Free reference
Plain-language definitions for the statistics terms you actually run into — each with the formula, how to read it, a worked example, and exactly how Stratum computes it. No jargon walls.
Jump to the terms ↓ · Free calculators → · Stratum's conventions →
Three measures of center — when they agree, when they diverge, and which to report.
How spread out the data is — and why the sample version divides by n − 1.
The 25th, 50th, and 75th percentiles, the box of a box plot, and outlier fences.
The shape numbers: which way a distribution leans and how heavy its tails are.
The bell curve — what it is, the 68–95–99.7 rule, and why so many methods assume it.
How many standard deviations a value sits from the mean — and how to read it.
What a p-value actually measures, the 0.05 threshold, and how to read it honestly.
A range for the true value — what "95% confident" does and doesn't mean.
How big the difference is, not just whether it's significant — the number p-values miss.
How tightly two variables move together — from −1 to +1, and the causation trap.
The share of variation a model explains — and why a high R² isn't always good.
Where a statistic admits more than one reasonable definition, Stratum picks one and writes it down. The conventions page is the single source of truth — quantile method, missing-value handling, number formatting, and reproducibility.
Every term here is a live feature in Stratum. Load a CSV and the means, quartiles, p-values, and correlations compute themselves — with a chart to match.
Stratum