Scientific misconduct, its problems and its solutions

There’s no denying that the current scientific institution has some great flaws, but that’s not to say it should be thrown out or replaced by some pseudoscientific alternative. The very core of science holds up by its very nature, and most scientists agree that the current flaws should be exterminated.

The one solution that we shan’t do, is to make science a monolithic process. Science must always be an open process where anyone can participate – no matter their degree, background, ethnicity or university. We shall not fall for the tempting, simple solution of only letting a selected few participate, instead there are much better ways, two of which are suggested in this graphic by clinicalpsychology.net and two of which are my additions:

  1. Make the journals publish all raw data (not necessarily in the physical journal where it takes up place, but on their website where it doesn’t disturb those less interested).
  2. Don’t just blame the researcher, blame the journalist and the journal. As I said, anyone should be able to participate, and it’s the job of the journal and the peer review process to distinguish the good scientists from the lousy ones.
  3. Publish all studies, positive or negative. A great problem in today’s scientific literature is that the interesting studies (i.e. positive ones or groundbreaking ones, not those that confirm what we already know) have a much greater chance of getting published. Obviously – the journal wants to sell copies, and so they want groundbreaking news. Sadly, this makes for statistical flaws, where confirming results on topics such as homeopathy seem much larger than they actually are. The solution to this is to release all studies of quality. Sure, journals can continue to release whichever they want in their physical edition, but there’s no harm in also releasing additional, less interesting articles on the web.
  4. Open up to the public. Most people are not students or researchers, but many who aren’t are still interested in science. They too want to confirm for themselves what works and what doesn’t, what’s true and what isn’t. Most journals today require amounts around €25 for borrowing an electronic edition of an article for a few days. An article – i.e. maybe 20 or 30 pages of material. Few if any would pay this amount just to check up on something. Open up, and release all publicly funded articles to the public. With current internet technology, it wouldn’t cost anything to allow free sharing of all articles, while still making money by selling physical copies or through state funding.

Related posts: