Tim Kaiser

Yeah imagine treatments that are effective but we don't know how they work. This would be a first in humankind, that never happened before.

Jetzt kann ich die Karriere eigentlich beenden, wir wurden bei MAITHINK X zitiert

Können wir das mal klären, @georghenning.bsky.social?

On the way to the Minerva-Gentner Symposium, which more than makes up for the conferences we missed this year! I will present a way to test psychometric network models for robustness against unobserved confounding. www.uni-trier.de/universitaet...

Wenn Kürzungen im Bereich Bildung/Familien damit zusammenfallen, dass gesellschaftliche Probleme auf Familien abgewälzt werden (PISA-Ergebnisse kommen auf einmal dadurch zustande, dass Eltern Bildung nicht so schätzen wie anderswo, usw.) hilft evtl. inference to the best explanation.

Looking forward to the @causalscience.org on Nov 4th-5th. I'll be speaking about mechanism design with applications to people analytics. Hope to continue the discussion there. "Front-door identification often looks impractical because mechanism design is hard.... #causal-inference #bayesian

🚨 Paper Alert 🚨 Our new study on the impact of a single small zombie trial on fluoxetine's overall efficacy estimation is out. With great colleagues @floriannaudet.bsky.social @richlyus.bsky.social and Gert van Valkenhoef 🧟‍♂️🧟‍♀️🧟 A longer 🧵 (1/n)

Yesterday I signed my contract at Julian Rubel's @julianrubel.bsky.social lab. What a great team to work with!

My #awesomelist has been getting quite some traction and different people have been adding lots of new resources they wish they knew when starting their scientific career. Have a look (and contribute) here: github.com/helenahartma...

If you need to digitally sign PDFs from time to time and don't want to install a slow, bloated 2 GB Adobe application, Okular is a nice alternative (<500 MB). It can also fill and edit PDFs - Windows, Mac and Linux. okular.kde.org/de/download/

One does not simply walk into Kassel!

I think this phenomenon can only be understood as a dramatic, horrific worsening of a preexisting situation.

New challenger approaching twitter.now#top

The 3yo just spooned a piece of runny Epoisses cheese. Might as well apply for French citizenship now. (Her, not me)

The kids these days are usually first of all "the kids " (age effects), maybe also a bit "these days" (period effects) but rarely ever "the kids these days" (cohort effects)

Posted by @eikofried.bsky.social on, well, LinkedIn

Agreed. People seem to not like the fact that the meta-analytic estimand is the *distribution* of effects. The mean effect is often meaningless and therefore prediction intervals should be seen as the primary statistical result of a meta-analysis.

Bluesky mag ja tot sein, aber für die dümmsten takes zu Danger Dan ist es nach wie vor einfach eine unersetzliche Quelle

Es macht mich einfach nur noch fassungslos. Die Psychotherapie soll offenbar eine Selbstzahlerleistung werden- niemand von den PsychotherapeutInnen wird ohne Wissen, was man verdient, eine Praxis aufrechterhalten können mit KassenpatientInnen. Und das ist es: genau das ist politischer Wille.

Ein paar hitzebedingte Posts: - Gestern ist der Unterricht wegen Hitze ausgefallen - Neben der Schule ist eine der größten Hitzeinseln Berlins, die Bismarckstraße. Hier wurden schon Temperaturen um die 60 Grad gemessen, dafür erzeugen die Autos zusätzlich Lärm, Abgase und Unfälle.

My reading of Hayes (2004) is that psychotherapy jumped the shark during the second wave

In addition to a hiring PhD in ~October & hopefully some PhDs and postdocs ~February, if anyone would like to work with me as part of eg a Marie-Curie or something similar, please do get in touch. Topics include depression, classification, measurement, systems, time-series, & theory building.

Meine Tochter hat sich einen Dino gekauft der so guckt wie ich, wenn das Wort „Verfahrensvielfalt“ fällt

Wie viele Suizide verbindet ambulante Psychotherapie pro Jahr in Deutschland? Meine Schätzung basierend auf der besten verfügbaren Evidenz: tkaiser.science/3mnossp7dzc2r

Did someone say “explanatory fiction”?

Incredible! Scientists receive a rare standing ovation after announcing that a new treatment can double the survival time for pancreatic cancer, the deadliest cancer which was once considered “undruggable.” When we fund and support science, we can achieve anything! (Video: Dr. Wish Dhillon)

Another paper showing why conformal prediction deserves serious attention in self-driving. The paper is: “Uncertainty-Aware Vision-based Risk Object Identification via Conformal Risk Tube Prediction.” The core idea is exactly the right one. In driving, risk is not a neat binary label.

"Our results showed that a three-cluster solution best fit the data"

Still remember the crushing moment as a young researcher when I realised that proving something through evidence had damn near no impact on social policy unless it was politically useful to believe it.

Ich möchte - wenn ich darf - für #PTBS Patientys eine besondere Empfehlung für die #SchönKlinikBadBramstedt aussprechen. Da werden sie geholfen! Vielen Dank!🫶

Einmal Bahn gefahren...

K1 und ich sind uns einig: Feuerschwanz hätten mehr Punkte geholt. #ESC #Eurovision

Some evidence of decreasing depression rates among adolescents from the latest National Survey on Drug Use and Health. 60% of teens with depression reported receiving mental health treatment (31% taking meds) www.samhsa.gov/data/sites/d...