ianhussey.mmmdata.io
Senior lecturer in Psychology of Digitalisation at University of Bern. Chief recommender @error.reviews. "Jumped up punk who hasn't earned his stripes." All views a product of my learning history.
415 posts
2,866 followers
756 following
Prolific Poster
Conversation Starter
comment in response to
post
Cheering for the RA trope, it’s always funny as it is enraging www.the100.ci/2024/12/18/r...
comment in response to
post
comment in response to
post
Absurdly small SD for the range
comment in response to
post
pubpeer.com/publications...
comment in response to
post
Transparency in research be like
comment in response to
post
comment in response to
post
These tend to only remain to be interesting when they are engineering questions (‘we managed to get LLMs to do X by doing Y’) and not psychological mysticism questions (‘wow, the thing trained on human biases exhibits human biases’).
comment in response to
post
Given your critique is about poorly researched work, I would say this is more than ironic.
comment in response to
post
You claim is that it’s one of the most robust findings, this work severely questions this. Not having read the Cochrane review of the work in the area you’re citing is not a good thing.
comment in response to
post
This seems important enough to be in the blog?
comment in response to
post
Nice to see the question asked, but the impression and the actuality are often disconnected.
I’m surprised we haven’t see a “Many-Experimenters” project yet, like an analogue of many analysts that simply attempts to recreate materials from methods sections and observes issues and heterogeneity.
comment in response to
post
If I am reading this correctly, this version with a rational x-axis provides evidence for a different conclusion than the paper seems to offer:
A (slight) majority of experts say that there is any evidence at all for only one of the 26 claims investigated.
bsky.app/profile/rube...
comment in response to
post