To give an idea of the status quo, the customary tactic in animal social networks is to not only aggregate across time but also to divide out the sample size by using various ratio outcome variables. If we could get ppl to stop using ratio variables, that alone would be a victory.

And, huge thanks to Sofia M. Pereira & Judith von Nordheim for the paintings!

This project directly grows from our previous work on causal inference for animal social networks (journals.plos.org/ploscompbiol...). Certainly not necessary to have read it, but might make the whole thing more digestible

Together with @danielredhead.bsky.social, @rmcelreath.bsky.social , and @primbehavecol.bsky.social!

We’d love to hear what you think! Please do not hesitate to reach out either here, or at [email protected].

www.biorxiv.org/content/10.1...

We then develop a Bayesian stats model that takes raw behavioural observations as input, and can recover the generative model’s parameters. We finally illustrate how to use our models to ask causal questions (eg effect of sex on behaviour)

Instead, we first develop a generative model describing what pairs of individuals (dyads) do over time. Dyads differ in how long they stay in, and how they transition between behavioural states. This results in varying sequences of behavioural interactions that are structured like empirical data.

Typically these data are aggregated over time & summarised as ratio indices to create SNs, losing obs-level info & ignoring inferential uncertainty. They are then often plugged in regression models that don't directly reflect the causal process of interest—opening the door to confounding & ambiguity

To test causal theories about why animal societies look the way they do, we often rely on observations collected as time-series. For example, with focal-animal sampling, observers follow individual animals and record what they are doing, and who they’re interacting with, in continuous time.

I still haven't read it properly, but perhaps that could help:

bsky.app/profile/chel...