100 Empty Worlds
See how often 'significant' happens with zero real effect.
watching five random, meaningless worlds turn 'green' out of 100 at p<0.05 is the fastest way to feel, not just be told, what a p-value actually promises and doesn't
What goes in, what comes out
The sim generates 100 side-by-side 'null worlds' — fake datasets with a genuinely random split and no real effect — and runs the same significance test on every one. The learner watches how many of the 100 boring, effect-free worlds still light up 'significant' by chance alone, then compares that count against their own real dataset's result.
pick a threshold and sample size, run the null-world batch, then compare against a real dataset
a grid of 100 worlds with false-positive ones highlighted, and a running false-discovery count
number of simulated worlds (20-500), significance threshold (0.01/0.05/0.10), sample size per world
Data L1-L2 stats-intuition drill; trivia deck seed for cost-tradeoff and spot-the-consequence items
Concepts click when you open the machinery.
Three labs are already live and free — the same hands-on style this playground brings to its module.