I recently spoke to a graduate coordinator of a mathematics PhD program I was planning to apply to. I asked about admission, requirements to complete the degree and such. The program coordinator ended up telling me that usually they only pass 50% of the people in the program in the qualifying requirement, and the rest either get kicked out entirely or can finish with a masters. At first I thought this only happens maybe once in a while or something, but I asked for a clarification and he indeed stated that they have made that as their threshold since "not everyone deserves to earn a doctorate"?

This is a US program. Is this attitude here common? If it is, I'd much rather apply to European programs. I don't want to get kicked out because a program made a mistake in choosing me. I thought programs only admit students they believe are able to successfully finish the program. Apparently I was wrong. Is this the norm or even fairly common? I am confused.

  • 4
    I know of engineering programs that kind of do this, although many only want a masters to begin with.
    – Jon Custer
    Jun 24, 2021 at 2:23
  • 54
    I genuinely like your "a program made a mistake in choosing me" attitude.
    – sleepy
    Jun 24, 2021 at 8:09
  • 4
    So, you currently hold a bachelor? Quite a few european phd programs require a masters. So you can take the US and if you exit gracefully with a masters you can take the EU program. If you have a masters already it the US program might be too slow (did they comment on the usual time? 5 years?)
    – lalala
    Jun 24, 2021 at 11:41
  • 4
    Sounds like a lightweight version of the Google hiring gauntlet. Jun 24, 2021 at 15:44
  • 2
    Not everyone that fails comprehensives at one institution fails to get a PhD from another. Even from a higher ranked place. People fail comps for a variety of reasons, burn out, say.
    – Buffy
    Jun 24, 2021 at 15:45

5 Answers 5


I am quite surprised to hear that a serious math grad program in the U.S. still operates in this fashion.

Yes, decades ago, this was somewhat the style, sadly.

Our/my program has not operated this way for decades. Many other top-rated places that formerly did operate this way have changed, so that, yes, admission (with funding) is a vote of confidence.

So, no, so far as I know, such an approach is hugely anomalous in the U.S. In particular, there's no reason for anyone to subject themself to such a game. Go where people have confidence in you, rather than are skeptical. Srsly.

  • 17
    "there's no reason for anyone to subject themself to such a game." People do it to get a visa. Jun 24, 2021 at 1:25
  • 3
    Additional information: this program in the US news ranking was ranked between 70 and 100, so I am not talking about a top ranked program or anything of that sort. Still, it's a very decent school I think, so I was very surprised. Jun 24, 2021 at 1:38
  • 23
    Yes - at that level in the rankings (which is quite low), its a lot less surprising, because those programs frequently have difficulty recruiting qualified graduate students, and view their graduate programs as giving a second chance to students who really want to go to graduate school but don't have undergraduate records showing they are capable. Jun 24, 2021 at 2:01
  • 23
    Worth noting that this is less exploitative than just running a normal masters program, this way you get a free masters degree. Jun 24, 2021 at 12:31
  • 5
    @polygonlink1 My department is in this range and it's true that a lot of our grad students don't get a PhD (for a variety of reasons). But even in this tier I don't think it's normal to put quotas on how many students can pass qualifying exams. And I would say A. Woo's comment is partly true: we first recruit students we think can succeed, and then when we need more students we take chances on students who have less of a proven track record.
    – Kimball
    Jun 24, 2021 at 16:30

Some universities need more teaching assistants, but cannot afford to pay for good ones. So they recruit unqualified teaching assistants as PhD students, and then kick them out when they fail their qualifying exams.

This is not an ethical practice and you should not enroll in a PhD program that does this.

At good quality universities, it is common for most students to pass their qualifying exams. At other universities, it varies.

  • 1
    "Calculus fodder." These programs tell themselves that they're "giving everyone a chance." Jun 24, 2021 at 15:40
  • 7
    Too pessimistic. I can't imagine a meeting of deans and such in which such a policy would be suggested.
    – Buffy
    Jun 24, 2021 at 15:50
  • 14
    Good quality universities get so many applicants that they only need to accept good quality students. They will have a high pass rate even if their teaching quality is poor, since good quality students will have already learned the most important life skill - how to teach themselves,
    – alephzero
    Jun 24, 2021 at 18:47
  • 2
    Yes, it might be hard to imagine a meeting of deans where this rationale would be proposed when suggesting a policy of recruiting more unqualified students, but I can definitely imagine it being mentioned in a discussion about whether to continue recruiting such students, in a place which already does recruit them.
    – kaya3
    Jun 25, 2021 at 23:10

I know that some physics PhD programs were notorious for taking on extra students (knowing that they'd likely fail the qualifying exam) and just using them for the cheap TA labor and giving them a master's degree after 2 years.


I can't speak for all disciplines, but I do not think that that program's harsh attitude is the norm in the US. I have heard from numerous sources (though perhaps they're all citing the same base source) that only about 50% of PhD students manage to finish their PhDs[1], but it was my understanding that that was mainly due to people dropping out, particularly after the PhD Candidate stage, since writing the dissertation is the hardest thing. Note that the warnings in the source I linked to are about things like picking a problematic dissertation topic, expecting too much handholding from other people, etc. I had not ever heard that it was mainly due to programs being too brutal regarding whose work they accept or reject. (At the same time, those things are connected: If you pick a problematic advisor and a problematic topic and expect other people to hold your hand more than they are willing to, you won't produce work they can accept).

There can also be issues of being able to afford being a student, such as being distracted by needing to work full-time while doing it, so of course you also want to consider how well funded you would be or would not be.

This says the PhD failure rate in the UK is only 19.5%[2], so perhaps you do have a better chance in the UK. However, each US program is different, so you would do well to ask about the failure rates of the ones you're interested in, and also ask why people are failing. (Are they submitting work and it's rejected or do they not even submit the work?) The program you're interested in is telling you "We're terrible," so I would believe them and avoid that program, but don't dismiss all US programs based on that.

  1. Example source: https://dissertationgenius.com/the-six-laws-of-phd-failure/#:~:text=To%20give%20you%20a%20dose,over%20the%20past%20three%20decades.

  2. https://www.discoverphds.com/advice/doing/phd-failure-rate

  • I suspect the second reference refers to the completion of the PhD after completing the masters. The question talks about a program where people need to first do the masters and then the PhD. Jun 25, 2021 at 17:42
  • Oh, does it? (I don't see that in the question.) And what is the norm in the UK - to separate MA and PhD, or no? At any rate, now I'm re-looking for stats for the US that make it clear if they're talking about PhD programs that encompass the MA as well or not, and they tend... not to specify. I just keep seeing the 50% figure again and again. (Here files.eric.ed.gov/fulltext/ED580853.pdf and here theatlantic.com/education/archive/2016/07/… and more.) Jun 25, 2021 at 18:36
  • The Atlantic is not so clear, I agree. When someone says they are in a PhD program in the US, sometimes that means they just completed their BA and hope to get a PhD. Jun 25, 2021 at 18:44
  • Yeah, it's my understanding that in the US there are PhD programs meant for people who currently have a BA and want a PhD (you get an MA along the way but it's not DESIGNED for people to quit at that stage unless something goes wrong) vs. programs that would make BA-holders apply for the MA, finish it, then apply to the PhD program. My program was like the latter. It was stressful because I had to choose between them--whom I liked but who technically had accepted me only for the MA--vs. other programs that had accepted me for the (MA +) PhD. Jun 26, 2021 at 21:22

It's not math, but UC Berkeley Operations Research (IEOR) has a similar policy. About 50-67% pass the preliminary exams at the end of their first year of the MS-PHD program, and can proceed to the PhD program. I think you get the MS at this point, whether you pass or fail. A small number get a "conditional" pass and have to take the test the next Spring if they want to continue. Students cannot get a PhD there without getting the MS first (from the same program).

I think it's a matter of tradition and hard-ass pride: the faculty went through hell in their careers, and that seems like the right way to them. I'm pretty sure the official rationale is that the best way to evaluate whether a student is a promising candidate for doing PhD research is by putting them through the preliminary exam process. They want their PhD students to write high-quality dissertations to maintain the reputation of their department, and also because they will (ideally) be investing substantial time in thesis supervision.

I personally think it's inhumane, or at least excessively stressful, and that they should get rid of this policy, but the faculty are pretty set in their ways. I don't think the math or other engineering programs at UC Berkeley have this policy.

I'm not sure whether the nefarious financial incentives mentioned in other comments and answers hold here or not, but I don't think they're a primary consideration. (Let's just say I'm closely acquainted with some of the faculty.) And surely the explanation for this practice, in this case, is not that it's an academically substandard program.

  • 3
    At some point in the past, the UC Berkeley Math Dept did also have a harsh filter after the first years, in which a substantial fraction of the grad students were channeled out of the program. I do not remember what the official description of the process was, and I only saw it "at a distance". Jun 26, 2021 at 22:30

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .