Free tools

Free Confidence Interval & Margin of Error Calculator

Estimate confidence ranges for sample proportions using standard confidence levels and sample sizes.

Copy-paste outputs

Win high-intent buyers from ChatGPT, Gemini, Claude, Perplexity, and AI Overviews before your competitors do.

One operating layer for monitoring, measurement, content action, and technical cleanup.

AI Visibility TrackingCompetitive RankingSentiment by ModelSource CitationsAI Overviews TrackingPrompt MonitoringAI Visibility TrackingCompetitive RankingSentiment by ModelSource CitationsAI Overviews TrackingPrompt Monitoring
Content GapsAI InsightsAdvanced AnalyticsData CopilotBlog GenerationUGC CampaignsLLM CouncilContent GapsAI InsightsAdvanced AnalyticsData CopilotBlog GenerationUGC CampaignsLLM Council
Shopping IntelligenceCrawler MonitoringGEO OptimizationMulti-Brand ManagementShopping IntelligenceCrawler MonitoringGEO OptimizationMulti-Brand Management

Tool 01

Confidence Interval + Margin of Error Calculator

Estimate confidence ranges for survey and conversion proportions.

Confidence Interval + Margin of Error Calculator
Estimate confidence ranges for proportion-based samples like survey or conversion data.
Result summary

Observed rate

45.00%

Margin of error

±2.81%

Lower bound

42.19%

Upper bound

47.81%

Required sample size for roughly ±5% precision at 95% confidence: 385

{
  "confidenceLevel": 95,
  "sampleSize": 1200,
  "successes": 540,
  "observedRate": 0.45,
  "marginOfError": 0.02814835696803634,
  "interval": [
    0.42185164303196365,
    0.47814835696803637
  ],
  "standardError": 0.014361406616345072
}

How it works

Confidence Interval + Margin of Error Calculator: methodology and worked example

How this tool computes its result

Computes a Wald confidence interval for a single proportion: p̂ = successes / sample size, standard error = √(p̂(1−p̂)/n), margin = z × standard error, with the interval clamped to [0,1]. The z-score comes from a fixed lookup table for four confidence levels (80/90/95/99%). It also reports the classic minimum sample size needed for a ±5% margin at the selected confidence level, computed as ceil(z² × 0.25 / 0.05²), where 0.25 is the worst-case p(1−p) value at p = 0.5.

Worked example

With the tool's defaults — sample size 1,200, successes 540, confidence 95% — p̂ = 540/1200 = 0.45, z = 1.96, standard error = √(0.45 × 0.55 / 1200) ≈ 0.01436, and margin = 1.96 × 0.01436 ≈ 2.82%. The reported interval is roughly 42.2% to 47.8%. The "required sample for ±5%" figure computes ceil(1.96² × 0.25 / 0.05²) = ceil(384.16) = 385 — the well-known survey-sizing benchmark for ±5% margin at 95% confidence.

When not to use this tool

This only implements the Wald (normal-approximation) interval for a single binomial proportion at four preset confidence levels — it has no interval for means/continuous data (which would need a t-distribution), no two-sample comparison, and the Wald approximation itself is known to perform poorly for very small samples or proportions near 0 or 1.

Common mistakes

  • - Entering a success count greater than the sample size (e.g. 150 successes out of 100) — this is explicitly validated and blocked with "Success count must be between 0 and sample size." rather than silently computing a rate above 100%.
  • - Trusting the interval at small sample sizes or extreme proportions — the tool always uses the normal (Wald) approximation, never switching to an exact method like Wilson or Clopper-Pearson, so intervals can be misleadingly narrow when the sample is small or p̂ is near 0 or 1.
  • - Assuming the "required sample for ±5%" number reflects your actual observed rate — it always uses the conservative worst-case p = 0.5, so if your true proportion is far from 50%, the real number of respondents needed for ±5% may be smaller than the 385 figure shown.

Ready to dominate AI search visibility?

Track where your brand shows up in AI answers, close the content gaps that cost conversions, and stay visible across ChatGPT, Claude, Gemini, Perplexity, and Grok.