How did you decide?

[Bayesian1701]

I am going to have to make a decision now on where to go. I am interested how others have or will make their decision.  I have a spreadsheet to rank based on various things (stipend, research fit, completion rate, location etc) and I am deciding how I want to weight things.  How I feel after visits is going to be a big factor for me because I want a great environment.  For those of you already attending how did you make your decision? 

Also if you have thoughts on my personal decision for which program is best for me as an aspiring academic wanting to do social science Bayesian stats between Virginia Tech, Missouri, Texas A&M, UT Austin, and Baylor I would like that as well. I don’t have an official offer at Baylor but I am assuming I will get in (my visit went well and I think I will get in), and that my Virginia Tech unofficial acceptance with funding will be finalized.  I am visiting all of them and will seek guidances from professors at my institution but I wanted more opinions.  

[bayessays]

I'll talk about some important factors and fit them into your specific situation.

Research fit:  For someone who is 100% sure they want to do Bayesian statistics, UT Austin is about as good of a place as you can go, besides maybe Duke.  It's really not even close.  TAMU has some great people doing Bayesian statistics too, but it's not the singular focus of the department in the same way.  Mizzou and VaTech are also Bayesian focused, but they can't really compare to the professors at UT.  I'd encourage you to keep an open mind of what you'd like to do  Writing a PhD dissertation in statistics is going to be very math-heavy, and while your thesis might have some applications to social science, you're almost certainly not just going to be able to continue a project you worked on as an undergraduate.

Prestige/rank:  This really fits into research fit, and it's hard to separate the two - how many well-known professors are there that match your research interests?  You'll have no problem getting a job working with many of the top people at UT or TAMU. Mizzou and VaTech have some good Bayesians too, but not in the same quantity and you don't want to put all your eggs in one basket.  Look at the professors who match your research interest, and see how many of them have published recently in top journals like JASA, JRSS, Biometrika, Annals, etc.  If you know you want a faculty job in a PhD-granting statistics department, it's in your best interest to go to UT or TAMU.

Department environment:  Going to UT and being in a cohort of 5 is going to be very different than being in a large cohort at TAMU.  It will be easier to form relationships with faculty in the smaller department, and you'll get more personal attention.  On the other hand, there are less choices for professors to work with, and there are less classmates to form relationships with.  Another thing I would consider is the student demographics.  In a department like Mizzou, it's likely going to be 95%+ students from China.  The sad truth is that it's easier to form friendships with other domestic students, and your first couple years will be easier if you are friends with your classmates.

Location:  You're going to be spending 5 years of your life in this place.  Do you want to spend it in Austin or College Station?  Most people don't even apply to TAMU despite its rank because they can't imagine living there.  This is a personal choice. 

Competitive atmosphere/how hard is the program: I'm not saying you should be choosing the easiest program, but seriously keep this in mind.  A student at a top school might spend 10+ hours a day doing homework, and worry about passing an incredibly difficult qualifying exam after their first year.  For instance, UT has a qualifying exam covering basic first year Casella&Berger level probability (and similar level topics) while TAMU has a more advanced qualifying exam covering measure theory.  Figure out how many (domestic) students get through their quals.

Stipend/Cost of living:  If it's close, this can help make the decision easier.  Yes, it'd be awesome to get the $40k a year fellowship at VaTech over the $20k a year stipend at UT, but keep in mind the big picture.  Personal choice.