49

The importance of a recent PhD’s GPA was recently dismissed in another question (What does an industry recruiter want to know my Ph.D. GPA for?).

Why isn’t GPA considered relevant for graduate study?

Controversial Post — You may use comments ONLY to suggest improvements. You may use answers ONLY to provide a solution to the specific question asked above. Moderators will remove debates, arguments or opinions without notice. See: Why do the moderators move comments to chat and how should I behave afterwards?

11 Answers 11

84

Simple: a graduate degree (esp. PhD) is about research. GPA doesn't measure research ability to any extent.

  • 10
    I would add that most PhD students consider classes and grades a distraction from their primary goal -- research. I put in the bare minimum effort to pass my classes in grad school. – Thomas Apr 6 '18 at 19:57
  • 1
    @xuq01 - in my university in the US the way a graduate student does research was 'graded' and a grade was given. One's thesis advisor obviously gives the grade for the research work done for the current semester. You had to maintain 'B+' to ensure the research assistant stipend was continued else you would be made a TA. So I would say say GPA had some value in my case – gansub Apr 7 '18 at 12:58
  • 3
    I've always regarded GPA and other "convenience" stats are next to useless at any level. They don't help find talent at all in my experience. An abomination for the use of accountants and administrators, IMO. But at PhD ? If their supervisor doesn't already know what's going on, then no GPA score will fix that system. – StephenG Apr 7 '18 at 13:50
  • 1
    @gansub Interesting. Do they give real grades or do they almost always give an A? – xuq01 Apr 7 '18 at 19:28
  • 1
    @Thomas true as that may be, I've lost track of the number of mid and late career PhD physicists who've said they wished they'd spent more time on their grad school courses. – DanielSank Apr 9 '18 at 4:58
54

The PhD is a ~6 year degree, and most classes are in the first year or two. As others have said, the classes are really incidental; the PhD is about doing original research. There is also the question of normalization: GPAs must be interpreted in terms of the subject, the quality of the school in that subject, and the student's specialization.

Given this, the graduate GPA is only useful internally, to flag students who are doing unsatisfactorily in the coursework, or to help advisors choose students -- advisors have the necessary context to interpret course grades. In contrast, some bozo in HR will not have the context to decide whether a GPA of 3.5 from Harvard in Music is more impressive than a GPA of 3.8 from NC State in Chemistry.

  • 9
    Downvote for needless dumping on HR (they have an important role and often do do the research to find out the meaning of those numbers); but otherwise the substance of this answer I agree with. – Richard Rast Apr 6 '18 at 13:23
  • 17
    @RichardRast Not any HR, cag51 mentioned some bozo HR. I think there is nothing wrong with the statement. – padawan Apr 6 '18 at 13:59
  • 8
    @RichardRast You had to admit, it is a lot more difficult than "which number is higher" and almost can't be boiled down to a formula. And you're right that HR provides a very important role in any organization, but that many times they're so focused on the wrong things that it leads them to make the wrong decision. I once had a manager who told HR "Stop filtering resumes for these positions, or I'm going to start posting them myself so you can't filter them." because he went through the reject pile and found 10x the number of qualified candidates than he officially received. – corsiKa Apr 6 '18 at 14:06
  • 16
    @RichardRast -- even if well meaning, I don't see how HR could possibly do the research to find out the meaning of these numbers. Things vary not only across the hundreds of schools and thousands of programs, but over time as well. Hence, my statement that anyone in HR who tried to rank candidates by grad GPA is likely a bozo -- though obviously there will be exceptions, like anywhere else. – cag51 Apr 6 '18 at 15:30
  • 6
    Not just the quality of the school but the culture of the school/department. Some places do A=acceptable, B=unacceptable, others may do A=great, B=acceptable/good, C=unacceptable. Absent a letter explaining the culture of games at the school, it's hard to judge the GPA. It's hard enough in undergrad, but you have ~40 data points in multiple departments to smooth stuff out. In a PhD program, there may be just 12 or less taken (especially if students have actual P/F option as my university let us do for up to two or three classes) – guifa Apr 6 '18 at 22:14
21

One more point: in some cases, a graduate course may be simply an opportunity for students to attend a series of advanced lectures on a particular topic. There may not be any homework, exams, papers, or coursework at all. In such cases a grade will be assigned more or less arbitrarily, perhaps just an A for every student who showed up to most of the lectures and/or asked some pertinent questions. So the grade really doesn't indicate anything about the student's knowledge or skill, and this will tend to make the GPA less meaningful.

11

I would like to extend xuq01's answer a bit:

Simple: a graduate degree (esp. PhD) is about research. GPA doesn't measure research ability to any extent.

For those who wonder why, here is my take.

  • Grades are mostly relative. PhD programs usually have too few students all studying unique subjects so they can't be graded relative to each other.
  • Most PhD programs are different and hardly comparable even within one field or one university. Without standardized curriculum, grading on absolute scale is pretty much impossible.
  • Formal grades are given by someone who is superior. By definition, a successful completion of a PhD program makes you an independent peer researcher with no one on top to "grade" you.

So even if a university provides PhD GPA, it will at best be interpreted as some in-faculty fraction-measure to track early-stage performance.

  • I really don't see why formal grades have to be given by someone who is "superior". Every single NBA player is "superior" in basketball skills to me, yet I feel qualified to evaluate their performance relative to each other. Similarly I feel confident that I could sit with a PhD candidate and come up with a reasonable assessment of the originality, utility, and value of said candidate's research while conceding the superiority of said candidate in the research domain. – emory Apr 7 '18 at 14:02
  • 1
    @emory superior in a hierarchy, not necessarily superior in knowledge or performance. You can rate NBA players all you want, but they're not going to listen to you, they're going to listen to their coaches. – barbecue Apr 7 '18 at 17:33
  • @barbecue I was unaware that PhD students exist outside the hierarchical organizations that incarcerate the rest of us. Do their supervisors, dissertation committees, deans, and funding agencies know about this? – emory Apr 9 '18 at 12:26
  • @barbecue NBA players don't (nor shouldn't) listen to me, but they listen to their coaches, managers, and team owners whose basketball skills while superior to mine are vastly inferior to that of the players. – emory Apr 9 '18 at 12:28
  • @emory I think a formal grade by definition has to be assigned by a superior. Otherwise it is not a formal grade. The issue here is the definition of the word "grade". It is pretty strictly embedded in GPA across all academia and multiple countries. It can't be stretched without discrediting the GPA itself. – Arthur Tarasov Apr 9 '18 at 12:36
9

In the case where you do have a GPA during your PhD, it's frequently irrelevant and largely the result of some arcane bureaucratic requirements.

After my wife defended her thesis and requested her transcripts, she was surprised to discover that she had been taking 12 units of a class called something like "continuing research". Her university had automatically enrolled her and her advisor had actually given her a letter grade in the course every three months for the past four years.

He gave her an A every quarter, because that's what was required for the university to continue paying her tuition.

7

The primary purpose of a PhD is to create an independent researcher and add to the existing knowledge in a given field. The secondary purpose may be to train the doctoral student on the 'tools of the trade', which includes practical knowledge of experimental/modelling/other techniques, as well as the fundamental knowledge needed to use those tools.

Completion of the primary objective is judged by a thesis (or equivalent research output). The secondary objective may be judged, in some part, by exam scores/GPA. Therefore, the GPA counts less than the actual research output. This is why some PhD programmes waive off coursework altogether. Similarly, completion of PhD is based on thesis acceptance, not coursework completion.

This is in contrast to say, undergraduate study, where the primary objective is to teach the fundamental knowledge. Any research work is generally intended to acquaint the student with research methodology, not to generate actual output. Therefore, the GPA is more relevant as a primary metric.

It may be further argued that even in undergrad (1) GPA is not the best measure (2) A low GPA student may turn out to be an excellent researcher later. These may be true, but until a better, standardised system is devised, the existing system stands.

5

My take on this is the fact that most reputable schools will dismiss students with low GPA. But even if GPA is sub par, if a graduate student gets to the point that an unbiased committee will award this highest degree, proficiency has been demonstrated. Is a PhD from A with GPA 3.2 really different from a PhD from B with a GPA 3.5? Probably not in any meaningful way.

5

That GPA from the PhD is irrelevant is not entirely true.

If you proceed with a career in the industry, other things count more, like for example just the fact that you could focus for so long on a specific goal, had the ambition, discipline, persistence and self-organization skills to finish quite an extensive, exhausting and more often than not frustrating project successfully. This is quite a statement about you as personality from the perspective of any head-hunter in industry. The more or less subjective perception of the supervisor of how well did you do your research however is naturally not the priority.

Things look a bit different, if you intend to proceed with career in academics. Then the GPA (or the "evaluation" of your PhD as we have it in Germany) suddenly becomes important. How important also depends on the research group, research field and the region where you are, but in general, it is more often really important than not. In Germany (and most of the EU) for instance it would be extremely hard to apply for research grants and funding (especially as a young, just starting-trough researcher) if the evaluation/grade of your PhD is anything but "with excellence". In some fields, it could kill entirely the chances for career in academics or in a specific research group (especially if it's closely affiliated with the supervisor who gave the grade).

So, it depends a lot. And yes, there are situations, where it is quite important:)

  • 4
    The German grades for doctoral degrees are a very different thing from the US GPA. In particular, the German grades are given based on the dissertation, while the US grades are given based on the anciliary coursework. – Arno Apr 7 '18 at 12:57
  • The UK has no grades, it hasn't stopped them applying and being awarded grants yet, – Marianne013 Apr 8 '18 at 23:19
5

FWIW,

I remember hearing the same thing when I did my Ph.D., "grade point is irrelevent" and I was starting to act on that. Had a buddy tell me "BS" and I then decided to get a 4.0. And this from a guy with Harvard capability but always an underperformer (3.3. in HS, college, nuke school: which are progressively harder...but still NEVER to my capability).

I still did publications and talks and lab work and the rest of the stuff. But that 4.0 helped me get a LOT of interviews, get into McKinsey, etc. So...I would just be a little cagey when people tell you the GPA is irrelevant. Yes, you need to "bring it" on research also. But get good grades too. It's not that hard. Doesn't hurt anything. And looks great on a resume.

4

The answer to this question is highly dependent on the way the school's course structure is set up. At my institution, a GPA would be meaningless. At the majority of institutions I have been associated with or worked with (all in the United States), the setup has been very similar:

In our first year, we take 4 core classes and must get a B or higher (3.0 grade points) or drop the program. After this, we may take 2-3 optional electives but these courses have no formal examinations and simply require giving presentations. By the way, you can sometimes get out of these courses or get infinite extensions on projects if you are in the middle of writing a grant or paper revisions. The rest of the time, for the purposes of tuition, we are enrolled in a 3-9 hour credit course called "Research Problems" which is automatically awarded an A unless your adviser intervenes (it's 3-9 hours to fill the remainder of credit hours so the student is taking a total of 9 credit hours). "Research Problems" corresponds to the fact that after the first year, you are just doing research. The vast majority of students are much more likely to drop the program than they are to continuously get poor grades for Research Problems so we can assume they would receive A's.

Why is the GPA meaningless? I took 14 semesters to graduate for a total of 126 credit hours. The classes made up roughly 20 of my total credit hours. Since the lowest grade I could make on these courses was a B (3.0), the remaining 106 credit hours would be A's (4.0). This amounts to a 3.84 as the lowest possible GPA I could have had. But this is not in any way reflective of the student's actual performance in their classes which is what the GPA is really measuring. If a recruiter asked me for this, I would assume they have no idea how graduate programs are structured or it was just a formality.

What a recruiter should ask for, if it matters (I can't see why it would), is the grades the applicant received in a class. I would still argue that this doesn't really tell anyone much of anything. In my candidacy examinations, the committee received a copy of my transcript beforehand so they could see the grades I made in my first year classes so they could ask question to test if I still had gaps in these areas. They would not have let me pass if they felt I was not ready. I could not have published any first author papers if I was still struggling with an introductory level class.

As a side note, having trained many undergraduates, the correlation between research aptitude and grade point average is not nearly as tight as one might think.

2

In a PhD program you don't take many classes, and 90% of your efforts are focused on research (the other 10% is split between TA'ships and classes). This is because you are being trained for a specific purpose, i.e., to do research. The ability to create, carry out, and report a research project is what you are there to perfect. Classes are good supplements to hone this process, but are mostly just a distraction. Similarly, evaluating a PhD on grades is like judging a carpenter on how quickly he can work out math problems, or recall how many teeth are on a saw. If he can frame a house to code, do those things really matter? Sure, they help him get there, but you look at the product he has been trained to produce when assessing his talents.

Generally, there is only one time when GPA really matters (and I say this loosely). This is when applying to grant agencies, in my case, NSERC.

Tangibly, GPA can play a big role in seeking employment. If an employer asks for your PhD GPA, this is an indication that you are in the wrong building, talking to the wrong person. Leaving immediately can save you some potential hassle down the road!! :D

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.