46

As a PhD student, I am expected to do research on my own, possibly assisted by my supervisor or colleagues. Like any researcher, I sometimes spend days looking for a small piece of information, sometimes just a single sentence, that answers a question whose lack of an answer was preventing me from moving forward. Sometimes, it was enough to stumble upon the right article after several days to get the answer directly, so it didn't require any real research work, except for browsing the internet.

What is absolutely wonderful about Stack Exchange is the ability to simply ask such a question and get the answer, without doing anything. Without doing anything ... That's not really true. It's not even true at all! Asking a question on a specific subject requires quite a lot of work, which in my opinion can be summarised in a few steps:

  1. Research to see if the question has a direct answer to be found (a few hours of research via various articles, or even a few days)
  2. Establish a developed context so that anyone with basic knowledge can access the question
  3. Formulate the question in the most understandable and educational way possible (this step is not so obvious: putting words to a question is not easy and requires serious knowledge of the subject!)
  4. Pay attention to the comments and answers to either edit the question and improve it, or to judge whether the feedback offers us a real answer to our question or not. In the latter case, it is then necessary to state how the answer does not answer the question.
  5. Once a tangible answer has been received, do some research to verify it and then accept it as the official answer to our question.

All this work is, in my opinion, absolutely necessary, both out of respect for the people who are going to read the question and try to answer it and for oneself in order to obtain the much-desired answer to our question. But despite the work that I believe a serious question should represent, I can't avoid feeling uncomfortable when I ask a question on Stack Exchange and the answer unblocks my research.

Is it ethical to use Stack Exchange for research, following these steps carefully?

I would like to make it clear that I have no intention of making the people who answer my questions work for me. When I ask a question on the site, my only goal is to get an answer that helps me get unstuck, not to expect someone to do a scientific research job for me. I see Stack Exchange more like an interactive service that tells you what has already been done.

Peter Mortensen
  • 836
  • 7
  • 7
Baloo
  • 585
  • 1
  • 4
  • 7
  • 70
    Part of research was always to walk around and talk with folks about where you were stuck (and talk to them about where they were stuck). Just nowadays you can 'talk' with many more people at once, and increase the likelihood of getting an on-target answer to a good on-target question. – Jon Custer Feb 22 '22 at 15:06
  • 8
    You might be interested in this Q&A: I have stack-exchanged through my undergrad math program. Am I likely to succeed in mathematics PhD programs? – Allure Feb 23 '22 at 02:24
  • 1
    (Ab)using SE for you own purposes is what it's for. "I am expected to do research on my own [and not pay someone to do it so that the piece of paper I'm paying them for might actually mean something]." By the time you graduate that paper had better at least mean you know how to use the internet.... That it's SE doesn't matter. Did they say you can't use the internet, like how they warned me that I might not have a calculator in my pocket 24h a day? What if the internet is down? Then life as we know it is over and that paper will be good for starting fires. – Mazura Feb 23 '22 at 21:13
  • 1
    Follow up question: "Hi guys, here's my dissertation problem. Please write out several chapters including history of the field and suggest several problems for future research. And double space it. ASAP, please!" – B. Goddard Feb 23 '22 at 22:08
  • 6
    I always see questions like this and suspect they are just reputation farming. Is there any controversy on this matter whatsoever? Does anybody say using SE during your PhD is unethical? If such a person exists, would they be found on this website? – ZachMcDargh Feb 24 '22 at 18:34
  • @ZachMcDargh I don't care about reputation farming, I care about ethical. Nobody said it was unethical, I just asked a question. – Baloo Feb 24 '22 at 19:22
  • "Is it ethical" on this site always goes straight to, is it plagiarism, which leads to, is it properly attributed, the answer to which gives you the answer to the first question. – Mazura Feb 24 '22 at 20:38
  • 1
    https://academia.stackexchange.com/questions/182664/is-it-ethical-to-get-some-help-from-stack-exchange-for-my-research-as-a-phd-stud –  Feb 25 '22 at 14:47
  • Here's a question I'd like to throw back at you: Would you use information found on StackExchange? And if you would, does it really matter who asked the initial question? – AVee Feb 25 '22 at 16:34

3 Answers3

85

Yes, it is ethical to seek, accept, and use help in a dissertation. The only difficulty in accepting help from others, and especially on sites like this, is giving acknowledgement to those who provide critical help (insight) into your work, since many of us are anonymous here.

But students have, since time immemorial, sat around a table (mead, tea, coffee, ...) and chatted about issues in research.

The general issue is one of plagiarism, if you attribute something to yourself that came from another. In mathematics a conversation of a few minutes can be enough to earn an authorship position. The same is likely true in other fields.

But "research" doesn't need to be just archaic texts retrieved from dusty library shelves or the 75th page of a Google search.

Of course your personal contributions to a dissertation need to be sufficient to convince advisors and reviewers of your advance of the art and science of your field.

And, asking the right question can be an important element in research.

Buffy
  • 363,966
  • 84
  • 956
  • 1,406
  • 26
    I would consider it acceptable to cite or acknowledge a anonymous or pseudonymous source (such as an answer on a StackExchange site) by the identifying information presented - so it would be okay to acknowledge "The answer given by user 'Buffy' on the StackExchange question Blah Blah Blah helped by ..." – Alexander Woo Feb 22 '22 at 15:19
  • 13
    Though it is better (when available) to click the "cite" button; the links that the result includes are useful if user 'Buffy' changes usernames, and the question is renamed from Blah Blah Blah to Yada Yada Yada. –  Feb 22 '22 at 15:25
  • 11
    I found in the old days that I rarely got the answer from chatting, but instead got nudged on to a different path or made to realize some alternative approach. Sometimes that was because I had to explain the problem to someone else, and that act of explaining it (rubber ducking) clarified my issues. I still think talking is better than posting a question on SE (which I generally don't do for technical stuff anyways). – Jon Custer Feb 22 '22 at 15:27
  • 3
    @AlexanderWoo Don't you think it might create a lack of credibility to quote a question asked on stackexchange? It doesn't seem very professional at first sight. Unless you are saying this with regard to the acknowledgement section? – Baloo Feb 22 '22 at 15:27
  • 9
    @tomweelen, as in many other situations, a quick explanation that suggests verifiable truths, or "more official" sources, can be very, very useful. Some writing-style purists are opposed to mentioning "intermediate" sources that were actually necessary to find the critical resource, but I think that is a very unfortunate and dishonest pretense! It's ethical to use MSE, and it is important to acknowledge (not just vaguely, but specifically) how it was useful... so other people know how you succeeded! :) – paul garrett Feb 22 '22 at 22:55
  • 7
    @tomweelen I have seen a number of papers refer to (cite) a mathoverflow post. It's not much different from citing a webpage, which is often done for data, computations, etc. It's certainly not quite the same as citing a peer-reviewed journal paper, but it's not particularly unprofessional. – Kimball Feb 23 '22 at 01:41
  • Ok I see, thanks for your opinion! – Baloo Feb 23 '22 at 08:24
  • 2
    I was acknowledged by a paper as "Dikran Marsupial", and was perfectly happy with that. IIRC the authors did ask for my real name (which isn't a secret) and I gave it, but was happy for my pseudonym to be used instead. I did once consider publishing a paper under that name, as it wasn't on my area of genuine expertise and I didn't want my academic credentials to imply expertise that I only had by proxy. For me the primary issue is not to claim credit for other peoples work, which doesn't require the real name be given. – Dikran Marsupial Feb 23 '22 at 08:45
  • 3
    @AlexanderWoo: Even better, save the question (and its answers) in the Wayback Machine and include a link to the archived version of the question as well as a direct one to the question itself. – Vikki Feb 23 '22 at 08:48
  • 2
    @tomweelen there have been journal papers (in good journals) that stemmed directly from discussions on stack exchange sites (such as the one I mentioned above). – Dikran Marsupial Feb 23 '22 at 08:48
  • 1
    There's a big difference between citing and acknowledging. You acknowledge people who were helpful without necessarily saying how they helped (the help might have been purely personal); you cite documents that you are using to support the argument presented in your thesis. A casual internet post will often be helpful, but it will rarely by citable, because it's unlikely to be authoritative enough. – Michael Kay Feb 24 '22 at 16:46
  • 1
    "Of course your personal contributions to a dissertation need to be sufficient to convince advisors and reviewers of your advance of the art and science of your field." This is a very crucial point. It's unethical to hide from your advisor which contributions are coming from SE, and if you SE your way through too much of your project your advisor may have legitimate concerns about whether you've satisfied the requirements of a PhD and/or whether they can write you a strong recommendation. So it's perfectly fine to ask SE questions occasionally, but it can't be your only approach! – Noah Snyder Feb 24 '22 at 22:47
  • The other issue, of course, is that since SE is not any kind of academic or other authoritative source, everything you get from it should be confirmed, either through your own research and/or actual authoritative sources. – RBarryYoung Feb 25 '22 at 17:45
3

I would say, yea it’s ethical with acknowledgment. When I was in grad school, we had a large, rolling dry erase board in our “lab” (cubicle area) that I would write equations on about my dissertation subject and sometimes a better mathematician/engineer than me would wander over and ask what was going on. So, I’d have to erase the board and start over and explain. Usually, about halfway though, the lightbulb in my head would turn on and I’d have the solution. I’d continue the explanation until I got past the part where I’d been stuck and applied my new idea. If the watcher agreed, I’d keep going until we got to an end. Then I could go off and start implementing this in code. Sometimes we’d find an error and I’d have to backup and start some part over. The positive moments often led to acknowledgements or coauthorships in a resulting paper. Others in the lab would do the same.

You can think of SE like that whiteboard, but maybe more critical. You can put things here (in the form of a question) and see if it’s a reasonable question and maybe get some guidance on it.

Bill Barth
  • 48,733
  • 6
  • 112
  • 194
2

I think the more general issue is citing a source as supporting evidence when it hasn’t gone through peer review. While the stack overflow process suggests a type of peer review it is not required and referees may have no formal affiliation. So, YMMV.

HEITZ
  • 10,070
  • 3
  • 24
  • 41
  • 14
    Well, yeah, unverified/unverifiable remarks by anonymous people on the internet are not good sources. :) But remarks on Math Stack Exchange and other sites often give very solid citations, in addition to "showing how to do it", so that the thing is reproducible/verifiable. – paul garrett Feb 22 '22 at 23:58
  • 2
    Is "cutting" supposed to be "citing"? The question says "I would like to make it clear that I have no intention of making the people who answer my questions work for me... I see StackExchange more like an interactive service that tells you what has already been done." So they aren't asking for original research, they're asking for references to existing research. – Acccumulation Feb 23 '22 at 06:42
  • 10
    Citation is not to just provide supporting evidence, but to acknowledge that the idea came from someone else and that the author is not claiming credit for it. So if it was the only source, it would have to be cited anyway. "personal communication" (i.e. "they told me") citations have been in the literature for a very long time and they obviously are not peer reviewed. – Dikran Marsupial Feb 23 '22 at 10:09
  • 8
    The idea that a source is reliable if it is peer reviewed is fundamentally wrong headed IMHO, it is a very low hurdle and I would venture that the majority of peer reviewed papers are wrong to a non-negligible degree. You have to read the source to know if the citation is reliable. The citation is there to allow you to investigate the reliability of the source if you wish to do so, it is not itself an indication of reliability (other than that the author presumably thought it was reliable) – Dikran Marsupial Feb 23 '22 at 10:11