>Percentage of papers reporting a support for the tested hypothesis plotted against year of publication and divided by discipline of journal (SP Space Science, AG Agricultural Sciences, BB Biology & Biochemistry, CH Chemistry, CM Clinical Medicine, CS Computer Science, EB Economics & Business, EE Environment/Ecology, EN Engineering, GE Geosciences, IM Immunology, MB Molecular Biology & Genetics, MI Microbiology, MS Materials Science, NB Neuroscience & Behaviour, PA Plant and Animal Sciences, PH Physics, PP Psychiatry/Psychology, PT Pharmacology & Toxicology, SO Social Sciences, General).

In 2007, 85% of published scientific papers affirmed their hypothesis. Some fields have had entire years without a single negative result, and this isn't JUST going on in silly fields like sociology. Oh no, this is molecular bio, chemistry, micro bio, materials science, and physics.

This is not the same thing as research failing to replicate, but it almost certainly is the cause: publishers are playing Texas sharpshooter and ignoring negative results. The rare false positive comes up in someone lab, and ends up becoming the first attempt that's published.