What is the best degree to become statistician after PhD in another discipline?

#1
I recently completed a PhD in the social sciences with a heavily quantitative focus in my research. During the PhD I realized my true calling is in statistics with a focus on methodology. So I decided not to pursue a career in my PhD field and am instead looking to go back to university and retrain to become a professional statistician. However, I did not take any mathematics courses at university so I cannot jump into a masters degree right away and may have to do a sequence of undergraduate units first.

What course sequence would you recommend I undertake so that I could become a professional statistician capable of independent research on statistical methodology and theory?

If anyone here has been in a similar situation transitioning into a statistics career after a PhD in another field, I'd love to hear about your experiences.

Thanks, and hello everybody, this is my first post on the Talk Stats forum. :wave:
 

spunky

Smelly poop man with doo doo pants.
#2
have you considered maybe a program on quantitative psychology?

or educational measurement?

those are usually geared towards social scientists with limited background in mathematics.

i kinda like the general requirements that my university asks for in terms of getting accepted into their MSc in Statistics. they're basically:

An introductory course in statistics, and one in probability;
A 3rd or 4th year course in statistical inference;
A 3rd or 4th year course in regression analysis;
at least 3 of the following courses: introductory stochastic processes, 4th year probability, 3rd or 4th year mathematical statistics, design of experiments;
preferably two other 3rd or 4th year courses in statistics or computer science.

if you take those as a general guide, i'm sure you're gonna end up with a good undergraduate-like background in statistics to be ready before graduate studies.
 
#3
have you thought about working towards published methodological briefs in the field of your interest?

The truth is, the cutting edge stuff is almost always a "learn on your own" kind of thing anyway. My course work had nothing to do with learning R or some of the advanced methods I use. For example, my work in education recently caused me to use a boot strapped time series analysis with a bias corrected and accelerated (BCA) confidence interval. Now that my research is complete, I plan on writing up and submitting a methodological brief.

I know at first its who you are on paper that matters for the first job, but then afterwards its mainly what you do in the field that determines what you do.
 
#4
What course sequence would you recommend I undertake so that I could become a professional statistician capable of independent research on statistical methodology and theory?
How many stat classes have you taken already? I think maybe you should have a good applied stats background under your belt? To answer your question though, I am sure you heard of the Central Limit Theorem? Look up online how to prove the Central Limit Theorem and see then if it is your cup of tea. You want to do research on the statistics itself, or do applied statistics? I think from your question I understood you want to do the non applied route? If so then check out the proof I just mentioned.

Check out Rutgers University Statistics Department based in the School of Arts and Sciences for some good information. I think many traditional Statisticians tend to have undergrad Math Degrees. Look up the Prof's CV and check out what they did! If you dig around on Rutgers website enough, I am sure many of your questions can be answered. Rutgers is a very good school and you should use it as a good resource for all your Math and Stat related needs. They even just came out with a Data Science program too.

as an aside: I used to be a declared applied Stat major with a Psychology minor, but after failing Calc 3 and barely passing Calc 2, I decided I was not mathematically prepared enough for it. Hence I had to switch out to Accounting as it was more practical for me and involved numbers. I wish I would have known about Statistics earlier and focus more on the mathematics side of things earlier in my college career! Stats seems like such a good and hot field to be a part of!
 
Last edited:

noetsi

Fortran must die
#5
It likely depends as well on whether you want to teach after your doctorate or do applied work. I think that classes in statistics commonly don't teach what you will actually do in an applied setting, although of course that depends on what type of applied work you do.

Of course with the work you have already done, by the time you get another doctorate you may simply retire anyway :p
 
#7
Thanks for the replies. FWIW, I have begun a Bachelor of Science degree. My first major is Mathematics and Statistics, and my second is Data Science. I think this combo will help me work on my weaknesses which is Mathematics and Statistical Theory and programming skills. There was another course sequence for a "Quantitative Methods" major, but this seemed too weak on mathematics and would better suit someone who has little or no experience as a research worker. I'm hoping to finish the degree in 2 years then begin working.