34

I graduated 3 years ago and defendeded my master's thesis without any problem. Since recently, I am planning to go back to academia, so my former advisor and I are working together to publish my master's project.

While doing so, I've recognized coding mistakes that affect results statistics. Thankfully, it does not change the whole implication and main findings but it changes some statistics numbers and minor results (which are not essential to the research hypothesis).

I don't know what to do. It was a mistake that I did not recognize at that time. I cried over and over for two days, and could not sleep soundly due to nightmares. I am panicking.

I've heard that I cannot make errata for my master thesis. (I don't know the reason why) What should I do?

9
  • 43
    People do mistakes. that's a general truth....And I am not talking specifically about your issue....
    – mike
    Commented Jul 23, 2023 at 14:07
  • 6
    Good to find errors, which lead to improve
    – Mike Liu
    Commented Jul 24, 2023 at 2:09
  • 11
    You've learned a valuable lesson that sometimes a bit of space from a project gives you fresh eyes upon your return 🙂. Rest well tonight knowing that you'll apply that lesson to your ongoing studies, and if it means anything to hear it from a random internet stranger: you're ok! Commented Jul 24, 2023 at 7:06
  • 3
    Smile and wave
    – polfosol
    Commented Jul 24, 2023 at 12:17
  • 8
    With all due respect, unless your masters thesis is something far beyond your average masters thesis... Nobody is ever going to read it. If you're really bothered, there's nothing stopping you putting up an errata section on your personal website and listing it under that. Commented Jul 25, 2023 at 10:10

5 Answers 5

81

This is not unusual. It is good you detected it and now know more about the issue than you did before. The university has examined the thesis and decided you made sufficient achievements to receive your award. There is no need to go back and make changes to it.

If you have a personal research web page, or some other similar online media you could write blog about it that contains the update and errata. This way when people look at future publications they can read the errata, or you can cite it whenever you cite the thesis work.

The detection and documentation of the improvement shows your academic and research diligence and rigour.

How common is it: When I made my thesis only typewriters were available, so I decided to digitise it using OCR. I was shocked at the number of typos, grammar and spelling faults that were made. It was shameful! How we have progressed with our modern tooling!

5
  • 33
    In other words, the degree is a statement that OP is capable of conducting research, not a statement that whatever results the OP demonstrated is true Commented Jul 24, 2023 at 1:46
  • 2
    @AgnishomChattopadhyay, pretty much how it is with masters degree mills these days. I didn't think my mine was given that much a a run through as it was too much work for the examiners (I had extra time on my hands for reasons and really went to town on the rigor) Commented Jul 24, 2023 at 4:31
  • 9
    @AgnishomChattopadhyay I second that statement, as that's what theses are about, at least from what I was told. My own master's thesis was a good example, I set up a statistical model in order find a result - not only could I not find a result, but I realized whole model was flawed and effectively unusable in that state. Since I became aware of the issues quite late, I extended the discussion of my own approach and critically reflected all shortcomings, potential improvements, the whys etc. I ended up having a superb grade even though the result was "this approach is crap, please never use it"
    – kopaka
    Commented Jul 24, 2023 at 9:36
  • 4
    @kopaka That result is a good result... And if the academic world didn't have such a bias against that boring result. (which to be fair is the most often result of scientific study. 301 different ways not to make a light bulb...) There world be a lot less papers with massaged data published.
    – Questor
    Commented Jul 24, 2023 at 20:52
  • 2
    @AgnishomChattopadhyay It seems OP's results were not 100% correct, but close enough to be useful.
    – gnasher729
    Commented Jul 25, 2023 at 16:59
28

The first thing you need to do is calm down. Deep breaths, go for a walk and appreciate nature.

Correct the errors, discuss them with your supervisor and make sure the published article based on the project is correct (perhaps add a note to the published article pointing out the mistakes in the earlier version in the thesis and explaining how they have now been fixed).

No-one will read your MSc thesis unless you have a personal website and you put it on there.

1
  • 5
    +1 for the walk and nature appreciation. I stand four square behind that counsel.
    – Trunk
    Commented Jul 24, 2023 at 16:10
14

Congratulations on finding and fixing your errors.

Publish the correct numbers in the paper you are working on.

If you and your adviser agree, you can note that this paper is based on work you did in your Master's thesis, with updates and corrections.

In either case no one will ever look at your Master's thesis.

11

If you cannot create an erratum, this means that your master thesis is not properly published - or is only "lightly" published. This means that you are really only worried about having obtained your thesis by mistake. This is however not the case. Your evaluators will have looked at your capability to write a thesis and found you not wanting.

You talk to your former advisor after correcting the errors and after going through your code once again. Since your mistake(s) appear(s) not to be material to the outcome of the research, your plans for publication can go ahead.

You are certainly not the first researcher who published (even in your case, you did not really publish) results based on errors and you were lucky in that it did not affect the outcome of your research. Especially in CS, errors in published work are a thorny issue because so much is published in conferences that do not allow errata either.

Other than learning about your capability of making mistakes despite trying your best and being open to your former advisor and now collaborator, there is nothing you can do.

1
  • It sounds like the OP almost certainly has the right to publish their corrected thesis somewhere so there is something they can do. Just update the thesis and make it available (for example on arxiv or a personal website).
    – Kvothe
    Commented Jul 26, 2023 at 17:10
2

Master and PhD theses are meant to be your first contributions to the research community and everybody knows that these contributions are primary aiming for training young researchers. Nobody will judge you for mistakes like the ones described in the question.

Steps ahead:

  1. Talk openly to your advisor about your bugs and how they affect the results.
  2. As the publication is based on your thesis, you have to cite your thesis (otherwise it might be close to self-plaritism). The citation could be a sentence like "this paper is based on the first author's master thesis [Tube2020] but covers additional aspect, more polished results, and refined analysis." The last statement can be more explicit ("fixed minor issues in the analysis").
  3. Enjoy your master thesis and your published paper!
2
  • 1
    can "minor issues" be more specific? fixed minor calculation errors? Commented Jul 24, 2023 at 10:27
  • 1
    @user253751 Sure, depends on what the mistake was and how it affected the results. I can image some mistakes are so severe that warning other peers justifies an article of its own.
    – usr1234567
    Commented Jul 24, 2023 at 11:16

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