Data & Science with Glen Wright Colopy

The episode contains a handful of genuinely interesting ideas - probability models as 'fake worlds,' counterfactual reframing of distributions, the semiotics-statistics bridge - but they surface slowly amid long tangents about academia, publishing politics, and education philosophy that add little informational value for a practitioner.

“you can think of these, um, probability models, okay, as fake worlds. Now, because it's an abstraction, you can learn everything you want about that world”

“the biggest predictor is where they went to graduate school. Well, that's what I learned to do in graduate school. So that's how I'm going to do your analysis”

The Peircean semiotics frame applied to statistical modeling is genuinely uncommon, and the Friston free-energy principle connection to statistical reasoning is an unusual angle, but much of the episode recirculates ideas already established in Gelman/Greenland circles without extending them in a notably contrarian direction.

“if organisms survive they've actually somehow figured out how to implement uh, scientific statistics”

“you started out a long time before I took statistics. I studied something called semiotics, which they call the theory of science, or better today, theory of representations”

O'Rourke is a credible practitioner with real exposure to elite researchers (Rubin, Cox, Gelman, Greenland) and genuine applied history in clinical meta-analysis and regulatory work, but he is not a widely recognised name and the conversation does not surface work done at significant scale or impact.

“When I first started in statistics back in the mid-80s, I was at, uh, the University of Toronto, and I started working with some clinical researchers and I worked out for them how to do meta analysis”

“Don Rubin told me is that whenever you're learning a new area in statistics, go back to the very early papers”

The conversation is overwhelmingly abstract and philosophical; concrete anchors are limited to anecdotes (mid-80s Toronto, Duke teaching, a researcher insisting on ANOVA) with almost no quantitative data, named studies, or dollar figures to ground the claims.

“the whole university, all the statisticians and biostatisticians were very upset at me and thought I was being a complete fool for showing people how to do meta analysis”

“if the model is correct, the confidence in low coverage is exactly 5%”

The host demonstrates genuine intellectual range - invoking Hume, KL divergence pedagogy, and cross-referencing prior episodes - but consistently lets the conversation drift into long mutual monologues about academia and publishing without pressing the guest for harder evidence or productive disagreement.

“And on the issue of the future orientation. So for example, we have, um. Many people are familiar with the problem of induction that David Hume talked about”

“Do you think science has become too statistics focused?”