Biostats PhD?

[Nibor6000]

I could use some advice.

I have an MS in nutrition and work in public health (chronic disease prevention). Last year I started thinking about doing a PhD in public health - most likely in epi. I decided to start taking the core public health classes to get my feet wet. Well, I'm not in love with epi, but I adore biostats. I have two semesters of undergrad calculus. Some PhD programs actually find this to be enough of a foundation, so I will be focusing on applying to those programs.

The big question: what are biostats PhD programs looking for? I have no research experience involving biostats. I just know that I really enjoy manipulating data and want to learn more about statistical methods - I read stats textbooks in my free time! I have a 3.9 undergrad GPA and a 4.0 graduate GPA. GRE scores: 780V/730Q/5.0A. Do I stand a chance of getting in with that low of a quant score? Should I do another MS in Biostats to strengthen my application before applying to PhD programs? Seems a bit of a waste since I already have an MS albeit in a different field. What else could I be doing to improve my chances?

I await your sage responses! Thanks!

---

Members don't see this ad.

 

[GoodmanBrown]

I could use some advice.

I have an MS in nutrition and work in public health (chronic disease prevention). Last year I started thinking about doing a PhD in public health - most likely in epi. I decided to start taking the core public health classes to get my feet wet. Well, I'm not in love with epi, but I adore biostats. I have two semesters of undergrad calculus. Some PhD programs actually find this to be enough of a foundation, so I will be focusing on applying to those programs.

The big question: what are biostats PhD programs looking for? I have no research experience involving biostats. I just know that I really enjoy manipulating data and want to learn more about statistical methods - I read stats textbooks in my free time! I have a 3.9 undergrad GPA and a 4.0 graduate GPA. GRE scores: 780V/730Q/5.0A. Do I stand a chance of getting in with that low of a quant score? Should I do another MS in Biostats to strengthen my application before applying to PhD programs? Seems a bit of a waste since I already have an MS albeit in a different field. What else could I be doing to improve my chances?

I await your sage responses! Thanks!

Hi Nibor. I feel fairly qualified to answer as I have a Masters in stats (and was in a PhD program). I also have a few friends from the program getting their PhDs now.

So, to start off, I think you're in really good shape in terms of applications. I think the 730 should be pretty good for most schools, though if you think you could get higher, I would probably take the GRE again. I got a 780 and had a professor at a good program tell me that was good enough to get into most high-end schools, FWIW. The deal with the GRE is that the quantitative scores are bimodal. So, you've got a big hill at ~600-620 for all the history and English majors in college, but then you've got a big hill at 770-800 for all the math and engineering majors. So, a bit higher might not hurt.

In terms of pre-reqs, my school required 3 semesters of calc, linear algebra, and a basic stats course. Real analysis will really help too because it lets them know you can cut the high-end stuff. You're welcome to apply to schools that require less, but they may not be as prestigious. The prestigious schools usually have more funding also, so if you take your time and get some other classes under your belt, you will probably be a happier person during your years in school.

As for research, don't sweat it. I'm sure it won't hurt, but you'll do plenty of that in your program, so it's not a big deal. I didn't have any, nor did most of the people coming into my program.

Here's the most important part of my post, whatever you do, apply to PhD programs. The way it works at my school is the way I think it works at many: If you apply to a PhD program, you get way more funding opportunities (aka, they pay your tuition and pay a stipend for you to study there.). This is upfront payment for the eventual fame they hope you'll bring back on the school. You'll have to earn an MS degree before your PhD anyhow, so there's no point to applying to an MS program with the intention of applying to the PhD program later, and it will only hurt your funding chances.

Also in terms of the MS: I'm sorry to rain on your parade, but no school I know of will let you jump straight into PhD studies without the Masters first. A Masters in nutrition just won't cut it. Some (like my school) will award you a Masters on your way to a PhD, and some will just let you work straight toward the PhD. But expect 2 years of Masters level classes and then about 2 years of PhD level classes, then about 1.5-2 years of research for your dissertation. You're looking at 5.5-6 years before you get your PhD. It's a long road, I know.

Hmm.. Can't think of any other advice. Hope it helps some. Feel free to PM if you have other questions, or post them here, and I'll check back.

 

[Stories]

Here's the most important part of my post, whatever you do, apply to PhD programs. The way it works at my school is the way I think it works at many: If you apply to a PhD program, you get way more funding opportunities (aka, they pay your tuition and pay a stipend for you to study there.). This is upfront payment for the eventual fame they hope you'll bring back on the school. You'll have to earn an MS degree before your PhD anyhow, so there's no point to applying to an MS program with the intention of applying to the PhD program later, and it will only hurt your funding chances.

This is particularly good advice because even if a school thinks you're not cut for the PhD program, but would want you on as a MS student, they will often defer you there.

The biggest piece of advice that Goodman didn't give is work on a cogent personal statement. That piece of writing is the single most important document in your application because of its focus on what your research experience is and where you want to do your research. Good ideas and a strong match with faculty will get you accepted.

 

[GoodmanBrown]

This is particularly good advice because even if a school thinks you're not cut for the PhD program, but would want you on as a MS student, they will often defer you there.

The biggest piece of advice that Goodman didn't give is work on a cogent personal statement. That piece of writing is the single most important document in your application because of its focus on what your research experience is and where you want to do your research. Good ideas and a strong match with faculty will get you accepted.

I'm of two minds on this. First off, I definitely didn't put exact research thoughts on my personal statement because I had no idea what stats research was. So, my personal statement was totally devoid of research aims.

However, I have a friend who was applying to history PhD programs, and Stories's advice was definitely true there. You had to identify a specific professor or two and write your PS with specific references to their research and how you wanted to work within their scope. So, I don't know what to do with the personal statement.

Lastly, have you thought of visiting a few schools? The PD of my program was a great guy, talked with me for about an hour, and I think it really helped my app to meet him and talk through the program's strengths.

Finally, if you're interested in biostats, you may want to explore the differences between schools that are more focused on theoretical stats vs. applied. For instance, UNC-Chapel Hill has a fairly theoretical focus while its neighbor NC State has a more applied focus. Depends on where you want to work after school.

 

[Nibor6000]

This has been very helpful and much appreciated. If anything else springs to mind, or anyone else has some thoughts, please let me know.

Thank you!

 

[Stories]

However, I have a friend who was applying to history PhD programs, and Stories's advice was definitely true there. You had to identify a specific professor or two and write your PS with specific references to their research and how you wanted to work within their scope. So, I don't know what to do with the personal statement.

This is definitely true of epi. Even if you're not sure exactly what you want to research, it wouldn't be a bad thing to identify broadly what you're interested in.

Goodman: I'm kind of curious to know what your PS consisted of? Why did you leave the PhD program? Just didn't need it anymore for your career goals?

 

[GoodmanBrown]

This is definitely true of epi. Even if you're not sure exactly what you want to research, it wouldn't be a bad thing to identify broadly what you're interested in.

Goodman: I'm kind of curious to know what your PS consisted of? Why did you leave the PhD program? Just didn't need it anymore for your career goals?

My personal statement was mainly an explanation of how I got interested in statistics in general. I was a math major in college and had only had one stats and one probability course, so I explained how I'd come to decide on graduate school in stats.

One note here is that my biostats masters was within the overall statistics department. My school didn't have a school of public health or even an epi program, so I was applying directly to a stats department. As such, people's research interests were "stochastic processes," "group sequential methods," "semiparametric methods," etc. Stuff that I didn't have a damn clue what it meant. So, I think you really need to research your schools and decide what will work best for your PS for each school.

Finally, to answer your PhD question, I decided to make a career change. I originally thought I wanted to do biostatistics in an academic setting, but decided eventually that an MD was more what I was looking for. I still do research with my biostats knowledge, but as a long-term career, I'm going to do research in the context of being an MD instead of a straight statistician.

 

[bigpharmD]

Finally, to answer your PhD question, I decided to make a career change. I originally thought I wanted to do biostatistics in an academic setting, but decided eventually that an MD was more what I was looking for. I still do research with my biostats knowledge, but as a long-term career, I'm going to do research in the context of being an MD instead of a straight statistician.

that seems like a big change from an academic statastician...good luck

 

[Stories]

My personal statement was mainly an explanation of how I got interested in statistics in general. I was a math major in college and had only had one stats and one probability course, so I explained how I'd come to decide on graduate school in stats.

One note here is that my biostats masters was within the overall statistics department. My school didn't have a school of public health or even an epi program, so I was applying directly to a stats department. As such, people's research interests were "stochastic processes," "group sequential methods," "semiparametric methods," etc. Stuff that I didn't have a damn clue what it meant. So, I think you really need to research your schools and decide what will work best for your PS for each school.

Finally, to answer your PhD question, I decided to make a career change. I originally thought I wanted to do biostatistics in an academic setting, but decided eventually that an MD was more what I was looking for. I still do research with my biostats knowledge, but as a long-term career, I'm going to do research in the context of being an MD instead of a straight statistician.

Ahh, so clinical trials it is. Good luck in your endeavors!

 

[GoodmanBrown]

It's funny you should mention that because that's what I actually did for a year after graduating. I definitely hope I never have to do that again. I mean, good things come out of clinical trials certainly, but I definitely felt like rules were too readily bent to give a ****ty new drug a positive spin...

No, I'm more hoping to be an academic attending and do large-scale retrospective studies. Stuff like the NSQIP.

 

[statsguy]

I could use some advice.

I have an MS in nutrition and work in public health (chronic disease prevention). Last year I started thinking about doing a PhD in public health - most likely in epi. I decided to start taking the core public health classes to get my feet wet. Well, I'm not in love with epi, but I adore biostats. I have two semesters of undergrad calculus. Some PhD programs actually find this to be enough of a foundation, so I will be focusing on applying to those programs.

The big question: what are biostats PhD programs looking for? I have no research experience involving biostats. I just know that I really enjoy manipulating data and want to learn more about statistical methods - I read stats textbooks in my free time! I have a 3.9 undergrad GPA and a 4.0 graduate GPA. GRE scores: 780V/730Q/5.0A. Do I stand a chance of getting in with that low of a quant score? Should I do another MS in Biostats to strengthen my application before applying to PhD programs? Seems a bit of a waste since I already have an MS albeit in a different field. What else could I be doing to improve my chances?

I await your sage responses! Thanks!

I am second year currently working towards my PhD in Statistics. When I was applying for grad school, I considered going the Biostats PhD route, but ultimately decided against it - but I remember the application process very well. Unfortunately I disagree with some of the replies in the thread.

First of all, the General GRE score is pretty much useless and irrelevant. As long as you score above a 700 or so, you're fine. A high Math subject test score might help at some big name universities. Don't even bother with it unless you were a Math major in undegrad and can devote a ton of time studying for it.

With regards to requirements, I highly doubt Calc I and II (that you took years ago) is enough to get you in. Every school I looked at required at least Linear Algebra, most required two semesters of Real Analyis, with the rest STRONGLY recommending it. In addition, all schools required a standard two semester sequence of Prob & Stats. Obviously, the more courses, the better.

Your GPA is good, of course, but graduate schools really only care about your Math and Stats GPA, which will be problematic for you. Research experience is VERY helpful, ESPECIALLY for PhD admissions, where independent research is an absolute requirement.

My biggest reason for going against the Biostats PhD route was that it required an MS in Statistics or Biostatistics, which much of the time was unfunded. Most biostats/stats jobs can be done by a Stats PhD anyways. Hell, I know a Math PhD with a MS in Stats who works for the University of Chicago as a biostatistician.

Considering the amount of brilliant US/Chinese students that struggle at the PhD level, no way in hell you'd make it, even if you got in, with your background. You need a better foundation. What Goodman said is pretty much spot on, except for one thing: lots of schools do not fund the MS and wont even let you apply to the PhD program unless you have an MS in Statistics. Thus, you may be better off applying to a decent MS program. If you get into one, do VERY well, make sure you butter up at least 3 profs for letters of rec down the road, and try to get a research gig. Otherwise take a bunch of Math/Stat prereqs and reapply the following year.

For the Biostats PhD, you want to go to a reputable school (does not have to be top 5, but should be at least "good") with highly regarded faculty, connections to industry, and lots of money for research. It may be a long road, but a 30+ year career doing what you love is absolutely worth it. Plus, about half of the students in my department started their Phd Work were older than 25. One lady was 38 when she started! That's one unique thing about Stats grad programs.

I really wish you the best of luck. Stats/Biostats is a wonderful career with lots of job opportunities and excellent pay/prestige etc... and I think you are making an excellent choice. Just don't rush it

 

[Nibor6000]

Thanks for the advice. Much appreciated.