I'm largely ignorant about modeling, but I happen to be close to statisticians, and they always find it weird that psych are more worried about justifying their final models than reporting their decisions along the way. (This isn't an argument, just a comment.)https://twitter.com/JCSkewesDK/status/1091658368694013957 …
-
Show this thread
-
I mean focus on the (predicted) result, not the process.
2 replies 0 retweets 3 likesShow this thread -
Replying to @nataliadutrapsy
Cognitive/computational modelling isn't only/exclusively about prediction though, right? Models don't have value only as predictive tools. Some modelling is about furthering understanding on a mechanistic level, for example.
1 reply 0 retweets 3 likes -
Replying to @o_guest
Hi, it was meant as a comment about stats modeling and its use in non-math psych fields. I mean, even if people claim they aren't doing cog/comp modeling, there is no reason (I think?) to not be open about their decisions - models, parameters etc. 1/
1 reply 0 retweets 0 likes -
Replying to @nataliadutrapsy @o_guest
I usually refrain from commenting on these topics bc my knowledge is average, and I can't really use technical terms or smth like that. I guess I tend to sympathize with inductive and descriptive approaches bc I studied behavior analysis in the past. 2/
1 reply 0 retweets 0 likes -
Replying to @nataliadutrapsy @o_guest
Imo, what got us in this mess was that people made analytical decisions to find arbitrary p values, instead of what was adequate to their question and data. This and the really odd culture of omiting procedures and decision making processes in the papers. But I digress
3/31 reply 0 retweets 3 likes
I think we are on the same page here both in that I now understand your perspective better and in that I think we agree. 
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.
