🐋 Menopause is an evolutionary mystery, but a new study on Orca whales is changing that. Researchers including Dan Franks from Biology at York found whale species experiencing menopause live around 40 years longer than those which don't: bit.ly/3TCbtqz #YorkResearch

Very pleased to see our paper taking a comparative approach to understand the evolution of menopause in toothed whales published today in nature . Couple of slides below talking through the key results but all the details here: nature.com/articles/s4158…

Contrary to what you might read, the BiSON approach can indeed include confounders. Also, BiSON (and STRAND) doesn't require the network to be any more well-sampled than previous animal SNA approaches. In fact it does better with less data than previous approaches (they're Bayes)

Postgrad recruitment time is approaching. As a working class kid from a deprived area, there’s no way I would’ve done a PhD under the current system (if I’d even done a BSc). Things I do to try to mitigate some of the systemic obstacles… 0/7

Bayesian FTW

Out today 'Postreproductive female killer whales reduce socially inflicted injuries in their male offspring' Led by Charli Grimes (not on twitter?).

Complete with data, code, DAG.

shorturl.at/grzT0

This is the new frontier in social network analysis. Super exciting to see it out in the world. Congrats Dr Jordan Hart!

I can say from experience that the R package is very user friendly, so be sure to check it out for your next network analysis!

Another wonderful study from this hugely talented group: Charli Grimes Darren Croft University of Exeter Lauren Brent 🐬Michael Weiss🐋 Dan Franks and Whale Research
Killer whale moms protect their sons from fights with other whales | Science | AAAS science.org/content/articl…

Brilliant to see BISoN now roaming out in the wild! 🦬 BISoN is a fully Bayesian approach to social network analysis that captures and propagates uncertainty through the entire stats workflow: besjournals.onlinelibrary.wiley.com/doi/full/10.11…

Great to see this finally out! We've necessarily spent some time showing why the SRI and permutations are often unsuitable for SNA, and it's feels good to be part of a solution.🦬

I spent far too long today messing with Quarto formatting options to make my thesis (which I haven't written yet) look pretty, and far too long tonight writing about it, but I hope this helps anyone who wants to make nice PDFs from Quarto in future: cameronpatrick.com/post/2023/07/q…

1/n It is ready! In this post, I walk through how to use renv, Docker, and GitHub Actions to automate a workflow that makes your R projects reproducible across time and space 🤖🤓.

The goal is to help you avoid the age-old problem of *dependency hell*:

haines-lab.com/post/2022-01-2…

Don’t let ChatGPT analyze your data.

(I look forward to 20k likes and getting roasted in the replies!)

'The really freaking part is when you hear about this going on in medical science'

From smbc-comics.com/index.php

Rishi Sunak justifies cutting university funding because they're 'full of people who don’t vote for us anyway.' mirror.co.uk/news/politics/…

Here are some ways of making your study replicable. (No, it’s not preregistration or increasing the sample size!) statmodeling.stat.columbia.edu/2023/06/22/her…

The last episode of #succession is just perfection.

DML is widely used in industry for causal inference. I often see it applied blindly and the results are dubious. Read this if you’re working on observational causal inference/DML