Formal & Natural Sciences
Reasoning under uncertainty — the discipline most often invoked and least often applied correctly.
7 fields · 66 topics · 94 curated works
Drawing conclusions from samples, and knowing how much to trust them.
Foundations & Overviews · Estimation Theory · Hypothesis Testing · Confidence Intervals · Likelihood Theory
Probability as degree of belief, updated by evidence.
Foundations & Overviews · Prior Selection · Posterior Inference · Markov Chain Monte Carlo · Variational Inference
The difference between correlation and cause, formalised.
Foundations & Overviews · Potential Outcomes Framework · Directed Acyclic Graphs · Confounding & Adjustment · Instrumental Variables
Structuring an experiment so the answer means something.
Foundations & Overviews · Randomised Controlled Trials · Factorial Designs · Blocking & Randomisation · A/B Testing
Fitting structure to data without fooling yourself.
Foundations & Overviews · Linear Models · Generalised Linear Models · Mixed Effects Models · Time Series Analysis
Doing statistics at a scale where the arithmetic matters.
Foundations & Overviews · Monte Carlo Methods · Numerical Optimisation · Simulation Studies · Bootstrap Computation
The failure modes — worth more practical attention than most of the theory.
Foundations & Overviews · p-Hacking & Garden of Forking Paths · The Replication Crisis · Base Rate Fallacy · Simpson's Paradox
No prior grounding assumed.
A Power Primer
Jacob Cohen · 1992
Gives working effect-size conventions and a lookup table for statistical power, arguing most published psychology studies were underpowered to detect…
+14 more at this level
Assumes you know the vocabulary.
Why Most Published Research Findings Are False
John P. A. Ioannidis · 2005
Shows that under realistic assumptions about power, bias and prior probability, most published positive findings will not replicate.
+30 more at this level
Primary sources and full treatments.
Bayesian Data Analysis
Gelman, Carlin, Stern, Dunson, Vehtari & Rubin · 1995
The reference for applied Bayesian work, and unusually candid about model checking and the ways a posterior can be confidently wrong.
+47 more at this level
12 of 94 works
Jacob Cohen
Gives working effect-size conventions and a lookup table for statistical power, arguing most published psychology studies were underpowered to detect the effects they claimed to be testing for.
link checked 17 Sept 2026Veronica Czitrom
Shows with worked examples that changing one factor at a time while holding others fixed misses interactions a factorial design catches for the same or fewer runs, so OFAT is not even the cheap option it is chosen for.
link checked 17 Sept 2026Elise Whitley & Jonathan Ball
Walks through the four quantities every sample-size calculation trades off against each other — effect size, variability, significance level and power — so a stated sample size can be audited rather than taken on faith.
link checked 17 Sept 2026Hans Rosling (TED)
Shows with animated development data that the rich-world / third-world split stopped describing reality decades ago, and that aggregate categories hide the within-country variation that matters.
link checked 17 Sept 2026Manish Kumar Goel, Pardeep Khanna & Jugal Kishore
Walks a non-statistician reader through reading a Kaplan-Meier curve and the log-rank test used to compare two of them.
link checked 17 Sept 2026David Diez, Christopher Barr & Mine Çetinkaya-Rundel
A free, open-licensed introductory statistics textbook that builds estimation and hypothesis testing from first principles for readers with no calculus background.
link checked 17 Sept 2026Rob J. Hyndman & George Athanasopoulos
Teaches forecasting as a practical workflow — decompose, model the remainder, validate out of sample — with the code needed to carry out each step.
link checked 17 Sept 2026Hadley Wickham
Defines a tidy dataset as one where every variable is a column, every observation a row and every value a cell, and shows most data-cleaning pain comes from a fixable violation of this.
link checked 17 Sept 2026Hadley Wickham & Garrett Grolemund
Teaches the whole analysis cycle — import, tidy, transform, visualise, model — as one workflow rather than as separate tools.
link checked 17 Sept 2026Carl T. Bergstrom & Jevin D. West
A free university course teaching readers to spot the specific statistical and visual tricks - misleading axes, cherry-picked baselines, spurious correlations - that make quantitative claims look more rigorous than they are.
link checked 17 Sept 2026Miguel A. Hernán
Argues that hedging a causal question behind associational language does not make the analysis more rigorous, it just hides the causal assumptions being made from the scrutiny they need.
link checked 17 Sept 2026Xavier A. Harrison, Lynda Donaldson, Maria Eugenia Correa-Cano, Julian Evans, David N. Fisher, Cecily E. D. Goodwin, Beth S. Robinson, David J. Hodgson & Richard Inger
A practitioner's walk-through of when a mixed-effects model is actually warranted, and the misspecifications, pseudoreplication chief among them, that make one necessary.
link checked 17 Sept 2026