Setting realistic expectation as a Fall 2025 Applicant

[swannyy]

Hi. I'll be applying to PhD programs in Statistics and Biostatistics program this fall. I finished my undergrad in 2019 and have been working since. I worked in a pharmaceutical industry for a few years in a research setting under supervision of Statistics PhDs, and recently moved to work as a data scientist / statistician at an academic institution. I recently completed an online Master's degree in Statistics, which I started with an intention of applying for PhD program afterwards. I'm trying to get an idea of which schools someone of my caliber should consider applying to set realistic expectation. I am generally looking for programs with opportunities and high interest in Bayesian Statistics.

Undergrad Institution: UIUC
Major: Math, Statistics (Double Major)
GPA: 3.66/4.0 (Math Major GPA: 3.69, Stat Major GPA: 3.)
Type of Student: Domestic Asian Male

GRE General Test: (these are old expired scores, so will need to take them again. I don't quite remember the numbers and these are guesses)
Q: 166
V: 156
W: 3.5

Grad Institution: Texas A&M
Concentration: Statistics (Online Masters)
GPA: 4.00

Research Experience:

• I don't have a publication of my own from the first role, but am included as a second other in 3 papers, several posters, 1 book chapter, and have my own R package (albeit not statistical but visualization).

Awards/Honors/Recognition: Graduate with Highest Distinction in Stat, Distinction in Math

Pertinent Activities of Jobs:

• course assistant for a statistical programming class in undergrad
• part of a undergrad research group for a semester
• worked in a pharma industry doing bioinformatics research for 3 years
• statistical consultant for researchers at a university
• developed programming / statistical workshop for students

Letters of Recommendation:

• I have 3 managers from my previous job who I have had good relationships with willing to write LoF. 2 with PhD in Statistics and 1 in bioinformatics. One person has had solid contributions to Bioinformatics
• 1 professor from my master's curriculum who is willing to write one for me.
• 1 professor from my undergrad statistics course. Might also be less strong, but we've also talked outside of the course setting that they could write a decent one.

Math/Stat Grade: 

• Math: Calc (B), Abstract Lin Alg (A-), Intro to Abstract Alg (A-), Diff Eq (B+), Linear Programming (A), Real Analysis (A), Nonlinear Prog (A-), Numerical Method (A)
• Stat: Probability (A), Biostat (A+), Bayesian Analysis (A), Time Series (A), Survival (A+), Machine Learning course (A)

 

Programs Applying:

Separating Statistics and Biostats program, some programs I'm considering are:

Statistics: Texas A&M University, UIUC, Purdue, University of Iowa, University of Wisconsin-Madison, Ohio State University, University of Conneticut

Biostatistics: Rice University, University of Florida, University of Michigan, University of North Carolina, University of Wisconsin-Madison

 

I didn't have the greatest start to my undergraduate years, but it certainly got better as I took more advanced courses.

Though I have a background in math, I understand that the statistics PhD programs are usually math heavy, and a professor I've spoken to suggested that I consider programs that lean more on the applied side (like biostatistics); that is not to say they don't involve any math at all. The latter would be more in alignment with my previous experience. 

Generally should I view all of these universities as reach schools?
 

[bayessays]

I think the list is pretty reasonable, and nothing sticks out as an "extreme reach."  I might take a few of the more mathematical programs (Purdue, UW-Madison Stats) and replace them with some safer biostat programs (if you're actually interested in biostats, of course).  What do you want to research?

[swannyy]

I think your suggestion makes sense, and I am open to doing research in Biostatistics. I did research in Bioinformatics (genomics), and it wasn't exactly my cup of tea, but there are other applications in biostatistics that still intrigue me. I'm interested in Bayesian methodology and inference the most. Spatial statistics, imaging, and data integration are also interesting to me.

Judging from reading recent candidate's dissertations, it seems like most "Statistics" PhD programs are rather mathematical than applied, and only Biostatistics programs are the "Applied Statistics" programs. For this reason I'm also looking into Data Science PhD programs, but given the novelty of these programs I'm skeptical if that would be the right move in the long run.

[bayessays]

I think a data science PhD from a reputable school would be 100% fine on the industry job market, but might be slightly tougher on the academic job market (but not necessarily - if you were able to research with a great statistician and write in stats journals, I think you'd be fine).

Statistics programs are definitely more mathematical, but there is a broad area of overlap in the both in terms of researching methodology without a ton of proofs.  If you read a paper in Annals of Statistics and want to do research like that, then you definitely need to go to a (good) statistics program.  Similarly, if you want to do to less theoretical math, you should probably go to a biostatistics program.

I wouldn't get too tied up on the "biology" part of biostatistics - most people do not do genetics/bioinformatics.  Biostatistics is so broad that you'll get basically any area of research.  Maybe not astrostatistics - but they'll use the same spatial statistics techniques to analyze MRI images or something instead.  You're always going to be working on new problems in the future anyways, so you can use the clinical trials techniques and get a job doing experiments for FAANG instead, etc.

There is a slight difference in culture at some schools though - if you view yourself as a mathematician, you'd probably fit in more with the liberal arts culture of a stats department moreso than a department that's in a school of public health/medicine.  Probably not a big deal for most people, though.