Confidence interval calculator

Interval that contains the true value with a given probability

Confidence interval — range that contains the true parameter (e.g. proportion, mean) with a given confidence level (e.g. 95%).

FAQ about confidence interval

Parameters

Percentage in the sample with the characteristic of interest

Number of respondents surveyed

Assumptions and limitations

  • Formula is for proportion, not mean
  • Simple random sample assumed
  • n × p and n × (1−p) should be > 5
Confidence interval
Enter data

In SurveyNinja, NPS, CSAT and more are calculated automatically. Create a survey in 5 minutes and get real-time analytics.

Start free

Data is not stored — calculation runs in your browser.

Confidence interval examples

1 Proportion in sample

p = 60%, n = 400, 95% confidence
95% CI ≈ [55%, 65%] — with 95% probability the true proportion in the population lies in this range.

2 Narrowing the interval

Increase sample size
Quadrupling n roughly halves the CI width. Plan a larger n for a more precise estimate.

3 NPS promoter proportion

"9–10" proportion = 50%, n = 200, 95% CI
CI for promoter proportion ≈ [43%, 57%]. NPS is computed from proportion difference; for full NPS build CI separately or via bootstrap.

4 Small proportion (rare event)

p = 5%, n = 400. CI is asymmetric due to 0 boundary
95% CI ≈ [3%, 7%]. For small p, Wilson score or exact binomial interval is better.

5 99% confidence level

p = 50%, n = 500. Wider interval at 99%
99% CI is wider than 95% (Z = 2.576 instead of 1.96). Used when minimizing the risk of missing the true value is more important.

6 Comparing two proportions

Group A: 60% (n=300). Group B: 45% (n=300). Build CI for each proportion
CI A: [54%, 66%], CI B: [39%, 51%]. Intervals do not overlap — difference is statistically significant at α = 0.05.

Frequently asked questions

Automate your metrics

In SurveyNinja, NPS, CSAT and more are calculated automatically. Create a survey in 5 minutes and get real-time analytics.

Start free