ab-test-calculator

Calculate statistical significance for A/B tests by comparing variant conversion rates using Chi-square, Z-test, or T-test. Input visitor and conversion counts for both groups to determine if observed differences are meaningful or due to random chance.

Estimate required sample size for an A/B test using power analysis features. Provide expected baseline conversion rate and minimum detectable effect to calculate the visitor count needed for statistically valid results.

Bayesian analysis calculates the probability that one variant outperforms another. This method evaluates underlying conversion rate distributions to give a direct probability percentage for decision-making.

The calculator supports multiple variants in A/B testing. It compares several groups simultaneously using the same statistical testing frameworks to evaluate conversion rate differences across all tested variations.

Computing confidence intervals and power analysis requires scipy, numpy, and statsmodels. These libraries handle the underlying statistical distributions and accurate computations for conversion rate intervals.

Use Chi-square tests for categorical conversion data comparing frequencies between variants. T-tests are applied when comparing the means of continuous metrics across test groups to determine statistical significance.