Recently, I was doing a project in computer science (involving modes of computation) that had a very specific time limit. Unfortunately, due to software problems, I was forced to obtain my data by hand. I went to substantial lengths to ensure the correctness of this data, but am worried that my credibility and the credibility of the data will be significantly decreased. What should I do to maintain credibility?

  • 1
    Can you elaborate on what exactly was done to "obtain my data by hand"? Timing experiment running times with a stopwatch? Subjective evaluation of output pictures by whether they looked right? Observation of user behavior in a paper notebook? Something else? Details may matter here. – Peteris Feb 22 '19 at 23:31
  • @Peteris Nothing quite as abhorrent as that, no. The data was obtained by following a model of the proposed system; the main difference was that it was being done by hand rather than by an automated program. – user124 Feb 23 '19 at 0:03

There is probably little to do but be honest about how the data was obtained. If it holds up under review then you have no issues. The past is the past.

However, if you have a chance you can try to obtain a better data set under more controlled conditions. This might be the core of a follow-up publication, either validating or invalidating the earlier result.

You learn something from just about every study. But if you are honest, it is hard to complain.

| improve this answer | |
  • Unfortunately, it will be several months until I would be able to get a better data set (and even then, I’d have to add something new). I will be sure to be honest about it, though. – user124 Feb 22 '19 at 23:21

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.