This conversation has gotten very far from where it started tho (textbook study & tweet about teaching). Psychologists don't need statisticians to take over teaching intro stats to undergrads. More complex things, of course ppl shouldn't do or teach stats they don't understand.
-
-
-
I don't know. I think it would be nice to have stats people teach undergraduate stats. It all depends.
-
At my uni stats profs teach undergrad stats to all majors. I think it’s good for our students. Though ofc not everyone is a good teacher, that’s aside from area expertise.
-
I've known cases where it has worked out terribly. Generally in the UK service teaching to other departments is low status and historically offloaded to the worst teachers (you save your best teachers for your own students).
-
I think the problem there is academics’ overall approach to service and terrible incentive structures. My institution assigns best stats teachers to service teaching and that pays off.
-
That is refreshing. Generally what matters is how something is taught less than who - you can appoint psychologists with good stats skills and be successful also. Generally you need people with a foot in both camps (e.g., statisticians with good knowledge of the application area)
-
My 2 cents (as somebody that teaches stats). I don’t think you need a PhD in stats to teach undergrads in psych depts. However, just having a psych PhD doesn’t equip you either. I think it’s a specialist role and I think depts should look to appoint specialist stats lecturers
-
I also don’t think you can seperate stats teaching and phil of sci. I think phil of sci (and phil of psych) teaching is missing in a lot of depts and the ideal (I think, without a hint of bias
) is somebody that can combine stats and phil of sci in the context of psych. - 4 more replies
New conversation -
-
-
They don’t need theory, they don’t need computational modelling, philosophy, experimental psychology, … sounds like inbreeding to me or is it perhaps self-preservation? Btw, who are they? I mean, are they really representatives of the subfield?
-
I think it's all up to interpretation.
End of conversation
New conversation -
-
-
True but Economics At least until like super recently
-
Many fields can have problems.
End of conversation
New conversation -
-
-
Cognitive bias in psychology studies could be defined as a meta-issue?
- 1 more reply
New conversation -
-
-
It isn't just stats but experimental design as well. My PI has a saying he likes (which I don't fully agree with) "if you need to rely on stats to show your experiment was significant, then you need to design a better experiment".pic.twitter.com/lwJgaDbJoP
-
The saying has been attributed to E. Rutherford and is a double-edged sword. You need a good knowledge of stats to take stats out of an experiment (i.e. render inference trivial). Then you can do it.
-
I would agree and so does he, he just annoyed when overanalyzing of data very clearly different occurs.
End of conversation
New conversation -
-
-
You all might be interested in our project that launches later this summer: a prediction market for replications of studies in the social and behavioral sciences. Follow us for updates!
#butwillitreplicateThanks. 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.