if so, then why can we 'peak' at our BFs as part of standard pre-registration procotol? #confusedhttps://twitter.com/ChrisFiloG/status/826785233102106627 …
-
-
No b/c opt stopping (under freq model) is just one of several forms of bias / p-hacking
-
I mean if you use Bayes...
-
Bayes is just as susceptible to oth forms of researcher bias that prereg also prevents
-
Optional stopping *does* affect BF & para estimation http://www.timvanderzee.com/sequential-sampling-with-bayes-factors-effects-on-error-rates-and-parameter-bias/ …
-
Better alternative is using Cred Int width for seq sampling
-
And keep number of peeks low; always introduces bias, afaik.
-
I don't think so but I'll defer to experts ->
@JeffRouder@richarddmorey - 8 more replies
New conversation -
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.