47

I did my thesis using corporate data (I used to work there, they don't give their information freely or sell it), data is from 2009 to 2018.

I collect the data, analyze it, everything normal. However, after my thesis was approved and submitted, I saw that the company has modified some of their data from 2009 to 2018, historical data from the past was changed (they have valid reasons) but I am worried that if someone tries to verify the source of my information, they will find different data and think that I commit fraud or data manipulation.

My supervisor and many people told me that nothing wrong will happen because my research was in a specified period of time when data was presented like that, they also told me that after a thesis approval, no one verifies data source. They also told me that the verification of data source is done before thesis approval.

However I am still worried, because it is not common for a company to modify data from the past.

My thesis wont be published because the used data is privated, so they are allowing me to not publish it.

Any ideas of what should I do?

  • 108
    Nothing. Adviser is correct. No one is going to check your data. It might be interesting to see if your results match up on the new data set, but no one is going to care. In the absolute worst case that someone does care you can simply point out that they revised the data. – JoshuaZ Jan 27 at 21:38
  • 23
    Go find another project to worry about - that one is dead and sorted... – Solar Mike Jan 27 at 21:50
  • 50
    You can always attach a short note saying what you just wrote here, namely the analysis was done on a data set that latter was corrected by the company, but it has not implication on your conclusions. This kind of anomalies are only problem when a researcher changes the data on purpose. – Greg Jan 28 at 4:21
  • 14
    Dont trust random people on the internet over your adivsor... Especially in this case, when he is correct ;) – Lot Jan 28 at 12:39
  • 5
    The fact that you are still concerned about this after your supervisor’s and our assurances makes me concerned for your anxiety levels. If you are having other symptoms of anxiety I would encourage you to seek out mental health resources. – Dawn Jan 28 at 17:32
93

Revoking a degree is rarely done and, then, only for the most extreme reasons such as explicit dishonesty and such. Any results in any thesis are subject to revision as new information becomes available that was not present in the past. It doesn't mean that the work was wrongly done, but only that what is known has advanced.

Since you re-did the experiment and came to the same conclusion, you may be able to publish something based on the new results and, when citing your unpublished theses, mention that the conclusions there were verified with new data.

19

Rest assured that few people read theses (PhD or otherwise), and as Buffy says, degrees never get revoked for outdated data sources so your degree is safe.

However, if you are concerned about people reading your thesis and not being able to reproduce your results - publish an updated version (say, on ArXiv if you're in a rush, or in a journal/conference if you want the paper to be peer-reviewed), explicitly referencing the thesis and emphasizing the fact that your results are on updated data. This is actually good practice that would help future researchers who may care about your work, and save them the trouble of trying to recover your result.

  • 1
    update: my thesis was only to get a engineer degree, it is not a PHD or master. My thesis wont be published. Any advice now? – Carlos Varas Tello Jan 28 at 15:25
  • CarlosVarasTello: BE/BSEE? You should mention that in your question. Personally I have never heard of someone trying to reproduce a BE/BSEE thesis. Anyway, don't worry. – smci Jan 29 at 8:32
  • It's an unpublished thesis for my engineer degree. – Carlos Varas Tello Jan 30 at 16:08
11

I did my thesis using corporate data

Remember, though, that your research was not the data, but its analysis and conclusions you drew. And since you "did the experiments again and the results match up" - then the thesis is perfectly valid not just at its time of publication, but now as well.

but I am worried that if someone tries to verify the source of my information, they will find different data and think that I commit fraud or data manipulation.

This should not be an issue if your citations were accurate. You should have included the physical document, or the on-line resource, you obtained the data from, and indicated a date of publication (at least a month of publication anyway). If the company now presents different data - it published this data at a later date. edit: If the company does not indicate when publication happened, or if the data is not available from a single source, then you should have included a paragraph (or a small section) in your the thesis explaining how you obtained/collected your data - and when. Specifically, dates of downloading data from internet URLs.

Like others suggest, an addendum/errata to your thesis and/or any paper you've published based on the data, mentioning the change, is a good idea. If the changes had affected your conclusions, then it would have been very important (IMHO).

My supervisor and many people told me ... However I am still worried,

You'll have far worse things to worry about in life - so don't worry about this minor issue :-)

because it is not common for a company to modify data from the past.

How do you know that? I'm not sure that's true. Also, it's very common for companies to not archive past data, and in that case as well, a researcher's work may not be thoroughly-verified.

My thesis wont be published because the used data is privated, so they are allowing me to not publish it.

Definitely publish it! Academic work should, and I might venture to say must, be shared and made public. The least you could do is censor out the private data and keep all the other parts of your work.

Any ideas of what should I do?

Take a deep breath and relax!

  • I included a screenshot of how the data was presented during my research, I compared to the current data and it shows and obviously update. Is that valid? – Carlos Varas Tello Jan 28 at 18:46
  • I would say the most important thing is an explanation about when, how and why the data changed; and the claim that you have verified that the same analysis and conclusions apply; beyond that it's all a courtesy to the reader/viewer. – einpoklum Jan 28 at 19:01
  • there wont be any reader/viewer. Thesis is not going to be published. – Carlos Varas Tello Jan 28 at 19:03
  • @CarlosVarasTello: Oh, you mean in your thesis... well, in that case, I'm not sure I would add something like a screenshot - that begs for "where is the data this shot is based on?" - sometimes less information is better. But whatever your advisor thinks would work, should work. – einpoklum Jan 28 at 19:05
  • the screenshot was taken from the company system, I included it in order to show that my source is valid and I am not fabricating anything. – Carlos Varas Tello Jan 28 at 19:12
5

In addition to @Buffy's answer:

  • Revoking a degree is such a serious matter that the revoking committee would have to prove positively that you did falsify data. I.e., in your case, they'd need to show that the data were not as you said it was at the time of your thesis writing.
    Which, obviously, they can not.

  • Nevertheless, it is a good idea (take home message for future) to keep a copy of your raw data if at all feasible*. If you are not the owner of the data, you can

    • ask the owner for permission to keep a copy for the explicit purpose of being able to show original data for your thesis/paper should anything be questioned.
    • if they don't agree, ask them to keep a copy available in case such a request comes in.

    IMHO, such points should ideally be part of the thesis contract between you, company and university. But from the science point of view, while of course public open data is nicer, granting access to the data only after an NDA is signed does satisfy requirements to be able to answer questions about your work and possible further questions about the original data.

  • Also, I think @Greg's comment about a note about the correction occuring afterwards is a good idea where practicable: you won't change already deposited copies at the library - but if you have your thesis online, I'd note it there. If you write a publication now, I'd probably use the corrected data and say that thesis [citation] was done on an earlier version of the data set discussed in the paper.


* Not even all scientific institutes allow that when you leave them. I have a former employer (research institute) that insists that leaving employees do not take any data with them without written permission (which is not given feely). They do promise to take care of archiving the original data and paper lab books, though.

  • 1
    Another case where it might be difficult to keep a copy of the data is when there are privacy concerns. – Anyon Jan 28 at 15:04
  • @Anyon: good point. The paperwork that is necessary to the allow the student/researcher to work with the data at all would be a good point where also the decision who archives and how could be included. Approaching data management plans, are we? (I'm trying to generate/write down here advise for future me ...) – cbeleites Jan 28 at 15:12
  • update: my thesis was only to get a engineer degree, it is not a PHD or master. My thesis wont be published. Any advice now? – Carlos Varas Tello Jan 28 at 15:25
  • @CarlosVarasTello: still all apply kind-of. However, I'd not make too much hassle: ask them politely (email leaves an almost-paper trail ;-) ) whether you should keep a copy in order to have your thesis archive complete, or whether they prefer to archive the original data. If they say no to both, and anyone ever asks, show the email exchange: that shows that you really did all you could legally do - if more records should have been kept, that's not your fault. As for (3), put a yellow sticker with the note into your copy (in case you hand it to someone at some point). Enjoy your degree! – cbeleites Jan 28 at 15:34
  • 1
    Re: when an institution can or can't revoke a degree, yes it is a serious matter, but the rules of what an academic institution needs to do in order to revoke a degree are going to vary. – De Novo Jan 28 at 21:26
1

I think you're worrying too much. Add a footnote under the data and/or somewhere else saying "Data as supplied at time of writing" (or as of XX/YY/ZZ) and leave it at that.

  • Thesis has been already submitted, no changes are allowed – Carlos Varas Tello Jan 28 at 18:53
0

If it'll help, imagine that you'd graduated 20 years ago doing a thesis on, say, IBM's sales data. IBM sees there was a problem with their historical data in archives, and publishes a corrected set of sales data. Do you think the university would revoke your degree from 20 years ago?

No. Of course not. Because when you wrote your thesis, you were using the information you had at the time, that by all appearances was correct. And the degree was awarded because your thesis came to a justifiable conclusion with the dataset you worked with. The data may have changed 20 years later, but that has no bearing on the work/skill/etc that was put into the thesis.

  • And how university verifies that the data has been corrected instead of not being manipulated by me? – Carlos Varas Tello Jan 30 at 19:34
0

Carlos, I am going to ask you a simple question, because it is very confusing that you continue asking for something, for 'advice now'.

At the same time, I've thought that an amount of misunderstanding here has been because of different native languages. You are writing in English, but I think you are from Peru.

So, is the problem you ask about actually that your thesis 'won't be published', as you've said several times? In other words, are you looking for a way it can be published?

A native speaker would understand [won't] to mean absolutely not. If you said [can't], then the understanding would be that if a barrier were removed, it could be published.

If you wanted advice to remove a barrier, then you would say 'can't be published because of [the barrier], so please suggest how I could remove [the barrier]'.

If what I describe is not your question, then another possibility is that you are worried that someone will attack your thesis, if they don't like something you might say or do because of what you learned in writing it, the analysis. Thus you worry the attack could have some consequence on your accomplishment, your degree.

This is where the note I suggested could give you position of better confidence, where your professors write as you indicate they say, that your thesis is valid, on data you used that was valid at its time, and sign it.

I called the note a 'certificate', only because it verifies truth, not because it is any special printed kind of paper from any government. It's just truth anyone would understand.

I hope that answers your question about what I meant in my answer above.

As far as what would adequately protect you if there is a challenge, you must surely use your judgement there. In any culture, there are things which can be risky in some way to do. Then we decide, and certainly may decide not to do them, if the risk is too high.

In that case, we move on in our life, and take up another task which is more appropriate to do, no?

I wish you well, Carlos, and hope you come soon to a path that will relax you about this -- whether we answering here understand or not isn't the first importance.

  • yeah you're correct. In my country Peru nobody cares about thesis after they are submitted (except for plagiarism). Also, it is logic that the company can do whatever they want with their data, I must feel happy that God give me the degree. I think that I need to do more exercise, hang out, visit family, etc. – Carlos Varas Tello Feb 2 at 16:16
  • great, that soiunds very good. When we work very hard to accomplish something, then it's not unusual that our vision gets narrowed in. Going back out in the world of family, friends, nature, and all of spirit fixes that -- and we learn another thing ;) You enjoy, Carlos, and others here smile also – narration_sd Feb 2 at 20:47

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.