Causal inference with observational data and unobserved confounding variables

https://doi.org/10.1101/2024.02.26.582072

A great introduction to this subtle but not so difficult topic. Ecologists miss opportunities to do causal inference. There are powerful tools that are mostly unused, but most will look very familiar. Often it just takes a bit of hard thinking and one or two tweaks to your model formula!

#rstats #ecology #research #CausalModelling #statistics

#correlation without #causation via #ColliderBias

a spurious relationship between the dependent (Y) & independent (X) variable can occur, when each effect a 3rd variable (C, #collider), and this 3rd is included in the linear model lm(Y~X+C)

#Rstats #CausalModelling #dataviz #ecology

Time for an #introduction.

I'm a cellist turned sociologist, researching #openscience as part of the Open and Reproducible Research Group (https://orrg.eu) in Graz.
My main research interests are #openaccess and #reproducibility of research, as well as expanding my methodological toolbox with #Stan and #CausalModelling.

I'm part of the #TwitterMigration and the social experiment proposed by @briannosek on actually moving academic twitter over to the fediverse.