Friday, August 28, 2015

Failure to Replicate

link
Aarts et al. describe the replication of 100 experiments reported in papers published in 2008 in three high-ranking psychology journals. Assessing whether the replication and the original experiment yielded the same result according to several criteria, they find that about one-third to one-half of the original findings were also observed in the replication study.
Important result...though, of course, not exactly a stunning one.
(h/t: S. rex)

2 Comments:

Blogger The Mystic said...

We had a server cluster once whose software failed to replicate.

We trashed it and got new software.

Psychologists be warned.

6:05 PM  
Blogger Pete Mack said...

In principle, roughly 5% of studies like this should be irreproducible, since pretty much all of them use standard statistical methods. 40% is crazy high.

6:57 PM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Home