Profile avatar
mattansb.bsky.social
Statistics lecturer | Freelance statistical consultant & research analyst | #rstats dev @easystats.bsky.social home.msbstats.info (He/Him)
634 posts 2,957 followers 291 following
Regular Contributor
Active Commenter
Posts 13 Comments 37

Man I love {posterior} rvars. #rstats

submitted 3 days ago • 0 comments

#statsmeme

submitted 3 days ago • 1 comment

#statstab #337 Confidence intervals and tests are two sides of the same research question Thoughts: Comment describing the connection between NHST p-values/test and Confidence Intervals (CI). #NHST #ConfidenceIntervals #pvalues #frequentist #estimation doi.org/10.3389/fpsy...

submitted 3 days ago • 0 comments

#statsmeme

submitted 4 days ago • 0 comments

It's often inconceivable to students that not only are there many different ways to analyze the same dataset, but also that there are many ways to analyze the same dataset to answer the *same* research question. #stats

submitted 4 days ago • 2 comments

submitted 4 days ago • 5 comments

A group of students and I asked ChatGPT and Gemini to explain the difference between conditional, joint, and marginal prob. We were just testing and no intention to learn. None of us (n =6) got correct responses. Then we each asked for a Venn diagram. This one was the most interesting.

submitted 4 days ago • 4 comments

This Thursday I will give a talk on (causal) mediation analysis -- happy to have finally worked out what I want to tell people. And happy to pilot a new mode of slide sharing: just putting them on my website (juliarohrer.com/resources/)

submitted 5 days ago • 13 comments

Um, sorry, but does anyone want to interpret credible intervals as confidence intervals? Why...? #stats From Wikipedia's entry on Bernstein–von Mises theorem (the Bayesian Central Limit Theorem) en.wikipedia.org/wiki/Bernste...

submitted 4 days ago • 1 comment

Could refer to this absolute banger of a paper title again today. ✌️ A title so good, you don't even have to read the paper anymore. doi.org/10.1177/1094...

submitted 5 days ago • 5 comments

I like this paper: not only does it review of all the reasons standardized coefficients are bad, actually, It also then goes on to show how test statistics can be used as measures of unique contribution/ effect size! - something that just sounds so so wrong! #stats doi.org/10.1080/0003...

submitted 6 days ago • 9 comments

"Gotta hit those KPIs bro. Words/minute read, words/minute written. It's all about that efficiency." I always saw the academic endeavor as a slow roast over a gentle fire, not a pressure cooker...

submitted 6 days ago • 0 comments

Remove a letter, ruin a movie: The Goo, the Bad, and the Ugly

submitted 6 days ago • 0 comments