Ulrich Schimmack

Sorry guys, Mastadon didn't work for me.
Find me on bluesky.

@ulrichschimmack.bsky.social‬

Psychology often is unable to resolve disputes with empirical research. Structure of affect is one example. I wrote this review over 15 years ago. I think it is still relevant and the issues have not been resolved.

https://replicationindex.com/2023/04/25/the-relationship-between-positive-affect-and-negative-affect-its-complicated/

#FalsePositiveRisk in #Medicine

We provide an empirical test of Ioannidis's prediction and find the false discovery risk is well below 50%. Now as citation friendly preprint.

https://arxiv.org/abs/2302.00774

#OpenData

When the article states that the data are available on request and you ask for them.

Thank you!
We are still in the process of studying these data. Hence, unfortunately, I cannot give you an answer at present time.

#NerdHumor #Statistics

https://xkcd.com/2726/

Forget about your H-Index.
What really matters is that you publish one article that makes a lasting contribution. Here Stroop who published only 4 papers. and has an H-Index of 4, but still 700 citations a year today.

Now this is what I call junk science.

https://psyarxiv.com/uy7ph/

#PsychologicalScience (the journal) after the replication crisis.
A z-curve analysis
https://replicationindex.com/2022/10/20/princeton-talk-about-z-curve/
The % of significant (p < .05, z > 2) results has not changed much (ODR = 75% vs. 71). However, the actual power to get these results has increased a lot (EDR = 21% vs. 68%). As a result, the false discovery risk has decreased from 20% to 2%.
Look out for the 2023 replicability rankings coming soon.
https://replicationindex.com/2022/01/26/rr21/

An unfair disadvantage for German researchers. They now are forced to make their file drawers public. Every university could mandate this as part of IRB review.

Happy New Year

Happy New Year

#PsychologicalDeception

Despite the obvious replication crisis in psychology, psychologists lack a consensus about the cause.
1. Some say it is normal (e.g., Strack, Stroebe, Fiedler).
2. Some say it was too easy to get significance (DataColada)
3. Some say the replication studies are shoddy (Gilbert, Baumeister).
4 Some say nobody did replication studies before (Nosek).
5. I say failed replications were common and hidden.
What do you think?

#ChatGPT on #Twitter vs. #Mastodon
It might be right, but I wasn't even sure it would know about Mastodon. A small win.

#SocialPriming #FalseConsensus #Whitewashing #ReplicationCrisis

#ChatGPT believes in social priming. A surprisingly forthright opinon for the AI.

Not surprising. Textbooks and review articles still whitewash the replication crisis. Psychology as a whole still lacks credibility. Social media are deceptive because a small minority here knows better.

#Twitter #Musk #Abuse

Maybe it is not surprising that a White guy from South Africa prefers his workers to sleep where they work rather than to work where they sleep.

https://www.cnn.com/2022/12/07/business/nightcap-twitter-work-culture/index.html

#AntiWokeScientists are complaining about the underrepresentation of conservative voices at universities. They ignore the fact that science is intrinsically progressive, questioning tradition, finding new ways, exploring, and challenging existing beliefs. Flat-earthers, anti-vax, life-starts at conception are outdated fringe ideas that don't belong at a university. That is the truth. Don't abuse truth to sell your ideology.

#ElonMusk #Neuralink

Musk's Neuralink accused of unethical animal tests.

In related news:
The first human experiment that Musk performed on himself seems to be working so far with the side-effect that he no longer experiences emotions.

https://www.reuters.com/technology/musks-neuralink-faces-federal-probe-employee-backlash-over-animal-tests-2022-12-05/