Free Power & Sample Size Calculator
Estimate how many participants you need before launching your study. This online calculator supports one-sample and two-sample mean comparisons using standard normal approximations.
What this power sample size calculator online helps you do
If you are planning an experiment, A/B test, pilot, thesis project, or clinical comparison, one of the first practical questions is: How many observations do I need? This is exactly what a power sample size calculator online is for.
A well-powered study helps you avoid two expensive outcomes:
- Too small a sample: You miss true effects because your study is underpowered.
- Too large a sample: You spend unnecessary time and money recruiting participants.
The calculator above gives a fast estimate using commonly accepted formulas for mean comparisons and normal critical values.
Key inputs explained
1) Effect size (Cohen's d)
Effect size is the expected signal strength. For mean comparisons, Cohen's d is the mean difference divided by the standard deviation. Larger effects require fewer participants, while smaller effects demand larger samples.
2) Alpha level
Alpha is your false-positive threshold (Type I error). Setting alpha to 0.05 means you accept a 5% chance of flagging an effect when none exists.
3) Statistical power
Power is the probability of detecting the effect if it is truly present. Power = 0.80 means an 80% chance of finding the effect at your chosen alpha.
4) One-sided vs two-sided test
A two-sided test checks for differences in either direction and is usually more conservative. A one-sided test assumes direction in advance and generally needs fewer participants.
5) Allocation ratio and dropout
If groups are not equal in size, you can set an allocation ratio. You can also add expected dropout so your enrollment target remains realistic.
How the calculator estimates sample size
For two independent groups with standardized effect size d, the calculator uses:
n1 = (1 + 1/r) × ((zalpha + zpower) / d)2, where r = n2/n1.
For one-sample mean tests it uses:
n = ((zalpha + zpower) / d)2.
Results are rounded up to whole participants and then optionally inflated for expected dropout.
How to use this calculator in practice
- Choose your study design.
- Enter an effect size based on prior studies, pilot data, or minimum meaningful difference.
- Set alpha and power according to your field norms.
- Choose one-sided or two-sided testing.
- Add expected dropout and click calculate.
You will get required group sizes and enrollment targets adjusted for attrition.
Quick interpretation tips
- If required n is very large: your assumed effect may be too small, or power too strict for current resources.
- If required n is surprisingly small: double-check if your effect size assumption is too optimistic.
- If budget is fixed: calculate achieved power separately and document this limitation.
Common mistakes to avoid
- Using effect sizes from overhyped or underpowered previous studies.
- Ignoring dropout in longitudinal or behavioral studies.
- Changing the primary endpoint after planning sample size.
- Using a one-sided test without a strong directional rationale.
Who should use a power sample size calculator online?
This tool is useful for students, data analysts, product teams, social scientists, healthcare researchers, and anyone planning hypothesis-driven experiments. It provides a practical starting point before running full software-based analyses.
Important note
This calculator provides planning-level estimates. For confirmatory research, regulated studies, clustered designs, repeated measures, non-normal outcomes, or complex models, consult a statistician and use specialized software.