Scott's Mixtape Substack

Remix release day

scott cunningham — Tue, 25 Aug 2026 08:01:03 GMT

Well today is it. Today is the day I start classes at Baylor, after being gone a year to Harvard, and today is also the day that my new book Causal Inference: the Remix, comes out in the United States.

Why should you get this book? Is it better than the Mixtape? If so why?

So, first off, you can still get the mixtape. The online mixtape has just moved he…

How and why I am using hooks (part 1)

scott cunningham — Mon, 24 Aug 2026 12:30:06 GMT

I am, as readers know, enthusiastically using agents for empirical research. And probably you’ve noticed over the last several months low key frustration at the same time. A feeling that something is not quite right, but that I cannot quite put my finger on what that is, creating some anxiety in me. But it’s not just anxiety — because I cannot settle,…

“Memory is the residue of thought” and fixing that when working with AI agents

scott cunningham — Thu, 20 Aug 2026 10:05:55 GMT

Yesterday was day one of a two day new faculty orientation at Harvard Kennedy school. Readers may recall that I did a one year stint at Harvard’s “government” dept for 2025-2026, and that now I am doing another one year at Harvard, but this time at Kennedy, which is “the other gov school”. All in scare quotes because I genuinely find it a little confusi…

Online version of The Remix is now here

scott cunningham — Mon, 17 Aug 2026 07:40:18 GMT

My new book is called Causal Inference: the Remix and it comes out August 25th. Like my first, its publisher is Yale University Press. And Yale, as with the first, has been gracious enough go agree to there being a free online version. And you can find that here.

I wanted to get something up now as soon as possible, so it’s still a work in progress. But…

Long overdue closing tabs on a Saturday morning

scott cunningham — Sat, 15 Aug 2026 12:35:37 GMT

Oh what fool I’ve been. I don’t even remember when my last “closing tabs on a Saturday” was but it was long enough for this to happen. 251 links!

Well I’m not going to post 251 links on here. But I am going to work backwards to 20-40 of them and just try my hardest to close a hundred more. So wish me luck! And wish me luck next week as I travel to Waco …

Which stages of research production should we let AI Agents loose

scott cunningham — Fri, 14 Aug 2026 10:19:49 GMT

Six months ago, it was both original and true to say that AI agents were capable of autonomously, start to finish, producing a manuscript of contemporary quantitative empirical research. Many of us had either done it ourselves or seen it done on videos by someone else, but you could prompt Claude Code or Codex to come up with its own research question,…

The Art of Loving

scott cunningham — Thu, 13 Aug 2026 12:15:45 GMT

On Monday night I attended Olivia Dean’s concert at the Garden in Boston. Me and my friend had floor seats, and once the stadium was nearly full, I remarked to them that I couldn’t remember the last time being in such a place where I was so far outside of the distribution with respect to demographics. If I had to guess, the average age was probably 25-…

Will AI revive the Labor theory of value

scott cunningham — Tue, 11 Aug 2026 12:12:52 GMT

I teach history of economic thought at Baylor every year. I asked to have it handed down to me because of this strong desire I have towards community and story. My connection to community and story is deeply rooted in some combination of my personality but also my spirituality. And history of economic thought fits into it because basically without inter…

My strategy for talks on agents for research

scott cunningham — Mon, 10 Aug 2026 14:05:30 GMT

My preference for teaching new concepts in statistics and causal inference to people is like this.

Discuss history of thought
Lead with narrative
Show numerical examples
Maybe something then interactive like a video
Then and only the get into the technical rigor

This has worked well for me in both university courses and workshops, though it does slow thin…

How to improve understanding with Claude Code

scott cunningham — Fri, 07 Aug 2026 11:15:20 GMT

Paul Goldsmith-Pinkham posted on his substack feed the other day a link to this interesting blog post by Geoffrey Litt called “Understanding is the new bottleneck”. I encourage you to read it. It’s short but sweet as the old kids used to say.

I thought it was interesting because it edged me a bit towards a bumpersticker-style thing I’ve been saying the…

Demographic differences in online dating

scott cunningham — Thu, 06 Aug 2026 11:56:18 GMT

I recently found some interesting surveys over at the Pew Center about online dating in America. They are dated somewhat as several of the surveys appeared to have been run during Covid, which I think is an independently interesting era for studying marriage. A lot of things happened during Covid. There was increased worries of infection, quarantines,…

Episode 16: Landing the Plane

scott cunningham — Tue, 04 Aug 2026 13:59:08 GMT

After 16 episodes of working with Claude Code to study the effect of abortion clinic closures on marriage rates using a natural experiment, so to speak, in Texas called "House Bill 2" and continuous difference-in-differences with a treatment of "change in distance", Caitlin and I start our descent. Caitlin shares her own thoughts about returning a poisson estimator, and we continue to wrestle with the implications of differences in econometric estimators and interpretation, as well as what we are willing to live with. Plus a lot of discussion about the implications of AI for our lives as researchers manages to creep in its pretty little head!

Hidden Curriculum (August 5th and 6th)

scott cunningham — Fri, 31 Jul 2026 15:29:40 GMT

Greetings again! One more nudge before we get going: the Hidden Curriculum workshop with Mark Anderson and Dan Rees is now less than a week away — Wednesday and Thursday, August 5–6, 2026, running about four hours each day starting at 7pm Beijing / 8pm Tokyo (7am EDT). If you’ve been on the fence, this is the moment to grab a spot.

I wrote at length a fe…

Prose is not constraints and other things I learned from Paul Goldsmith-Pinkham's NBER talk on AI in empirical research

scott cunningham — Thu, 30 Jul 2026 10:24:56 GMT

By mid-2026, I think many researchers understand that coding agents like Claude Code and Codex have dramatically reduced the marginal cost of producing empirical research. And of those who understand this, a growing proportion of them understand that the cost of verification has not fallen proportionally with it. The former seems to have fallen and req…

Sex recessions, rise in casual sex, online dating, declining trust, screening and other changes in US sex patterns

scott cunningham — Tue, 28 Jul 2026 14:04:51 GMT

I always wanted to be the Kinsey of economics who specialized in sex research. But, I sort of went down narrow side roads and then more or less stuck at those narrow topics for decades. The economics of internet mediated sex work and then later decriminalization of sex work were the two dominant ones, though I also wrote two papers on abortion policy, …

Which research mistakes matter for human capital?

scott cunningham — Mon, 27 Jul 2026 10:53:24 GMT

I had a coding error once in a paper that I only discovered during the R&R stage. The second I saw it, my stomach sank. I immediately recognized the kind of error — a classic Stata one — and pretty much could see how bad it was going to be for me. And sure enough, fixing it changed the results from significant and interesting to not significant and no…

Books are here!

scott cunningham — Thu, 23 Jul 2026 09:45:38 GMT

Well well well, what have we here. I left the apartment yesterday, walked downstairs and saw 50 copies of my book waiting for me all boxed up.

Countdown to the Remix

scott cunningham — Wed, 22 Jul 2026 12:01:15 GMT

I’ve been a little quiet about the new book coming out. It’s called Causal Inference: the Remix, and it comes out August 25th with, like before, Yale University Press. It’s around 800 pages long, and similar and different than the Mixtape, which always be for my baby.

I will struggle to call it the Remix as a result of my affection for the Mixtape brandi…

Stale code, deprecation, and the canon

scott cunningham — Tue, 21 Jul 2026 15:45:15 GMT

I think I first noticed I needed a structured workflow with using Claude Code when during a long project the analytical sample went from having N observations to N+k observations. It had no explanation as I had been consistently verifying results, but it was something about the inter-temporal results, not the contemporaneous results, where I just notic…

Hidden Curriculum Workshop (August 5-6th)

scott cunningham — Mon, 20 Jul 2026 11:33:32 GMT

Greetings! I want to share that we have a workshop coming up in a couple weeks called “The Hidden Curriculum“ taught by Mark Anderson and Dan Rees, and I think there are great reasons for you to consider coming. We’ve done this workshop many times now, not counting its embedding in the CodeChella Madrid conferences we’ve held so far, and Mark has been t…