iiinova — where your digital self keeps evolving

The presentation we can all aspire to

a five-panel meme of vince mcmahon getting progressively more excited

panel 1: Our next presentation is on <complex topic>
panel 2: 20 min presentation. slide 1/41
panel 3: "so here's the results" slide 8/41
panel 4: "thanks for coming to my talk" slide 19/41
panel 5: every question answered by the extra slides

After what feels like forever, I've gotten to do some statistical modelling and two package-related things stuck out to me: {glmmTMB} makes parallel processing very easy {marginaleffects} is SO useful! Plots of predicted probabilities seem much easier to explain than a table of odds ratios

1140

Reposted by Ben Harrap

Amstat-American Statistical Association @amstatnews.bsky.social · 29/09/2026

Last call! Nominations for the 2027 International Prize in Statistics close Oct. 1. #STATSSKY #STATISTICIAN Recognize a breakthrough contribution in statistics or data science: statprize.org/nominations/

026

Reposted by Ben Harrap

Nicola Rennie @nrennie.bsky.social · 27/09/2026

You can use rel() to specify the size of individual elements relative to the base size. Means if you increase the base size, then you don't need to readjust sizes of each individual element again

162

Ben Harrap @bharrap.bsky.social · 26/09/2026

Forthcoming readxl update to read these as list cols @hadley.nz @jennybryan.bsky.social?

0110

Ben Harrap @bharrap.bsky.social · 24/09/2026

Getting the last speaker in a conference session to be the session chair like

0191

Reposted by Ben Harrap

ViCBiostat @vicbiostat.bsky.social · 23/09/2026

Seminar announcement Dianne (Di) Cook: Practical Tools for Data Exploration in Biostatistics 15 October, 4-5pm AEST via Zoom Further details: www.vicbiostat.org.au/event/practi... @visnut.bsky.social

0169

Ben Harrap @bharrap.bsky.social · 22/09/2026

When were your flabbers last gasted?

140

Reposted by Ben Harrap

Danielle Navarro @djnavarro.net · 20/09/2026

Please, I am begging you. Do not use dplyr::ntile() in any serious statistical analysis in which you want to group continuous observation into quantile-based bins. Despite the name it doesn't actually do that, and the thing it actually does is very dangerous when accuracy matter. #rstats

blog.djnavarro.net

The trouble with ‘ntile()’ – Notes from a data witch

Please, please, please do not use ‘dplyr::ntile()’ in an analysis that will be submitted to a regulatory agency. It’s secretly an ill-conceived SQL tool that doesn’t do what you probably want it to, a...

68831

Ben Harrap @bharrap.bsky.social · 19/09/2026

I'm really enjoying the implications of "Authentic BUT Australian"

A packet of tortillas that reads "cali mex tortillas. Authentic foods but Australian made"

2142

Reposted by Ben Harrap

samuel mehr @mehr.nz · 18/09/2026

the longer I'm a prof the more certain I become that each university operates in a wildly different way to the rest. most sensible, then, is to adopt a strategy of radical advice-seeking ask for help! early and often! academics love nothing more than to put off work to help out junior colleagues!!!

2619

Reposted by Ben Harrap

Michael Friendly @datavisfriendly.bsky.social · 18/09/2026

Vision science: The difference between Rod Stewart and Cone Stewart finally clarified

Two photos of the singer Rod Stewart, left: B/W labelled Rod Stewart; right: color, labelled Cone Stewart

1467

Ben Harrap @bharrap.bsky.social · 18/09/2026

> Please complete your staff feedback survey > This is your personalised link, don't share it with anyone > No one can identify responses

171

Ben Harrap @bharrap.bsky.social · 17/09/2026

Roll up, roll up! Get your tickets for the Australian Data Science Network 2026 Conference before early bird pricing ends tomorrow! adsnconf2026.netlify.app

adsnconf2026.netlify.app

Australian Data Science Network Conference 2026 – ADSN2026

26th & 27th November 2026, Australian Bureau of Statistics House, Canberra

062

Reposted by Ben Harrap

Jon Harmon (he/him/his) @jonthegeek.com · 16/09/2026

I made it through my #PositConf2026 talk! The Design of Everyday #RStats Functions. Slides here: stbl.wrangle.zone/slides/every... The underlying 📦 {stbl} is on CRAN!

stbl.wrangle.zone

The Design of Everyday R Functions

5548

Ben Harrap @bharrap.bsky.social · 14/09/2026

Where do all the average philosophers like to hang out? At the why bar

010

Ben Harrap @bharrap.bsky.social · 11/09/2026

Overall I have a good time on social media but it bothers me how often I come across people taking needlessly strong positions on whatever topic I get that controversy drives engagement, but it's tiring to see. Most AI takes are like this

260

Ben Harrap @bharrap.bsky.social · 10/09/2026

Turns out that music for slaying demons to is also good for coding to youtu.be/Bed8NPkVHRQ?... Fun fact I learned yesterday: Mick Gordon lives in Melbourne, which is obviously why the DOOM soundtrack is very cool

youtu.be

Mick Gordon - 04. Rust, Dust & Guts

YouTube video by Mick Gordon

000

Ben Harrap @bharrap.bsky.social · 09/09/2026

This has to be the most bizarre junk mail I've received in my life Is the Venn diagram of junk-mail readers and people who would install Linux not just two separate circles? What's going on here???

A folded sheet of A4 junk mail, extolling the virtues of Linux mint

1101

Reposted by Ben Harrap

Grant @syncingsinking.bsky.social · 08/09/2026

Now that Bluesky seems to have automatic thread numbering, a short thread on the probability of winning the Monty Hall problem if you stick.

2235

Reposted by Ben Harrap

Aki Vehtari @avehtari.bsky.social · 08/09/2026

I'm looking for doctoral students, postdocs and research fellows to work with on Bayesian Workflow. Flexible starting time

02115

Reposted by Ben Harrap

Dariia Mykhailyshyna @dariia.bsky.social · 08/09/2026

This workshop is in two days so don’t forget to register ! Details: bit.ly/3wBeY4S Please share! #AcademicSky #EconSky #RStats

065

Reposted by Ben Harrap

Per Engzell @pengzell.bsky.social · 08/09/2026

Causal inference is the art of knowing exactly why your estimate isn't causal

611418

Ben Harrap @bharrap.bsky.social · 08/09/2026

Recently in my Intro to R workshop, I had a brief section on error messages, debugging, and how to solve your own problems. An attendee asked why LLMs were at the bottom of the list — here's a few thoughts on why benharrap.com/post/2026-09...

benharrap.com

Limited Learning Mode: Why LLMs are my last resort – Ben Harrap

2216

Ben Harrap @bharrap.bsky.social · 08/09/2026

For this episode of Ben Actually Reads The Documentation, we've discovered the documentation for dplyr's by/.by argument. dplyr.tidyverse.org/reference/dp... There's so many functions it works with — mutate, summarise, filter, slice Avoids writing group_by() and ungroup() so often #rstats

dplyr.tidyverse.org

Per-operation grouping with .by/by — dplyr_by

There are two ways to group in dplyr: Persistent grouping with group_by() Per-operation grouping with .by/by This help page is dedicated to explaining where and why you might want to use the latter....

2153

Ben Harrap @bharrap.bsky.social · 05/09/2026

Why not both? (I'm disgusted that this works)

76412

Ben Harrap @bharrap.bsky.social · 05/09/2026

I reckon this has got to be my top read this year "On interpreting the causal effects of race"

062

Reposted by Ben Harrap

Karen Lamb @kelamb.bsky.social · 05/09/2026

I'm still in shock from winning the 2026 Eureka Award @eurekaprizes.bsky.social for #mentoring! Congratulations to the other finalists- thank you for your incredible work! Huge thanks to @nicolewhite.bsky.social, David Price and Ximena Camacho for the nomination and support. @mischhub.bsky.social

022

Reposted by Ben Harrap

Vincent Arel-Bundock @vincentab.bsky.social · 04/09/2026

Big news! 🎉 𝚖𝚊𝚛𝚐𝚒𝚗𝚊𝚕𝚎𝚏𝚏𝚎𝚌𝚝𝚜 1.0.0 for #Rstats is out. It’s a big number and it feels like a big step. I wrote a blog on the challenges of interpreting statistical models, SPEED, cool new features, the future, and a 5 year package development and writing odyssey. arelbundock.com/posts/margin...

1642097

Ben Harrap @bharrap.bsky.social · 03/09/2026

Today as I trialled my intro to R workshop I explained RProj files, directories, and absolute vs. relative paths Of the people in the room, whose education ranged from no tertiary quals to PhDs, every single person learned something about directories Good thing I included that content 😅

2352

Reposted by Ben Harrap

Clément Canonne @ccanonne.github.io · 31/08/2026

We don't go on strike "for fun." This is a disruption, yes: for students, colleagues, and us. I'm taking a day of unpaid leave which I'll spend at a picket line to discuss with passers-by why it matters, and why we do it for the University and our students. Join us @nteunion.bsky.social! #USyd

1122

Ben Harrap @bharrap.bsky.social · 31/08/2026

Short-form video content idea: Random Walks The hosts walk and talk statistics/data science in scenic locations Bonus points if it's hosted by Monty and Carlo (expand your social media offerings @posit.co @hadley.nz?)

181

Ben Harrap @bharrap.bsky.social · 29/08/2026

I'm staying in Richmond over the weekend. I've never spent time east-side and my impression so far: not enough greyhounds

120

Ben Harrap @bharrap.bsky.social · 26/08/2026

How To Write An ABC News Article Without Getting Sued Written by Al Edge and Ed Lee

040

Reposted by Ben Harrap

tj mahr 🤘 @tjmahr.com · 25/08/2026

annoyed this never occurred to me. (it's a method for bootstrapping repeated measures data. text via Deen & de Rooij, 2020, doi.org/10.3758/s134...)

Balanced bootstrap The balanced bootstrap can be used to ensure that every individual appears exactly B times in the bootstrap samples, in contrast to randomly drawing bootstrap samples from the parent sample. Davison and Hinkley (1997) show that the balanced bootstrap results in an efficiency gain. For unbalanced longitudinal data, where some subjects have more measurements than others, the balanced bootstrap ensures that the average size of the bootstrap samples equals the (subject) sample size N. In the balanced bootstrap, rather than simply drawing at random, a matrix is made with B copies of the numbers 1 to N. This matrix is vectorized, randomly shuffled, and turned back into a matrix of size N × B (Gleason, 1988). Each of the columns of this latter matrix gives the indices of a single bootstrap sample.

2306

Ben Harrap @bharrap.bsky.social · 24/08/2026

Only a handful of Australians on the list and, well, I'm not surprised to see some of them there...

000

Ben Harrap @bharrap.bsky.social · 23/08/2026

The initial randomise function made a call to the Google Fonts API every time, but I knew people would have fun with the button (I did!) So I set all the fonts to be called on page load to avoid hammering the API 😅

050

Ben Harrap @bharrap.bsky.social · 23/08/2026

🎉New tool for RevealJS slide theming🎉 You can play with: - background colours - base font size - colour and typeface for text and code - aspect ratio It displays contrast info, to help keep things accessible, and outputs the corresponding scss! benharrap.com/tools/quarto... #rstats #Quarto

A screenshot of a revealjs slide theme editor. An example slide is presented with several options to change colour schemes, typefaces, and font size

3378

Ben Harrap @bharrap.bsky.social · 21/08/2026

No no, may I instead direct your attention to Q It seems to be the market research equivalent of SPSS, except the price tag is eyewatering at $4600/yr 😵 It can "Automatically find the most interesting tables"! At least it supports #rstats? www.qresearchsoftware.com/features#Tab...

qresearchsoftware.com

Features

Use the latest machine learning and statistical techniques. Smart systems and automation for popular methods and full support of the R language.

052

Ben Harrap @bharrap.bsky.social · 18/08/2026

Please, if you're thinking of writing your own "Yes But He Did Actually Plagiarize" blog post, read this thread first

011

Reposted by Ben Harrap

Olúfẹ́mi O. Táíwò @olufemiotaiwo.bsky.social · 16/08/2026

Plenty to reflect about re: the Arday affair, but I'm going to repeat this one thing. The playbook the right used here is effective even though it's not especially sophisticated, as it relies on the audience to do the dirty work. I've been calling it "Trojan horse propaganda", let's go with that. 1/

5763172017

Reposted by Ben Harrap

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 14/08/2026

For enjoyers of this thread on reasons not use ratio outcome variables in statistical analysis, I have a short blog post from 2022 about why dividing an outcome by something like population density doesn't always control for population density: elevanth.org/blog/2022/09...

elevanth.org

There Are No Magic Outcome Variables

I'm very sorry but I am going to write about statistics and causal inference again. I'd much rather be doing science. I'll make it brief. I was reading a preprint about the relationship between kinsh...

13811

Reposted by Ben Harrap

Aki Vehtari @avehtari.bsky.social · 13/08/2026

All four books I've co-authored are freely available online for non-commercial use: - Bayesian Workflow at avehtari.github.io/Bayesian-Wor... Links to other three books are in the quoted post 👇 (too many books to fit in one post!)

avehtari.github.io

Bayesian Workflow book: Website – Bayesian Workflow book

Website for the Bayesian Workflow book by Gelman, Vehtari, McElreath, et al. — case studies, code, and exercises in R and Stan.

8283133

Reposted by Ben Harrap

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 13/08/2026

These ratio outcome variables are a plague in the sciences. If your analysis treats 1/2 and 10/20 as the same evidence, then you have discarded uncertainty in the outcome. Model what you measure, not what you construct. Use the denominator as an exposure instead.

612237

Reposted by Ben Harrap

Amstat-American Statistical Association @amstatnews.bsky.social · 12/08/2026

Nominations are still open for the 2027 International Prize in Statistics. Help recognize influential contributions that have advanced statistics and data science. Deadline: Oct. 1 statprize.org/nominations/

016

Ben Harrap @bharrap.bsky.social · 12/08/2026

The Statistical Society of Australia has a new President... No, it's not Randy the greyhound (although he would've done a fine job) It's me! If you want to know a little more about me, I have the blog post just for you :) benharrap.com/post/2026-08... #statssky #datascience

benharrap.com

Get to know the new President of the Statistical Society of Australia (it’s me) – Ben Harrap

13778

Ben Harrap @bharrap.bsky.social · 10/08/2026

I watched someone stop briefly as they entered a coffee shop, to look at a poster "is that AI?" they muttered They didn't give any more time to the poster, they didn't care to understand its message or art It had that vibe about it and that was the first and only thing they considered

030

Ben Harrap @bharrap.bsky.social · 10/08/2026

The LLM writing 'quirk' I've adopted in the last few years, if you can call it a quirk, was not from LLMs directly, but the discourse on em dashes as indicators of LLM-generated text It made me actually learn the difference between hyphens, en dashes and em dashes, and start using them correctly!

170