4

Although rarely admitted or communicated, in scientific research it's quite common to change the originally formulated question once insurmountable obstacles have made an answer difficult or impossible. After having invested so much in research, the scientist is wondering, which which other questions he could ask by using the material he has collected/created.

So my question is: is there literature on this phenomena, that problematizes the pattern and possible (systematic) impacts profoundly? Any hints from philosophy, science theory, science history or any other discipline are highly appreciated.

  • It seems that you are interested into the individual researcher action... On a more general aspect, regarding the scientific community program of research, you can find something similar in Kuhn's concept of Paradigm shift. Also related Bachelard's Epistemological break concept. – Mauro ALLEGRANZA Dec 6 '19 at 13:44
  • The principal philosophical concerns are the problems of induction and abduction which is inference with uncertainty and inference to the best explanation, so skepticism and Hume are relevant. Daniel Dennett in his Intuition Pumps and Other Tools for Thinking brings the idea of using intuition as a philosophical technique to enhance understanding to improve our metaphysical presuppositions. – J D Dec 6 '19 at 16:01
  • Oh, and my boilerplate: Welcome to SE Philosophy! Thanks for your contribution. Please take a quick moment to take the tour or find help. You can perform searches here or seek additional clarification at the meta site. – J D Dec 6 '19 at 16:02
  • On the title question see SEP Challenging the Rationality of Science. But I do not see how the post is related to it. Changing the original research formulation in accordance with new evidence is justified on perfectly epistemic grounds that have little to do with pragmatic co-determination of theories. – Conifold Dec 6 '19 at 17:42
  • @Conifold That was my ham-handed work. Distinction duly noted. – J D Dec 6 '19 at 20:13
4

Interpreted narrowly, your question seems related to the problem of (data) fishing, where someone investigates hypothesis after hypothesis on the data until getting statistical significance on one (without correcting for the number of hypotheses considered), so that in all likelihood it was just a fluke. This is a well understood problem.

Interpreted more broadly, the phenomenon is not always undesirable. A mathematician might set out to prove one theorem, fail, but be able to turn the failed attempt at a proof into a correct proof of a different theorem that is perhaps even more interesting. (If we care only about the natural sciences, well, something very similar might happen for a theoretical physicist.)

I imagine that there is a way to frame the question so that it is broader than just data fishing yet still narrow enough that it's always undesirable, but I'm not sure whether this has been studied before.

| improve this answer | |
  • 1
    In Six Sigma (Total Quality Management) project mathematics there is an obligation to stay true to the original parameters. Wandering outside of these is termed 'project drift' and is to be avoided. Problems occur when the 'drift' tends to distort the original assumptions and problem description. This may or may not be involved in your research question. CMS – Charles M Saunders Dec 6 '19 at 17:34
  • 2
    P-hacking explained by xkcd: xkcd.com/882 – Adam Sharpe Dec 6 '19 at 20:14

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.