Why is the Statistics PhD program at Northwestern so selective?

[Bonferroni_Correction]

Hey guys! I'm a current junior in college hoping to apply to Statistics PhD programs this fall (for admissions in 2020). So far, I've been looking into programs to apply to that I have a reasonable chance of getting into, and something that has helped me a lot is when graduate schools post admissions statistics for admitted students by program. For instance, Duke posts their admissions statistics here, which shows that they generally admit about 10% of applicants with an average undergraduate GPA of 3.7 (in recent years). I also found the admissions statistics for Northwestern, and I was shocked to see that the 5-year average acceptance rate was only 5%! In the most recent year on the document, there were 178 applicants and only 6 who were accepted (barely more than a 3% acceptance rate)! 

That struck me as especially strange, since Northwestern Statistics is only ranked 55th in Statistics according to U.S. News. Even Duke, which is ranked 12th, is apparently less selective than NU. I know that part of the reason for the low acceptance rate is that NU has a small program (they have a class size of 5 every year according to the document) and that a lot of students may be applying to the program due to the prestige of the Northwestern name rather than program quality. I also heard that the U.S. News rankings tend to give a lower rank to smaller programs (like NU and Yale) even though these programs might be good programs. However, I feel like this doesn't fully explain how the 55th ranked program accepts only 3% of applicants! Is the program at Northwestern better than what its ranking would suggest? Looking at their recent placements, they actually seem pretty good (a fair mix of industry and academia, with industry placements at Google, etc.). I am interested in the program due to its location, but I don't know if spending almost 100 dollars to apply would be a good investment given how selective the program is. The University of Chicago and Northwestern seem to be the two universities with Statistics PhD programs in Chicago, but I heard that UChicago is exceptionally hard to get into. I didn't anticipate NU to be so selective as well. Can anyone elaborate on the program quality of Northwestern Statistics?

Edited April 8, 2019 by Bonferroni_Correction

[Stat Assistant Professor]

Some programs are just very small and hence very selective. For example, NYU Stern's PhD program in Statistics (ranked #61 by USNWR) has only 6 PhD students total. 

I don't think the rankings are necessarily about program size. UPenn Wharton also has 5 PhD students in each cohort (for a total of 25 Statistics PhD students), but they are highly ranked, with many famous professors like Tony Cai, Ed George, Eric Tchetgen Tchetgen, Dylan Small, etc. It seems to me that the USNWR rankings mainly measure the perceived reputation of a program (i.e. responses to surveys sent to academic statisticians), and reputation is assessed in large part by how many famous faculty there are and what journals they are publishing in. Looking at Northwestern's departmental website, it seems as though only one faculty, Han Liu, is consistently publishing in the top statistics/ML journals and conferences like Annals of Statistics, Biometrika, ICML, and IEEE Transactions. A lot of the other faculty seem to publish in more 'niche' areas like education/behavioral science journals or bioinformatics journals -- these faculty may indeed be very good at that, but they may not be as well known to the statistics community as a whole, and the PhD graduates may be more likely to take academic positions in departments that are not specifically statistics (e.g. I saw on the Placements page that one of their alumni is now faculty at UPenn's Graduate School of Education).

Edited April 8, 2019 by Stat PhD Now Postdoc

[Bonferroni_Correction]

On 4/8/2019 at 7:54 AM, Stat PhD Now Postdoc said:

Some programs are just very small and hence very selective. For example, NYU Stern's PhD program in Statistics (ranked #61 by USNWR) has only 6 PhD students total. 

I don't think the rankings are necessarily about program size. UPenn Wharton also has 5 PhD students in each cohort (for a total of 25 Statistics PhD students), but they are highly ranked, with many famous professors like Tony Cai, Ed George, Eric Tchetgen Tchetgen, Dylan Small, etc. It seems to me that the USNWR rankings mainly measure the perceived reputation of a program (i.e. responses to surveys sent to academic statisticians), and reputation is assessed in large part by how many famous faculty there are and what journals they are publishing in. Looking at Northwestern's departmental website, it seems as though only one faculty, Han Liu, is consistently publishing in the top statistics/ML journals and conferences like Annals of Statistics, Biometrika, ICML, and IEEE Transactions. A lot of the other faculty seem to publish in more 'niche' areas like education/behavioral science journals or bioinformatics journals -- these faculty may indeed be very good at that, but they may not be as well known to the statistics community as a whole, and the PhD graduates may be more likely to take academic positions in departments that are not specifically statistics (e.g. I saw on the Placements page that one of their alumni is now faculty at UPenn's Graduate School of Education).

Thanks for the explanation. Do you think there are any major disadvantages going to one of these lower-ranked but very selective programs (such as Northwestern, NYU, or Yale)? I know there are higher-ranked programs that are much larger and hence can accept more students, but I hope to move to Chicago for family and personal reasons, and I unfortunately don't think I am competitive enough for the Statistics PhD program at the University of Chicago (but I will still apply on the off chance that they'll accept me). If I ultimately have to decide between a higher-ranked larger Statistics program (such as Iowa State or NCSU) vs. a lower-ranked smaller Statistics program located at a 'prestigious' private university (such as Northwestern), what would be the disadvantages in choosing the more selective lower-ranked program? Would choosing to attend these smaller lower-ranked programs limit my options career-wise if I want to keep my options open (since I'm still undecided if I want to go towards industry or academia)? 

[biostat_student]

I got an offer from Northwestern and can confirm that they do have great placements (at least in terms of industry). Just the Northwestern name itself is very helpful. Both the students and staff were very friendly on my visit. I can't advise whether you should apply or not, but here were my reasons for not going (all are based on personal preferences):

• Very small PhD cohort
• Department building is basically a big house instead of its own building
• Coursework seemed rather traditional (e.g. Survey Methods being a required course that gets covered in the qualifying examination)

All the above completely depends on your preferences and end goals. Honestly they could all even be advantages depending on what you're looking for. These were just my observations. I still think its a great department, wonderful location (both life-wise and for career opportunities), and were very accommodating and personalized. If you do apply and get in, they offer a first-year fellowship to some students which you should ask about. Good luck with the process!

[omicrontrabb]

@Bonferroni_Correction 

It mostly depends on how interested you are in academia. If you want to become a professor at a top stats department, where you go matters a lot. It's generally difficult to get a professorship at a school ranked substantially higher than your PhD program/PostDoc. Higher ranked schools have more well known professors who would help you land post-docs, etc. It's possible to work with a well known researcher at a lower ranked school, but going to a school with the goal of working with one specific (in demand) professor is pretty risky.

If you want to go into industry, it doesn't matter nearly as much. Basically every ranked stats PhD program seems to place a lot of people into large banks/big pharma/big tech. That being said, super prestigious companies like Google/Amazon/Goldman Sachs/etc. seem to recruit more heavily from the top schools.

[bayessays]

Small department with a prestigious name and location. Their placements are inflated because Hodges places people at great Education departments.  Never look at industry placements for PhD programs - anyone with a brain and a PhD can get an industry job if they want one.  I would not recommend it unless you need to be in Chicago, or really want to work on educational statistics with Hodges. 

Disclaimer: This is just one story, and may not be indicative of a systemic issue, but I would not recommend applying to the department.  They told me I would be admitted with a fellowship if I wanted to attend and then, after my campus visit (NOT an interview), told me they were giving it to another student.  I had almost already told U Chicago to take me off their wait-list at this point.  I'd be wary of what it's like to be a student at a place that is so willing to screw someone over without an apology. 

[Stat Assistant Professor]

On 6/14/2019 at 11:43 AM, Bonferroni_Correction said:

Thanks for the explanation. Do you think there are any major disadvantages going to one of these lower-ranked but very selective programs (such as Northwestern, NYU, or Yale)? I know there are higher-ranked programs that are much larger and hence can accept more students, but I hope to move to Chicago for family and personal reasons, and I unfortunately don't think I am competitive enough for the Statistics PhD program at the University of Chicago (but I will still apply on the off chance that they'll accept me). If I ultimately have to decide between a higher-ranked larger Statistics program (such as Iowa State or NCSU) vs. a lower-ranked smaller Statistics program located at a 'prestigious' private university (such as Northwestern), what would be the disadvantages in choosing the more selective lower-ranked program? Would choosing to attend these smaller lower-ranked programs limit my options career-wise if I want to keep my options open (since I'm still undecided if I want to go towards industry or academia)? 

For academia, most departments at R1's (with the possible exception of the very top ones) don't care so much about where you got your PhD, they care mainly about your ability to do high-impact work. SLAC's will care about your teaching as well.

That being said, there is certainly a correlation between doing a PhD or a postdoc at a higher ranked department and getting high-impact work done -- higher-ranked departments do tend to have more productive and famous faculty, as another poster above mentioned. And more productive/famous professors means you have better chances of getting good postdocs or publishing in good journals. It's still possible to get a good academic job with a PhD from a mid- or lower-ranked PhD program (one of my cohort got an Assistant Professor job at University of Minnesota with no postdoc, and my PhD advisor has former students working as professors at TAMU, Duke, etc.), but it might be a little bit harder. But if you do good work and publish in good journals and ML conferences, your chances in academia should be good, irrespective of your PhD alma mater.

Edited June 22, 2019 by Stat PhD Now Postdoc

[DanielWarlock]

 

On 6/22/2019 at 11:31 AM, Stat PhD Now Postdoc said:

For academia, most departments at R1's (with the possible exception of the very top ones) don't care so much about where you got your PhD, they care mainly about your ability to do high-impact work. SLAC's will care about your teaching as well.

That being said, there is certainly a correlation between doing a PhD or a postdoc at a higher ranked department and getting high-impact work done -- higher-ranked departments do tend to have more productive and famous faculty, as another poster above mentioned. And more productive/famous professors means you have better chances of getting good postdocs or publishing in good journals. It's still possible to get a good academic job with a PhD from a mid- or lower-ranked PhD program (one of my cohort got an Assistant Professor job at University of Minnesota with no postdoc, and my PhD advisor has former students working as professors at TAMU, Duke, etc.), but it might be a little bit harder. But if you do good work and publish in good journals and ML conferences, your chances in academia should be good, irrespective of your PhD alma mater.

Professor Han Liu is very good. I read some of his papers about dimension reduction. He collaborated with another chinese guy at Princeton a lot -- So essentially studying with Prof Liu is comparable to studying with the other guy at Princeton. But if you go back to your home country, say China, to find a non-research job in industry, then all they know is Princeton (some may not even know Princeton!) and when you say you are from Northwestern it's like you did your PhD in wild west shooting buffalos (They call foreign universities they don't know "chicken university"). Your parents are embarrassed to talk about your school and your relatives and friends think of you as a failure, studying in your 30s at chicken university unmarried living on minimum wage. You can say it is all about science and not opinions of other people etc. But the truth is that it is very hard to land a faculty job no matter what and you will likely return to China after PhD to find a non-research job.

One exception would be that you are truly genius and do manage to do great contribution to science. But the price to have that proposition tested out is enormous. How do you know you will not be a mediocre PhD if you are not even good enough for some of those PhD programs as a college student?  One of my classmates (from Peking U Math!) at my Harvard master program was actually admitted to Northwestern PhD and didn't go. He would rather come to Harvard, pay tons of money, studying with less competent people such as myself, just so that his education background will be recognized.  

Another exception is that you absolutely cannot live without doing your research. Your love to your field is so crazy that the addiction cannot be contained and you don't care about anything just so that you can keep pursuing it. Then fine. You do it at chicken U or whatever. But this is very very rare even among the best scientists. Mostly, people work like crazy just so that they can prove they are the best. Perelman, probably the most powerful mathematical mind there is, retired from mathematics after he proved the conjecture, so did his predecessor Hamilton. More mundane examples are profs slow down research (sometimes greatly) after being granted a tenure. Students think they are "interested in" doing some area of research but when they actually do it, they do it out of necessity to get their degree and that becomes unbearable. I witnessed a stats PhD at Harvard broke into tears one day while saying "I hate this shit so much". My old boss dropped out of Waterloo PhD just because the psychological burden was too great and to watch his peers went on to live great life in industry broke him. He said "PhD was my darkest hours". So don't go to PhD unless you are absolutely sure you do it at place you are happy with. You sacrifice a lot and are doing mankind a favour by pursuing PhD, especially at a place with little recognition beyond academia. 

 

 

Edited October 26, 2019 by DanielWarlock

[Stat Assistant Professor]

15 hours ago, DanielWarlock said:

 

Professor Han Liu is very good. I read some of his papers about dimension reduction. He collaborated with another chinese guy at Princeton a lot -- So essentially studying with Prof Liu is comparable to studying with the other guy at Princeton. But if you go back to your home country, say China, to find a non-research job in industry, then all they know is Princeton (some may not even know Princeton!) and when you say you are from Northwestern it's like you did your PhD in wild west shooting buffalos (They call foreign universities they don't know "chicken university"). Your parents are embarrassed to talk about your school and your relatives and friends think of you as a failure, studying in your 30s at chicken university unmarried living on minimum wage. You can say it is all about science and not opinions of other people etc. But the truth is that it is very hard to land a faculty job no matter what and you will likely return to China after PhD to find a non-research job.

One exception would be that you are truly genius and do manage to do great contribution to science. But the price to have that proposition tested out is enormous. How do you know you will not be a mediocre PhD if you are not even good enough for some of those PhD programs as a college student?  One of my classmates (from Peking U Math!) at my Harvard master program was actually admitted to Northwestern PhD and didn't go. He would rather come to Harvard, pay tons of money, studying with less competent people such as myself, just so that his education background will be recognized.  

Another exception is that you absolutely cannot live without doing your research. Your love to your field is so crazy that the addiction cannot be contained and you don't care about anything just so that you can keep pursuing it. Then fine. You do it at chicken U or whatever. But this is very very rare even among the best scientists. Mostly, people work like crazy just so that they can prove they are the best. Perelman, probably the most powerful mathematical mind there is, retired from mathematics after he proved the conjecture, so did his predecessor Hamilton. More mundane examples are profs slow down research (sometimes greatly) after being granted a tenure. Students think they are "interested in" doing some area of research but when they actually do it, they do it out of necessity to get their degree and that becomes unbearable. I witnessed a stats PhD at Harvard broke into tears one day while saying "I hate this shit so much". My old boss dropped out of Waterloo PhD just because the psychological burden was too great and to watch his peers went on to live great life in industry broke him. He said "PhD was my darkest hours". So don't go to PhD unless you are absolutely sure you do it at place you are happy with. You sacrifice a lot and are doing mankind a favour by pursuing PhD, especially at a place with little recognition beyond academia.

Nearly all of the international PhD students at my program (which is well-regarded but by no means considered an "elite" school) ended up staying in the U.S.A. and going to work in industry or doing a postdoc after graduation. My program was over 70% international students, and the ones that went into industry post-PhD had no difficulty getting jobs at good companies like Amazon, Google, Wells Fargo, JPMorgan, etc. For industry, it doesn't matter that much where your PhD is from, as long as it is from a school with some name recognition (which would include most of the flagship state schools in the country and schools like Northwestern). It is quite difficult for international students to get these jobs without a PhD in a STEM discipline (whereas domestic students can often get these jobs with only a Masters or a Bachelor's), but with a STEM PhD from *any* decently reputable program, it is significantly easier for them.

For academia, it is a little bit harder to move up in the ranks, but not impossible if you publish in good journals/conferences, work with good postdoc and PhD advisors, and network with the top people in your field (it is highly advantageous to have a famous professor be familiar with your work and write you a letter of recommendation). But still, most people should not expect to land a job at an "elite" program -- even the majority of PhD graduates from top schools like Stanford, Berkeley, Harvard, etc., will end up working as professors at large state schools or small liberal arts colleges if they choose to stay in academia. There are only a finite number of jobs at "top" programs, so the chances of ending up as a professor at one of the elite programs tend to be minuscule for most people, unless you are a true rock star.

Edited October 26, 2019 by Stat PhD Now Postdoc