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.
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:
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.