I often get so confused when people on Twitter talk about the “false discovery rate” when discussing multiple comparisons... turns out some folks are actually talking about the *familywise error rate*, not the procedure by Benjamini & Hochberg.
-
Show this thread
-
Replying to @dailyzad
Health Nerd Retweeted Health Nerd
Speaking of, did you see this?https://twitter.com/GidMK/status/1150974885822226434?s=19 …
Health Nerd added,
Health NerdVerified account @GidMKPeople, I cannot even Every single cell in this table represents a separate t-test between placebo and a sage extract for various cognitive factors. Bolded = p<0.05 No control for multiple comparisons Conclusion of the paper? Sage extract is effective
pic.twitter.com/0ZuAr6B2QnShow this thread1 reply 0 retweets 1 like -
Replying to @GidMK
yes, supplements are notorious for spurious findings and bad analyses... if trying to control FWER for planned comparisons, Holm-Bonferroni, FDR or multilevel modeling...
1 reply 0 retweets 2 likes
Replying to @dailyzad
Nah just say that you planned your analyses so you don't need to do anything
7:55 PM - 16 Jul 2019
0 replies
0 retweets
2 likes
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.