Acknowledgments
This book is built with bookdown by Yihui Xie and the R Markdown community. The wider R ecosystem — base R, the tidyverse, and the many domain-specific packages cited in the body — is the foundation on which every chapter rests.
The original lab and tutorial material that inspired this volume was authored under
the CTTIR/courses and CTTIR/tutorials repositories. Those repositories are credited in the Preface and at the points of use.
The decision-tree wizard hosted at cttir.github.io/tutorials/decision-tree/
draws structural inspiration from the University of Zurich’s
Methodenberatung (methodenberatung.uzh.ch).
Drafting assistance from Anthropic’s Claude was used during preparation; all editorial decisions, errors, and omissions remain the author’s.
Inspiration
The structural and pedagogical model for this book draws heavily on Mathias Harrer, Pim Cuijpers, Toshi A. Furukawa, and David D. Ebert’s Doing Meta-Analysis with R: A Hands-On Guide — both the book and its open-source repository at https://github.com/MathiasHarrer/Doing-Meta-Analysis-in-R. Their thorough, transparent, and beautifully laid-out bookdown setup — sidebar grouping, semantic callout boxes, the citation block, the dark/light toggle, MathJax-rendered formulas, downloadable BibTeX/RIS citations — set a clear standard for what a reproducible R-based research book can look like. Where the present book mirrors their conventions, it does so deliberately and gratefully. Where it diverges (colors, typography, individual icons), the divergence is cosmetic; the underlying craft is theirs to credit.
Use of LLM tools
Portions of this book were prepared with assistance from large language
model tooling for narrowly defined, non-authorial tasks: copyediting,
prose smoothing, LaTeX/Markdown formatting, scaffolding of boilerplate
files (CI configs, CSS, build scripts), code refactoring, and citation
lookup against verified DOIs. The tools used were the latest Mistral
Le Chat model run locally via Ollama and the
ollamar R package — all
inference performed on local hardware, with no data sent to third-party
services — and GitHub Copilot inside RStudio for code completion.
All scientific claims, methodological choices, data analyses,
interpretations, and conclusions are the author’s own. No text generated
by an LLM was incorporated without review and revision by the author.
LLM tools were not used to generate, fabricate, or paraphrase cited
sources; every reference was verified against its DOI, arXiv ID, or
ISBN (see scripts/verify-citations.R).