Two-Proportion Z-Test Calculator — Free Online | EverydayCalcSkip to main content
everydaycalc.

Find calculators and file viewers

Press Esc to close · / or Ctrl+K to open

Home›Math & Scientific›Two-Proportion Z-Test Calculator
Math & Scientific

Two-Proportion Z-Test Calculator

Compare two independent binomial success rates with a pooled two-sided Z test and expected-count diagnostic.

✦ INSTANT CALCULATION

Two-Proportion Z-Test Calculator

YOUR RESULT
—

ⓘ Calculated locally in your browser. Results depend on the values and units you enter.

Interactive mathematical graph

The axes, points and dimensions are recalculated from the values you enter.

What changes if one input changes?

41: 0.104058693475897 p-value · 45: 0.0284596376527878 p-value · 50: 0.00389254974145229 p-value

How this calculator works

For two independent groups, the null hypothesis is pA=pB. The pooled proportion is (successesA+successesB)/(totalA+totalB). The Z statistic divides the difference in observed rates by the pooled standard error; the two-sided p-value is from the standard normal distribution. All counts must be nonnegative integers, totals positive and successes no greater than totals. The tool deliberately rejects groups with fewer than 5 expected successes or failures under the pooled model because a normal approximation may be unreliable. Use an exact method for sparse observations; a p-value alone is not evidence of a causal effect.

Worked example

45 successes out of 100 vs 30 out of 100 gives a 0.15 difference and a two-sided p-value of about 0.028.

Engine-checked example

This example uses the values shown below, not your current entries.

Group A successes
45
Group A total trials
100
Group B successes
30
Group B total trials
100
Result for the sample inputs0.0284596376527878 p-value

Frequently asked questions

Does this work for A/B conversion testing?

Only when observations are independent and the expected success/failure counts are sufficient; repeated exposures require other methods.

Is it a one-sided or two-sided test?

Two-sided: either direction of a difference is considered.

Step-by-step explanationStandard Deviation vs. Standard Error: What Changes With Sample Size? ?Standard deviation describes variation in observations. Standard error describes uncertainty in an estimated mean.

Results are informational and assume the stated inputs and formulas. Double-check critical financial or engineering decisions with a qualified professional.

Add this calculator to your website

Share a working calculator on your blog or website. Choose a fixed size or let the height fit the content.

Fixed height

Works on websites that accept iframe HTML.

Responsive height

Automatically adjusts the height. Your website must allow scripts.

Open calculator preview ↗

Powered by EverydayCalc