I'm a relatively new PhD student, and I'm faced with a new (for me) situation. Before, I always had grades and the likes to tell me how good or bad I was doing: if I was at or near the top/bottom of my class, if I passed my exams or not... were all more-or-less objective indicators to tell me how I was doing.

But now that I've started to do research, I have none of that anymore, and it's quite discouraging. It's not that I need constant validation to boost my self-esteem, but rather I'm afraid that I'm not working hard enough, not learning enough... and that it's all going to crash down on me at the end ("welp, your funding has run out and you've produced nothing good enough, no PhD for you"). I sometimes find results, but they always feel rather insignificant. And the fact that I used to be good as a student whose only job was to learn about well-established topics from good professors doesn't mean I'm good enough at research, either.

I've read several Q&A's here (How should I deal with discouragement as a graduate student?, “I've somehow convinced everyone that I'm actually good at this” - how to effectively deal with Imposter Syndrome) and they've been helpful, but simply telling myself that maybe it's just the impostor syndrome talking isn't helpful – maybe I'm really an impostor, too...

Of couse I can ask my advisor from time to time, but if I start asking every month I don't think he's going to appreciate it, and if I ask that bluntly he may not want to hurt my feelings and tell me the truth. So how do I, a PhD student, can evaluate how well I'm doing?

  • 4
    If you are genuinely looking for objective criteria, then submissions such as posters, conferences, papers etc. can be long-term targets. Grades are usually obtained on a semester-wise basis, but you might not be able to have publications every semester (of course, that depends on your field). I would think that having short-term targets such as "finish X this week", "make setup for Y by month end" should help quantify your work. Commented Jan 6, 2016 at 17:04
  • 2
    If you are an experimentalist, than there should be some plan for what experiments to do, what you expect to see, and if things are actually getting done. Note that seeing something unexpected (and not bogus) is kind of the exciting part. But, measurements get done, data gets analyzed, slides updated, there is something to talk with your adviser and colleagues about. If more theoretical, well, there is a reason I didn't go in to theory...
    – Jon Custer
    Commented Jan 6, 2016 at 17:18
  • 1
    Number of papers written + number of years left. Larger number is better. Commented Jan 28, 2016 at 14:09

8 Answers 8


The reality is that it’s impossible to evaluate how you’re doing in an entirely objective manner once you progress beyond the stage of doing problem sheets with well-defined answers. Once there’s no objective right and wrong, it becomes a question of what is valued and who assesses it. For example, people try to assess PhD students in terms of all kinds of things:

  • How many papers they’ve published.
  • Where their papers are published.
  • How many people have cited their papers.
  • How much impact they’ve had on their field.
  • How often they’re invited to give talks.
  • Where they’ve worked.
  • How successful their collaborations have been.
  • If they teach, how successful their own students are.

The reality is that all of these things measure something, to some extent. Different people will value different things, and assess them in different ways.

If you want to go into academia ultimately, and you’re asking because you want to know what the people who will assess you for academic jobs will value, a good place to look is at academic job specifications - they’re often very detailed and make it quite clear what they’re looking for. You can also ask your supervisor, who will probably be able to give you a good idea of what’s involved.

If you want to go into industry, they may be looking for very different things - again, you can find out by looking at the job specifications, and the skills of the people who work at the relevant places.

Ultimately, this is a good point in your life to realise that now you’ve moved beyond standard taught courses, you’ll never again have the same kind of “objective” outside feedback that you once did. Thus far, you’ve been optimising towards someone else’s objective function; from now on, you have to find your own. Other people can tell you what they value, or even what they think you should value, but no one can decide what you value but you.

  • 2
    +1. It's really a short-term/long-term process. First, decide what you want to have done in, say, one year. (Get your supervisor's input on that.) Then set up a plan how you actually want to reach that goal, with SMART weekly action steps, as per @varun's suggestion. Finally, track whether you are meeting your weekly goals. Adapt if you feel over- or underwhelmed. Commented Jan 6, 2016 at 19:22

As PhD student, your supervisor is the key. He is probably the person that will have a higher stake at your final evaluation as a PhD candidate, both scientifically and bureaucratically. Therefore, he is the key person to talk to.

Of couse I can ask my advisor from time to time, but if I start asking every month I don't think he's going to appreciate it

Of course, but that does not mean that you cannot evaluate how he reacts to your meetings. As scientist (and any other carrear path), you are expected to understand what other people think of your work. This starts during PhD and will continue to happen on your career.

It is also by his judgment, and the judgment of your peers, that you will start to be able to auto-evaluate your performance. And again, this is also expected from you at a later stages of both scientific and non-scientific careers.

and if I ask that bluntly he may not want to hurt my feelings and tell me the truth. So how do I, a PhD student, can evaluate how well I'm doing?

If you suspect that he is not telling you the truth, then you have to fix that. Either on your side (e.g. he doesn't feel you can take the truth), or on his side (he is not able to tell you the truth), having this fixed is something that both will profit from.

Disclaimer: I am a PhD student and this is purely knowledge that I gathered as a PhD student.

  • 8
    +1 for If you suspect that he is not telling you the truth, then you have to fix that.
    – JeffE
    Commented Jan 6, 2016 at 21:38

Only as a partial answer: most grad students (in mathematics, in the U.S., my experience...) do not fully appreciate that their very admission is a strong vote of confidence from the admitting department. Further, although it would be wonderful if for any given person everything "clicked" and they were incredibly lucky and did something super-amazing ... this would be a ridiculous standard. Even at the most elite places, for all but a tiny fraction, a PhD is an apprentice-ship. What is a successful apprentice? Well, rarely glamorous, ...

Bottom line is that if you stay engaged, and still care about your studies/work, you are almost surely "successful". Successful enough... because now it would be non-sensical to think about "beating" other grad students, both because the notion is far too vague, and because the population of grad students in a reasonable program is heterogeneous in many senses that make comparisons unreasonable. Not to mention that as one is selected into ever-more-elite groups, "being the best" or even "being in the top half" (if that had a sense) becomes ever less likely (and/but, mercifully, ever less meaningful).

So, again, if you are staying engaged and working, you are doing fine. Period. Yes, your advisor should be able to confirm that. Yes, you may disturb your advisor by asking once-a-month... but that's probably ok. Your advisor may not have the insight/communication-skills to respond appropriately, which raises the possibility of misunderstanding... thereby upsetting you...

That is, once people have managed to get into graduate programs in the first place, they have adequate talent/capacity to finish PhD's, if they care and work hard. For most people, "lifetime accomplishment" is very hard to predict at the grad-school stage, yet that is what we should care about. Often, the most "obedient" apprentices give a good impression, but will not have the energy or imagination to do much on their own. Meanwhile, apprentices who had some difficulty making themselves "stay with the program" eventually make very valuable and original contributions.

(Paper-count, grades, all that stuff, ... is pseudo-objective, sure, but only loosely correlated with real-life success, ...)

So how does this answer the question? The answer is that you're doing fine, if you are working hard and staying engaged. Any attempt at finer pseudo-objective evaluation is probably nonsense. Even having trouble with "preliminary exams" (or whatever they're called locally) results far more often from misunderstanding (and mis-communication) than from incapacity. And I'd be remiss to not remind you that the gossip/opinions of other grad students is very often not merely inaccurate, but very wildly so. Even faculty often do not directly communicate the belief system that is the true basis for action, but, rather, quote standard professional mythology.


In universities and in general in life, an external sincere evaluation is the strongest tool you can have to assess if you're doing well or not.

Many PhD programs include an yearly internal evaluation of the student's work, but sometimes it's not enough. My suggestion is to ask an informal external evaluation to someone outside your institution. He/she might be someone who knows you a little or someone you don't know in person, but he/she must be someone expert in your field.

Once you found someone who is expert in your domain and is available to informally review your work, you should write down a technical report about your activities since you started your PhD program, and then send it him/her asking to evaluate it.

If he/she's a serious person, in a couple of weeks he/she will reply to you including his/her assessment of your work, and some suggestions on which directions to take.

If you were able to find more than one informal external reviewer, that would be even better.


  • 2
    This is an interesting idea in principle, but I wonder how effective it would be in practice. Have you had experience with people who actually do this? I would wonder whether a random external expert is going to have the time or motivation to evaluate the work of random grad students. I would also wonder about political considerations. If the external person gives an opinion of the work that is much higher or lower than the advisor's, that seems likely to lead to tension between the two of them - and academic research communities are often small... Commented Jan 7, 2016 at 6:22
  • 2
    So I'd wonder whether the external person would really feel able to give an honest opinion. Commented Jan 7, 2016 at 6:23
  • I must say that I would not agree to doing such external evaluations of grad students at other institutions than my own. For one thing, the "external, objective" aspects of grad students' work (in math, my field) do not typically reflect their eventual course... The key things are intangibles, and that's why everyone has trouble "objectifying" what's going on... Commented Jan 7, 2016 at 23:32

In my experience, it is very hard. If you do not already have self-confidence in your progress and skill development (and the fact that you ask, indicates you do not), you will need to accumulate experience to find out how to judge yourself; whether you are over-optimistic or undervaluing your work. A supervisor can help, but need not. They may think in different categories.

What you could do is, that, after accumulating a bunch of results, you ask your prof whether they can constitute a paper, e.g., for less dominant results, a conference. That externalised feedback is likely to be better and more specific than if you ask about your own progress.

Don't worry too much about impostor syndrome - it is the hallmark of conscientious students and signals that you care; it's ok to take that into account, but don't let it take over. Also, don't forget, if you compare yourself to the stars, you are not going to win - there are absolutely brilliant people out there which are able to do amazing things. So, forget them, and compare yourself to what you want to achieve and your progress in view of that, of course, with regular feedback from the scientific community, via talks, papers and the like. On the PhD level that's all that counts.


To expand on Stuart's list of what different people may find interesting

  • What courses have they taken and how fast did they complete them? How relevant are they?
  • (If teaching) How good they are at teaching, grading exams and producing teaching material?
  • Have they had main responsibility of any particular course or lab?
  • Have they participated and gotten experience writing applications and documentation?
  • Have they gotten practice co-supervising graduate and/or undergraduate theses or (individual or group) student projects?
  • (How well) have they learned to make presentations / explanations of their work tailored to a particular audience?
  • Have they acquired practical know-how of independently working with the tools they use, and are those tools useful in other fields or industries outside of academia?
  • Can the results be applied to solve any problems in their own, or other fields or industries?

Your research has an ultimate goal. You divide the long way to the goal to X stages. Then you complete each stage, moving from one to another. Your success can be assessed as Y/X*100% where Y is the number of successful steps.

  • 3
    Can you please expand a bit on how you think this might be possible in general? There are a lot of unknowns in research.
    – jakebeal
    Commented Jan 7, 2016 at 15:35
  • I don't understand your question. When you get a grant you get it for some final purpose. Also, you present your research plan to the money-giver. Full-fulling the steps in the plan shows your advancement. Usually nothing goes by plan so you change it accordingly. But again you remain with the plan which you are going to complete. Also, negative results are valuable results, too.
    – rlib
    Commented Jan 7, 2016 at 15:44
  • 4
    "Usually nothing goes by plan so you change it accordingly." And thus "3 of 7 stages" isn't very meaningful to a graduate student with little experience in how to evaluate such things.
    – jakebeal
    Commented Jan 7, 2016 at 15:57
  • It was meaningful for me. But researchers are different, yes.
    – rlib
    Commented Jan 7, 2016 at 16:29
  • 3
    Only low-level, automatable non-research "research" can be done so mechanically, and evaluated mechanically, I suspect. Commented Jan 7, 2016 at 23:34

As a Ph.D it really is simple: 1) Have a plan when to graduate and communicate that expectation to your advisor 2) Figure on publishing in referred journals three times - use these as your progress milestones 3) Don't get married 4) Don't have children 5) Live in the lab/library

Notes: In re: 1) if your advisor cannot give you a timeframe - GET OUT FAST 2) Different disciplines have different expectations. If you are publishing, you will graduate and you will have what you need to get a job. 3) & 4) Dissertations are very jealous mistresses 5) If you ain't doing the work; the work ain't getting done.

And remember most of all: You are the only one who cares if you finish. To your advisor you are cheap labor.

Ed Ph.D, PE

  • 2
    This is altogether a bad advise. To your advisor you are cheap labor OP never mentioned a remotely close statement leading to this.
    – Sathyam
    Commented Jan 7, 2016 at 23:03
  • 2
    This may be how you got your PhD, but to say this applies as a general rule for everybody is terrible advice. (Even with your disclaimer with respect to #2.)
    – ff524
    Commented Jan 7, 2016 at 23:20
  • This opinion is merely an incompetent rant, and is very bad advise, in any case. Commented Jan 7, 2016 at 23:33

You must log in to answer this question.

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