Is going to a famous university necessary for a science career
Does attending a famous University make you more famous?
There is no doubt that specific elements of your CV will support your high-class reputation. For example, if you have published a high impact factor paper recently (e.g., in NATURE or SCIENCE) as a junior scientist or several high impact papers during the last 5 years as a senior scientist, this will be noted by your colleagues. Typically, this will be mentioned when you are introduced as a speaker during a scientific meeting or when there is a press release. Similarly, it will be mentioned when you have worked in the lab of a famous scientist and/or when you have worked in a famous university. Thus, there is a certain value in attending an elite institution to promote your reputation.
It does not matter where you get your college/university degree!
It is not easy to find hard data. However, there are a few studies that suggest that it does not matter economically where you get your college/university degree. In a landmark study by Alan Krueger and Stacy Berg Dale and a follow-up study it has been shown that it does not have a significant effect on your earnings whether you attended an elite university such as Harvard or not. Thus, there is no “Harvard effect” when you compare graduates 20 years later. However, there seems to be a positive effect on career development for students from more disadvantaged family backgrounds. In a thoughtful analysis, Derek Thompson summarized it as follows:
“Research suggests that elite colleges don’t really help rich white guys. But they can have a big effect if you’re not rich, not white, or not a guy.” It is important to note that these studies have a US American perspective and do not focus on life sciences. In addition, economic outcome is only one of many important parameters to judge the success of a career. Especially in science, many if not most people are not primarily focused on earnings. Instead, they strive for a meaningful job and want to create an impact on science and society. Read more here: What is “a successful career” in science?
Forty years ago, elite colleges offered a demonstrably higher level of education.
Nowadays, many universities and colleges provide a high level of education at much lower costs. The reputation of the place where you got your degree is also elusive. It may play a role when you apply for your first job, but 5 to 10 years later it counts much more how you have performed in your previous jobs. Thus, if you chose well you may get a high level of education also at less expensive places. It does not seem to hurt to go to an elite college, and it may be a nice label of excellence on your CV, but the added value is not easy to pin down
Should I do my PhD or postdoc at a highly ranked University?
Labels such as “a famous university” or “an elite college” are very vague and heavily influenced by clever marketing to form the public opinion, smart people have developed University rankings which strive to be more objective. A typical example is the Times Higher Education World University Ranking.
To address whether you should do your PhD (or postdoc) at a highly ranked institution, it is critical to understand how university rankings are generated. Mark Bennet provided a very helpful analysis whether university rankings matter for PhD studies. Briefly, he lists three points you should consider when using University rankings:
Firstly, – no surprise – the research performance is relevant for a young researcher.
Secondly, the teaching performance may be relevant, for example, the ratio of PhDs to undergraduate degrees awarded and the total number of doctorates awarded.
Thirdly, the rankings help to find the most highly ranked institutions in selected countries.
Thus, you may discover elite institutions not only in the USA but also in many other countries such as the UK, Germany, the Netherlands, Switzerland, Canada, China, etc. International education is also an important parameter. Especially in times of rising nationalism, it is an important parameter whether an institution is open to international students (= foreigners). Mark Bennet concludes that University rankings are useful, especially when you take a closer look at the ‘sub-rankings’ summarized above that might matter for PhD students and postdocs.
However, a reader who critically analyzed an early version of this text pointed out that the arguments I am advancing in this article may be only made by Europeans. “I have never heard any American or Asian seriously doubt the importance of institutional rankings. Rather the exact opposite is abundantly evident at Harvard Yard which is teeming year-round with Chinese pilgrims.” Probably, he is right.
The biggest problem with University rankings is that they use selected metrics and describe the performance of a University in very general terms. It does not tell you whether your potential future supervisor is excellent or not as a supervisor or as a scientist. It does not tell you whether your supervisor is a great person and supports you, or whether he/she is an authoritarian sadist who will traumatize you for the rest of your life.
You can have a great or a lousy research stay at a highly ranked University
It is much more essential to select a PhD or postdoc position based on the performance of your future boss as a supervisor and as a scientist. Briefly, if former and current PhD students and postdocs report that they experienced the environment and the supervisor as very stimulating, it is a good sign. Unfortunately, you cannot exclude that their supervisor has instructed them to tell you fairy tales to convince you to come. If the publication list of the supervisor is very promising and the younger lab members have appropriate positions in the author’s list of previous publications (first or last or equally contributing first or last author), it is another good sign. If there is a high turn-over of young researchers who do not finish their papers and/or thesis, this is a red flag. Read more here: Pressure or pleasure – what produces better science?
There are many advantages of doing a PhD or postdoc in a great lab
Above I have been somewhat critical about focussing too much on elite institutions. I am convinced that you profit much more when concentrating on working in a great lab. This great lab can be in an elite institution or in a less prestigious institution. A successful research stay in a great lab has many advantages.
- You learn how to think like an excellent scientist.
- You learn how a high-performance lab is organized.
- You learn how to publish in journals with high impact factors (at least you get to know the best practices that may work).
- You may be trained in high-end technologies or even (co-)develop a new breakthrough technology.
- You may stay in contact or collaborate with your colleagues for a long time after leaving the lab.
- You may leave with one or two excellent publications in high impact journals.
Thus, there are many reasons to find the perfect lab to do your PhD and postdoc.
But they all tell me that I can *only* become a professor when I have done my postdoc at an elite institution!
Will it help me to get hired as a professor?
There is no doubt that deans and selection committees like to hire people who come from elite institutions because it seems to increase the value of their own institution. In psychology, this is called association fallacy. Thus, honor by association means that someone or something must be reputable because of the organizations that are related to it. However, if another candidate has a better publication list and has raised more grant money, it will not help you.
Besides, especially at smaller or regional institutions, there may be more chances for young researchers who have not published Nature or Science papers and who did not come from an elite university. Read more here: Do I need Nature or Science papers for a successful career in science? and Should I choose a big or a small university?
Should I take a loan to finance a stay at an elite institution?
One of my readers asked me whether he should take a loan to finance the stay at a US elite institution or attend a less famous (but still excellent) European university. My very, very personal opinion is that taking a loan to finance your PhD or postdocs at an elite institution is a bad move. Economically, you may put a lot of debt on your shoulder for many years with an unclear return of investment. However, you can apply for multiple grants to circumvent this problem.
Yes, you may build your reputation, and whenever you give a talk you may be introduced with “… he worked in [famous university] in the lab of [famous researcher]…”. Yes, you may work in a fantastic environment with a lot of money and a lot of intellectually stimulating high performing scientists.
However, there are genuine downsides to attending elite institutions, most importantly high housing costs. Living in a typical university city such as Boston may be extremely expensive, especially if you have small children needing daycare. However, if you are a high performer, always consider free or fully funded PhD programs that may reduce the financial stress.
If in doubt – do *not* do it. It is much more useful to choose a great lab that fits perfectly with your skills and expertise in a less expensive institution and to perform on an excellent level. If you want an industry job or any other non-academic job, it is questionable whether a stay at an elite university is worth the money and the effort. If you want to become a professor, you will be primarily judged by your publication list and the grant money you have raised. Read more here: How to become a professor?
Besides, especially at smaller or regional institutions, there may be more chances for young researchers who have not published Nature or Science papers and who did not come from an elite university. Read more here: Do I need Nature or Science papers for a successful career in science? and Should I choose a big or a small university?
Conclusions
In conclusion, there are not many good reasons to study at an elite university. There are strong arguments to do your PhD or postdoc at an elite institution, but the return-of-investment is difficult to be measured. However, in my very personal opinion, it is much more effective to choose a great lab that fits perfectly with your skills and expertise and to perform on an excellent level. If this happens to be at an elite institution – even better.
Recommended reading
The following articles may also interest you:
- Should I study medicine, dentistry, or life sciences?
- Should I choose a big or a small university?
- Do I need nature or science papers for a successful career in science?
- Free PhD programs versus fully funded PhD programs
- How long does it take to complete a doctorate?
- Does medical school ranking matter?
- Do I really have to work abroad as a scientist?
- How to choose the best postdoc position?
- Should I quit my postdoc?
- 10 simple strategies to increase the impact factor of your publication
A reader wrote: “Since I started at Harvard I’ve received weekly emails requesting me to apply at nearby pharmaceutical and biotech companies with starting salaries around 100k USD. So I think these are a great plan B if my academic applications are not successful. Also, I’ve noticed that people from Harvard Medical School labs (at all levels from technicians to group leaders) move to higher-paying positions in the industry at a rapid rate. This is something I have not seen anywhere else. This is anecdotal, but I suspect companies such as LinkedIn and Glassdoor would have plenty of data to quantify the Harvard effect.
Nevertheless, in your post you could probably use stronger language to discourage people from taking loans for a PhD or Postdoc in the US. Elite universities are not charities but (exploitative) big business. Taking loans for PhD/PostDoc in the sciences is a bad idea and fully unnecessary since there is a lot of funding to support research globally.
I read this manuscript that analyzes the effectiveness of EMBO longterm fellowship application peer review. It contains some relevant data:
Talent Identification at the limits of Peer Review: an analysis of the EMBO Postdoctoral Fellowships Selection Process
https://www.biorxiv.org/content/biorxiv/early/2018/12/08/481655.full.pdf
“Fine discrimination among high-quality applicants is not straightforward and may lead to decision bias [16]; “when there is no objective basis for choosing one qualified candidate over others, people naturally fall back on subjective preferences. A selection committee might consciously or unconsciously favour certain research topics, groups of people or even individuals” [15; p. 7]. Social indicators analyzed include the prestige of former and future supervisors, the prestige of former and future scientific institutions where the candidate has worked or will work and the effect of the country of nationality of the candidate as well as the countries where they have worked and plan to work in the future.”
Institutional prestige measured using a performance-based institutional ranking as a proxy (see Methods) did reveal a small effect (ANOVA, P=0.29439 for the PhD institution and P=0.032588 for the host supervisor institute).
But notably only applicants from high ranking institutions qualified in the first place:
“most preselected candidates obtained their PhD in highly ranked institutions and plan to continue with their careers in also highly ranked host institutions”
The authors come to the conclusion that peer review is effective in distinguishing poor applications from good ones but fails at ranking high-quality applications (they propose a random allocation of resources which I suspect is not going to happen). As a junior scientist you’re repeatedly participating in competitions like this and this is exactly where institutional prestige translates into a permanent advantage. Yet the weak association implies you need to accrue every possible advantage on top of that and to participate in the maximum number of competitions possible.
Interestingly they identify this major myth:
“Also interesting to note is the fact that despite the still pervasive misuse or JIFs as a proxy for individual performance and scientific quality in evaluation processes, we did not find any significant association between the JIFs of the first author publications published by candidates at the time of application (as a graduate student) and their future career progression.”