Teaching data science as a human practice.
We build curriculum, textbooks, and open resources that prepare students and practitioners to design, evaluate, and govern data-driven systems in the public interest.
The textbook
Human-Centered Data Science: An Introduction
Data science is a sociotechnical practice: every dataset reflects human choices about what to measure and what to leave out.
The first textbook in human-centered data science, written for courses in data science, ethics, and HCI. It is the foundation for the lab's open teaching materials below.
Open-access toolkit
The HCDS Pedagogy Toolkit
An open curriculum companion to the textbook, for anyone teaching data science, ethics, or HCI. Each module is a ready-to-use unit that connects social theory to computational practice.
New modules, mapped to later chapters of the book, are released over time. Star or watch the repository on GitHub to hear when they land.
Modules available now
- 01 What is Human-Centered Data Science?Introduces the HCDS framework, drawn from Chapter 1.
- 02 Foundations and a Critical Lens on the LifecycleApplies a critical, interpretive lens across the data science lifecycle.
- 03 Data Practices and Responsible ModelingData practices, responsible modeling, and visualization, worked through a real dataset.
- 04 Problem Framing and Scoping NewDesigning meaningful research questions and scoping a project, using the COMPAS, Gender Shades, and FairFace cases.
Every module includes a slide deck, a Python notebook, and an instructor guide.
What each module includes
Slide deck
About 30 teaching slides
Key concepts, learning outcomes, and embedded activities to use or adapt in class.
Code notebook
A worked example in Python
A Jupyter notebook (.ipynb) built on real-world data, for class or as a take-home exercise.
Instructor guide
Step-by-step teaching notes
Timing for the slides, additional readings, and a quiz for individual assessment.
How to participate
We invite you to “Star” or “Watch” the repository on GitHub to be notified as new modules are released.
How to cite
If you use the toolkit in your teaching or research, please cite both the toolkit (for the materials) and the textbook (for the methodology).
-
The toolkit: code and slides
Guha, S., Chui, V., Silver, M., & Irfan, M. (2026). HCDS Pedagogy Toolkit. GitHub. https://github.com/uofthcdslab/hcds-pedagogy-toolkit. doi:10.5281/zenodo.18092859
-
The textbook: methodology
Aragon, C., Guha, S., Kogan, M., Muller, M., & Neff, G. (2022). Human-Centered Data Science: An Introduction. MIT Press. https://mitpress.mit.edu/9780262543213/
Questions or feedback on the materials? Email hcds.uoft@gmail.com.