Sample Size Calculator

Find the right survey sample size for any confidence level and margin of error.

Confidence + margin of errorSurvey sample estimatesUpdated May 2026

Smaller margin means larger sample.

Use 50% if you are unsure.

Finite population correction is applied if entered.

Used to estimate how many people to invite.

Live estimate · Finite population correction · Response-rate adjustment

Required sample size

385

Completed responses needed.

Invites needed

385

Based on expected response rate.

Large population n₀

385

Adjusted sample size

385

Confidence level

95%

Z-score

1.96

Margin of error

5%

Estimated proportion

50%

Formula used

n₀ = (Z² × p × (1 − p)) ÷ e²

(1.96² × 0.5 × 0.5) ÷ 0.05²

At 95% confidence and 5% margin of error, you need about 385 completed responses.
Using 50% response distribution is conservative because it usually gives the largest required sample size.
A smaller margin of error needs more completed responses.
A higher confidence level needs more completed responses.
Population size is treated as large or unknown, so the large-population estimate is used.
At a 100% response rate, you may need to invite about 385 people.

Sample Size Formulas

Large Population Sample Size

n₀ = (Z² × p × (1 − p)) ÷ e²

Finite Population Correction

n = n₀ ÷ (1 + ((n₀ − 1) ÷ N))

Response Rate Adjustment

Invites Needed = Required Completed Responses ÷ Expected Response Rate

Margin of Error

e = Z × √((p × (1 − p)) ÷ n)

Variable Explanations

n

Adjusted required sample size.

n₀

Sample size for a large or unknown population.

Z

Z-score for the selected confidence level.

p

Estimated proportion or response distribution.

e

Margin of error as a decimal.

N

Population size.

Response rate

Expected percentage of invited people who complete the survey.

What Sample Size Means

Completed responses

Sample size is the number of completed responses or observations needed.

Precision

Larger samples generally improve precision.

Confidence

Higher confidence means more certainty but requires more responses.

Margin of error

Smaller margin of error means tighter estimates but larger samples.

Bias warning

Sample size does not fix biased sampling.

Research planning

Useful for surveys, research studies, quality checks, and market research.

Worked Examples

95% confidence, 5% margin, 50% proportion

Formula: n₀ = (1.96² × 0.5 × 0.5) ÷ 0.05²

Rounded-up result: 385

Common large-population survey estimate.

99% confidence, 5% margin

Formula: n₀ = (2.576² × 0.5 × 0.5) ÷ 0.05²

Rounded-up result: 664

Higher confidence requires more responses.

95% confidence, 3% margin

Formula: n₀ = (1.96² × 0.5 × 0.5) ÷ 0.03²

Rounded-up result: 1,068

Smaller margin of error increases sample size.

Finite population N = 1,000

Formula: n = n₀ ÷ (1 + ((n₀ − 1) ÷ N))

Rounded-up result: 278

Finite population correction lowers the required sample.

Estimated proportion 30%

Formula: n₀ = (1.96² × 0.3 × 0.7) ÷ 0.05²

Rounded-up result: 323

A proportion away from 50% often needs fewer responses.

Response rate adjustment

Formula: Invites = 385 ÷ 0.40

Rounded-up result: 963 invites

Invite more people if not everyone responds.

Why 50% is conservative

Formula: p × (1 − p) is largest at p = 0.5

Rounded-up result: Largest sample

Use 50% when unsure.

Invalid margin of error

Formula: 0% margin

Rounded-up result: Invalid

A zero margin of error would require an impossible sample size.

Confidence Level, Margin of Error, and Population Size

Confidence level

Controls how confident the estimate is.

Margin of error

Controls how precise the estimate should be.

Estimated proportion

50% is usually the most conservative choice.

Population size

Matters most for smaller finite populations.

Finite Population Correction

Large-population sample size assumes the population is very large. Finite population correction adjusts the required sample when the population size is known. The correction matters more when the sample is a meaningful share of the population. For very large populations, the adjusted sample is close to the large-population estimate.

Common Use Cases

Customer surveys
Market research
Employee surveys
Polling education
Product feedback
Academic research planning
Quality control sampling
Website and user research
Research planning estimates

For A/B tests, use a dedicated power calculator when you need power, effect size, baseline rate, and minimum detectable effect.

Common Sample Size Mistakes

Using sample size to fix biased sampling.
Forgetting to round up to a whole response.
Treating invited people as completed responses.
Assuming population size always changes sample size a lot.
Using 0% margin of error.
Choosing a confidence level without understanding the tradeoff.
Ignoring expected response rate.
Confusing sample size for proportions with sample size for means or A/B tests.

Understanding Your Results

Required sample size

Completed responses needed.

Rounded-up sample size

Practical whole-number target.

Z-score

Confidence-level multiplier.

Margin of error

Acceptable range around the estimate.

Finite population adjustment

Smaller sample needed when population is limited.

Invites needed

Estimated outreach count based on response rate.

Frequently Asked Questions