What Can You Do With a Degree In Statistics? – A Follow Up

Posted to Statistics  |  Nathan Yau

This past Friday, Columbia University stat graduate students hosted a symposium on careers for students in statistics. Kenneth Shirley, a stat post doc, was nice enough to write this guest post about the conference so that we can all learn from it. There were two panels – academic and industry – including representation from Google, AT & T, and Pfizer.

Yesterday’s conference at Columbia about career opportunities for Statistics Ph.D. graduates was a great success. It was organized by the graduate students in Columbia’s Stats department and advertised on the web here:


Andrew Gelman made some opening remarks, and then there were two panel discussions, each with five professional statisticians. The first panel consisted of academic statisticians, and the second panel consisted of industry statisticians. Here are some comments I found interesting.

1. Andrew mentioned in his opening remarks that wherever you go (as a Statistics Ph.D.), you will be a teacher. It may be lectures to undergrads in college, or lessons to co-workers in your company, but you will always have to teach in some capacity, because statistics are useful, and understanding statistics is a relatively rare specialty. I guess the point is don’t try to duck your teaching responsibilities as it’s part of being a statistician in any context. Good point, I thought.

Eric Bradlow echoed this opinion and pointed out that being a bad teacher is a miserable experience (Eric is a very good teacher, so maybe he’s just imagining what it would feel like), so you might as well work to get good enough at teaching so that it’s tolerable. A few of my lectures have been duds, so I personally know the feeling…

2. Rebecka Jornsten suggested that doctoral students on the job market avoid padding their CVs with a million working papers, and also suggested printing out papers and attaching them with the job application. This seems like a pretty fair and common-sense rule: If you list it as a working paper, it should be something that is written up, and could be printed out and sent if requested; otherwise, call it a “research interest” rather than a working paper.

3. Everyone (industry/academic/other) agreed that it’s a good idea to attend conferences, present work, and be social. Industry professionals encouraged job applicants to negotiate for conference travel and independent research time as part of their job.

4. All of this career talk reminded me of Xiao-Li Meng’s statement that he posted for his candidacy for ASA President (elections are going on now). He stresses that the supply of young talent in statistics appears to be less than the future demand for statisticians (good to know that we’re in demand), and argues that we should encourage high school and college students to pursue statistics through good teaching at introductory levels, and through good professional compensation. All the statements by candidates for various positions are worth reading, and can be found here:


On the whole, the conference got me excited about our profession. Hopefully the Columbia students can follow up on this with future events – it was an all-around success.

You can also find more comments on the symposium at Andrew’s blog. Thanks, Kenneth!


Graphical perception – learn the fundamentals first

Before you dive into the advanced stuff – like just about everything in your life – you have to learn the fundamentals before you know when you can break the rules.

19 Maps That Will Blow Your Mind and Change the Way You See the World. Top All-time. You Won’t Believe Your Eyes. Watch.

Many lists of maps promise to change the way you see the world, but this one actually does.

Life expectancy changes

The data goes back to 1960 and up to the most current estimates for 2009. Each line represents a country.

Think Like a Statistician – Without the Math

I call myself a statistician, because, well, I’m a statistics graduate student. However, the most important things I’ve learned are less formal, but have proven extremely useful when working/playing with data.