I’ve just got back from a tiring but awesome couple of days at the Science Online London conference (which–conflict of interest declaration–I helped organize). I’ll be writing up my speaker notes for the final panel tomorrow, I hope, but in the meantime I’ve just been challenged by Ruth Seeley on twitter to explain the scientific method:
I’d LOVE it if ONE scientist would take on the challenge of CLEARLY explaining the scientific method. Wikipedia sure doesn’t.
Martin Robbins took up the challenge straight away, but in the interests of clarity (and non-jargon) I’d like to see suggestions that might replace the Wikipedia entry. As far as I’m concerned, these explanations should be lay-readable; understandable to a high school student, say.
Please feel free to have a go, and then, seeing as the Wikipedia article is the first Google hit for ‘scientific method’, let’s edit the bloody thing to something more like. And I know commenting here puts some people off, so feel free to comment at the BioLOG, where you don’t have to sign up for an account…
I mentioned to Ruth that I think part of the problem is constitutive to Wikipedia and similar user-edited media. I’ve become frustrated by Wikipedia entries in my chosen field of neuroscience. Many of them are not wrong, they’re just awfully strange ways of explaining things that should be put much more simply.
Worse, when I’ve tried to refine or correct such entries, I’ve been met with stiff resistance by both the Wikipedia elite (Wikileet?) and the authors themselves.
If I were to take the right combination of drugs to give me the confidence to edit the Wikipedia entry on Scientific Method, I would approach it this way.
I would first introduce the topic with a short, punchy definition. The existing article has something close to this.
I would next define the term operationally using examples from each of the major fields of science. (By the way, I would avoid using Watson’s “discovery” of the structure of DNA as an example, unless we want to make his unauthorized use of Franklin’s data part of the canonical method.) I would give examples of the scientific method on a field-by-field basis.
One problem is that the “scientific method” is not a unitary concept; what constitutes an acceptable experimental design in (for example) psychology is not exactly the same as what would constitute an acceptable experimental design in even the closely related field of clinical neurology, or molecular neuroscience.
Finally, I would wrap up with a summary that shows how all these disparate methods fit under the umbrella of “the” Scientific Method.
I would also add that I have always been a Kuhnian in a world of Popperians so my view of the scientific method is quite different than many other scientists’.
I wrote a detailed blog post on the topic of scientific skeptics paying close attention to Wikipedia articles and editing them.
The biggest suggestion I can make to avoid/deflect resistance from other editors is to find good sources that back up the changes you make, and footnote every change you make using them. (Use the “ref” tag to mark your footnotes, as explained in the blog post I linked). Simple edits of wording can be argued endlessly. But if you can show that some credible authority stated something, you get much less hassle.
Jim, yes there are certainly different ways of doing things, and these should all be in the definition (or definitions, if you prefer). Excellent point about the Franklin data.
Good advice, Tim. Thanks.
There’s no such thing as THE scientific method. It’s never worked that way and never will. Scientific inquiry is too much a creative process that proceeds by luck and intuition for it to be captured in a 3-step procedure. It would be more honest to talk about scientific ethics or rules of conduct that frame that process, procedures that have proven to be useful and are thus widely spread, or maybe general terminology (what is a “theory” what is a “prediction”) etc. I wrote some words on that here:
What is a scientific prediction?
Models and Theories
Sabine, those are exactly the points that need to be made in lay terms.
Yes, Sabine. That’s what I was trying to say. You put it much better.
Problem. Trial. Error. Reflection. Trial. Success? Reflection. Trial…
Hah! Lovely, although I might want a ‘publish’ step somewhere.