I don’t like the idea of providing just recipes, although many students that can see they will use stats in their dissertations want to have some recipes. Emphasizing the concepts with example applications seem to make most sense to students.

]]>Really!

]]>Most of the people I know who did that course who became researchers in stats actually did something else and sort of “fell into” statistics almost by accident (be it through needing more classes and accidentally taking a fun stats course, or by job opportunity).

I became a statistician because my functional analysis (at the time) wasn’t up to developing mesh-free methods for systems of reaction-diffusion equations and statistics seemed like a nice excuse to compute matrix functions.

I had to re-learn all of the content of the intro course, which, for my purposes was completely useless.

That being said, it’s a question of target audience. At least where I did my study (QUT), the 101 stats course is a large “service teaching” course not strictly aimed at mathematicians. I am told that those who were never likely to become a research statistician quite like the “recipe-based” approach.

]]>