Sample Size Calculator Online
Find the survey sample size you need for a chosen confidence level and margin of error. Your numbers stay on your device.
The Sample Size Calculator runs entirely in your browser. The confidence level, margin of error, population size, and every computed result stay on your device and are never uploaded to ArrayKit.
Open the Confidence Interval Calculator
About Sample Size Calculator
Sample Size Calculator finds how many respondents a survey needs to hit a target confidence level and margin of error, using Cochran's formula for estimating a population proportion. Enter a confidence level (or pick a 90/95/99% preset), a margin of error, and it returns the minimum sample size, the z-score behind it, and the proportion assumed. Turn on the finite population correction when the total audience is known and bounded — for example a company's 4,000 employees — to shrink the requirement further, or set an expected proportion if a prior study already gives you a rough split. It is built for researchers, product and UX teams, and students planning a survey, poll, or A/B test who need a defensible sample size before collecting data. Every calculation runs in your browser — no survey parameters are uploaded.
Features
- Computes the required sample size from confidence level and margin of error with Cochran's formula
- One-click presets for the common 90%, 95%, and 99% confidence levels, or any custom value
- Optional finite population correction when the total population size is known and bounded
- Optional expected proportion input, defaulting to the conservative 50/50 split when unknown
- Shows the exact z-score, corrected sample size, and proportion used behind the result
- Copies a plain-text summary of the confidence level, margin of error, and required sample size
- Handles very small pilot studies and very large population sizes without losing precision
- Runs entirely in your browser — survey parameters are never uploaded
How to use the Sample Size Calculator
- Enter your desired confidence level, or pick a 90/95/99% preset
- Enter the acceptable margin of error as a percentage
- Optionally turn on a known population size or an expected proportion
- Read the required sample size and copy the summary
Example
Input
95%, ±5%
Output
Sample size: 385
At 95% confidence and a ±5% margin of error with an unknown 50/50 proportion, Cochran's formula needs 385 respondents.
Common errors & troubleshooting
- Margin of error typed as 0.05 instead of 5, producing an enormous sample size. — Enter the percentage form, e.g. 5 for a ±5% margin — the Sample Size Calculator already converts it internally.
- Confidence level typed as 0.95 instead of 95. — Enter the percentage (95), not the decimal probability; the field expects 0–100.
- Required sample size looks far larger than the total number of people being surveyed. — Turn on 'Limit to a known population size' and enter the total population — the finite population correction reduces the requirement for small, bounded groups.
- Result changes a lot after switching the expected proportion. — 50% is the most conservative (largest) sample size; only enter a different expected proportion if a prior survey or pilot study already supports it.
Frequently asked questions
- What is a Sample Size Calculator?
- It is a tool that applies Cochran's formula, n = z² · p(1 − p) / e², to work out the minimum number of survey responses needed for a chosen confidence level and margin of error, optionally correcting for a finite population.
- How does the Sample Size Calculator handle a known population?
- When you turn on 'Limit to a known population size' and enter the total, the calculator applies the finite population correction, which reduces the required sample size for small or bounded groups such as a single company's staff.
- What proportion should I use in the Sample Size Calculator if I don't know it?
- Leave the expected proportion off. The calculator then assumes 50%, which is the most conservative choice and yields the largest — and therefore safest — sample size.
- Does the Sample Size Calculator upload my survey numbers?
- No. The confidence level, margin of error, population size, and every result are computed locally in your browser and are never sent to ArrayKit or anywhere else.
- Why does a 99% confidence level need a bigger sample than 95% in this calculator?
- A higher confidence level uses a larger z-score, which widens the formula's numerator, so the Sample Size Calculator returns a larger required sample to be more certain the true value falls within the margin of error.
- Can the Sample Size Calculator be used for a customer survey with 10,000 people?
- Yes. Enter your confidence level and margin of error, then turn on the population size option and enter 10,000 so the finite population correction is applied to the result.
Related tools
All ArrayKit tools