Correlation between prestige of graduate school program and success as a scientist?

[N3UR0]

I'm curious what other people's perceptions are. How strong is the connection between graduation from 'top-tier' research schools like Hopkins, UCSF, or Harvard, and a successful career as a scientist? How does the name of the program you complete your PhD at compare relative to how much you thrived as a young scientist during grad school, irrespective of the program? What matters the most during your PhD years in beginning a successful career in research?

Thanks in advance!

 

[NeuroMetro]

I would like to hear any honest thoughts on this as well.

[Neuro PolarBear]

What matters most during a PhD is not what school you go to, but who your mentor is. I'd recommend Ben Barres' excellent article on picking a PhD mentor. You want someone who is both a good scientist  (well published, well-funded, etc.) that is also an excellent mentor. Scientists that are both can certainly be found at many different types of institutions, and are not just concentrated at the ones you listed. In fact, some of those labs at the top institutions are so big that maybe they're better for doing a postdoc since it could be harder to be mentored by a super well-known scientist who's not as involved in your project. Or maybe that would work better for you!

Your mentor's name and reputation is what will help you get published much more than what school you're at. Maybe if you decide to pursue a career outside of academia, people might place more weight on the institution that you attended, but this is certainly fairly misguided-in the same ways that impact factor isn't a good job of evaluating the quality of the work you publish from your thesis.

[jordy_protein]

My dad is a physics professor at a top research institution. He attributes most of his success to his Ivy League education (and his mentor, of course). He thinks that I should go to the mostly highly ranked institution I get into (provided I interview and feel that it's a great fit as well).

And, if you think about it this way - top schools have more potential mentors who are top in their field and well-known. You can't lose by going to a top-tier school.

[biotechie]

8 hours ago, Micecroscopy said:

 

I agree with part of this.... however, I would sub the word PhD for post-doc in all of this. The most important thing is, if you want to stay in Academia, that you post-doc in a great lab. Other than that, so long as you are a productive grad student, your PIs reputation won't matter as much.

I'm a third-year PhD Student. I did 4 years undergrad research, then a MS, and I also spent a lot of time talking to people who ran this PhD gauntlet ahead of me. Through this, talking to PIs, and my experience until now, here's my opinion on the matter. Nearly all of it pointed me to joining the lab of a fairly new PI as a PhD student (as one of their first three students) and then joining a big-name lab where you can really run with a project and prep for your own lab. I rotated in bigger name labs, and the younger PI won out.

Your PhD is the time when you need to learn to be a scientist. You need to learn to not only do the bench work and improve your current skills, but you also need to learn how to run a lab, organize your research, write grants and papers, give presentations, and manage lab personnel and teach students. Most of us do rotations, so we get to see how the PI is going to be in the lab so we can make the right choice. Here are the criteria I used to pick my lab and why the new PI won out for me over a more established one. Remember we're all different; this is what worked for me, but your needs/wants as a student may be different.

1. Direct mentorship and funding: I wanted a PI who was going to be available to me when needed but would let me be independent when I could be. Established PIs that have big names are notoriously busy and may not have the time needed to give you the attention you will need early on in your PhD education. New PIs have not been thrown onto 20 different committees and will have time to work with you if needed. I could talk to my PI multiple times a day if I needed to. If you don't need a lot of one-on-one to teach you the research, then you can work more collaboratively to get epic amounts of research done. New PIs also have startup funding to start their lab and generate data to get funding, so they should have money to fund you if you can't get a fellowship. On top of this, your success as a graduate student (publishing and ultimately graduating) is essential for them to get tenure. Thus you becoming a good scientist also helps them become successful. The PIs I've talked to always seem to cite one of their first few students as their most successful, and I think the highly collaborative atmosphere and these goals really help those students succeed. This is also the only time that funding isn't do-or-die for a couple years so they can really delve into the science.

2. Research Project: The biggest thing that increases time to PhD is starting with a poorly thought out project with weak supporting data (and/or poor mentorship). I wanted a project that looked like it had a couple of directions it could go in and that I could get some supporting preliminary data for. I got this as well as a strong "back-up" project. My PI made it clear to me from day 1 of my rotation which projects were available to me and how far he saw them going. He had them well thought out, but left lots of room for me to make my own choices for direction within it. This was another benefit of the young PI. He had very clear goals, but was just getting started, so directionality was not set in stone. 

3. Student Track Record: This is harder with new PIs as you may be their very first official graduate student. However, this one had mentored many undergraduates during his PhD and then graduate students (unofficially) and undergrads during his post-doc. He gladly gave me information on where these students ended up and I was even able to meet one of them during my rotation. I also got to observe this professor working with and teaching another rotations student. His teaching style was always patient with us, but he quickly learned how both students learned and altered his teaching method individually for each of us. This is important in the beginning while you're still learning the field and methods of the lab, and this is something you can really observer during rotations. If you can learn really well from the PI, you stand a good chance of doing well in the lab.

4. Lab skills: New PIs are still sharp on their skills from post-doc. They'll teach you directly, and I find that this is very helpful, especially when you're working up the data together.

4. Ultimate career goals: I want to be a PI and run my own lab one day. What better way to learn than by being part of a new lab getting set up? My PI got his keys about 2 months before I rotated. When I talked to him about the rotation, his lab was still empty. Since then, I watched and helped set up a lot of the lab. I helped establish cataloging systems for samples, talked to vendors, etc. I'm also getting to help train new people in the lab and I get to watch how my PI interacts with collaborators and hires new people. He was very recently in my shoes, so he also makes sure I get to see the outcomes on things he's been writing and he makes me write things on my own (fellowships and papers, etc). He also tries to let me take the lead on small projects and small collaborations. Finally, he has me out presenting posters and small talks as often as I can, which benefits me in the experience, but is also highly positive for the lab. By being here for this part of his journey, I'm learning a lot that will help me when I get there myself. I am lucky enough to have PI that cares about his students, allows me to freely voice my concerns, and takes time to help me work on my career development as well.

Finally, I highly recommend choosing your lab based on lab environment and how well you mesh with the PI and lab personnel over the project or lab/institution ranking itself. You're going to be spending 5 years there... don't make it a miserable and unproductive 5 years. That can lead to burn-out or drop-out. You have to fit well in the lab, and you have to be happy. Don't just choose a young PI or a big-name lab because someone tells you to. We can only offer you our opinion based on our own experiences. Obviously I've been very lucky so far, but my PI is also new and he does make some mistakes. However, I am very happy in my lab and it has been very productive so far. I have no doubt I will be abel to graduate and get a good post-doc.

[ruckaround]

This is a topic I'm very interested in as well, and something that I think people are particularly bad at talking about. I agree with a lot of what @Micecroscopy and @biotechie said. Certainly the most important factor in success is not your institution but rather your mentor. There seems to be little disagreement about that. I also agree that in many ways, prestige likely matters more for one's post-doc than one's phd.

However I think that to presume there is not a meaningful correlation between elite institutions and success is misguided. To some extent the most important part of this discussion is an agreement about what "success" means. I'll go out on a limb and suggest: 1) success at landing TT faculty positions. 2) funding availability 3) publishing in high IF journals [just for the hell of it, 4) Nobel prizes] I'm not sure I agree with how the system is setup (ie, the presumed importance of Cell/Nature/Science pubs, etc) but I think it's hard to dispute that these things matter in real ways. (It's especially hard to dispute a Nobel as success :p)

Regarding 1) http://www.slate.com/articles/life/education/2015/02/university_hiring_if_you_didn_t_get_your_ph_d_at_an_elite_university_good.html

"The data revealed that just a quarter of all universities account for 71 to 86 percent of all tenure-track faculty in the U.S. and Canada in these three fields. Just 18 elite universities produce half of all computer science professors, 16 schools produce half of all business professors, and eight schools account for half of all history professors."

"The top schools generate far more professors than even just slightly less prestigious schools. For example, in history, the top 10 schools produce three times as many future professors as those ranked 11 through 20."

While this study didn't look specifically at biology, it's hard to presume our field trends differently.

Regarding 2) There is plenty of information about funding, the following are just two random google results:

http://www.genengnews.com/insight-and-intelligence/the-top-50-nih-funded-universities-of-2014/77900233/

https://sunlightfoundation.com/blog/2012/09/13/nsf-funding/

While there is certainly a degree of intricacy here, I think it suffices to say that the funding situation at a top 10 or top 20 school is almost always better than at some misc. mid-sized state school.

Regarding 3)

One example: http://www.natureasia.com/en/publishing-index/global/

Not hard to imagine how point #2 contributes directly to this.

Regarding 4)

https://en.wikipedia.org/wiki/List_of_Nobel_laureates_by_university_affiliation

---

Interesting thought regarding undergraduate affiliation: http://qz.com/498534/these-25-schools-are-responsible-for-the-greatest-advances-in-science/

 

I certainly do not mean to say that no one outside of these "elite" schools does good science or achieves success, or vice versa that everyone at those schools is successful and a great scientist. However, I think it is abundantly clear, at least by these metrics, that there is a strong correlation. I haven't thought as much about it, but it seems safe to presume there is also a correlation between attending one of these schools and success outside of academia as a biology phd.

Like I said before, I'm very curious about this idea and I think people are bad about talking about it. It's very awkward. It's likely not how we would like things to be, yet it certainly seems to be the state of affairs.

 

Curious to read others thoughts!

[eeee1923]

I agree with what's been mentioned already but I will say that the more elite institutions tend to have the most funding which manifests in better equipment, more cutting edge research and collaborations at other similar institutes. This means that PI's at more prestigious institutes tend to have contacts at other prestigious institutes which can help one when they are looking for possible postdoctoral positions. 

[NeuroMetro]

@PlanB, thank you, that article gave me some useful ideas for screening potential mentors. I appreciate it.

[aberrant]

I generally agree what @biotechie said, and about this statement:

On 12/22/2015 at 2:49 PM, Micecroscopy said:

Your mentor's name and reputation is what will help you get published much more than what school you're at.

I would doubt that your mentor's name and reputation can help you get published much more -- if you don't have the data, their name means nothing  I know for a fact that a "big" PI can at least get you an interview for a post-doc position at another "big" PI, top-tier schools or not. And I know for a fact that your school reputation doesn't mean as much as outsiders thought for a post-doc position. Ultimately, it is about your quality of work (also suggested by another professor at HMS -- your personality, your eagerness and willingness to learn and accomplish tasks, and how challenging your PhD project was [this prof gives a thumb down to "easy" PhD projects, when a student took over a well-established project and publish the work(s) with "little effort"]).

 

On 12/22/2015 at 3:03 PM, jordy_protein said:

My dad is a physics professor at a top research institution. He attributes most of his success to his Ivy League education (and his mentor, of course). He thinks that I should go to the mostly highly ranked institution I get into (provided I interview and feel that it's a great fit as well).

And, if you think about it this way - top schools have more potential mentors who are top in their field and well-known. You can't lose by going to a top-tier school.

That observation is highly field-specific, besides the fact that you are assuming everyone is competitive enough to get into top-tier programs (I differentiate "programs" from "schools"). I know enough PIs move from top-tier institute(s) to "relatively lower tier" schools because of monetary reasons, and authority in a specific department. So a "non-top-tier" school doesn't necessarily mean they have less successful PI (and/or mentors). To echo what biotechie says (in a way), under ideal scenario, you want to work for a big fish, regardless the size of the ponds. That fish better matched your personal research interest, and hopefully, is a good mentor.

[biochemgirl67]

I think we know the answer to the question already; indeed, prestige has an impact on success.  It's why we're all reaching for the stars on our applications.  However, I think it's worth noting that in this age of limited funding, incoming grad students are probably more concerned with a department's prestige or the quality of work coming out of a particular program than the overall institution.  For example, the University of Tennessee is not prestigious.  It just isn't.  However, their microbiology department plays host to some of the national experts in marine microbiology.  If you were an aspiring scientist interested in those things, you may choose UTK over some other more canonically prestigious place because the program has better research coming out of it.

 

The elite institutions are elite for a reason.  I think we can all agree that it's important to go to a good school but that the definition of "good" can actually vary.  And finally, a name of an institution will not help to get you published.  It's all about the validity of your work.  Well, weird sporadic politics aside.

[ruckaround]

8 hours ago, biochemgirl67 said:

I think we know the answer to the question already; indeed, prestige has an impact on success.  It's why we're all reaching for the stars on our applications. 

Perhaps the more interesting question then becomes... "How does prestige impact success." I think we're all in agreement that going to a brand name school certainly does not take the place of doing great science. But presuming that one does do great science, in what ways do y'all think going to a elite institution or top program helps enable one's success?

Or, is that great science more efficiently/successfully achieved in such a program?

Edited December 30, 2015 by ruckaround
Second question

[biochemgirl67]

2 hours ago, ruckaround said:

Perhaps the more interesting question then becomes... "How does prestige impact success." I think we're all in agreement that going to a brand name school certainly does not take the place of doing great science. But presuming that one does do great science, in what ways do y'all think going to a elite institution or top program helps enable one's success?

Or, is that great science more efficiently/successfully achieved in such a program?

Money, facilities, collaboration.  I don't go to a prestigious undergrad.  It's a good school and the faculty were all educated at top 10 schools, but my department and university and not top 10.  This compares easily to Harvard.  There was so much more money there.  The department I worked for had excess money so they bought everybody a backpack.  If you wanted testing of any sort you just did it.  You could buy the fancy equipment for ease of use.  Our flow cytometer broke and a part was ordered that day.  At my home institution, we are greatly limited to the testing we can do due to expenses.  I have to engineer set ups rather than just buy the expensive plastic pieces.  It's made me an inventive scientist, but it was also nice to be somewhere where the money and funding wasn't such an issue.  Elite institutions foster great science at a high rate of speed in part due to the money, I think.

[Eigen]

You also have to be really careful in how you describe success. 

Working at a top-10 program that has money to throw at every problem really only prepares you for work at another top-10 program, and only if you will be successful enough to have money to throw at every problem. 

If you want to expand your faculty search to lower ranked schools, lots will appreciate you being able to be productive without piles of money to throw at problems, either through creative solutions or being able to build/repair instruments on your own. 

Similarly, while top institutions have a ton of funding, they also focus more on post-docs and PIs research progress then developing you as a grad student into an independent scientist. It can be really easily (relatively) to be lost in the shuffle, or be only valued for your use as a pair of extra hands in the lab with no particular interest in what the work is doing for your career. 

[TakeruK]

3 hours ago, Eigen said:

You also have to be really careful in how you describe success. 

Working at a top-10 program that has money to throw at every problem really only prepares you for work at another top-10 program, and only if you will be successful enough to have money to throw at every problem. 

If you want to expand your faculty search to lower ranked schools, lots will appreciate you being able to be productive without piles of money to throw at problems, either through creative solutions or being able to build/repair instruments on your own. 

Similarly, while top institutions have a ton of funding, they also focus more on post-docs and PIs research progress then developing you as a grad student into an independent scientist. It can be really easily (relatively) to be lost in the shuffle, or be only valued for your use as a pair of extra hands in the lab with no particular interest in what the work is doing for your career. 

I just want to say (although in a different field) that I am at a top 10 program and this post encapsulates all of the concerns that a student at a top-10 program really needs to think about when it's time to pick which school/lab to go to as well as what to specialize in when you graduate!

So many different people---grad students, postdocs, and faculty---warned me about what Eigen wrote in the last paragraph when I was considering this top-10 program for my PhD. I would say that if you let yourself be relegated as another pair of hands (or another "cog in the giant research machine" etc.) then all of the advantages of being at a top-10 school disappears. So for those who are considering top programs, think about this when you figure out where you would fit into the department. It's important!

And for the second issue---the one where a student is only successful because they have all these resources---it's something that many of my fellow students and I have thought about and worked on. It's a real concern at top schools and the good faculty point this out to us right away and make sure we are developing skills that are useful beyond institutions with a lot of resources. For an example: a student who had to work really hard to get the most science out of their second-tier equipment is going to be able to learn and develop a lot more than a student who works at a place where the equipment is easy to use and always works. Of course, I'm generalizing here, but being at a top school means you have to work extra hard to prove to future job committees that you aren't just successful because you had all these resources, but that you would be great without the resources too. Just something to keep in mind as well.

[St Andrews Lynx]

Here's some v. anecdotal, small sample-size observations of mine, based upon the difference between an Ivy League/Top 10 (IL) and a Top 50 (T50) school.

• IL has a better class of invited speakers. More likely to invite for seminars top-ranking American & international professors from similar top-tier schools. As a result, the grad students & postdocs are exposed to more cutting-edge research, big names and the kind of presentations that reel in awards, prestige, etc. Grad students who then meet with or go out to lunch with the invited speakers learn how to act like a big name, make great connections, etc.
• IL research groups have more group meetings. Not just research updates but they seem more likely to give literature presentations or thematic subject intros (a lot of them are posted online) or "synthesis challenges" if you're in the synthetic chemistry subfield. Thus as an IL grad student you may get training in a more diverse set of skills and have more opportunities to strengthen/develop your scientific knowledge.
• There exists "department wide job-seeking intelligence". The students in one IL department have a well-publicised strategy for submitting postdoc applications to maximise the chance that the target PI will read their materials. It was pretty clever and widely adopted. I've not heard about this sort of thing where I study. 
• Good universities are often located near (i) other good unis (ii) companies. Think Boston or the Bay Area. Which means that it's easier for recruiters to come to the big name universities or for you to gain exposure to invited talks at nearby universities too, etc. 

 

That said...

• Not all IL labs are "success factories". 
• If you work for a well-respected PI at a T50 school and work hard, you can graduate with a CV that's just as good (if not better) than if you went to an IL. 

 

[Eigen]

Just to respond to your point 1:

I'm not even at a top 50 school.... But I've met a number of nobel laureates as speakers in the time I've been here, as well as had lunch with heads of groups in some really nice national and international labs. 

One difference at a lower school is that I get to meet with (have a significant solo meeting) with every speaker I'm interested in meeting. Show them my work and network with them. I'm guessing the same chance doesn't exist at the IL schools to the same extent.

Speakers depend a lot more on the connections of the faculty than the prestige of the school- well respected scholars and well connected scholars bring in good speakers. 

[Edotdl]

2 hours ago, Eigen said:

One difference at a lower school is that I get to meet with (have a significant solo meeting) with every speaker I'm interested in meeting. Show them my work and network with them. I'm guessing the same chance doesn't exist at the IL schools to the same extent.

 

Out of curiosity, why do you think the same chance doesn't exist at IL schools to the same extent? I don't see why there would be a difference besides maybe fewer students competing for their time.

[TakeruK]

1 hour ago, Eigen said:

I'm not even at a top 50 school.... But I've met a number of nobel laureates as speakers in the time I've been here, as well as had lunch with heads of groups in some really nice national and international labs. 

One difference at a lower school is that I get to meet with (have a significant solo meeting) with every speaker I'm interested in meeting. Show them my work and network with them. I'm guessing the same chance doesn't exist at the IL schools to the same extent.

Speakers depend a lot more on the connections of the faculty than the prestige of the school- well respected scholars and well connected scholars bring in good speakers. 

Just wanted to speak to this point because it was one of the biggest difference between a top 10 school (my current PhD program) vs. top 200 school (my MSc program). I think it really depends on the "non-top-10 school". At my MSc program, it was in between (as in, a 2 hour train ride from) two other very prestigious programs and we were only able get "big name speakers" for one seminar per year. Sometimes, if we were lucky, maybe two in the same year. And even when we do this, the speakers rarely stay overnight--they often arrive from a train at the first school at say 9am, the seminar is at noon, there is a 1pm-2pm lunch with the speaker. There may be a dinner with the speaker, but they might choose to take a train to the next school that evening around dinner time. In any case, 1-2 grad students might get a meeting, but the professors tend to monopolize the speaker's time.

At the top-10 program, we get "big name" speakers practically once a month. And, students are always prioritized in arranging meetings. The program is also able to ensure the speaker stays overnight (school pays for accomodations) and the speaker spends two meals with students. Also, student input at the MSc program was pretty limited ("send us some name suggestions and we'll see what we can do"). At my top-10 program, the third year grad students take full responsibility for choosing speakers, inviting them, arranging their schedule, and hosting them. I think this is great practice for building our network. I did this last year and it helped opened new connections for things like getting invites for me to speak at other institutions (nothing confirmed yet since it's too far away but a bunch of people indicated that they would be happy to host me in the future). 

I do agree that connections with the faculty do help bring in good speakers (at my MSc program, all of the good speakers came because they were a collaborator of a faculty member) but the school brand name plays a big role too. Many speakers that we invite are very happy for the chance to speak at a "big name school". And, I think putting the invite power in the hands of the students gives us a leg up in building our own network (I mean, "Would you like to speak at <top 10 school name>?" is a pretty good opening :P)

[Eigen]

3 hours ago, Edotdl said:

Out of curiosity, why do you think the same chance doesn't exist at IL schools to the same extent? I don't see why there would be a difference besides maybe fewer students competing for their time.

That's exactly it. Most IL schools in Chemistry/Biology have around 200 graduate students. Some of the lower ranked schools are 50 or less. 

I can get a half hour to an hour private meeting with almost any big name I am interested in meeting- and that's huge. 

It's not just a group dinner or group lunch, it's a large section of private face time. And that's definitely played a role in those people remembering me down the road in a much more personal manner than if it had been a group event.

I'm sure some of it depends on the school and how it's arranged, but it's one of the complaints I here from my friends at IL schools, as well as students of mine now at IL schools.

You are right that a lot of it depends on the "non top 10 school". My department brings in ~ 30 speakers a year, most of them from good programs/well known in their field, and at least a handful a year that are in the top 1-2 names in the world in a given area. They're all in for at least a day, most over a weekend, and so there's lots of time for interaction. 

[TakeruK]

10 hours ago, Eigen said:

That's exactly it. Most IL schools in Chemistry/Biology have around 200 graduate students. Some of the lower ranked schools are 50 or less. 

I can get a half hour to an hour private meeting with almost any big name I am interested in meeting- and that's huge. 

Ah, I see a big difference that explains why my experience is different! At top schools in my field, we have "large" populations, which is like 20 planetary science students. I think the biggest program in the country has ~35 students. At small schools, planetary science is part of another program and there are like 4-5 of us. So, the difference in "large" and small is tiny, which means there's no disadvantage in the larger size (I also usually get 30-60 minutes of private meeting time with any visitor and we have visitors for multiple colloquia / professor group meetings per week!).

Also, at smaller schools, being too small is also bad because it means the being 4 or 5 planetary science students in a bigger department of, say, 50 students means that the majority of colloquia visitors are doing other things. I guess there is a sweet spot, about 20-50 students (in my opinion) as the ideal size where you are big enough to have a good community but small enough to get individual attention

[hippyscientist]

Program size is definitely going to impact networking opportunities. In my area of expertise, most departments have maybe 15 PhD and MS students maximum and in each sub-discipline that might shrink to 3. As such, you get a lot of 1:1 with PIs, visiting lecturers, and a sizeable portion of grants for research.

However, it differs slightly at my current school. One thing that has been touched upon in this thread is the disparity between high ranking programs and overall university ranking. My current program is #1 for my discipline, but the university is something like 800 odd on the QS rankings. We get much lower funding than someone at a higher ranked QS university in the same discipline, but our work is respected more due to the department reputation. So funding and resources =/= success, it just aids it. However, what makes my department stand out is that we have access to an unparalleled pool of subject participants - my research area focuses on elite athletes at the minute and I have access to half of 5 GB Olympic squads. It's swings and roundabouts, playing that balancing act between the research you want to do, and the research you can afford to do. But again, money doesn't equal prestige, you just gotta get more creative.

Equally, when I go to conferences in the UK, people talk to me about my MS programme because of its reputation, and I guess my own to a certain extent (my BSc project was ambitious to say the least and my department are still talking about it 3 years later). The networking opportunities in the UK are fewer than those in the US, and I'm curious to see the differences and similarities when I do my PhD (yes, positive thinking I will get a program admit!!). 

Sorry I think I rambled there but I hope it contributes to the discussion in some way!