Causal Inference: What If (opens provider site in a new tab)
Hernán and Robins' standard text on causal inference from observational data. Free to download.
Probability, inference, and regression — the reasoning under every data claim.
15 picks · 12 free · 13 providers · curated by Rowland Kuru
Editor's intro
Statistics is the discipline of not fooling yourself with data. It underpins data science, machine learning, medicine, and every news story about "a new study," yet most people only meet it as a box-ticking required course. This feed collects resources that teach statistics as a way of reasoning — what a p-value does and doesn't mean, why sample size matters, and when correlation is worth taking seriously.
If formulas put you off, start with the visual and interactive picks; the intuition matters more than the notation at first. Then work through one proper textbook with its exercises. The intermediate section splits between probability theory and applied modeling; the advanced section is about causal inference, which is where statistics meets real decisions.
Start here
Khan Academy's Statistics and Probability course is thorough, patient, and free, and it covers everything a first university statistics course would — from describing data through probability, sampling distributions, confidence intervals, and significance tests. Each short video is followed by practice problems with hints, so you find out immediately whether an idea has landed. Work through it unit by unit and don't skip the practice; that's where the understanding comes from.
Hernán and Robins' standard text on causal inference from observational data. Free to download.
Partner Some links on this page are affiliate links, marked Partner. Commissions fund the site at no cost to you. Read our affiliate disclosure.
Distributions, sampling, confidence intervals, and hypothesis tests — with intuition before formulas.
The Open University's free short courses on data and statistics. Clear, self-paced, and well structured.
A free, peer-reviewed introductory textbook used in hundreds of courses. Excellent exercises.
Brown University's visual introduction to probability and statistics. Beautiful, interactive, and short.
Josh Starmer makes p-values, distributions, and regression understandable. Great alongside any course.
A compact course on descriptive statistics, inference, and regression. A good-value refresher.
University of Michigan's applied statistics, taught through Python notebooks. Good for analysts who code.
Probability in depth, regression, and Bayesian thinking. Expect real mathematics.
A friendly, modern introduction to Bayesian thinking with R.
MIT's balanced course covering frequentist and Bayesian methods, with complete class materials.
Joe Blitzstein's famous probability course. Hard, beautiful problems; lectures free online.
Allen Downey teaches statistics through Python code rather than formulas. Great for programmers.
Duke's thorough treatment of inference, regression, and Bayesian statistics, using R.
Causal inference and principled modeling, for people making decisions from data.
Hernán and Robins' standard text on causal inference from observational data. Free to download.
Gelman, Hill, and Vehtari on regression done carefully. Practical, skeptical, and free to download.
McElreath's Bayesian course changes how you think about models and causation. Lectures free.