Doing a meta analysis right now. Someone had a txt file of their data. No variable names. No codebook. :)))))

Interesting new bot-detection strategy in online samples:

academic.oup.com/jcr/advance-...

Also the same problems would also apply to Rmarkdown/quarto too no?

The international boycott of ChatGPT has absolutely exploded.

2.5M people have already joined, and this is just the beginning.

Go to quitgpt.org.
Cancel your subscription, delete your account.
And tell at least one person why.

My op-ed: www.theguardian.com/commentisfre...

Very cruel bunch indeed 😤

Quantitative psychologists, why did no one tell me a SINGLE Monte Carlo simulation condition was going to take me HOURS to run...

This looks great! I also look forward to when we can “hide” the sidebar panel to create more space in the main view.

Wow! Congrats, Grace!

Also: this is limited to observed models.

Not true for SEM! So interactions themselves are not the problem, per se.

2. Prolific and other users are now providing tools that effectively detect LLMs.

Combined, we are better off using Prolific than turning to the public who can infinitely take our surveys.

1. At least you know the account is under a real person. You have to show official documentation and intense verification.

It also means users can’t take the same survey twice. Reducing the need for LLMs even more.

And second

Jumping on this to say, we are probably safer with Prolific that other solutions.

I think too many people are creating fear.

Yes bots can do this, but it also requires some tech savvyness that the average person cannot fully perform yet.

Platforms like prolific are great because…

Especially in this economy

GitHub!!!

100% wish there was a more DSEM version of network models

But do they support, Word Online yet?

This but that one TikTok videos asking, “so who’s gonna do it? I’m not saying I want it. But, we all want it… so who’s gonna do it?”

I must confess, I do this all the time when I publish something BUT haven’t used AI. I just feel like it’s the easiest way to engage with someone.

Boy would I love to work this job…

Could you elaborate on reverse causality? I was under the impression that DSEM are quite robust its both statistical and causal inferences. To be clear, i am very (very) skeptical of the whole social media = distress (like many scientists).

But this paper does seem to provide compelling evidence.

Really interesting work, Grace! Will have to give this a read!

Big fan of your work. I would also suggest considering LinkedIn. More academia people are heading there, and plus it’s a more professional environment.

I would like to answer your question, but I lost you on the second part of your question. I am a clueless football person 😭

As in do latent interaction folks!

Solution: SEM and Bayesian models

No-I feel so emotional tonight. There is hope on the horizon. 😭

We already know that lagged effects in CLPMs are likely to be upwardly biased, but just how easy is it to find significant effects? Way too easy. I tested CLPMS in 100 randomly selected pairs of correlated variables and found significant effects in 98 of them. New preprint: osf.io/preprints/ps...

Especially after Ta-Nehisi Coates schooled him? I mean man needs to look himself in the mirror.

Glad of these changes, love that they look more and more like github