I actually have my own story about dodgy p-hacking. I was an undergraduate psychology student, and the tutor was talking about CBT for gambling addictionhttps://twitter.com/GidMK/status/967942288415457280 …
-
Show this thread
-
It turned out that this type of CBT was ineffective overall - based on a study the tutor was lead author of - but was effective in SOME populations
1 reply 0 retweets 1 likeShow this thread -
On further examination, it appeared that CBT was effective for - get this - Asian women in one specific age group
1 reply 0 retweets 1 likeShow this thread -
The entire paper was published based on this one significant finding, among 30-odd statistical tests that were not controlled for multiple comparisons
1 reply 0 retweets 1 likeShow this thread -
When I asked if this finding might just be due to statistical chance, the tutor paused, uncomfortable, and said "could be" We then moved on to discuss the implications of the study
2 replies 0 retweets 1 likeShow this thread -
This was just one small class of undergrad students in one university, but based on the work by Wansink it's hard not to see a pattern of p-hacking in psychology
1 reply 0 retweets 1 likeShow this thread
I would guess that 99% of the time it's unintentional, but it does highlight the importance of getting an independent statistician to do the stats for your studies
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.