Stats program by tiers?

[biostat_prof]

If you are applying for a PhD program (and particularly if you are interested in academia), the reputation of your dissertation adviser is far more important than the reputation of the program where you did your PhD. Thus, I always tell students not to obsess over rankings, because you would be much better off attending a lower-ranked school and working with a star adviser than attending a higher-ranked school and working with a mediocre adviser. And you also have to consider whether or not a department is strong in the areas in which you are interested. To give a couple specific examples, Michigan's biostat program is a fantastic option if you are interested in genetics, but it's not such a great option for most other areas. Likewise Harvard's stat department is small, but it's definitely the best in the world for missing data and a couple other areas. But it would be a mistake to attend Harvard's stat department unless you are interested in one of the research areas where they are strong.

 

That said, it can be advantageous to attend a higher-ranked department, particularly if you're not sure about your research interests when you start grad school (which in my experience is true of a high percentage of students). Typically the higher-ranked departments tend to have strong faculty in a wide variety of research areas. They also tend to be better-funded, so there is less of a concern about having your funding run out after a couple years. Nevertheless it would be a mistake to choose a PhD program purely based on rankings given that there isn't a huge difference between many of the top programs. My advice would be to carefully examine the research interests of the faculty at each school. It's also worth considering funding, quality of life, and things like that. Rankings would be very low on the list of things that I would consider.

 

Having said all that, if you want my feedback on the rankings that people have proposed earlier, I would put UNC in the top tier for biostatistics. Maybe Michigan as well, although they are hard to rank due to being so heavily focused on genetics. But most people say that UW/Harvard/Hopkins/UNC represent the top tier of biostatistics and sometimes Michigan depending on who you talk to. All of the four aforementioned schools have large departments with diverse faculty research interests so practically any student should be able to find a good adviser at any of those schools. After that, I would say that the research interests of the faculty and availability of funding should be more important than rankings, because many of the remaining departments are strong in a couple areas but very weak in other areas.

 

As for statistics, I'm less familiar with the gossip about the quality of the various statistics programs. That said, even among the top-ranked schools, one should consider the research interests of the faculty and how they align with your interests. Some departments have the reputation of being more theoretial (e.g. Berkeley, Chicago) whereas others are more applied (e.g. Stanford, CMU), for instance. But I'm basing this largely on gossip that I heard years ago, so take it for what it's worth. I recommend that you carefully research the faculty research interests of each department you are considering.

[5151]

If you are applying for a PhD program (and particularly if you are interested in academia), the reputation of your dissertation adviser is far more important than the reputation of the program where you did your PhD. Thus, I always tell students not to obsess over rankings, because you would be much better off attending a lower-ranked school and working with a star adviser than attending a higher-ranked school and working with a mediocre adviser. And you also have to consider whether or not a department is strong in the areas in which you are interested. To give a couple specific examples, Michigan's biostat program is a fantastic option if you are interested in genetics, but it's not such a great option for most other areas. Likewise Harvard's stat department is small, but it's definitely the best in the world for missing data and a couple other areas. But it would be a mistake to attend Harvard's stat department unless you are interested in one of the research areas where they are strong.

 

That said, it can be advantageous to attend a higher-ranked department, particularly if you're not sure about your research interests when you start grad school (which in my experience is true of a high percentage of students). Typically the higher-ranked departments tend to have strong faculty in a wide variety of research areas. They also tend to be better-funded, so there is less of a concern about having your funding run out after a couple years. Nevertheless it would be a mistake to choose a PhD program purely based on rankings given that there isn't a huge difference between many of the top programs. My advice would be to carefully examine the research interests of the faculty at each school. It's also worth considering funding, quality of life, and things like that. Rankings would be very low on the list of things that I would consider.

 

Having said all that, if you want my feedback on the rankings that people have proposed earlier, I would put UNC in the top tier for biostatistics. Maybe Michigan as well, although they are hard to rank due to being so heavily focused on genetics. But most people say that UW/Harvard/Hopkins/UNC represent the top tier of biostatistics and sometimes Michigan depending on who you talk to. All of the four aforementioned schools have large departments with diverse faculty research interests so practically any student should be able to find a good adviser at any of those schools. After that, I would say that the research interests of the faculty and availability of funding should be more important than rankings, because many of the remaining departments are strong in a couple areas but very weak in other areas.

 

As for statistics, I'm less familiar with the gossip about the quality of the various statistics programs. That said, even among the top-ranked schools, one should consider the research interests of the faculty and how they align with your interests. Some departments have the reputation of being more theoretial (e.g. Berkeley, Chicago) whereas others are more applied (e.g. Stanford, CMU), for instance. But I'm basing this largely on gossip that I heard years ago, so take it for what it's worth. I recommend that you carefully research the faculty research interests of each department you are considering.

Great, thanks. Although I would say that Stanford and Berkeley are strong in both theory and applied field. 

[Newton01]

Stats and biostats departments should really be ranked separately. 

 

For stat, my rankings would be:

 

TIER I

Stanford; UC - Berkeley; Harvard; Chicago;

 

TIER II

UW - Seattle; CMU; Duke; UW - Madison; UM - Ann Arbor; NC State;

 

TIER III

Wharton; Cornell; Columbia; Minnesota; UCLA; UNC; Yale;

 

For biostat, I would say:

 

TIER I

Harvard; Johns Hopkins; UW-Seattle;

 

TIER II

Minnesota; Michigan; UNC;

 

TIER III

UC-Berkeley; Penn; Emory; Columbia; UCLA; Brown; UW-Madison;

I know this is an old topic, but I was wondering if anyone had any input or updates.

 

Also, I've seen a lot of admissions to schools that are similarly ranked by US News/NRC (understanding that rankings are certainly not the be-all end-all) to some of the schools being discussed -- for example Ohio State/Penn State/Texas A&M/University of Florida. Any input on where those would fit in?

[Karoku_valentine]

What about Univ Minnesota, Twin Cities? Is it really that high?

[Goliath15]

I would say CMU is very strong in Stats, especially since Machine Learning is merging with the field.

 

My rankings for Stats (with top to bottom gradient in each tier):

 

Tier 1

Stanford, Berkeley, CMU (Stats + ML), Harvard

 

Tier 1.5

Columbia, Duke, Washington, Wharton, Chicago

 

Tier 2

Michigan, UW Madison, Yale (tiny program)

 

Tier 2.5

UCLA, UNC, NC State, Cornell, Rice

 

Tier 3

Penn State, Ohio State, Minnesota, and a lot of others

Edited March 13, 2015 by Goliath15

[stats_applicant]

Tier 1+: Stanford, Berkeley

 

Tier 1: Wharton, Harvard, Chicago, CMU

 

Tier 2: Washington, Wisconsin, NC State, Duke, Michigan

 

Tier 3: A bunch of others

[Stat Assistant Professor]

IMO, Statistics doesn't have the same pedigree issues that a lot of other academic disciplines do (although certainly it can't be denied that someone with a doctorate from Stanford, Berkeley, Chicago, Wharton, Harvard, etc. might have an edge when applying to jobs within these same top-tier schools). A well-known advisor, publications in good journals, and a prestigious post-doc can definitely go a long way in getting a job at an R1 institution, if not at one of the aforementioned top-tier schools.