Stat/Biostat PhD Application Profile Eval - Fall 21

[permanent_gradstudent]

Type of Student: Domestic White Male

Undergrad and Graduate Institution: Large State School (~Top 20 Public, Top 60 USNews General Ranking)

Undergrad Major: Neuroscience
GPA: 3.45 (BSc)

Grad Degree: Biostatistics (MS)

GPA: 4.00 

GRE General Test: 165 V/ 168 Q/ 5.5 AWA

 

Programs Applying: Statistics/ Biostatistics PhD 

 

Research Experience: 

2.5 years as data analyst/statistician in academic research lab. One first-author publication, two "co-first-author" publication (so second-author), another second-author, and a a handful of 3-4 poster presentations including one oral presentation with a "top undergraduate research" award at a local/institutional research symposium. All of this research very epidemiological/clinical (e.g. clinical prediction model of outcome, large cohort studies), no formal math or statistics work (outside of applied data analysis).  

In addition to statistician position, currently collaborating with a clinical lecturer to build out a Python course by aggregating python resources, simulating data and creating assessments for course projects 

 

Awards/Honors/Recognitions: 

Full tuition undergraduate scholarship; Two fellowships/honors programs in undergraduate (mildly competitive but completely unrelated to statistics); Small/partial tuition graduate scholarship; Summer 2020 scholarship to attend UW's SISMID; 

 

 

Letters of Recommendation: 

Collaborative clinical lecturer from Python work mentioned above (Clinician, so would it be better to identify a third writer with a quantitative PhD rather than clinical degree?) 

Professor with whom I've worked with in research outside of coursework

Thesis supervisor and previous professor

 

 

Math/Statistics Grades: 

Undergraduate: Multivariate Calculus (A), Data Science (A), Experimental Design (B), Applied Regression (C)

Graduate: Applied Regression (A), Statistical & Probability Theory (A+), Health Data Science (A), Python for Data Analytics (A), Mathematical Methods for Statistics (A+), Machine Learning (A), Biostatistical Methods (A) and currently Linear Models, Statistical Learning, Neural Data Science

 

 

Programming/ Software skills: R, SAS, Stata, Python, SQL, MatLab (basic)

 

Note: I'm worried about my relatively poor and less-quantitative undergraduate performance (in general and in the few statistics courses I took). My recent academic performance has been much improved, and I'm hoping that some of the research experience/background might be able to moderately compensate. I'm also unsure if my graduate coursework is sufficiently rigorous wrt math/theory courses to be competitive for any statistics programs. 

Schools: 

Currently casting a wide net with applications to both Stat and Biostat programs, but mostly as I am unsure how to best evaluate how competitive my application is. I'm currently assuming that my school list may be optimistic and should include a few more realistic options, but would greatly appreciate any feedback!

 

In no particular order:

 

Carnegie Mellon - Stats 

Washington (Seattle) - Biostatistics

Emory - Biostatistics

Johns Hopkins - Biostatistics

Yale - Biostatistics

UNC - Biostatistics

Brown - Biostatistics

Vanderbilt - Biostatistics

Virginia Commonwealth University - Biostatistics

Virginia Tech - Statistics

Duke - Biostatistics

 

Edited September 17, 2020 by permanent_gradstudent

[bayessays]

Your math background is too thin to get into any ranked PhD program in biostatistics. All worthwhile programs require a linear algebra course, at a bare minimum, which you don't seem to have. You need to take more math. 

[permanent_gradstudent]

1 hour ago, bayessays said:

Your math background is too thin to get into any ranked PhD program in biostatistics. All worthwhile programs require a linear algebra course, at a bare minimum, which you don't seem to have. You need to take more math. 

Sorry if my course section isn't very descriptive, the "Mathematical Methods for Statistics" was a combined section of some calculus and the majority of linear algebra, and the Linear Models course in which I am currently enrolled is a PhD level course of applied linear algebra

[StatsG0d]

16 hours ago, permanent_gradstudent said:

Sorry if my course section isn't very descriptive, the "Mathematical Methods for Statistics" was a combined section of some calculus and the majority of linear algebra, and the Linear Models course in which I am currently enrolled is a PhD level course of applied linear algebra

The Mathematical Methods for Statistics course is not a substitute for a standalone, proof-based linear algebra course taken within a math department, and since your linear models course only covers applied linear algebra, this would also not satisfy admissions criteria. 

It seems you have not taken any proof-based math courses. I think most departments would automatically reject you for this. Maybe some lower-ranked biostatistics departments who are only focused on applied work would accept you, but I still think, at best, you would be a marginal case.

At a minimum, you should take a proof-based linear algebra course in a math department and Analysis I. If you add on Analysis II, you give yourself a tremendous boost and would be competitive for biostats programs outside Washington and stats programs outside the top-15. Add on something like complex analysis or abstract algebra, and you will have a good shot almost anywhere.

You should note that admissions for stats/biostats are much more competitive than they were a half decade ago when I applied. With COVID-19, probably more people are going to be applying for grad schools making it even harder.

Edited September 18, 2020 by StatsG0d
grammar

[permanent_gradstudent]

I appreciate the detailed feedback from both of you! I know there are opportunities to take Analysis I and a more thorough Linear Algebra course, and possibly a later Abstract Algebra course. 

It certainly sounds like it would be to my advantage an applicant (and statistician) to enroll in some additional theory coursework and build these foundations for a more competitive Fall 2022 application