About
Who made this site, how, and what it is not.
What this site is for
It is for medical students who have taken introductory biostatistics but have not yet run an analysis themselves. The aim is twofold: to read the statistics section of a clinical paper — knowing what each number guards against and which cell should make you stop — and to run the same analysis yourself.
The site belongs to the med-study-rpg family and shares its pixel-art visual language, but it is teaching material: reading and looking things up are the main actions.
Author
WLK, a medical graduate from Taiwan, with a medical background. The author sets the topics, the chapter structure, the choice of datasets and the acceptance criteria.
How the content is made
The prose, figures and code on this site are generated with the help of AI (Claude); the author sets the topics and structure and is responsible for review. Before a page goes live it is reviewed again by a separate, independent AI, and the chip at the top of every page shows its current review status ("Draft", "Independently reviewed, not yet spot-checked by a human", "Human reviewed").
Every figure and every statistic on a page is computed by an R script from a public dataset, and the page reads the script output directly rather than retyping it. The R and Python code on the pages is executed by an automated check.
Errors are still possible. If you see one, please use "Report a problem on this page" at the bottom of that page.
Disclaimer
This site is for teaching only. It is not medical advice, nor statistical consulting for any particular study. Every example uses a public dataset and contains no patient records; conclusions drawn on these pages are teaching demonstrations.
Editorial stance
Statistical significance is worded strictly: a non-significant result is written as "no difference was detected" or "not statistically significant", never as "the groups do not differ". Claiming equivalence requires a non-inferiority design with a margin set in advance.
Every example dataset ships inside an R package and is one line of code away; no data access application is required.
Content is licensed CC BY-SA 4.0; pages adapted from other open material list their source and licence in the page footer.