Review and improve the introduction to causal inference chapter

[mbcann01]

Overview

In the Fall of 2023, I moved over a bunch of stuff from PowerPoint slides (nearly) verbatim. I was in a rush, so I told myself to move it just move it over and improve it later.

Go back, reread, and improve. PowerPoint doesn't always translate perfectly to book format.

2023-10-05: I wonder if I shouldn't greatly simplify this chapter. Just introduce some of the key terms and ideas for now. Focus on creating DAGs and SCC models in R. Later, I can add more information about what causality is.

Left off at

2023-10-06: I cut a TON of stuff out of the chapter. For the sake of clarity, I thought it better to slim things down and keep them simple. By simple I mean, just focus on using R to create SCC diagrams and DAGs for now. Having said that, I would like to add all the theoretical stuff back into this part of the book when I have the time to do it right. 2023-10-05: Marked up the chapter for revisions. I'm too tired right now, but I think tomorrow I should just focus on making the intro chapter very simple and teaching students how to make SCC models and DAGs with R. 2023-10-04: Moved over the PP slides. There is still a lot of cleaning up to do.

Tasks

• [ ] Start with Simpson's Paradox (see Causal Inference in Statistics)
• [ ] Define cause, causality, causal inference, causal effect estimation
• [ ] Add more detail about the ladder of causation (see below).
• [ ] Define prediction in a more rigorous way. At some point, will prediction get its own chapter or part?
• [ ] Do we want to keep anything from the current "Causal questions" section?
• [ ] Do we want to keep Hill's considerations?
• [ ] If we keep Hill's considerations, we should add a paragraph or so on their history.
• [ ] If we keep Hill's considerations, we should remake the "hills criteria" table directly in R.
• [ ] We need to add a lot of information to the section on Rothman's model of sufficient and component causes.
• [ ] Make causal pies directly in R. Maybe use the causalpie package?
• [ ] Improve the flow of the RMSCC section.

[mbcann01]

Ladder of causation

Fall 2023: I started adding stuff about the ladder of causation, but ran out of time to do it right. I'd like to add it back in the future.

Judea Pearl's ladder of causation

In their 2018 book, The Book of Why, Judea Pearl and Dana Mackenzie do a masterful job of bringing the basics of causal inference, a topic of significant complexity for most people, to the masses. @Pearl2018-im A ladder is one of the metaphors they use to outline the hierarchy of causal thinking we humans (and many other animals) use to navigate the world. Pearl's ladder of causation has three rungs -- seeing, doing, and imagining -- each representing a step up in the power of our conclusions, but also a step up in the complexity of the process required to make such conclusions. For readers who are curious, we highly recommend this book.