4 ms·
Most published medical research is wrong... https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.0020124 https://journals.plos.org/plosmedici
by mgamache 4y ago
Most published medical research is wrong...
https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.0020124 https://journals.plos.org/plosmedicine/article?id=10.1371/jo...
- Retric 4y agoWrong is hardly the ideal metric here. Being less wrong than random can beat being correct and utterly trivial.
- mgamache 4y agoThe standard is a p value of .9 or .95. Not a flip of the coin.
- Retric 4y agoThat’s hardly the only way to be wrong. Suppose you made a mistake in the synthesis process and the dosage described doesn’t match the dosage revived by patients. Now the paper is incorrect but possibly still quite useful.