Why most published research is wrong

[Balog]

Slashdot article pointing out that studies are frequently quickly refuted. The "hotter" the field, the more likely to be wrong. I think the implications are kinda obvious.

http://science.slashdot.org/article.pl?sid=08/10/19/172254&from=rss

[Standing Wolf]

Me first, me first, me first!

[Physics]

 

[zahc]

I still like the meta-scientific argument showing that the LHC is too risky. The argument goes that all scientific papers that have investigated the probability of total protonic reversal from the LHC put the risk at many millions-to-one. However, one in several hundred scientific papers is later refuted, so the overall risk taking this into account is much higher. Worth thinking about any time a scientist gives you statistics.

[Balog]

Worth examining the metholody, sample size etc etc of any given study. And realizing that even obeying all guidelines science is still fallible. Appeal to authority (a scientist said it!) Is a very common logical fallacy in our culture today.