But what about the cat litter??!!

This year I am growing four types of pumpkin, including Atlantic Giants and Pacific Giants. 🎃

Epistemic Virtues for Science in the Age of Automation
statmodeling.stat.columbia.edu/2026/04/14/e...

arxiv.org/pdf/1503.01603 Until fairly recently, we’ve been more comfortable stating the limits/threats to inference, and far less comfortable being explicit about the assumptions that license our generalisations. Assumptions feel riskier: once written down, they can be scrutinised. #causalinference

Also...Experimental studies can also be misinterpreted. E.g., multi-stressor experiments often log-transform Y without considering how this influences the meaning of interaction terms in regression models: tethys.pnnl.gov/sites/defaul...

I've lost count of the number of times I've reminded students that science is value-laden: what scientists chose to study & how they study it is influenced by a whole series of other commitments.