This summer: A week of @JASPStats hybrid workshops:.- Discovering Statistics using JASP.- Bayesian Hypothesis Testing with JASP.- Machine Learning with JASP.- State-of-the-Art Meta-Analysis with JASP. (see full details at ECs for students available)

Why is this loss-loss?.- I'm going to take the flight anyway, just more inconvenience for me.- They could've sold my inbound way to someone else (They get paid twice!).- I would've bought one more inbound way (Potentially with them again - they get one more booking!). Well, gg. .

The genius behind @AirCanada 's inflexible booking system is causing a loss-loss situation. I booked a non-refundable return ticket for later this year. For unforeseen reasons, I want to cancel the inbound (but keep the return) WITHOUT a refund. Talk with an agent- NOT POSSIBLE.

RT @Nature: From today, all new submissions to Nature that are published will be accompanied by referees’ reports and author responses — to….

RT @OmidVEbrahimi: 📜 Our new article on urbanicity and mental health is just out:. We find no universal link between urbanicity and the pre….

RT @I4Replication: A re-analysis of Ciacci's (2024) "Banning the purchase of sex increases cases of rape: Evidence from Sweden" Journal of….

RT @JkayFlake: Robustness analysis is going to be a big part of the future of how we think about evidence synthesis and rigor - exciting ne….

Also, I want to acknowledge and thank the original authors who shared an extremely well-documented and fully reproducible script with data!.

See the full reply submitted for publication at

These results align with our previous re-analyses of nudging as a behavioral change technique (see e.g., , which also suffers from reporting issues (. Maybe it's time to start doubting these "simple behavioural change techniques"?.

This is not that surprising since the authors themself reported that “most trials (k = 66; 84%) had some concern of bias regarding selection of the reported result(s) given that preregistrations and prespecified analysis protocols were rare”.

In summary, we found that publication bias exaggerated the evidence in favor of SC-BCT to such a degree that once this bias is properly adjusted for, the effect disappears entirely. In fact, the data show moderate evidence against the presence of an effect.

A recently published meta-analysis in Nature Human Behaviour "found evidence supporting the efficacy of social comparison as a behaviour change technique in shaping behaviour in the desired direction". I was curious, so I re-analyzed the manuscript but funnel plots below say all

RT @I4Replication: Registration is open for 7 upcoming replication games: . Looking for researchers in econ/ecology….

Thanks to @RexDouglass, who shared the awful-looking funnel plots on his timeline, which prompted us to look deeper!.

See the full response just submitted for publication at

The present literature appears to be contaminated by publication bias; high-quality registered reports are needed to properly evaluate the effect of large language models in educational settings.

Importantly, our findings do not imply that there is no positive effect of ChatGPT, or other large language models on learning; in fact, our analysis reveals that there is “absence of evidence” rather than “evidence of absence”.

We re-analyzed the meta-analysis and found that the conclusion is almost entirely driven by publication bias. Both state-of-the-art and standard methods reduce the degree of the effect 2-3 fold. Moreover, the data no longer show statistical evidence for the main conclusions.

Looking for recommendations to finish my #statistics #mugs collection!

See and our blog for more details: