let's start a post-punk band called the Aggressive Denials

Thanks Tom! Hope you're well!

sure, dispose of proposals if you want to embed bias towards previous achievements, but to the extent CV and proposal strength correlate (in this sample), the power of this study to detect that proposals are not being selected on is reduced

(for example it doesn't tell you the counterfactual about what would happen in the CV & abstract & proposal condition if a terrible CV candidate submitted an excellent proposal, or if a excellent CV candidate submitted a terrible proposal)

and to detect whether the best proposals are selected, or which factors predict selection (e.g. CV & abstract vs CV & abstract & proposal) you need variation. If CV strength correlates r = 0.99 with proposal strength then showing the two methods select the same is less informative

But it means that there is less variance to explain, so we should be less surprised by a none effect. Both reality and study limitation

Tom Stafford (@tomstafford@mastodon.online) Attached: 1 image Do grant proposal texts matter for funding decisions? A field experiment https://link.springer.com/article/10.1007/s11192-024-04968-7 Simsek et al (2024) conduct a randomised expe...

In this thread I caution about overintepreting those study results mastodon.online/@tomstafford... ymmv, let me know!

my sympathies at this difficult time ;-)

Well, most precisely I am arguing that it very hard to tell if they are or not, and it certainly can't be done using the calculation performed by Schweiger ;-)

That's great to hear you read it and liked it - I always worry when I'm trying to represent someone else's work. I hope my enthusiasm for the paper came through - it's fantastic work

"you could - by careful choice of an existing scoring method from the literature - find any effect, or nothing, or the reverse of any effect you choose. This is bonkers. ... so extreme as to be farcical."

This is the sentiment we were hoping people would come away with!

w/@anniria.bsky.social

“…as we use AI models, let’s keep asking ourselves if we are knowledge generating or merely output generating. Are we adding information to the world, or just words?”

Zahra Arjmandi-Lari, Alexios Mantzarlis, Tom Stafford: Threats to the sustainability of Community Notes on X https://arxiv.org/abs/2510.00650 https://arxiv.org/pdf/2510.00650 https://arxiv.org/html/2510.00650

Bluesky wants me to upload my face or credit card details to read direct messages, which I'm not going to do. So if you sent me a DM, sorry.

The 1,000 neuron challenge A competition to design small, efficient neural models may provide new insight into real brains—and perhaps unite disparate modeling efforts.

Read my new piece in the @thetransmitter on @rougier's 1000 neuron challenge

www.thetransmitter.org/computational-neuroscien...

models, brains, and the value of scientific competitions as sites of integration!

#Neuroscience #CompNeuro

"...how psychological science is laundered to support grander narratives of human failing and irrationality. Normally this happens via popularisation in the mass media, but you can also catch it happening in the scholarly literature."

In a new @wonkhe.bsky.social piece, RoRI fellows Anna Butters and @tomstafford.mastodon.online.ap.brid.gy explore distributed peer review - and what it can (and can’t) fix: wonkhe.com/blogs/a-new-...

Thanks, that helps a lot to put it in context!

With the bonus incentive to achieve fluency in Danish within three years!

When less (communication) is more (collective intelligence) Team work, complex problems, and preserving diversity in the ecosystem of ideas

Monday boost for yesterday's newsletter, looking at whether we can discern general principles for how to connect teams to make them smarter:

tomstafford.substack.com/p/when-less-communicatio...

#Psychology #CogSci #CollectiveIntelligence

I believe it is. Did you read Returning to Rheims? The ending of that book expresses what i was thinking of when i wrote that?

Hi Robert - thanks for this!

is there any option to collect in person?

where can I pick up a copy in Sheffield please?

broken link?

separately, where can I pick up a copy in Sheffield please?

Anyone can say they are open-minded, but how can you prove it? We show how the 'Ideological Turing Test' can be used to measure mutual understanding across controversial topics (and the result suggest people who disagree may understand each other better than we expect) […]

Community Notes require a Community How we used a novel analysis to understand what causes people to quit the widely adopted content-moderation system

New newsletter! Understanding participation in the X Community Notes system, via my love of data visualisation and stronger causal inference methods!

#DataViz #CausalInference #FactChecking #CommunityNotes

open.substack.com/pub/tomstafford/p/commun...

I like how you roll!

yes, that was my immediate thought too!

there must be some kind of estimate you can make, but i don't know off the top of my head how to make it

sure, if you want to calculate the uncertainty! The original poll reports absolute figures, without uncertainty estimates, which is why the graph of the poll also lacks them