12

It is a known fact that reputation and lineage are a very important factor in academia. Master students need recommendation letters to get a PhD opportunity and the same for PhD students looking for a postdoc position. If a PhD supervisor knows someone in another institute or a company, he can very easily help his student get a postdoc or a good job in the industry. If your supervisors are prestigious in the community, you can start your career much more easily than your peers. Conversely, a mediocre PhD supervisor may become an obstacle for his talented student.

I am a PhD student in CS, and I am at the final stage of my PhD. However, I wonder whether it is better to abandon my PhD degree because of the mediocrity of my supervisors. Since a thesis will be stored in a database; once deposited, the PhD student's name will be permanently bound to his supervisors, and if the public perception of my supervisors is low, I'm concerned that I may have difficulty building my own brand.

To address my concern, I would like to know which of the following states is likely to make it harder to become a leading scientist?

  • having a PhD from mediocre supervisors in a mediocre lab, or
  • not having a PhD, and building one's name from scratch?
  • 4
    For me this is a typical example of a "PhD stress". Sleep over it a night, take a day off, have a vacation, etc. If you already have a thesis, most complicated part is done. I am quite perplexed about a single submitted paper though. But hey, theoretically you don't need papers to get a PhD, just write done a monolith of theory. – Oleg Lobachev Mar 23 '18 at 19:31
  • 3
    You might appreciate some of the answers to What makes someone deserving of a Ph.D.? If you're close to completing the degree, it's probably better to complete it. It's not up to you to determine whether it's a "good" dissertation or not; your committee (who are more experienced) will approve or not. If they're willing to approve it, why second guess them (especially if you're feeling less qualified)? – Joshua Taylor Mar 23 '18 at 19:36
  • 5
    If you fail to finish your PhD you can expect to explain why in a lot of job interviews, in the private sector or anywhere else. People will be genuinely interested to know why you invested so much of your life in something and then didn't finish; many will assume you were insufficiently competent. If I were hiring you I would definitely spend some time probing you about this to understand your motivations. "In my area, it is very easy to get a well-paid job." --- unless you already have a well-paid job, that is a bold statement. – Calchas Mar 24 '18 at 1:15
  • 2
    Voting to reopen. I can't imagine a situation in which it would be advantageous to abandon the PhD right before finishing. Therefore I don't think this is an "individual factors" unanswerable question. – aparente001 Mar 24 '18 at 4:38
  • 2
    @Otsirc - So far, it's not looking good for your question to be reopened. However, you could try posting at Meta. // Additional thought for you to consider. If Student A does a thesis with Professor B who is a hotshot with a big name and a large number of students, a solid thesis that falls short of being outstanding may garner a lukewarm recommendation letter from B. So, the opposite situation from the one you described can also bring problems. Bottom line, just keep your focus on doing good work and behaving respectfully toward your mentors regardless of your poor opinion of them, and ... – aparente001 Mar 25 '18 at 1:50
9

I strongly argue that a mediocre Ph.D. is better than no Ph.D.

The standard path to bootstrapping yourself up after a mediocre Ph.D. is a sequence of rising postdocs. If you're really good, then when you move to a new institution with a postdoc, you should be able to do some nice work and build your reputation. After a year or two of that, you can try to move to a better institution with better resources and more interesting projects and do yet stronger work, etc. Doing two or three postdocs is not unusual, and is one standard path for improving one's lot after finishing a Ph.D. Moreover, even if your institution and/or advisor is known for their mediocrity, that may not count against you: "They didn't do something very interesting, but how could they under Prof. Boring? It looks like they did well with the topic they had to work on, though."

If you are interested in being a high-profile researcher in industry or other non-university research institutions, the path looks much the same (in fact, some positions in industry and elsewhere are explicitly set up as postdocs). Once you move to "permanent" positions, however, you will likely want to take a longer period between moves, so that you aren't perceived as being unable to stick with a job. For that reason, term-limited positions like postdocs are likely to be better for rapid mobility.

For many research positions, however, you will simply be disqualified (either explicitly or de facto) if you do not have a Ph.D. A Ph.D. is a minimal bar that shows you are capable of scientific work of some sort, even if mediocre, and if you have dropped out of a program, many people will assume you found you could not do this. You might be able to get around this by going through the world or tech start-ups, DIY engineering, or other non-standard routes, but this is much harder and the (scientific) success stories you hear about from these are very rare indeed---these worlds tend to make impact in other ways.

In short: a Ph.D. is a much straighter path to scientific prominence, and there are good approaches to help overcome mediocrity in your origins.

  • I support your answer. One little quibble: "Doing two or three postdocs is not unusual" is very true in some fields but not at all true in some others. – aparente001 Mar 25 '18 at 14:07
  • @aparente001 The OP is asking specifically about computer science: once, this was very rare in CS, but these days is not terribly unusual (I know a number of people who have done multiple CS postdocs). – jakebeal Mar 25 '18 at 14:14
  • OP was asking specifically about CS but the question doesn't say that any more! You know OP is in CS (I'll be honest and confess that I forgot!). But the question as it got pared down doesn't mention that. I believe the intent of the paring was to make the question generalizable so as to reopen it. My personal opinion: if you soften that sentence a bit, I think your answer would be even stronger.. – aparente001 Mar 25 '18 at 14:22
  • @aparente001 The question still has tag:computer-science, which I think is fine. – jakebeal Mar 25 '18 at 15:02
  • The exact number of postdocs is not really important; we can simply regard them as steps toward a good permanent research job, either in academia or in industry. This answer could be more thorough by adding the scenario where I get my PhD (maybe plus a postdoc) and then move into the industry. – user90316 Mar 25 '18 at 15:04
23

Your argument for abandoning your PhD hinges on a few false premises.

P1) You will wear the mark of your PhD for your whole life.

That is not true. A PhD is, in principle, a degree given to someone capable of independent research. Whilst you do your PhD you may be associated with your supervisors. That's because you're a minnow in the sea and it is the easiest way for someone to know whether you are worth talking to (sorry about that). After your PhD, people will rarely if ever ask that. I say that as a postdoc.

P2) You will exchange your dignity for money if you renew your contract. WHAT? That's something internal to you, an external observer is unlikely to see it that way.

Moreover, if you do not intend to do a postdoc, none of these points you made will apply to industry people. They will just see that you have a PhD and they will think that's worth something.

  • 2
    I agree with this. I would just like to add that after finishing, even if you stay in academia, you can start your own tribe and stop feeling associated with the previous advisors/school. Being passionate about your research as you seem to be, you won't need to be under their influence, or any kind of relation. Furthermore, if postdoc is something you would consider, don't take it for denied just because what you saw in those meetings. – iled Mar 23 '18 at 21:40
  • Thanks for you two's reply. I have rewritten my post and make it clearer, hopefully. Maybe you have something to add. – user90316 Mar 25 '18 at 0:18
5

If you dislike it as much as you imply and you're actually 2-3 years away, then by all means find something better to do with those years. But you say that you're very close, so I'll assume 6 months in my answer.

If I had to do it all over again, I would have not gotten my PhD; it was too much of my life that could have been better spent. However I have absolutely no regrets about the final 6 months, as difficult as they were. I recommend that you finish. Tech bubbles come and go, but potential employers and investors will always pay more attention to you if you have that particular set of initials. You will also gain personal experience in finishing something big.

Finally, if you're concerned that your project will yield valuable IP that the university or your former advisers will lay claim to if you complete it as a student, talk to a lawyer. Otherwise, as others have said, the "tribe" reputation stuff is nonsense.

  • Thanks for your reply. Could you elaborate about the last sentence? BTW, I have rewritten my post (see the paragraphs below the horizontal line). – user90316 Mar 25 '18 at 0:32
  • Thanks for the rewrite. It sounds like my "if you're concerned..." comment doesn't apply here. I agree with others who have said building a reputation as a scientist without a PhD would be far harder than building one with a PhD from a mediocre group. – Eric Hirst Mar 26 '18 at 20:23
3

I do not quite understand what career path you have in mind, but a PhD in computer science is very valuable in industry. In the US you can expect 6 figures minimum. Obviously this is better if it's related to an in demand field, but really CS is close enough for many, and it shows your ability to do independent research.

On the personal side, it seems to me like you'd be happier bringing this chapter of your life to a close (another option could be transferring to a different university for a clean start) but if you quit this close to finish line I wager you'd regret it in the future.

  • "Transferring" is indeed an option, but I don't have in my mind the destiny university. – user90316 Mar 25 '18 at 0:24
0

I am an extremely imaginative person, and I can only imagine a few scenarios where leaving your PhD program (without transferring to another) would set you up with a better science "brand."

  1. If your advisor is not just mediocre but committing fraud or widespread plagiarism or advancing pseudoscience.
  2. If you left for an extremely prestigious job working for/with someone known for their rigorous science. Essentially, someone on the Stephen Hawking level sponsors you as not needing to trifle with completing your Ph.D. (Separate from that sponsor, though, you're unlikely to receive academic or government funding. And after the Theranos debacle, private investors will probably be asking for more credentials before funding "science.")
  3. If your advisor were a world-class jerk/abuser (think the Harvey Weinstein of academia) and mediocre, people might cold-shoulder you for the association. Even then, though, leaving the lab would say good moral things about you, but only neutral scientific things about you.

As commenters point out, you can make up for your brand. If you're almost finished, a postdoc is a good opportunity to "rebrand." If you're a few years from graduation, consider networking at conferences, especially where there's a graduate student track or special opportunities grad students can apply for. Those are potentially ways to make up for networking and advising opportunities you miss out on with a "mediocre" supervisor.

-1

You must complete your PHD. Sure there are lots of useless PHDs but employers hate quitters much more than overspecailised irrelevent PHDs .So if you do not finish it will count against you badly .If you finish it may count for you .Maybe you should not have started but that is water under the bridge .Given that you have started and are more than half way there you should finish.

  • FWIW, I agree with what you're saying but your style is pretty blunt ("employers hate quitters"). The "must" and "should" that you begin and end make your answer harder to upvote, though, because they're broad advice rather than an answer to the OP's narrow question, and such an opinionated answer (in my book) would need even more evidence and discussion of values (what principles/perspectives underlie the "should"?) than you've provided. – cactus_pardner Mar 26 '18 at 18:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy