sided, two-sample t-test, the calculator finds the smallest number per group that reaches your power, from the exact noncentral t distribution. Small groups need that precision: at an effect size of 2, 5 per group gives 79.1% power, so 80% takes 6.Effect size
The effect size d is the difference you expect divided by the standard deviation. Enter both, or d itself, from 0.1 to 10.Expected losses
Each group gets enough extra animals for the losses you expect: the number to analyze divided by one minus the loss, rounded up. With 10 to analyze and 10% losses, you request 12.Resource equation
For k groups of n animals, E = k(n - 1) should fall from 10 to 20. With 4 groups, that's 4 to 6 animals per group (Festing and Altman, 2002).
Questions
Common questions
- What if I have more than two groups?
- The calculation compares two groups at a time, so it gives the number for each pair you compare. If you'll compare several pairs, you may need a stricter significance level. Ask a statistician which test fits your design.
- Where does the effect size come from?
- From a pilot study or a published paper that measured the same outcome. Use the smallest difference that would matter biologically (Festing and Altman, 2002).
- When should I use the resource equation?
- For exploratory work, where no effect size can be estimated in advance. When you can estimate one, a power calculation is the better justification (Arifin and Zahiruddin, 2017).
- Does the calculator keep my numbers?
- Your numbers stay in this browser, so they're there when you come back, and we don't receive them.
More features
Also in the free account
- Exempt checkFour questions give the likely IRB review path, the exempt or expedited category and what to file.Check your study
- Risk heat mapRate risks for likelihood and impact and see where each one sits on a 3 by 3 map.Build a heat map
- FDA deadline calculatorCount the days for a Form FDA 483 response, a Field Alert Report or a records request.Count the days