It's time for statistics departments to start supporting their applied students

Statistics departments are failing their applied students. In this post, I have a lot of opinions and give two pieces of advice: statistics departments need to start supporting their applied students, and they need to hire applied faculty.

Rebecca Barter

I graduated with a PhD from UC Berkeley’s statistics department in December. My PhD dissertation consisted of three 100% applied projects (one of which was a piece of open-source software). This is, unfortunately, incredibly rare. Over the past few years, I’ve had a number of current and prospective statistics PhD students both at Berkeley and outside Berkeley get in touch with me to ask me how I made my way through a statistics PhD by working only on applied projects.

Which hypothesis test should I use? A flowchart

A flowchart to decide what hypothesis test to use.

Rebecca Barter

Many years ago I taught a stats class for which one of the topics was hypothesis testing. Many of the students had a hard time remembering what situation each test was designed for, so I made a flowchart to help piece together the wild world of hypothesis tests. While the flowchart isn’t pretty (if I made it today, it would be much more attractive), I feel like it might be useful for others, so here it is:

Understanding Instrumental Variables

Instrumental variables is one of the most mystical concepts in causal inference. For some reason, most of the existing explanations are overly complicated and focus on specific nuanced aspects of generating IV estimates without really providing the intuition for why it makes sense. In this post, you will not find too many technical details, but rather a narrative introducing instruments and why they are useful.

Rebecca Barter

Suppose, as many do, that we want to estimate the effect of an action (or treatment) on an outcome. As an example, we might be interested in estimating the effect of receiving a drug vs not receiving a drug on the incidence of heart disease. In an ideal futuristic world, we would take each individual in our population and split them into two identical humans: one who receives the treatment and the other who doesn’t.