Two-proportion A/B Test Calculator
Compare two independent success rates with an explicitly approximate two-sided pooled z test.
Loading tool…
How to use Two-proportion A/B Test Calculator
The null model pools successes across groups. Divide rate B−A by the pooled standard error; two normal tails give the two-sided p value.
- Enter group a successes, group a trials, group b successes, group b trials.
- Check the model, units, and supported limits before calculating.
- Run the tool, inspect the result, then copy or download it.
Good to know
- Whole successes within positive trial counts. Approximate p is withheld when any pooled expected cell is below 5. Independent randomized observations required; no sequential monitoring, causal, or winner declaration.
- Runs locally. Numeric inputs are bounded; floating-point output is approximate and rounded to 12 significant digits. Unsupported or non-finite calculations produce an error.
Common questions
How does Two-proportion A/B Test Calculator calculate its result?
The null model pools successes across groups. Divide rate B−A by the pooled standard error; two normal tails give the two-sided p value.
What assumptions and boundaries apply?
Whole successes within positive trial counts. Approximate p is withheld when any pooled expected cell is below 5. Independent randomized observations required; no sequential monitoring, causal, or winner declaration.
Keep going with related tools
All math →Proportion Confidence Interval Calculator
Calculate a bounded Wilson score interval from a success count and sample size.
MathProportion Sample Size Calculator
Plan a simple-random proportion sample from confidence, margin, and an explicit anticipated proportion.
MathChi-square Independence Calculator
Inspect a two-way count table with Pearson’s statistic, expected counts, and a conditional approximate p value.
MathTwo-proportion Experiment Power Planner
Plan bounded independent group sizes or estimate fixed-horizon power under a stated normal-arcsine approximation.
MathCommon-mix Stratified Rate Comparison
Compare two observed success rates after applying one explicit common stratum mix, retaining crude and standardized results.
MathConfusion Matrix Metrics Calculator
Calculate explicitly oriented binary or multiclass prediction metrics from a whole-count confusion table.