1

Imagine, you did data evaluation, published. When preparing your data for the thesis, you've realized that your approach was wrong. It does not change the main course of the paper and do not damage main conclusions, although it does affect the quantitative part of research. It is more than 1 year back published, so instead of erratum you decided that you put your updated data in your thesis and will do your best to address your mistakes in the next publication.

Question is how to formulate decent answers during the PhD defense if the committee asks why didn't you publish erratum/correction if you've made a mistake in your data evaluation?

Could I answer in a way "I've realized I've made a mistake and it will not happen again. The correction will not change main conclusions of the publication. I will make sure to address these points in the next publication concerning this topic to avoid any misunderstanding"

Maybe some other ideas?

Thanks in advance!

  • I'm not clear how it can be meaningful and at the same time not matter to the result. If you did not prove the result then you need to do so before your thesis will be accepted. If you have done so within the thesis, but not published it yet (as you seem to be saying), you may or may not be allowed to graduate depending on how demanding your department is regarding publication requirements for graduation. – A Simple Algorithm Aug 25 at 17:44
3

If I were on your committee, the key take-away I would look for is that the mistake will be corrected in the literature. Either an erratum or an explicit correction in a future publication would satisfy me, though in the latter case I would ask, "What additional steps need to be completed before this future publication will be available, and how long do you anticipate that may take?"

If the future publication is still quite far away (by the standards of your field), then I would recommend an erratum in the meantime, which should do no harm. I would not consider a long period of time since publication a very strong argument against providing an erratum. Here is one famous paper in my field that received an erratum after 3.5 years.

A promise that no mistakes will happen in the future would strike me as naive and unrealistic.

1

There is a world of difference between publishing correct data with correct results that are in an incorrect direction compared to data and results that are incorrect.

The former may well save someone else lots of time when choosing the direction they need to go, in fact, it may well be the correct direction for them.

So, in your defense, it's not a "mistake" , just a "blind alley" which meant you had to change direction to the method you subsequently used. All the people at your defense have had the same...

  • From the question I got the impression that the latter case had happened: The mistake (as LongWay describes it) affects the quantitative data/results without changing the general direction ("the main course") of the research. Of course, all the people at the defense probably will have had this happen to them as well. – David Schaich Aug 25 at 11:33

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.