Math & Scientific
Sample Size Calculator
Sample Size — it uses n = z²·p(1 − p) ÷ e² for the margin of error and confidence level you set. This page runs that same calculation on your numbers, in your browser: no signup, no uploads, nothing stored.
Use Sample Size Calculator to work out a sample size: give it your population size, confidence level and margin of error, press Calculate, and read off the required sample size — all worked out on your device.
About this calculator
Get sample size right and the rest of the task gets easier. It uses n = z²·p(1 − p) ÷ e² for the margin of error and confidence level you set. It matters when you need the exact relationship checked quickly rather than worked through by hand. For a 95% confidence level and 5% margin on a large population, the rule-of-thumb sample size is ≈ (1.96² × 0.25) ÷ 0.05² = 385 . The tool above does the same computation in your browser — free, private, nothing stored.
How to use
- Enter a positive whole-number population size, or leave it blank for a very large population.
- Enter the confidence level (for example 95).
- Enter the margin of error you can accept (for example 5).
- Press Calculate to see the required sample size.
Why use Sample Size Calculator
Fast
Calculations run in your browser. No round trip to a server.
Private
Your inputs never leave your device. Nothing is uploaded.
Free
No signup, no paywall, no ads-in-results, no watermarks.
Accurate
Uses standard, well-established formulas; check critical results independently.
Mobile-ready
Optimised for phones, tablets, and desktops alike.
Transparent
Honest disclaimers wherever browser limits or regional rules apply.
Common uses
Use it when preparing worksheets
Use it for verifying answers from a textbook
Use it for science-fair and project numbers
Use it for quick mental-math cross-checks
Sample Size in statistics and data coursework
Sample Size in research and report calculations
Technical notes
The form estimates a normal-approximation sample count for a population proportion using a fixed assumed proportion of 50%. Confidence levels use a predefined z table; this is not a power calculation for a difference, a mean or a case-control study. For a finite population N, it uses n0 ÷ [1 + (n0 − 1) ÷ N] before rounding upward. Population counts must be positive safe integers; unrepresentable calculated counts are rejected. The estimate does not adjust for design effects, nonresponse or bias.
Inputs are taken exactly as entered, so check for typos before trusting an answer.
All arithmetic runs in your browser using double-precision floating point, so extreme inputs may show tiny rounding differences.
FAQ
Does it apply a finite population correction?
Yes. If you enter a population size, the required sample is adjusted downward; leaving it blank assumes an effectively infinite population.
What proportion does it assume?
The form fixes the assumed proportion at 50%, which maximizes p(1 − p) in this proportion-estimation model. That does not guarantee a valid study design or account for nonresponse, clustering or bias.
What do confidence and margin of error control?
Higher confidence or a smaller margin of error both increase the sample size you need.
Does the Sample Size Calculator send my data to a server?
No. The Sample Size Calculator runs entirely in your browser, so nothing you enter is uploaded, stored, or shared.
Worked example
With 95% confidence, z = 1.96 and a 5 percentage-point margin, the large-population model gives n0 = (1.96² × 0.25) ÷ 0.05² = 384.16. Rounding upward gives 385. The form performs this calculation locally; it does not establish whether the normal approximation or sampling design is appropriate for a particular study.