Incorporating Causal Inference in Statistics Courses

Details of CanCOTS 2025 including this working group

1 About This Website

This Quarto website contains the final deliverable from one of the working groups of the first edition of the Canadian Conference on Teaching Statistics (CanCOTS), held in the summer of 2025 at HEC Montréal. CanCOTS is primarily a participant-driven working meeting that follows the format of a curated unconference. In this style of unconference, various working groups focus on priority areas of interest related to educational topics in statistics. All groups are expected to produce concrete deliverables addressing these priority areas, which may include exercise banks, curriculum guidelines, sample learning activities, and assessment items. Each working group has a leader who is responsible for overseeing collaborative efforts and shaping the final deliverable.

Image by manfredsteger via Pixabay.

CanCOTS 2025 began with an initial roundtable discussion that brought all participants together. During this session, each group leader presented their educational topic along with a proposal for the final deliverable. Following these presentations, participants split into smaller working groups based on their interests. During the working group meetings, participants—including the group leaders—discussed and brainstormed ideas to develop the final deliverable. Additionally, a comprehensive list of potential resources was compiled to ensure that the deliverable contains fair and scholarly content.

CanCOTS 2025 featured six different working groups, each focusing on a specific area of interest. One of these groups prioritized incorporating causal inference into statistics courses. As a result, the final deliverable for this working group was a curriculum guideline for developing a fourth-year undergraduate course in causal inference. At the start of the unconference, the specifics of the final deliverable were outlined as follows:

An overall course curriculum on causal inference (including lesson plan, learning objectives, and an overall structure of course content) with two fundamental pillars: experimentation and quasi-experimentation. A stretch goal of this working group would be incorporating data science-flavored topics such as A/B testing.

Nevertheless, during the group meetings, the specifics of the final deliverable were revised to go beyond experimentation and quasi-experimentation to include observational studies. Additionally, the brainstorming sessions focused more on statistical perspectives rather than a data science approach. As a result, the final deliverable included in this repository has the following specifics:

Our final deliverable is an overall fourth-year undergraduate course curriculum on causal inference (including a limited lesson plan, learning objectives, and an overall structure of the course) based on a general causal inference roadmap of eight stages: (1) research question, (2) causal model representing knowledge, (3) counterfactuals and causal parameter definition, (4) statistical model definition, (5) model fitting, (6) result interpretation, (7) data analysis report.

The above causal inference was inspired on the one used in the course Introduction to Causal Inference by Petersen and Balzer (2014).

2 Website Overview

  • Introductory Slides: These slides were preliminary before shaping them into our final deliverable.
  • Course Outline: A general course outline including course rationale, course description, general learning goals, intended student audience, course format, assessment criteria and grading.
  • Course Roadmap: The roadmap including the above eight stages with brief explanations.
  • Limited Course Lesson Plan: A limited, structured guide that outlines what will be taught during a specific class session or over a short unit.

3 License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Creative Commons License

References

Petersen, Maya, and Laura Balzer. 2014. “Introduction to Causal Inference.” https://ctml.berkeley.edu/introduction-causal-inference.