rude

also, we read the books and thought about them in the old college

Obviously the best decade for television (especially if you include afternoon re-run culture).

Great for music and Bears' Super Bowls.

Bad for research design.

Bonica shows a problem with the primary RD design for later periods, which Hall & Thompson agree with. They conclude we lack statistical power to conclude anything about changes over time. (Note: the second paper in my first post is not an RD paper).

www.andrewbenjaminhall.com/hall_thompso...

I think these by Hall et al are the most convincing:
www.andrewbenjaminhall.com/Hall_APSR.pdf

onlinelibrary.wiley.com/doi/10.1111/...

Just for the record, though, and not defending any of these analyses that belong in the 1980s, the well identified studies we do have suggest that there are positive electoral returns to moderation.

I think Kevin is likely referring to a scenario in which you are being chased by your enemies

Alternative headline: "Eagles' Arrogance Trumps Bears' Incompetence in Race for Third Seed"

You mispelled Ron Guidry

I'm taking credit for you reading fiction in 2025.

You are living a charmed life.

My whole mental model of you is broken by the absence of the Wellness reporting from this list.

I mean, I see what you are saying.

But I'll be damned if I'm going to hand it to Adorno.

Why?

If I want to discover that I don't really understand what's going on in my proof, I can just try to write the text explaining the result.

Or did you have in mind that I would try and succeed?

I mean, is it really worse than Automatic for the People?

I would be disappointed if you did...

Suppose, for the sake of argument, that all CS majors were 1st gen and all art history majors were from well-connected families.

Even if CS actually provides better job opportunities,we'd observe CS majors having worse job outcomes, despite the causal effect going the opposite direction.

That seems likely to be true.

But my point is, you cannot even tell the sign of the effect, let alone its magnitude, from this sort of evidence.