Statistical assumptions and model checking

[malcolmbarrett]

I've had an idea set seed in my mind for the chapter on evidence about statistical inference. We can't cover this topic in depth but we can clarify what we're assuming about population, sampling, model assumptions, taking evidence from many studies, etc.

Causal inference will likely fail without these things in place, but the relationship is like how regression can be used in any type of analysis, not just causal. Statistical inference can be used in all three approaches. I keep thinking "it's orthogonal" but that's not quite right.

Links:
https://www.tidyverse.org/blog/2022/11/model-calibration/
https://easystats.github.io/performance/articles/check_model.html

Might be better as an appendix