Gene-environment hypothesis again fails in meta-analysis. Prior for these kinds of claims should be taken as LOW. n = https://www.biologicalpsychiatryjournal.com/article/S0006-3223(17)31993-5/pdf …pic.twitter.com/CRig9ewGs0
-
I tend to be most skeptical of interaction effects across the social sciences. It seems that they are less likely to replicate, tho not sure if there is data on it.
Just so happens that I did a yuuuge study on that by looking at general importance of interaction effects across randomly chosen triplets of variables in a large dataset. Didn't analyze in detail yet, but generally interactions are usually not large/important.
-
-
Given that, this means that they are generally hard to detect with the usual small sample sizes in social science, and thus if detected by usual standard of p < .05 (with QRP), will generate fail to replicate. I need someone to collab with to finish this study.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Nice, looking forward to seeing it. Seems silly to think that we (social scientists) can understand interactions btw independent things that we can’t agree on nor measure well.
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.