Statistics

Significance and A/B testing, power and sample size, confidence intervals, effect sizes, and a chooser that routes you to the right test.

Statistics is where most quantitative work goes right or quietly wrong, so this section is built to be the deepest of the seven. It covers the everyday decisions: whether a difference is real, how large a sample you actually need, how wide your interval should be, and which test fits your data in the first place. The machinery behind those calls now runs end to end here: distribution tools that turn a raw value into a z-score, a tail probability, or a critical cutoff; the workhorse tests themselves, from the t-test to chi-square to one-way ANOVA; and regression with its correlation coefficient for when two variables move together. Paste a dataset into descriptive statistics to see it summarized, or start from a single question and let the test chooser narrow it down.

Every instrument names the test it runs and states its assumptions, so the number you report is one you can defend and show the work behind.

Test chooser

Start with the shape of your data

Pick the closest case. The chooser routes common designs to published calculators only, so every recommendation is a working page.

What do you have?
What are you trying to do?

Use for sparse 2 by 2 count data.

Recommended route counts · two independent groups

Compare the two observed groups

This route fits two independent groups with observed counts, conversion rates, or numeric means.

A/B significance test

Compare observed conversion rates or two independent means.

Other useful pages

Experiment Planning and A/B Testing

Analyze observed experiments, plan sample size, and connect effects to uncertainty.

Two-Way ANOVA Calculator: Full Table, F, and p-Value

Run a two-way or one-way ANOVA in your browser: sums of squares, df, mean squares, F, and p for every effect, with the balanced-design and interaction conventions stated on the page.

A/B Test Calculator

Compare two conversion rates or two means with a named frequentist test, effect size, and confidence interval.

Power and Sample Size Calculator

Plan sample size, power, or effect size for two proportions or two independent means.

Minimum Detectable Effect Calculator

Estimate the smallest conversion-rate lift an A/B test plan can reliably detect.

Bayesian A/B Test Calculator

Compare two conversion variants with posterior probability under a visible beta prior.

t-Test Calculator: One-Sample, Two-Sample, Welch, and Paired

Run one-sample, two-sample (Student or Welch), and paired t-tests from raw data or summary stats. Returns the t statistic, df, exact p-value, confidence interval, and Cohen's d.

Fisher's Exact Test

Run an exact fixed-margin 2 by 2 count test with odds ratio, p-value, and support table.

Chi-Square Test Calculator: Independence and Goodness of Fit

Chi-square test of independence or goodness of fit on any size table, with the expected-count table, each cell's contribution, degrees of freedom, the p-value, and Cramér's V.

Mann-Whitney U Test Calculator

Compare two independent samples with ranks, exact or asymptotic p-values, and effect sizes.

P-value in Excel

Use Excel T.TEST for a two-tailed unequal-variance t-test and compare a worked example to the canonical A/B test logic.