Practical techniques—blocking, factorial designs, power analysis, and validity checks—to design statistically robust experiments that yield trustworthy evidence for decisions.
A practical, decision-focused guide to blocking, stratification, factorial and split-plot designs, correct identification of experimental units and clustering, sample-size and power formulas, sequential and adaptive testing considerations, and concrete validity and sensitivity checks teams should use when stakes are higher.
A practical, decision-focused guide to statistical power: core concepts, simple closed-form formulas and worked examples for means and proportions, approximations for clustered and factorial designs, pseudocode for quick calculators, rules of thumb for baseline variability and minimal detectable effects, tradeoffs between runtime and detection threshold, and a short decision flow that explains when to use sequential or adaptive approaches.
A plain-language, practical guide to statistical power, sample size, minimum detectable effect, and realistic shortcuts for discovery teams. Includes clear formulas, worked examples for proportions and means, cluster adjustments, an annotated spreadsheet template, and a short decision checklist for pilots vs full power calculations.