"Schrödinger's" Inference

[will-ball]

Hi Lucy and Malcolm - first of all fantastic resource. I've already found reading this work useful so hats off for your efforts.

This isn't a particularly substantive point about content really, but rather a question of framing. The issue of ambiguous causal language & denials of causality in reporting observational studies is really important. I know it's often characterised as 'Schrödinger's' inference, but I think it's worth raising that Erwin Schrödinger isn't a benign individual to reference and it may be better to avoid doing so in future.

There's been growing recognition of his problematic behaviour in recent years, and academic institutions have taken steps to distance themselves from his personal legacy.

Of course I leave it up to you.

Thanks, Will

[malcolmbarrett]

Thanks for sharing this. I was unaware of this and know virtually nothing of his life. We'll have to think about this.

Of course, this is hardly an homage. It's a quote from someone else referencing his famous thought experiment. I don't know if there's a good way to convey that without a direct reference. If we can, though, I have zero attachment to using his name specifically.

Maybe something like "quantum" inference? I don't know a good way to refer to the particular problem in the thought experiment, though.

I guess another way is to leave the quote from the paper lie but change the section header to something like "casual inference" and use a literal description of this problem, e.g. a state of uncertainty as to the goal

[LucyMcGowan]

for what it is worth, I think I like your last option! I am obviously partial to the term "casual" inference but also think it fits nicely here. Makes sense that we can leave the quote as is, since that is what was said.

[malcolmbarrett]

It's an inexact analogy, anyway, so maybe that's better from a writing perspective as well, although I do appreciate having a term to succinctly refer to the problem of doing but denying causal inference

[malcolmbarrett]

"Causal Denial" 😆

[will-ball]

Thanks for taking this on board - absolutely happy for you to keep the quote and I take your point about this being different from naming a building after the man.

I sometimes liken it to 'having your cake and eating it too', although there's no snappy way to refer to that.

'Causal Deniability' perhaps?