Supplementary material

Stop adjusting for anything : How the backdoor and adjustment criteria empower causal human-centric research

Authors

Jian Pan

Matteo Favero

Andrew Sonta

Ruiji Sun

Julia Damrath

Marcel Schweiker

Published

25 November, 2025

Preface

This notebook is specifically designed to serve as a detailed guide for the examples with synthetic data that have been discussed in the paper titled “Stop adjusting for anything: How the backdoor and adjustment criteria empower causal human-centric research”. The primary objective of this notebook is to provide a comprehensive and practical explanation of the entire modelling process, along with detailed insights into the various methodologies and techniques employed in the process.

Our aim is to provide a valuable resource for researchers and practitioners at all levels of proficiency —both experienced and inexperienced. By providing a clear and practical explanation of the modelling methods and techniques used in the examples, we hope to enable researchers and practitioners to gain a deeper understanding of the subject.

However, it is important to note that this notebook is not intended and should not be used as a set of recipes to solve problems. Instead, it should be viewed as a map to guide researchers through the modelling process.

Contributors and license

This document was written by Matteo Favero and Jian Pan with feedback from Andrew Sonta, Ruiji Sun, and Marcel Schweiker.

This work is openly licensed via a Creative Commons Attribution 4.0 International (CC BY 4.0).

For attribution, please cite this work as (BibTeX citation):

@article{PJ2025114002,
  title = {Stop adjusting for anything: How the backdoor and adjustment criteria empower causal human-centric research},
  author = {Pan, Jian and  Favero, Matteo and Sonta, Andrew and Sun, Ruiji and Damrath, Julia and Schweiker, Marcel},
  journal = {Building and Environment},
  pages = {114002},
  year = {2025},
  issn = {0360-1323},
  doi = {https://doi.org/10.1016/j.buildenv.2025.114002}
}

The corresponding primary article can be found at https://doi.org/10.1016/j.buildenv.2025.114002.