26

I am a PhD student in computer science (theory). I am worried about my productivity. I try to do as much as possible. I used to take four courses per semester during my first two years of studying. Now I have started doing research. In the last 4 months, I have only read 2-3 proofs and one research paper. It took 2-3 weeks to go over each mathematical proof.

Question: How do I know I have done enough work in one semester? I mean is there any parameter to measure the work that I have done in the last semester? I have heard some students read only one research paper per one semester.

I can get feedback from my research supervisor, but the problem is he might say I did enough work just to keep my motivation high.

  • 8
    The most important point is not whether you have done enough work in one semester, but what are the strong and the weak points of this work. And there is only one person that can tell you about this, and it's your adviser. – Massimo Ortolano Apr 8 '18 at 17:07
  • 6
    You can never do enough work! No matter what you did, it will always seem like you could have been more effective, or like somebody else in a similar position did more... – einpoklum - reinstate Monica Apr 8 '18 at 22:00
  • 4
    In mathematics, it's quite common that almost all the research for a PhD dissertation is done within a single week, after a lot of repeated failure. In such a scenario, progress is pretty much impossible to measure. I think theoretical computer science can be much the same way. – Alexander Woo Apr 8 '18 at 23:38
  • 7
    Enough work ... for what? To what end? For your own satisfaction? To "beat all competitors"? To "impress your advisor"? To "succeed"? None of these quite make sense, which hints at the impossibility of really answering (or giving meaning to) the question. – paul garrett Apr 9 '18 at 0:05
  • 19
    I can get a feedback from my research supervisor, but the problem he might be saying you did the enough work to just keep my motivation high.Do not listen to the Impostor Syndrome. If your advisor tells you you're doing fine, then you're doing fine. – JeffE Apr 9 '18 at 1:34
30

Everyone works at his own pace; moreover, the pace can fluctuate a lot depending on the time of year, your personal life, and "position of stars in the sky". Even though the semester seems like a decent amount of time to average out those fluctuations, I don't think it is.

I would try to use the following criteria to estimate the term success:

  • feedback from your advisor
  • how did the term go compared to the original plan (yep, here I assume that you make the plans for your term/month/week and correct them accordingly. Hopefully, some milestones are discussed with the advisor as well)
  • feedback from your committee (in some universities, PhD students meet with their committee regularly or send them the progress report to hear their feedback)

I would certainly not recommend comparing your progress with other students because they are different, they have different goals, and you don't have a complete information about their progress either. It is very easy to get discouraged for no reason.

To sum it up, I would stress the importance of initial planning and correcting the plans throughout the term. Then, you will have a very good measure of your success. You will end up with a different question of "how to plan", but that is a totally different problem.

13

Quality is more important than quantity in research.

One solid proof that proves an important result likely matters far more than a bunch of smaller, less significant proofs. Similarly, one big research advance that shows long-term promise will matter more than a bunch of small experiments that don't open any new doors research-wise.

Also remember that most research does not succeed at first! I tell my students when they're getting started that research will likely be a bunch of failures punctuated by the occasional success. This is not to discourage them, but rather to get them to realize that the process is slow and winding. At this early stage, I want to get them familiar with the tools they need and the skills they must develop, rather than focusing on early breakthroughs.

So don't try to plan your research too closely—you should have some goalposts and markers in mind so that you're not a perpetual grad student, but you don't need to micromanage your work down to the day or week (other than keeping track of external deadlines). Focus on doing the best work you can, and talk to your advisor about your progress.

8

Worrying about whether you've been productive enough is not going to make you more productive.

I suggest that you diversify your goals for yourself. You mentioned two prongs to your work: reading papers, and writing proofs. See if you can find some additional prongs to add to your definition of "being productive." Here are a few ideas to get you started:

  • Get to know the publications in your area, and figure out what parts of each are of most interest to you.

  • Regularly browse the most recent issues of those publications. Gradually get better at skimming articles to get a general idea what they're about.

  • Develop a system for cataloguing your notes about your journal reading.

  • Attend department seminars. Try to gradually increase your level of understanding of these talks. When you feel ready, start to ask questions, either during the Q&A part at the end, or afterwards, more informally.

  • Form a study group with some fellow students.

  • Attend some thesis defenses.

  • Attend a conference.

  • Start figuring out what you think makes a good talk.

  • Start working on your writing and powerpoint skills.

I don't mean that you should necessarily try to do all of the above, and of course, the papers and the proofs do need to continue being in center stage. But a bit of diversification may help you worry less and produce more.

6

I'll take an unloved twist to this. I should warn you that the milage varies hugely, so instead of taking the numbers for granted, look at people in your area of research that you would assess as good and look at their output. You should aim at seasoned PhD candidates or early postdocs.

The number of papers per year

In my PhD area it was "good" (but not extremely, insanely, genius-level wonderful) to publish two papers in a year. This number, as I already warned, may vary. In life sciences in might be like one big paper in three years. Tenured professors can manage 10 in a year.

Not only the field, but also the experience, basically how long you are in the academia, is important. It is Ok to get, say, only one paper in the first PhD year. You should get more then one in your final PhD year, though. And, for the last time, your milage may vary, depending on your "tenure" and field.

Most importantly, if you have such a goal from carefully observing your surroundings, and you keep missing it, there is something wrong. Say, if your goal is two papers in a year, and you have none in total by the end of your second PhD year (of three years total, isn't it?), you might be in trouble.


I did not manage to see the answer by Thomas King before I posted mine. This is basically the same idea as his.

5

The concept of "doing enough work" has to be measured against some standard. Are you aiming to win the Turing Prize? Then you probably aren't doing enough work. Are you doing enough for your advisor to say that you are? Then by your own words you are.

So realistically, you have to determine at what standard to compare yourself to. There are two broad categories in which to measure: how much you need to do to advance professionally at some level (e.g. obtain a post-doc, get an assistant professorship, etc.), and how much you need to do in order to satisfy your self-esteem.

The first is primarily concerned with the materiality of living. Find a job with health benefits and be able to pay a mortgage. Whether or not you are focused on a tenure track job, you should really have a backup plan semi-worked out for if you leave academia. This can take some pressure off. For a job in academia, get an idea of what recent PhDs did to graduate and linearly interpolate for where you are at.

The second is trickier. Steeped in academia is the currency of prestige and recognition. One of its appeals is to be a revered expert in your field. Trading complements like "brilliant" or "genius" are something peculiar to academia. I would suggest you evaluate how important recognition is to you and how much of your anxiety on the amount of work you do is tied to envy of the success of others.

Again, to emphasize, both of these are value judgments that only you can make. Personally, I found I was too taken in by the prestige culture in academia which contributed to my health issues that required me to leave.

Finally, and this is something I cannot contribute to that much, is the question of efficiency. If you could be doing more work without changing much more than your approach, then in that sense you aren't doing enough. I.e., if you spend too much time on irrelevant details.

Most of the answers here have addressed this last issue. As such, my interest was primarily in calling to your attention my first two points.

3

My take:

Try to get papers done, get LPUs, etc. People will tell you it doesn't matter, but it does. Some people will say that the only thing that matters is some huge discovery but this is a very high risk/reward calculation. And it makes more sense for professors than for students. There are even professors who knowingly burn multiple students on very risky problems since they have tenure and going after the big prize makes more sense for them. Not in any way saying this is common or is your case...but you always need to watch out for yourself. It won't HURT anyone if you just churn out some mediocre product, especially at the beginning. It's better than getting skunked or spinning wheels.

I would also say that it is much better psychologically to be getting some sort of measurable accomplishment. Ph.D. can be pretty long and discouraging. Create ways to get some intermediate accomplishments. You also learn something about writing papers and doing research.

I have even used it as motivation when I was sick of something in the lab. Only had about 80% of the experiments done...so I just wrote the whole thing up as a paper (even including conclusions of what I expected to see). Once I had done all that, it was very motivational to just get the last 20% of the data done. I think I even had to change some of the conclusions, but so what. It was easy. That's what word processing is good for...easier revisions than back in the days of typewriters.

In case this sounds sleazy...it's really not. There is a good section in E Bright Wilson's Dover book (look it up) where he talks about the importance of writing up research. Or maybe it is Katzoff Clarity in Technical Reporting (look that up also...and read it...and do it). In any case, the basic point is that most research is government funded. And if you don't write it up, you are not just screwing yourself, you are hurting others (and the whole enterprise of getting more science done faster) because others have to do your work instead. This is even the case where funding is cut and only partial reports done...yes, you should try to aim for perfect work...but if you have to cut a project, write up WHAT YOU HAVE. There is always a journal for everything...and you would be surprised how WELL stuff gets accepted if you are just honest about things versus trying to act all cool or to gloss over gaps.

Now, some of this will vary from field to field. I did solid state chemistry (shake and bake). Which is like a dirty little secret for how easy it is to do the experiments AND to produce at least mediocre publications. But in any field, including SSC, there are ways to do LPUs and to publish a lot. And I find it will even motivate your work (even in theory) to push for pubs.

So...I totally feel ya on the loneliness and the miserableness of it. It really is not a great system of apprenticeship. All that said, you are the captain of your fate and if you push, you can get good things. There is a lot of freedom there too.

Good luck.

Get some publications!

3

You are doing this backwards. Instead of worrying about whether you are doing enough in a semester, manage your degree like a project. Set dates on long term goals, then put dates on intermediate goals. This is the only way to tell if you're falling behind

2

One way is to compare with other students, collect the set of students that have: 1) recently completed their PhD, 2) in your field, and 3) in yours or a similar department (at least within the same country). Take a few median PhDs in terms of output as your example and see how many papers they are publishing per year and at what venues. That will tell you what you have to aim for. For the shorter-term objectives (reading papers, proving theorems), you will get a better understanding after your first publication.

There are some shortcomings of this approach. Fields such as mathematics where only two papers published is the norm for a PhD (or fewer!) makes it hard to gauge how much work you need to do to move forward until it is potentially too late. But, apart from that I do think you should compare with others. It's a competition if you want to continue in academia, and even if not, your PhD will be obtained by meeting the expected standard.

Apart from this attempt at quantification, much of progress is qualitative and so you will have to talk to your adviser in terms of quality of work, etc.

1

Question: How do I know I have done enough work in one semester? I mean is there any parameter to measure the work that I have done in the last semester? I have heard some students read only one research paper per one semester.

First, if you need to know something, you have to read papers until you find the information. Consequently, the amount fluctuates.

Second, the amount of papers you read for fun or interest, might be not relevant for your work or subject. This is why, it has no influence on your performance.

Third, the comparison with your co-mates has no influence on your own work.

Fourth, it is impossible to compare real productivity. Productivity, is dependent on so many variables which you can not access, because you are not the other person. Maybe (but unlikely) it looks like someone is not productive for ten years and then he suddenly makes a big publication and wins a prize.

Fifth, thinking about this question, lowers your work related productivity.

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.