Lightnews — Scholar-powered news

Reposted by Cameron Patrick

Jeffrey Ross-Ibarra @jrossibarra.bsky.social · 2h

Heck no, I'd be a collider that was biased as heck, just to be a pain.

2 2

Cameron Patrick @cameronpat.bsky.social · 2h

if I was a confounder I would simply be easy to observe

Chelsea Parlett @chelseaparlett.bsky.social · 3h

My job would be so much easier if correlation was causation😔

1 2 16

Reposted by Cameron Patrick

Ben Harrap @bharrap.bsky.social · 3d

I clearly had a lot of thoughts/feelings from reading Terry's blog post, so I wrote one of my own

It ended up being a bit of a stream-of-thought post but I think it captures some of my worries

benharrap.com/post/2025-10...

#statssky #episky #academicsky

5 28 77

Reposted by Cameron Patrick

Dominique Baker @bakerdphd.bsky.social · 3d

Fleabag season 3 writes itself

Lindsay Beyerstein @beyerstein.bsky.social · 3d

Of all the clerics they could have popped, those ICE idiots had to hit Hot Priest.

Presbyterian minister David Black on CNN, who was shot with pepper balls while praying outside an ICE facility.

2 35 190

Cameron Patrick @cameronpat.bsky.social · 3d

even worse when it comes to natural language processing, is the "AI" parsing your free-form medical records doing a keyword search or an LLM or something in between? if nothing else, there's huge difference in potential privacy implications for something that's easy to run locally vs must be cloud

2 4

Reposted by Cameron Patrick

Grumpy Old Health Stats Dude @healthstatsdude.bsky.social · 3d

AI is gonna render the internet unusable inside a decade, so you’ll just be leading the way to those woods!

1 1 6

Cameron Patrick @cameronpat.bsky.social · 3d

This is cool, although slightly disappointed to see that car crashes were not one of the causes you examined in your list! (my hunch is that they've been so normalised they've become not-news?)

1 1

Cameron Patrick @cameronpat.bsky.social · 3d

I am also a "beg the question" pedant, thanks to my never-completed undergrad philosophy major

1 4

Cameron Patrick @cameronpat.bsky.social · 3d

i do not like this, not one little bit

Merriam-Webster @merriam-webster.com · 3d

Here’s a primer on ‘primer.’

It’s pronounced ‘PRIMM-er’ if you mean “a small book” or “a short informative piece of writing.”

It’s pronounced ‘PRY-mer’ if you mean “an initial coat of paint.”

7 1 20

Reposted by Cameron Patrick

Kate Laskowski @katelaskowski.bsky.social · 3d

So much this!! Code is so valuable and literally the thing that creates your results. When you include it in your paper it can be such a valuable resource to everyone (plus earn you a couple more citations).

Daniel Bolnick @danielbolnick.bsky.social · 3d

This:
www.nature.com/articles/d41...
Absolutely yes. And I'm proud that @asn-amnat.bsky.social has led the way on this with requiring code and data editors checking code for about five years now!

Stop treating code like an afterthought: record, share and value it

Scientists, research institutions, funders, libraries and publishers must all improve software practices.

www.nature.com

9 26

Reposted by Cameron Patrick

Grumpy Old Health Stats Dude @healthstatsdude.bsky.social · 17d

1 5 20

Cameron Patrick @cameronpat.bsky.social · 3d

I'd never seen this punction mark before;—but it was love at first sight

Olivia Guest · Ολίβια Γκεστ @olivia.science · 3d

these are something my student used and I was impressed en.wikipedia.org/wiki/Compoun...

Compound point - Wikipedia

en.wikipedia.org

1 9

Reposted by Cameron Patrick

disc horse bot @horsedisc.bsky.social · 7d

Men would rather game than go to therapy

2 10 27

Reposted by Cameron Patrick

Peter Tennant @pwgtennant.bsky.social · 3d

Just because an LLM can produce a report with various figures & charts doesn't mean it is good at statistics.

Because good statistics is not about producing code.

It's about deep knowledge of study design & conduct. In my opinion, 95% of all data science problems come from poor questions & design.

Terry McGlynn @hormiga.bsky.social · 4d

Y'all. I just got ChatGPT to do everything in R for this manuscript. I mean EVERYTHING. And it's all legit and reproducible. I'm shook.

How are we mentoring our trainees in statistics now? Who needs to learn coding in R line by line, and who doesn't?

scienceforeveryone.science/statistics-i...

Statistics in the era of AI

How do we mentor, teach, and do stats when AI can do so much of the work?

scienceforeveryone.science

13 24 100

Cameron Patrick @cameronpat.bsky.social · 3d

semicolons are awesome; people should use them more often

1 6

Reposted by Cameron Patrick

Ben Harrap @bharrap.bsky.social · 3d

It always makes me sad to see non-statistcians so readily and happily replace actual human experts with LLMs

I get that academia is a resource constrained environment, but it's doubly painful to see this kind of thinking at a time when universities are defunding their statistical consulting centres

Terry McGlynn @hormiga.bsky.social · 4d

Y'all. I just got ChatGPT to do everything in R for this manuscript. I mean EVERYTHING. And it's all legit and reproducible. I'm shook.

How are we mentoring our trainees in statistics now? Who needs to learn coding in R line by line, and who doesn't?

scienceforeveryone.science/statistics-i...

Statistics in the era of AI

How do we mentor, teach, and do stats when AI can do so much of the work?

scienceforeveryone.science

3 3 27

Reposted by Cameron Patrick

John A Dracula Cullen @nellucnhoj.bsky.social · 3d

Whenever Godot's not on stage, all the other characters should be singing, "Where's Godot?"

Lux "Nightmare" Alptraum @luxalptraum.com · 4d

TFW you paid $1400 to see Beckett’s most famous work without knowing anything about it

One Star Review of Waiting for Godot on Broadway
I recently attended Waiting for Godot on Broadway and spent over $1,400 for two Row C seats (103 and 104). I'm a longtime admirer of Broadway productions and even hold a season pass for Shea's Performing Arts Theatre, so I came in with genuine enthusiasm and high expectations. Unfortunately, this show was unlike anything ! have ever experienced —and not in a good way.
What I encountered was not the artistry, music, or emotional storytelling I usually associate with Broadway, but instead what felt like an endless cycle of nonsensical conversation between characters who seemed trapped in their own madness. I tried-truly tried-to find meaning, symbolism, or even a thread of emotional resonance. I stayed through the first half hoping the second would offer clarity. But by intermission, it was clear: this was a waste of both time and money.
Keanu Reeves is an actor I respect greatly, but I cannot fathom why he would agree to participate in such a disjointed, inaccessible production. His talent was lost in a performance that defied reason rather than provoked insight.
To anyone considering attending: unless you are drawn to highly abstract, nearly incomprehensible theater, I strongly caution you against this show. For the average, educated, thoughtful theatergoer, it is far more frustrating than fulfilling. In my opinion, this was the single most disappointing Broadway experience I've ever had - an unfortunate waste of money and, more importantly, of time.

9 13 140

Cameron Patrick @cameronpat.bsky.social · 3d

the more I think about it, the more I dislike the term "tech industry". Microsoft is a software company. Google and Facebook are advertising companies with a website hobby. Apple and Nvidia are electronics companies. and OpenAI is giant hype-emitting furnace fuelled by investor dollars.

5 37

Reposted by Cameron Patrick

Peter Ellis @freerangestats.info · 3d

The only thing that makes this slow is not being in parallel. The "R doesn't work with for loops" has become practically a superstition. Yes, vectorised operations are much better in R than loops, but plenty of operations (like this modelling) don't lend themselves to vectorisation.

1 2 6

Reposted by Cameron Patrick

Michael Hobbes @michaelhobbes.bsky.social · 3d

I love it when academics publish papers that are like, "I checked and I'm still right"
www.jstor.org/stable/24393...

In this paper I review research since 2000 and conclude
that it has vindicated this view

8 50 800

Cameron Patrick @cameronpat.bsky.social · 3d

... or is success a collider for the relationship between social media and KSAs?!

1 3

Cameron Patrick @cameronpat.bsky.social · 3d

... stats Kingdom of Saudi Arabia???

1

Reposted by Cameron Patrick

Alex Hanna-ween @alexhanna.bsky.social · 4d

As an instructor, I'd rather see your fever dream, No Doze-fueled 4AM essays written at an IHOP rather than anything generated by an LLM.

Hope this helps

Dane 🔜 NYCC AA L-34 @monkeyminion.com · 4d

I wrote a 15 page report on heraldic symbolism in medieval armor and weapon design for my art history class the night before it was due (8am class). Made up 90% of it (only found one book for reference) and got an A. GenAI could fucking never.

wrote 20 pages on Faulkner's The Bear four hours before final papers were due on trucker pills and coffee and cigarettes and got an A, fuck you.
You people couldn't hang with real slackers.
finn
wokeupchic • 4d
It's fuck Al till your homework due in 25 minutes

40 440 1.9K

Reposted by Cameron Patrick

Sevoris (he/him) @sevoris.bsky.social · 4d

Personally: my recipy for writing reports and suchlike has been to just keep notes. A mixture of outlines and digital whiteboards. That stuff is amazing. It‘s like learning for that report - anchors things just a tad more in your brain and it‘ll know way more about to put it into final form.

Alex Hanna-ween @alexhanna.bsky.social · 4d

As an instructor, I'd rather see your fever dream, No Doze-fueled 4AM essays written at an IHOP rather than anything generated by an LLM.

Hope this helps

Dane 🔜 NYCC AA L-34 @monkeyminion.com · 4d

I wrote a 15 page report on heraldic symbolism in medieval armor and weapon design for my art history class the night before it was due (8am class). Made up 90% of it (only found one book for reference) and got an A. GenAI could fucking never.

1 2 2

Reposted by Cameron Patrick

Grumpy Old Health Stats Dude @healthstatsdude.bsky.social · 4d

i ❤️ gtsummary

use it in every project

Sharon Machlis @smachlis.bsky.social · 4d

The {gtsummary} #RStats 📦 “summarizes data sets, regression models, and more, using sensible defaults with highly customizable capabilities. . . . [It] creates beautifully formatted, ready-to-share summary and result tables in a single line of R code!”
www.danieldsjoberg.com/gtsummary/

Presentation-Ready Data Summary and Analytic Result Tables

Creates presentation-ready tables summarizing data sets, regression models, and more. The code to create the tables is concise and highly customizable. Data frames can be summarized with any function,...

www.danieldsjoberg.com

1 1 14