Do you teach intro stats? Consider using or contributing short example videos from real researchers that show stats concepts here: osf.io/zc89h/ I used these last year and students had good things to say in my evals. I'll leave some example comments from them 👇👇

Very cool website, R package, and tutorial paper on preprocessing data from ESM / EMA studies. Website: preprocess.esmtools.com Preprint: osf.io/preprints/psya…

Finally, we have a hallucination leaderboard! 😍😍 Key Takeaways 📍 Not surprisingly, GPT-4 is the lowest. 📍 Open source LLama 2 70 is pretty competitive! 📍 Google's models are the lowest. Again, this is not surprising given that the #1 reason Bard is not usable is its

Labeling topics with LLMs is an incredibly efficient approach with BERTopic! I used a 4-bit quantized LLM here to perform the labeling of the topics. You can find the full Google Colab Notebook to reproduce this here: colab.research.google.com/drive/1DdSHvVP…

On my blog: Why pandas feels clunky when coming from R sumsar.net/blog/pandas-fe…

A history of adults blaming the younger generation [📹 History Hustle]

New work by Björn E. Hommel and me. We fine-tuned a language model to predict correlations between survey items. In our pilot, the out-of-sample accuracy for item correlations was .71, .86 for reliabilities, and .89 for scale correlations. We're planning a preregistered follow-up.

Navigating job opportunities as an #ECR is tough🥲 Check out "Psychology Academic Life and Job Market" via Gilad Feldman for transparency insights (link in🧵. #OpenScience advocates transparency & so should our job market! Interested in updating this resource? Volunteers needed!

For years I was doing hyperparameter tuning with XGBoost wrong. Then I stumbled upon this 2-step process. Here's how I hyperparameter tune in R (10X faster and better results). I wrote up a full article with step by step instructions in R. Here it is: business-science.io/code-tools/202…

PSA: All #rstats package on #cran will get an official DOI! This will facilitate bibliometrics and giving credit to R package authors. Registering all 20,000+ packages will still take a few more days. But the first couple of thousand are already live. Example:

🚨Do work in HR or HRV in human populations, see this new pub on reporting guidelines... Publication guidelines for human heart rate and heart rate variability studies in psychophysiology—Part 1: Physiological underpinnings and foundations of measurement onlinelibrary.wiley.com/doi/10.1111/ps…

This is still the funniest exchange of shots in the scientific literature in the 21st century

Logistic Regression clearly explained 👇🏻

New: use GRIM or GRIMMER to test reported statistics for consistency. Are these values even possible? Find out with the error detection webapp! Test many values at once, get summary statistics, and download your results. 🧵 1/5 errors.shinyapps.io/scrutiny/

I really dislike this view of statistics - as a bag of separate tests to be learned or deployed depending on a flow chart. Most of these are just linear models; why learn all these things when one model, a GLM, would do? Also, Continuous != Gaussian, which is implied below

1. In 2003, Kahneman published an article in American Psychologist describing the magic of collaborating with Tversky but also lamenting how current modes of scientific disagreement were unnecessarily hostile and unproductive. He hoped for a better way: Adversarial Collaborations

Please share your favorite examples of absolutely terrible graphs/figures (misleading, confusing, aesthetically abhorrent, etc). Need examples for a class lecture.

Check out our publication in Social Psychological and Personality Science! Can studies about violent video games and aggression be biased by demand characteristics? With two experiments, we show that demand characteristics can lead habitual gamers to respond less aggressively