Statistics, hands-on.

Roll dice, flip coins, spin roulette – and watch live how relative frequencies approach the true probability. Includes a gambler's-fallacy visualisation.

From medical testing to Gigerenzer-style frequency formats: why a 99%-accurate test often still produces mostly false positives – and how a confirmation test fixes that.

Normal, binomial, Poisson, t, chi² – move the parameters and watch density and distribution function react.

Draw samples and observe how often the confidence interval contains the true mean.

Visualise the null and alternative distributions, and interactively explore the critical region and the p-value.

Vary effect size, sample size, and significance level and read off the statistical power.

Pick any underlying distribution and watch how the sampling distribution of the mean converges to the normal as n grows.

Place or drag data points and watch the regression line and residuals update in real time.

When does a significant p-value point to a real effect? Connect prior probability, power, α, and bias.

Monte Carlo simulation after Rasch, Kubinger & Moder: how well do Student's t, Welch, Mann-Whitney, Yuen, and a permutation test hold the nominal α – and how much does their power differ?

Paste a CSV or pick a sample dataset and immediately compare Welch's t, Student's t, Mann-Whitney U, and Yuen's t with confidence intervals, effect sizes, and diagnostic plots.

Explore Cohen's d, r, odds ratios, overlap, and the common language effect size interactively – with Cohen's benchmarks set against the effects actually observed in psychology.