Robert (Bob) Kubinec

The 1 lesson from poli sci I would give ordinary folk:

Political ideologies *do not* give you a road map for what to do in office. A candidate's ideology is just one ingredient for what happens in office. Look at how they learn, what they've done so far, and most importantly, *who they represent.*

I'd be OK with this if we sent all that money straight to Ukraine.

Somehow I don't think that's going to happen, though...

Kevin Hassett: decent, amiable, authoritarian apologist. It doesn't take evil; it just takes spinelessness. Read the whole thing in today's Morning Shots: www.thebulwark.com/p/the-banali...

You want to overthrow democracy, fine. But you're not doing it in a Penske truck without facing the consequences.

💯

This line in the piece is exactly my experience with coding with AI (and I've had some notable successes)

Everyone I know selects on the DV so I wouldn't know, sorry

This is why many researchers only use OLS.

We're recruiting a new tenure-track faculty member to MIT Sloan!
apply.interfolio.com/170116

We're looking for someone doing research in quantitative marketing, broadly considered. Faculty in our group work on economics of privacy, mathematical psych, networks, and applications of ML & AI.

NEW JOB POSTING - #DataScientist / Senior Data Scientist at @stonelivinglab.bsky.social - come help us make a wealth of climate resilience and environmental data accessible to everyone from scientists to policy makers to the public! See employmentopportunities.umb.edu/boston/en-us... for details

Open Rank Professor

University of South Carolina is hiring 3 #polisky lines:

American Judicial/TT uscjobs.sc.edu/postings/192...
Open field/TT uscjobs.sc.edu/postings/192...
IR/Open Rank uscjobs.sc.edu/postings/192...

Happy to chat about the department/life here in Columbia. Please apply!

hmmm have they tried economy plus? the extra legroom would be helpful for bulky research designs

It's a great rule, although for me it's about 12 months.

That is precisely the point of the article above--assigning a causal interpretation to measurement models.

The point is that it doesn't matter which is a cause and which is an effect. It only matters if they are correlated with being an axe murderer.

In other words, I don't need to know the causal process that produces axe murdering, I just need to know that if I observe A, my posterior probability for B should increase (decrease).

The problem with forcing a causal interpretation on a measurement model is that you can measure well without knowing causality.

Suppose I want to measure being an axe murderer. Then 1) purchasing an axe (a cause) and 2) the presence of dead bodies in their car (an effect) are both valid measures.

Yes Trump is making America more authoritarian. But authoritarianism is a way of imposing preferred policies and keeping a preferred leader in charge--it doesn't inoculate against dissent or anger among the population.

Authoritarian regimes are still vulnerable, and smart dictators know that.

Trump's poor choices for cabinet members are coming home to roost. Allying with RFK Jr. may have helped him win the November election, but canceling vaccines (which are still broadly popular) while also cutting job growth & stoking inflation is not a great recipe for political success.

Health secretary RFK Jr. shuts door on U.S. investment in mRNA vaccine research Vaccine experts and people steeped in pandemic preparedness expressed horror at the news that RFK Jr. has shut the door in U.S. investment in mRNA vaccine research.

WHAT, WHAT, WHAT ARE WE DOING!!!
“Vaccine experts and people steeped in pandemic preparedness expressed horror at the news.” www.statnews.com/2025/08/05/m...

Law firms, universities, media companies, and now banks. If people don’t understand what’s happening to our country, it’s because they don’t want to.

And follow me for obscure puns about panel data statistics

This is the thing -- a lot of what people see as "caving" isn't necessarily caving per se but rather elites who are ideologically supportive of the MAGA agenda (or the parts of it they like) putting their resources to use on his behalf.

fighting gerrymanders can seem hopeless, but there is a solution. big moment right now for people to see districting and the two-party system as inherently linked, and a relic of our archaic electoral rules. congress has the power to ban partisan gerrymandering and establish multi-member districts.

This is a cool approach, of course, but some difficult measurement issues with satellite data. I'd be more interested in an approach leveraging something like SafeGraph or maybe survey data.

what's the smr paper?

The Norway standard deviations are on the metric system unfortunately

Age period cohort is so back

Yes I'm posting this from my phone, stop judging me

BREAKING: new study just dropped that shows compelling evidence that cell phone bans cause student grades to modestly increase.

Grade increases are largest among low performing students.