calculator of sample size

Why sample size matters

Sample size is one of the most important design decisions in research. If your sample is too small, your results will swing around due to random noise and your confidence intervals will be wide. If your sample is too large, you may waste time and budget collecting data you do not need.

A good sample size finds the practical balance: enough precision to support your decision, without unnecessary data collection. That is exactly what this calculator helps with.

What this calculator is designed for

This tool is built for proportion-based questions, where outcomes are represented as percentages. Typical examples include:

• “What percent of users are satisfied?”
• “What percent of customers prefer plan A?”
• “What percent of voters support policy X?”

It calculates minimum sample size for a target confidence level and margin of error, with optional finite population correction and response-rate adjustment.

The formula behind the calculation

For large populations, the base sample size is:

n₀ = (Z² × p × (1 − p)) / e²

• Z = Z-score from your confidence level (e.g., 1.96 for 95%)
• p = expected proportion (as a decimal, e.g., 0.50)
• e = margin of error (as a decimal, e.g., 0.05)

If your population size is limited, the calculator applies finite population correction:

n = n₀ / (1 + (n₀ − 1) / N)

Then it adjusts for non-response:

Invitations needed = n / response rate

How to use each input correctly

1) Population size

If your target group is very large, leaving this blank is fine. If your audience is small and known (for example, 2,000 employees), entering the population will reduce required sample size because each response represents more of the total group.

2) Confidence level

Confidence level reflects how often your method would capture the true value over repeated samples.

• 90%: lower sample size, less strict certainty
• 95%: common default in business and social research
• 99%: higher certainty, larger sample size

3) Margin of error

Margin of error is your desired precision. Smaller margins require much larger samples.

• ±5% is common for directional decisions
• ±3% is stronger for published findings
• ±2% or smaller can be expensive

4) Expected proportion

If you have no prior estimate, use 50%. This is conservative and produces the largest required sample. If historical data suggests a value like 20% or 80%, you can use that to lower sample requirements.

5) Response rate

Most surveys do not get 100% response. If you need 400 completed responses and expect a 50% response rate, you should plan to invite around 800 people.

Worked example

Suppose you are surveying product users:

• Population: 25,000 users
• Confidence level: 95%
• Margin of error: 4%
• Expected proportion: 50%
• Response rate: 40%

You will get a required completed sample and the number of invitations needed after non-response adjustment. This makes planning much more realistic for timelines and outreach.

Reference points (large population, p = 50%, response rate = 100%)

• 95% confidence, ±5% margin: about 385 responses
• 95% confidence, ±3% margin: about 1,068 responses
• 95% confidence, ±2% margin: about 2,401 responses
• 90% confidence, ±5% margin: about 271 responses
• 99% confidence, ±5% margin: about 664 responses

Common mistakes to avoid

• Ignoring response rate: completed responses are not equal to invitations sent.
• Using overly tight precision by default: ±1% often requires massive samples.
• Confusing confidence and accuracy: a high confidence level does not fix biased sampling.
• Skipping representativeness: random and balanced sampling is as important as size.

When you should use a different approach

This calculator is ideal for estimating a single percentage. If you are comparing two groups, running an A/B test, or modeling continuous outcomes, use statistical power analysis instead. Power analysis includes effect size assumptions and Type I/II error tradeoffs, which are not captured in a simple proportion calculator.

Final thoughts

A good sample size plan prevents weak conclusions and reduces wasted effort. Start with the decision you need to make, set a realistic precision target, and account for response rate from the beginning. With those steps, your survey becomes both credible and practical.