What this calculator does
Sample size is the number of completed responses a survey needs before its margin of error is meaningful. More confidence or a tighter margin requires more responses. A finite population reduces the requirement slightly. The invitation count is separate because not everyone responds, so the calculator divides by the response rate and rounds up.
How to use it
- Choose a confidence level. 95% is the common default.
- Enter the margin of error you can accept, in percentage points.
- Enter the population size, or 0 if it is unknown or very large.
- Enter the response rate you expect.
- Read required responses and invitations needed.
How it works
Example calculations
Example 1: 95%, ±5%, unlimited
Required responses are 385. At a 100% response rate, invitations are also 385.
Example 2: 90%, ±3%, unlimited
Required responses are 752. The tighter margin and lower confidence trade off against each other.
Example 3: 95%, ±5%, 1,000 population, 30% response
The finite-population correction gives 278 responses. At a 30% response rate, you need about 927 invitations.
What the result means
The margin of error is about the whole sample, not each subgroup. If you need to report a subgroup, calculate the sample size for that subgroup separately. A low response rate can introduce nonresponse bias even when the math says you have enough completes. The 50% proportion is the conservative choice; if you expect a result near 10% or 90%, the required sample may be smaller. Always round up and allow for unusable responses.
Common mistakes
- Using a 5% margin as 5 instead of 0.05.
- Confusing responses with invitations.
- Ignoring nonresponse bias.
- Applying the whole-sample margin to a small subgroup.
- Treating the estimate as exact rather than a planning target.
Assumptions and limits
- The sample is random or close to random.
- The population is large relative to the sample unless a finite population is entered.
- The expected proportion is 50%, the conservative default.
- The response rate is an average, not a guarantee.
- The margin of error is a 95% or chosen-confidence interval.
Frequently asked questions
What sample size do I need for a survey?
For a 95% confidence level and ±5% margin with an unlimited population, about 385 responses are needed.
Why does a bigger population not require a much bigger sample?
The sample-size formula has a finite-population correction. Once the population is large, the required sample levels off.
What is a good margin of error?
±5% is common for general surveys. ±3% is tighter and costs more responses.
How does response rate change the invitation count?
Divide the required responses by the response rate. A 30% response rate needs more than three times as many invitations.
Should I use 50% or an expected proportion?
50% is conservative. If you have a reliable prior estimate, using it can reduce the required sample.
Does this handle subgroups?
No. Calculate each subgroup separately if you need subgroup-level estimates.
Sources
2026 update
How to use the professional output
This page is not just a formula box. The result is designed to be checked, compared, and taken away.
- Professional worksheet. The result includes the inputs used, the method and scope, and a one-variable sensitivity check for finance calculators.
- A/B scenarios. Save two input sets in the browser and compare the primary, secondary, and breakdown figures side by side.
- Print / PDF and CSV. Print a branded worksheet, or export the year-by-year table when the calculator produces one.
The formulas and assumptions above still govern the number. The worksheet makes the estimate auditable; it does not make it a lender quote, tax return, or personalized advice.
Go deeper
