-1

We performed a very complicated measurement in the high-energy physics field and now as a PhD, I am doing the data analysis. The setup has two detectors, one of which broke. Therefore, we only have information from one detector. The thing is, both detectors are crucial to reach the final goal of the measurement. Instead, we have only one working but we know why the other one accidentally died during the experiment. So, I suppose this is neither wrong methodology nor it is a null hypothesis. As much as I hate the term, it is a failed experiment due to mistakes that could've been avoided.

The question is, how would one extract something publishable from such data? Would it be enough to state what I did and why it failed?

Improve this question
2
  • I think this is too specific for this site; consult with your research advisors/mentors/collaborators for how you might salvage these particular data, if it's possible at all. Stating what you did and why it failed is often perfectly reasonable for a course project for a student to be graded on, but rarely publishable unless the failure is very spectacular or interesting to the field. That's a bit unfortunate, I wish errors were more appreciated in the scientific literature.
    – Bryan Krause ♦
    Jul 20 at 14:25
  • 1
    When deciding whether something is publishable, try to answer the question "what does my article add to the knowledge of the world". The "finding" that a broken detector does not work does not add that much. So you need to look at whether there is something interesting in your data. For that I would talk to your advisors/mentors/collaborators, like @BryanKrause suggested.
    – Maarten Buis
    Jul 20 at 15:14

3 Answers 3

Reset to default
3

Think about why we publish to begin with: Because we want to add to the existing knowledge, not because the authors need it for their careers. As a consequence, what makes something publishable is if it provides for something that is new in the sense that it is either an observation no others have made so far, or that it adds evidence to something others have published but for which the evidence was not yet convincing (think: the precise mass of the Higgs boson). A paper that is solely written "because we made an experiment and we want to write a publication on it" is neither interesting to readers, nor does it add anything worthwhile to the body of literature.

Which way things lean for your specific case I don't know. It is possible that you can scrape enough evidence from your one detector to show something that was not yet known, even if the evidence is maybe not overwhelming because you don't have the second detector. In that case, it may be possible to publish the results. Or, it may be that there really isn't anything others haven't already observed. In that case, you're out of luck.

Improve this answer
2

Since both detectors are "crucial" you haven't really got anything and the experiment needs to be repeated (woe).

What you have is an anecdotal story about instrument failure and the necessity of carrying out experiments according to a pre-defined plan.

Perhaps you have enough data from the one instrument to be able to say that the data available doesn't seem to contradict the hypothesis, but not much else.

Sorry for your troubles, but lots of people have setbacks in their research. Math is full of it also. Yours may be expensive in money as well as time, but little in the world of scholarship is assured.

Whether anything can be published is up to the publisher. Since you can't accept or reject the hypothesis based on the failed experiment it seems doubtful.

Improve this answer
3
  • So if you can prove the hypothesis from the available data, would it be something? Knowing that the technique was already used before but with different apparatus.
    – Zen
    Jul 20 at 14:20
  • 2
    I can't judge the case, but it seems doubtful. At best you have poor evidence. And, FWIW, you don't "prove" a hypothesis in experimental work. You only give evidence, usually statistical.
    – Buffy
    Jul 20 at 14:25
  • Thank you for the correction.
    – Zen
    Jul 20 at 17:04
2

If the data are useless, too bad.

If the accidental death of the second detector was somehow interesting or instructive you might be able to write about it.

See What to do when you spend several months working on an idea that fails in a masters thesis?

Improve this answer

You must log in to answer this question.

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