Three pictures of cats. Left: A domestic cat with leg over head, tongue out, washing itself, label is "GLM" Middle: A tiger in same pose, label is "GLMM" RIght: many cats together in same pose, label is "FIXED EFFECTS"
No matter the size or number, cats are always cats
The entire Democratic strategy debate — "moderate" or move left? fight or compromise? — is based on a misreading of data. Voters don't think Dems are too progressive; they think Dems are weak. The actual math on what to do about this isn't even close!
New: www.gelliottmorris.com/p/new-poll-d...
I feel seen 😭 🤣
Another paper for frustrated methods nerds to put on the syllabus before their students' advisors strongarm them into ignoring it
Finally we can tell the LLM haters to back off: they *can* do cutting-edge social science.
90% of the time I feel like a god as I automate my reports using #quarto
The other 10% I wonder how I've managed to even turn on my computer as spend three hours trying to make the headings in my pdf bold
#rstats
Recent work has shown how vulnerable online survey research is to LLMs. Motivated by this, we examined our online Posner cueing data from Prolific. It's concerning. We now must carefully consider when (or whether?) online behavioral data can be trusted.
see our comment:
www.pnas.org/doi/10.1073/...
The Big 5 test is about twice as accurate as the Meyers-Briggs Type Indicator for predicting life outcomes, placing the usefulness of the MBTI test halfway between science and astrology.
Any psychologist will tell you, the Meyers-Briggs is mostly bullshit.
powerofusnewsletter.com/p/why-are-we...
Have you noticed how navigation apps include walking & waiting for public transit, but excludes parking & walking for driving? After being late a few times 😅, we finally did. We got curious: what if these apps account for parking?
Mental gymnastics meme. TOP: Regression mental gymnastics. The effect is significant...even when I adjust for...everything we have measured. BOTTOM: Causal Inference Mental Gymnastics. unmeasured confounds could explain the effect the sample might suffer from selection bias differential measurement error! SUTVA violation!
Almost done with my next lecture, so going back now to work on some new memes
#AcademicSky
I spend 1/2 my time asking people to peer review more, & other 1/2 complaining that filthy rich publishers should pay for our labour
These aren't incompatible complaints
Scientists DO need to review more (especially if publishing) Publishers DO need to pay reviewers
Me with my honors thesis project that was a failed replication of a very reputable study, in hand, when the crisis "started"
Good news alert: Just want to shout out Elana Meyers Taylor, the most decorated Black woman in the Winter Olympics, who won her sixth medal and first gold tonight, AND is also one of the only 8% of hearing parents who learn to sign for her deaf kid. Badass through and through.🥇
People form beliefs not only as individual agents, but as members of social groups.
Children (4-6 years old) who belonged to a group were more convinced by evidence that supported their ingroup’s belief (and were less convinced by evidence that opposed their ingroup): www.nature.com/articles/s41...
Obama: "At some point, you age out. You're not connected directly to the immediate struggles that folks are going through. I'm not making a hard and fast rule here, but I think Democrats do well when we have candidates who are plugged into the moment. To the zeitgeist. To the times."
I have continued to update my Data Consultants table. It currently includes 96 businesses. So if you are looking for help with data wrangling, visualization, analysis, reporting, dashboarding, training, and more, consider checking out these consultants!
docs.google.com/spreadsheets...
EL PASO, Texas (AP) — FAA says it has lifted temporary closure of airspace over El Paso, Texas, and all flights to resume.
Trust in leadership is essential to success
A new study found that PhD students who had greater trust in their graduate advisor finished their first year more motivated, higher in well-being, and more academically successful than those with lower advisor trust.
academic.oup.com/pnasnexus/ar...
dplyr 1.2.0 was released last week and since I use {dplyr} a lot in my work, I wanted to take some time to try some of the new functions.
This post provides some supplemental examples of the new functions, beyond what is provided in the new Posit materials.
#rstats
cghlewis.com/blog/dplyr_u...
A colorful data visualization of IMDb ratings for the first 11 seasons of The Simpsons. The chart uses a "small multiples" layout where each season (S1—S11) is represented by a vertical column. Individual episode ratings are shown as small white circles plotted against a vertical axis ranging from 5 to 9. Each column features a unique, vibrant color—starting with purple for S1 and moving through blue, teal, green, yellow, orange, red, and ending in grey for S11. A horizontal line in each column indicates the season average, while a unique "stepped" gradient effect creates horizontal bands of color that grow lighter toward the top, highlighting the density of the episode ratings.
"Vibe Coding" the old-school way 🌈
I was prepping a ggplot2 lesson when a stat_summary() error went sideways. Instead of fixing it, I leaned in.
Sometimes the best dataviz is found in the detours.
(Or at least, the most joy on a Monday ☕️)
#rstats #ggplot2 #dataviz #TheSimpsons
Open public resources and records are a key way for journalists, factcheckers to hold people in power accountable when they lie - now they're openly being destroyed
A graphic for Posit featuring a hexagonal sticker with a camel carrying boxes and "yaml12" text, over a blue wave background.
Announcing yaml12: High-speed YAML 1.2 for #RStats & #Python 🐫
Built entirely in Rust, get up to 2x faster in R and 50x faster in Python vs. defaults, tag evaluation (like code execution) is opt-in, and strict YAML 1.2 compliance for consistent data.
Learn more: tidyverse.org/blog/2026/01...
More excited for this than possibly any R package update ever! The amount of janky solutions I had to these problems... RIP the DataCombine package in particular.
dplyr 1.2.0 is out now and we are SO excited!
- `filter_out()` for dropping rows
- `recode_values()`, `replace_values()`, and `replace_when()` that join `case_when()` as a complete family of recoding/replacing tools
These are huge quality of life wins for #rstats!
tidyverse.org/blog/2026/02...
Yup - they started with immigrant journalists - Mario Guevara had a valid work permit and was arrested in Atlanta while covering the June No Kings protest. He spent over 100 days in jail (70 in solitary confinement) before deportation to El Salvador.
www.france24.com/en/americas/...
Really important reminder from Angie Holan that fact-checking is a useful tool in the fight against misinformation.
"Today, these programs are being dismantled, not because they didn’t work, but because powerful actors decided they didn’t want them to work."
www.poynter.org/fact-checkin...
A retired Minneapolis couple who had guns pointed at them by ICE in a church parking lot.
"They were obviously not trained at all. I've known many police officers in my life. These people were right off the streets."
"They had the professional demeanor of criminals."
ht: @paulgraham.bsky.social
Democracy doesn’t require perfect truth—but it does require something more fragile: independent voices. The “wisdom of crowds” depends on independence between judgments. If a single actor can speak through thousands of inauthentic accounts, the apparent consensus of the crowd stops being informative
