Medical Statistics with R and Python
Learn the statistics behind clinical and public-health research, then run every method yourself in R and Python without leaving the page.
Sign in once, then learn at your own pace
- Intermediate level
- 18 modules
- 34 lessons
- 8.0h of reading
- 69 knowledge checks
- Self-paced, learn anytime
- Lifetime access
About this course.
A complete, hands-on course in medical statistics built on the Kirkwood and Sterne spine, from describing data to survival analysis, meta-analysis, and study design. You learn each idea in plain language first, then run it live in the browser in both R and Python: no install, no setup. Every lesson carries worked clinical examples in a Malaysian and Southeast Asian context, common mistakes to avoid, and a question bank to test yourself. By the end you can take a messy dataset, clean it, choose the right method, run it, and write up the result the way a paper would.
What you'll be able to do.
- ✓Read and write data in R and Python, and get a clinical dataset into a tidy, analysis-ready shape
- ✓Describe data correctly: means, standard deviations, standard errors, the normal distribution, and the right summary for each variable type
- ✓Build confidence intervals and interpret p-values without the common misreadings
- ✓Compare groups with t-tests, ANOVA, chi-squared, risk ratios, odds ratios, and rates
- ✓Fit and interpret linear, logistic, Poisson, and Cox regression models, and adjust for confounding
- ✓Run survival analysis (Kaplan-Meier, log-rank, Cox) and analyse rates and person-time
- ✓Pool evidence with meta-analysis and reason with Bayesian methods at an introductory level
- ✓Choose the analysis from the study design, calculate the sample size you need, and judge the effect of measurement error
Who this course is for.
- ✓Clinicians, medical and public-health postgraduates who need to understand and run the statistics in research papers
- ✓Researchers and trainees who want to move from clicking through software to writing reproducible R or Python
- ✓Epidemiologists and biostatistics students who want one course covering the whole applied workflow
- ✓Anyone in health data who wants the concepts and the code together, taught with real clinical examples
What you'll cover.
18 modules · 34 lessons · ~8.0h read
010. Getting started: R and Python for medical data3 lessons
021. Foundations: variables and displaying data2 lessons
032. Describing data2 lessons
043. Estimation and inference2 lessons
054. Comparing means2 lessons
065. Linear and multiple regression2 lessons
076. Transformations1 lesson
087. Binary outcomes: risk, odds, proportions, the binomial2 lessons
098. Comparing two groups2 lessons
109. Confounding and stratification1 lesson
1110. Logistic regression and matched studies2 lessons
1211. Rates and the Poisson distribution2 lessons
1312. Standardization1 lesson
1413. Survival analysis2 lessons
1514. Statistical modelling2 lessons
1615. Evidence synthesis: meta-analysis and Bayesian methods2 lessons
1716. Study design, sample size, and measurement3 lessons
1817. Capstone: an end-to-end applied analysis1 lesson
Questions about this course
Do I need to install R or Python?
No. Every code block runs in your browser. R runs through WebR and Python through Pyodide, so you can run, edit, and re-run each example with nothing installed.
Do I need to know how to program first?
No. Part 0 starts from the very beginning: variables, vectors, data frames, and reading a table. If you have programmed before you can move through it quickly.
Why teach both R and Python?
R is built by statisticians and is the standard in much of medical research; Python is the general-purpose language of data science. Every method is shown in both, so you can work in whichever your team uses.
What background in statistics do I need?
None beyond school mathematics. The course builds each idea from intuition before any formula, then shows the formula and the code.
Is the course based on a standard textbook?
It follows the structure of Kirkwood and Sterne's Essential Medical Statistics, taught in our own words with original examples, and layered with the applied R workflow from the Epidemiologist R Handbook.