Diving Into Statistics: Better Late Than Never
Finishing my second course in biostats/epi at age 77 has not been easy.
Posted Nov 21, 2020
I confess to a weakness: I often buy a $2 Mega-Millions ticket when the prize is over $100 million. Otherwise, not being much of a gambler, my practical experience with statistics over the last 50 years has been limited to talking with patients about risks and benefits, trying to invest my retirement funds wisely, and writing peer-reviews for papers submitted for medical journals.
About two years ago, I realized that writing, "I'm not a statistician, but these numbers don't look right" was a weak-kneed excuse. A position of authority—as an advisor or reviewer—carries a responsibility. Either know enough to comment, learn enough to comment, or don't comment. So I did some careful searching for online courses and applied to the University of South Florida School of Public Health as an online student.
Online classes don't allow the quality of interaction that the classroom experience offers. But, as a student, I can set my own pace except for the one weekly lecture at 6 p.m. on Wednesdays. The courses have been rigorous; the homework assignments are demanding but instructive.
The real eye-opener has been the clarity of the frequentist statistical approach. This is the branch of statistics that deals with hypothesis-testing: means, standard deviations, confidence intervals, and p-values. And no, I'm not going to try to explain any of those terms. I have just two points. First, the message is that all those terms make sense; they are not beyond the grasp of mere mortals. Second, data drive more and more of our professional decisions and our lives every day. For ourselves and our patients, it's critical to learn how data are presented and interpreted.
If you are mid-career or beyond, consider refreshing your skills.