About

ABOUT THIS CURRICULUM

Why this curriculum exists

A standalone, open curriculum for people who would rather understand statistics than memorise recipes.

About This Tutorial Collection

This site hosts a four-course biostatistics programme written in R and Quarto and released under the MIT licence. The aim is to offer a generous, opinionated curriculum to the biomedical research community — the kind of curriculum the author wishes had been handed to him at the start of his own PhD. It runs from first principles through modern statistical learning, and it is freely available for self-study, for citation, and for adaptation into local teaching.

Three convictions run through every lab.

Statistics is thinking, not buttons. The hardest part of any analysis is not which function to call but whether the question, the design, and the data are good enough to answer it. Every inference lab therefore begins with a hypothesis and a plot, not with a line of code.

Reproducibility is the default, not a decoration. Every lab renders from a Quarto source, every package version is pinned with renv, every figure is generated by the same code that fitted the model, and every dataset is either built into R or simulated in the file that uses it. If the site cannot be re-rendered from a clean clone, it is considered broken.

If you cannot present it, you do not yet understand it. Every lab renders twice — once as a long-form article and once as a Reveal.js slide deck. Boiling a method down to eight slides is a reliable test of comprehension.

Who Is This For?

The intended reader is a biomedical researcher — a PhD student, postdoc, clinician-scientist, or junior PI — who works with data and wants to understand what they are doing rather than copy a recipe from a collaborator’s script. The material assumes some prior exposure to R and to undergraduate-level probability, but it does not assume a formal statistics background. It is also suitable as a backbone for graduate teaching: every lab is a self-contained Quarto file with an article and a slide deck, so a department can lift any week into a Friday seminar without rebuilding the material from scratch.

What You Will Find

• Four courses, each four weeks long, building on the previous one:
  • Course 1 — Foundations of Biostatistics with R
  • Course 2 — Regression, ANOVA & Model Diagnostics
  • Course 3 — Study Design, Longitudinal Data & Causal Inference
  • Course 4 — Modern Statistical Learning & High-Dimensional Biomedicine
• Twenty labs per course, each rendered as both a long-form article and a Reveal.js slide deck.
• A unified five-step inference template — Hypothesis → Visualise → Assumptions → Conduct → Conclude — applied across every lab where inference is the goal.
• Weekly cheatsheets summarising the syntax and decisions of each week on a single page.
• Appendices: a decision tree for picking a method, a glossary, a common-errors page, a research-workflow guide, and a writing-a-report guide.
• A unified reference list covering every source cited across the curriculum.

Technical Requirements

To re-render the site from a clean clone you need:

• R (≥ 4.3) with the renv package.
• Quarto (≥ 1.4).
• A working LaTeX installation is not required — the site renders to HTML only.

git clone https://github.com/CTTIR/courses.git
cd courses
Rscript setup_check.R
Rscript -e 'renv::restore()'
quarto render        # builds articles + slides
quarto preview

The full list of pinned R packages lives in renv.lock. See Get Started for a longer walk-through, including common installation pitfalls.

Author

R. Heller — author and maintainer.

Contributions, corrections, and suggestions are welcome via pull request or issue on the project’s GitHub repository.

References

Every citation used across the curriculum is collected on the unified References page. Course-specific reference subsets are also linked from each course’s index page.

Acknowledgements

Nothing on this site grew in isolation. See the dedicated Acknowledgements page for the full list of didactic sources, content references, tooling, and structural inspiration that made this curriculum possible — in particular the OCBE team at the University of Oslo and Chi Zhang’s teaching_mf9130e site, which provided the direct structural template.

Citation

If you use this curriculum in your own teaching or research, please cite it as:

Heller, R. (2026). Biostatistics Courses: A Four-Course Programme in R and Quarto. https://cttir.github.io/courses/

BibTeX:

@misc{heller2026courses,
  author       = {Heller, R.},
  title        = {Biostatistics Courses: A Four-Course Programme in R and Quarto},
  year         = {2026},
  howpublished = {\url{https://cttir.github.io/courses/}},
  note         = {CTTIR/courses, MIT licence}
}