Will Lowe | Lightnews

Light up
your news

About Privacy Terms Help

Will Lowe

Will Lowe

@conjugateprior.org

3.4K followers 750 following 10K posts

Señor Research Scientist, NPC at the Hertie School in Berlin 🇩🇪 via Princeton, Mannheim, Edinburgh and a bunch of other ivory towers that will probably be billiard balls and decorative boxes by the end of the decade.

Rome Statute appreciator. .. more

Computer science 32%

Political science 28%

more

Posts Replies Media Videos

Pinned

Will Lowe @conjugateprior.org · Apr 15

For Monty Hall problem aficionados: a #causal DAG, with explanation in the alt text.

Who knew that the M in M-bias stood for Monty?

A DAG for the Monty Hall problem.

G, the door Monty opens, is a function of your first door choice D_F and of where the car is C.

Conditioning on G, which happens because it determines your door choices when you have the opportunity to switch, induces collider bias between C and D_F, such that they are now less likely to be equal. (Why? because G cannot be C and G cannot be D_F. These are in the same 'direction' so, conditional on G, D_F is less likely to be C.)

You decide to switch conditional on G, which you just observed, and D_F, because that defines what you can switch to.

Conditional on G you now know D_F is less likely to be equal to C. So you switch.

Note: In the original problem, W is a deterministic function of C and D_S so it's slightly cheeky to put it in a DAG, but I'm going to do so anyway. The argument doesn't require it, but if you want to think informally about opening paths to W from D_F via C, it's helpful.

That makes this problem a variation on M-bias (or a clipped wing butterfly bias, if you prefer insects to goats).

Reposted by Cyrus Samii

Will Lowe @conjugateprior.org · 8h

Folks, some news. No, not that kind of news – what do you think this is, LinkedIn? It's this:

1. The applied causal graphs workshop deadline is 28th Feb. so get your abstracts in and hang out with us in Potsdam this May. Form and description is below

2. @dagophile.bsky.social is giving a keynote 🥳

Applied Causal Graphs 2026

applied-causal-graphs.de

Will Lowe @conjugateprior.org · 8h

You too! 🤜 🤛

Will Lowe @conjugateprior.org · 9h

In my experience there are no prizes for either. But maybe the ideal bet-hedging strategy is Claude Shannon's: ask a bunch of questions and then immediately answer them.

Will Lowe @conjugateprior.org · 9h

A perfectly good definition, but then the 'challenge' of asking a good question is a lot more like the challenge of "roll ten sixes in a row" (difficult because unlikely) and less like the challenge of "make ten free-throws in a row" (difficult because superior skill is required).

Will Lowe @conjugateprior.org · 9h

Could well be.

That retrospective quality though. It makes "the challenging thing is coming up with good questions" sound uncomfortably like "the challenging thing is to be born with the right parents" 🙃

Will Lowe @conjugateprior.org · 9h

Can't fault this answer.

ON WHAT THERE IS*

A curious thing about the ontological problem is its simplicity. It can be put in three Anglo-Saxon monosyllables :
"What is there?" 'It can be answered, moreover, in a word
— "Everything" — and everyone will accept this answer as true.

Will Lowe @conjugateprior.org · 9h

I don't think I disagree. At least this focusses on what questions are actually for (and admits that the answer is not "to be answered correctly"). Still, I fear there's not a lot of operational guidance to be gleaned from knowing that good questions are ones that help discover truth.

Will Lowe @conjugateprior.org · 9h

More Papal Choice theory, with a special guest appearance by Ken Arrow himself.

Vox Populi, Vox Dei, Vox Sagittae on JSTOR

Forrest Maltzman, Melissa Schwartzberg, Lee Sigelman, Vox Populi, Vox Dei, Vox Sagittae, PS: Political Science and Politics, Vol. 39, No. 2 (Apr., 2006), pp. 297-301

Will Lowe @conjugateprior.org · 10h

aha, a variety of

4. [barely suppressed excitement] this is one of the nine questions I prepared extra slides for.

Will Lowe @conjugateprior.org · 10h

Questions as a variably effective - a.k.a. variably good - tool for revealing (or offering) assumptions and forcing theory construction seems like a decently instrumental theory of questions to me, though it makes being a good question community-specific, retrospective, and unrelated to the answer.

Will Lowe @conjugateprior.org · 10h

That's kind of what I mean by 'out in the pragmatics'. When there's a shared understanding about what sorts of quantities are relevant to a topic then sure, you can pitch to adjust that set, but you have to do more than mention the other stuff, e.g. offer a theory that motivates now caring about it.

Will Lowe @conjugateprior.org · 10h

An off-the-cuff taxonomy of upstream reasons for 'great question'
0. omg I totally missed that (rare)
1. you have seen the thing that I am currently having trouble with (common)
2. you have reformulated the problem in a way that allows you to ask about something my formulation can't see (useful).

Will Lowe @conjugateprior.org · 10h

I'm increasingly thinking of this as, at best, metonymic in that the question itself has little directly to do with why I like a paper. In the same way as someone saying "that's a great question" is not usually commenting on the question, but on something far upstream of it, out in the pragmatics.

Will Lowe @conjugateprior.org · 10h

"The challenge is really coming up with good pictures"

Will Lowe @conjugateprior.org · 10h

[ahem] rumbled.

It was indeed stolen shamelessly from your post. I thought about reposting yours but if I remember rightly it was about something much more useful which would have revealed my Sunday navel-gazing for what it is.

Will Lowe @conjugateprior.org · 10h

Reasonable, but isn't being able to ask the question downstream of (and easier than) actually reframing the problem? Maybe there's no pressure to reframe until a question arrives.

Either way, whether a question is good is only learned later and is unrelated to the difficulty of coming up with it.

Will Lowe @conjugateprior.org · 12h

Eine Montessori-Schule an dem Syringenweg?

Berlin still not beating the allegations.

Will Lowe @conjugateprior.org · 15h

People say this quite often and I'm never sure know what to make of it. What's a good question anyway? One that's answerable? One that isn't? One that makes you ask more questions? (regress much?) One that creates other good things as a side effect that are somehow easier than finding the question?

"The challenge is really coming up with good questions." Guido Imbens, Economic Sciences Prize.

Will Lowe @conjugateprior.org · 15h

Eh, maybe?

The late 90s had some odd moments

Will Lowe @conjugateprior.org · 15h

I'm finding the whole premise of this Alternative Career of Andrew Eldritch highly implausible

Andrew Eldritch on stage

Will Lowe @conjugateprior.org · 15h

TIME NEDS, REFORM?

This MCU / Dr. Who crossover looks terrible.

Will Lowe @conjugateprior.org · 17h

uff, "that is"

Will Lowe @conjugateprior.org · 17h

Obviously I'm well out of area here, but I can imagine filling in Samii's steps 1,2 and 3 for normative questions also, albeit this formulation might look rather too much like the 'conceptual engineering' that are so inexplicably popular right now.

Will Lowe @conjugateprior.org · 1d

Everybody needs a sign to tap and this is mine:

"You have to give a shit!"

Anton Strezhnev @astrezh.bsky.social · 1d

Our job isn’t to be advocates - we’re hardly well-positioned for that - but like…you have to give a shit!

Reposted by Will Lowe

Anton Strezhnev @astrezh.bsky.social · 1d

Our job isn’t to be advocates - we’re hardly well-positioned for that - but like…you have to give a shit!

Reposted by Efrén O. Pérez, Will Lowe, Alex Braithwaite

Anton Strezhnev @astrezh.bsky.social · 1d

I keep coming back to @cdsamii.bsky.social essay on the “problem-centered” over “puzzle-centered” research paradigm and I can’t help but feel like so many problems with social science methods boil down to this cyrussamii.com?p=3682

The “problem solving” approach and social science methodology – Cyrus Samii

Will Lowe @conjugateprior.org · 1d

omg how did I not realize that was a Lenin line?

Will Lowe @conjugateprior.org · 1d

Exactly.

"Fewer, better, and higher paid" is my campaign slogan, perfectly calibrated to alienate all possible supporters. Right up there with "cost benefit analysis is a real thing and you should do it harder and in public"

Will Lowe @conjugateprior.org · 1d

This parenthetical remark.

When cost-benefit analyses, investment, and leadership decisions are no much more than rhetorical forms, floating in a sea of euphemism, it's not my waist I'm imagining tightening my belt around.

Reposted by Will Lowe, Federica Genovese, Nicolai von Ondarza

Ulrike Franke @rikefranke.bsky.social · 1d

The only comment I have on the Rubio speech is that I don’t want to spend too much time on the Rubio speech.
Nothing is changing with regards to US policy. We don’t need to do a line-by-line analysis.