When can I make assumptions in scientific papers?

Question

I would like to know if assumptions are accepted among the scientific community. especially in the field of theoretical computer science.

To be more specific, is it OK to say something like: "The behavior X might be explained by . . ." when you are not 100% sure that it is the correct explanation, or you are not able to explain it mathematically or statistically.

Edit: Indeed the possible explanation I'm talking about is not the main point of the paper, but a small part of results discussion that is, from my point of view, important for future investigations. Also, if I take the time to run some additional tests it would be off-topic and very time-consuming.

Answer 1

It depends on what you mean by "is it OK". If you write "The behavior X might be explained by...", it's certainly OK in the sense that nobody will be offended or automatically reject your paper. You are acknowledging some uncertainty, so there's nothing to object to (while there would be if you claimed greater certainty than your work justified). On the other hand, speculating about possible explanations is not as good as giving a compelling explanation. It's reasonable if this is not the main point of your paper, or if the speculative explanation is awfully clever. However, if the main point of your paper comes down to speculation that you just haven't investigated carefully enough to resolve, then it will probably not impress readers. You could end up with a paper that's not objectionable but still isn't very good.

Answer 2

This is not so much an assumption, but a hypothesis. If this is the bulk of the work, then it is not likely to be acceptable. But if this is part of the discussion of a larger piece of work, then it may well be acceptable. Your paper needs a contribution beyond the hypothesis.

Ultimately, hypotheses are fine in papers. Good questions drive research forward.