Scientific habits of mind, or the scientific process, can be succinctly summarized as the use of evidence and reason to arrive at conclusions. This sounds simple but it’s a demanding code: Take nothing for granted, form conclusions on the basis of careful observation and honest thinking, and be willing to modify those conclusions in the light of new experience.
Art Hobson, Professor Emeritus of Physics, University of Arkansas
I copied this here because I think this is a far better description of what science is, or should be, than the black swanning meme I keep hearing. By that I mean I get pretty tired of hearing the various versions of the "a good scientist tries to disprove their hypothesis" nonsense. Any research scientists in my audience believe that? This is why I've said all along that there was never a need for an alternate hypothesis to explain the obesity epidemic, and further that the alternate hypothesis put forth anyway was never formulated based on observations. Hobson continues:
In other words: Trust the universe, rather than tradition, emotion, authority figures, ancient texts, or popular vote. It’s a code that has worked surprisingly well for science. It might be science’s most important benefit.
I really like this part because it even takes a poke at so-called conventional wisdom. However many seem to have taken countering conventional wisdom for countering's sake to the level of Olympic sport. The "popular vote" rings true with me and can be seen in the comments sections of most posts by the likes of Eades circa 2008-2009. I don't think I need bother with the appeal to authority, only to point out the double standard with which this activity is often engaged in. Speaking of authority figures, however, the anti-establishmentarianism (is that a word?) rampant in the greater community continues to astound me. Lastly his admonishment of trusting emotion is quite fitting. Many seem to have way too much emotion invested when analyzing or discussing the science. How you feel about a study, or how the results of a study make you feel, should have no bearing on your consideration of the evidence as presented. In this regard, the whole n=1 thing is useful to finding what works for you, useful to share so that others might try it and it might work for them too, but totally irrelevant -- especially when unverified -- to discussions of science.