Hooray! See y'all next year in my hometown! 🐎 #positconf #rstats #htx

You're very welcome @vincentab.bsky.social P.S. Check out our preprint on an alternative to staring at coefficients: j-rohrer.github.io/marginal-psy...

Whoa—my book is up for pre-order! 𝐌𝐨𝐝𝐞𝐥 𝐭𝐨 𝐌𝐞𝐚𝐧𝐢𝐧𝐠: 𝐇𝐨𝐰 𝐭𝐨 𝐈𝐧𝐭𝐞𝐫𝐩𝐫𝐞𝐭 𝐒𝐭𝐚𝐭 & 𝐌𝐋 𝐌𝐨𝐝𝐞𝐥𝐬 𝐢𝐧 #Rstats 𝐚𝐧𝐝 #PyData The book presents an ultra-simple and powerful workflow to make sense of ± any model you fit The web version will stay free forever and my proceeds go to charity. tinyurl.com/4fk56fc8

At the risk of stretching the imagination beyond the limits of human endurance, the reader is asked to pretend that the author is interested in splines #rstats

Selectively Remove or Hide Legends in ggplot2 datavizpyr.com/selectively-... #dataviz #rstats

🐸🐸🐸

Ever stared at a table of regression coefficients & wondered what you're doing with your life? Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social) Preprint: doi.org/10.31234/osf... Website: j-rohrer.github.io/marginal-psy...

Lil’ Muncher (official name lol) update: 1) LM abandoned his post at 50% consumption but 2) Found a friend, LM2 3) And many more 😵‍💫 Consumption rates have become exponential and sadly all (visible) Munch Bunchers had to be evicted.

Little muncher < 24 hrs later: 1) still on the same lead (! I was surprised by this!) 2) leaf consumption at 50%

I’m supposed to hate this little cutie, because it will eat up my tomato plant but…. Come onnnnn look at it 🥹🥹🥹🥹

🚀 gratia 0.11.0 is out! Now has a paper in JOSS — please cite 📄 doi.org/10.21105/jos... Experimental parallel processing ⚡ New assemble() for building plots 🎨 Better support for complex families + new diagnostics 🧪 Lots of bug fixes + polish ✨ 👉 gavinsimpson.github.io/gratia/ #Rstats

Yes!

Reminder that all three books I've co-authored are freely available online for non-commercial use (and the fourth will be, too)

#rstats So chuffed! I printed the first PDF copy of my book to see what it might look like in print! It will be printed in full color! Visualizing Multivariate Data and Models in R On the whole, looks good, but lots of tweaking to do. It weighs in at ~440 pgs., so perhaps some cutting needed.

This plot has more than one *You had one job* kind of energy. ✨️ The prominent one, is using the same y-axis scale for both Median Home Price & Real Weekly Wages. Anyway, gently dropping my attempt to visualize the y-axis scaling mismatch: darakhshannehal.quarto.pub/scrollytelli... :) #dataviz

art

This take is so damn HOT

I don't remember the last time I was this emotionally invested

Time for one of my favorites:

One my favorite resources!! I am working on a review of how paired t tests are a unique case of mixed models.

AI filmmaking is peeing. It’s pooping. It’s throwing up. It’s a new kind of liquid coming out. It’s hated by the establishment. It’s what goes in the toilet. It’s wet. It’s what’s in the bathroom.

“Do we like Damien? We like Damien, don’t we? It’s all for him, we do it all for him. We just want him to look at us. We treat him well, sometimes he’s not very nice to us, but it is for him, it’s really all for Damien.”

To say that "the data show these vaccines fail" in regards to the mRNA plaform is unforgivable. For those of you that don't remember - this is the data from a SINGLE dose. In December 2020. So good I cried in grateful tears.

Oh no. And, seems like they never adjusted for it or considered it a candidate covariate? 😢 This is a really great reminder that: 1. Blind pre-specification & testing “baseline imbalances” can really miss the mark. 2. Confounding due to differential loss to follow up is real and bad lol.

Sometimes it's good to revisit the basics. Part 1 of a series on using the binomial distribution to interpret & power studies on rare events -- inspired by Martin Bland's write-up, "Detecting a single event". Spoiler: rare events are hard to find 😜 🔗: tinyurl.com/2hb9vh68 #rstats #stats101

I regret to inform you that she has, yet again, written more words than are necessary on a topic you don't care about

Humans vs Ants attempting the “PIANO MOVERS PROBLEMS” Prof. Ofer Feinerman and his team at the Weizmann Institute of Science made this evolutionary competition that asks the question: Who will be better at maneuvering a large load through a maze?

Eye opening read on how tech companies view us as renters and not buyers.

Rarely try to publish them, or journals rarely accept them?

If this happened to me. Sheeeesh. Joy maxed out. Time to pack it up, friends. Couldn’t get better.

#tidytuesday about colors from the xkcd color survey! Colors are always fun to work with. I looked at overlaps in descriptions of colors given by participants of the survey. code: github.com/mic-wypych/t... #rstats #dataviz #ggplot2

I can only imagine how "FUN" was to clean this #TidyTuesday dataset. This is my plot for the week. #R4ds #dataviz #ggplot2

*chef-kiss* so many times I've banged my head against my desk going "omg, nooo, not on all the facets!!!!!!"

#TidyTuesday week 25 📊 Exploring #measles incidence and laboratory confirmation rates in Africa (2024). Code: github.com/rajodm/TidyT... #dataviz #rstats #ggplot2

Better late than never — here’s my #TidyTuesday submission from last week's data! I built a #shiny app with #plotly to explore the @xkcd.com color survey results in both 🌈 RGB and HSV space🌈. 🖥️ App: jessgraves.shinyapps.io/xkcd-color-s... 🌐 Personal site: jesslgraves.github.io/apps/2025-07...

An excellent old post I just discovered by @statsepi.bsky.social on the commonly held misconception that randomization is meant to “balance” factors. (My schooling 100% propagated this myth & taught us to always test baseline factors across groups & if p<0.05 then treat as confounding! 😬💀🤡)

One year on, it's probably time to check to see how the study i critique here has been cited. (ICYMI) statsepi.substack.com/p/one-simple...

Ugh I just LOVE this type of shit nodaysoff.run

Rewatching Us (2019) and the beginning is so absolutely delectable

To do lists are just the chronicles of my shame

Fair point! This fig doesn’t scream “this is gonna cause bias”. Sex is a covariate here, but the imbalanced sampling forces it to become a confounder. In balanced designs, you don’t have to adjust for sex to get an unbiased age estimate (but it’s less powerful). In imbalanced designs you do.

I wrote a little post on a common issue: You accidentally got a biased sample & now your results might biased be too. How to recover? Can you?! 🔍 Confounding (yes, ice cream & crime is in there 😅) 📊 Model misspecification 💻 Code, plots, & takeaways tinyurl.com/4af2td83 #rstats #episky #stats