New working paper up on SocArxiv! osf.io/preprints/so.... I use the 1940 Census and linked mortality records in combination with an IV-design to study the causal effects of racial segregation on longevity. I show that segregation reduces both Black and White longevity.

So grateful for this office view!

Excited to post a new working paper with @instrumenthull.bsky.social and Michal Kolesár: arxiv.org/abs/2511.03572 Will post a thread on it soon, but if you're interested in judge/examiner designs, I think you'll find this guide very helpful!

A huge congratulations to DOCTOR Haowen Zheng, who defended a fantastic dissertation on how changes in family formation and in the spatial distribution of jobs in US affects who moves, where they move to, & who benefits most from moving. Haowen starts a post-doc at Michigan's Stone Center in Jan.

Other academics are fighting (sometimes literally) one another to publish Nature papers. But I am just sitting here writing abstraction layers for probabilistic programming frameworks.

New @pnas.org paper constructing subnational estimates of internet and mobile adoption by gender, including gender gaps, for 117 low- and middle- income countries from 2015 through 2025. www.pnas.org/doi/10.1073/...

The pronatalists have done an excellent job of using bad demography to convince the legacy media that low birth rates are a crisis. They are using bad demography to whitewash their sexism, racism, Christian nationalism, classism, & anti-LGBTQ sentiments. It’s why my colleagues & I wrote this:

Both the Brown and Columbia agreements require the universities to report the test scores and GPAs of applicants, admits, and matriculating students by race *and color,* among other attributes. Wait, what? What's "color" doing in there? A sociologist weighs in. 1/9

the destruction of chattel slavery is one of the great accomplishments in our nation's history and the reason conservatives hate celebrating it is because doing so legitimizes the black counter-narrative of the united states

I want to highlight 3 things from this excellent report: 1) Women's reproductive labor is too often the solution to "population problems." 2) Low birth rates aren't a rejection of children & parenthood. 3) Supporting families is a worthy goal in itself, even if policies don't raise birth rates. 1/

Sad that this has to be my first post here, but the gutting of the FSRDC program is disastrous for demographic research.

The honorable mention goes to Haowen Zheng, “Diverging Trajectories: Gendered Income Dynamics Pre- and Post-Family Migration.” 2/3

An example of the disruption to science caused by the NSF’s 15% overhead dictate: At the University of Washington, all newly awarded NSF grants are now on hold indefinitely with no advance spending allowed.

Just posted updated version of our DID textbook! We now have drafts of all chapters, including the one on general designs! Now you can tell your friends still on X that they are DID-outdated :-) Happy easter for those of you that celebrate it. papers.ssrn.com/sol3/papers....

Feeling thrilled to have successfully defended my M.A. thesis today to advance to Ph.D. candidacy!

Our Annual Review of Sociology article on conceptualizing and measuring structural racism is now available. Get it while studying racism is still legal. @tyson-brown.bsky.social @pahoman.bsky.social (also, lots of great articles are in this volume). www.annualreviews.org/content/jour...

The biggest project I've worked on for the last chunk of years was just published. It asks, how big are US Black-white lifespan differences? This might seem like a narrow question. I hope to convince you by the end that there are answers you didn't anticipate. And I hope some of them will move you.

Very excited to have this paper with @franciscaantman.bsky.social Bruce Weinberg, and Sgeng Qu out on the (positive!) long run effects of the @aeainformation.bsky.social Mentoring program. www.nber.org/papers/w33689

these people are literally trying to destroy america’s future

Helpful NYT article showing 30 exclusion restriction violations for studies claiming covid is an instrument for their preferred x variable nytimes.com/interactive/2025/03/09/upshot/covid-lockdown-five-year-charts.html

My rough guess is that about 1/3 of young Democrats identify as LGBTQ+. Keep that in mind when thinking about what issues folks are labeling a "distraction" - for young Dems especially, LGBTQ+ issues are not remotely niche. scatter.wordpress.com/2025/02/24/t...

Needing a six-figure salary to purchase a home means many Americans are cut off from the primary avenue for building intergenerational wealth. www.nbcnews.com/data-graphic...

In spite of everything, my morning commute to campus is peaceful and the snow is pretty .

It's Monday, so over at the blog I wrote about two great new papers that take a swing at what you can and can't do with sibling data. asocial.substack.com/p/two-great-...

New study finds “counties more exposed to coverage of BLM protests by a TV station owned by Sinclair, a conservative media conglomerate, were *less* likely to support Republican candidates…” Suggests this may be due to Sinclair “raising salience of racial tensions,” papers.ssrn.com/sol3/papers....

31 faculty members from the NYU Sociology Department released this statement about the forceful suppression of peaceful protest (including the arrest of faculty and students) by the NYPD on April 22. Free speech and political dissent at universities must be protected. Get the cops off our campuses.

The nice thing about computing spatial weight matrices is that their computation is so slow and laborious that you have built in break time to surf the web while you wait.

serious answer: Jay Inslee

The recent revelation that (x1+x2)^ == x1*x2 in R is astounding. Still learning something new about the language 5 years in.

Excited to share RSF edited volume "Disparate Effects of Disruptive Events on Children" www.rsfjournal.org/content/10/1 along w/ jenniebrand.bsky.social & jasonmfletcher.bsky.social Impact of disruption usually varies by individual & contextual factors. Estimating one single effect might miss a lot.

Dear DEMOGRAPHERS 🙌 Following my bibliographic exploration of a *subjective* selection of demographic journals I want to run a study to explore the question more objectively. Please consider participating 🤝 forms.gle/5Eh7VSTvRWa3... (feedback welcome) #demography #AcademicSky #ScientificPublishing

That's a wrap on the first semester of the Ph.D.! Now, to get to all the non-class articles I accumulated this semester for my winter break reading list.

Your reminder that less than half of one percent of undergrad students attend Ivy League Institutions

Some good news to keep me motivated during finals week! My abstract was accepted to the Mobile Apps and Sensors in Surveys workshop 2024! It’ll be great to head to DC and learn about cutting age real time data collection. I’ll also be at the CIPHER conference right after.

When home heating prices are lower, fewer people die each winter, particularly in high-poverty communities. That's the punchline of my paper with Janjala Chirakijja and Pinchuan Ong on heating prices and mortality in the US, just published in the Economic Journal. 📉📈 academic.oup.com/ej/advance-a...

I think the biggest value-add of replicating published journal articles, aside from verifying results and assessing validity, is the opportunity to get ideas for robustness checks and intuitive graphical representations of model assumptions.

After having to improve my code efficiency to handle problem sets better, I am convinced that the Frisch Waugh Lovell theorem is the most overpowered trick for statistical programming.

New @PNASnews article w @je_nobles shows large adverse impact of COVID infection on infant health.  Adverse impact disappeared ~1 year earlier in zipcodes with high vaccination rates, suggesting vaccines prevented thousands of preterm births. www.pnas.org/doi/10.1073/... Funded by  NICHD_NIH, NSF

Imagine there's millions of dollars in ad revenues and billions of dollars in valuation tied to some railroad tracks and if you let all the worst people in the world scream racial slurs for a year, a train will divert and destroy all of it forever. www.nytimes.com/2023/11/24/b...

I feel like the true generational divide during thanksgiving is to spatchcock or not to spatchcock.

🚨 New paper forthcoming in Demography 🚨 Drawing on the weathering hypothesis, we document staggering racialized gaps across 3 measures of biological aging, showing that these gaps are linked to inequities in life course social & economic exposures, inc wealth. read.dukeupress.edu/demography/a...

This may be a really boring take, but since its release, Halloween (1978) has been consistently one of the best horror movies in terms of subtlety, cinematography, and suspense even compared to the recent boom in quality horror movies in the last 10 years.

___|/¯¯ difference-in-difference ¯\_(ツ)_/¯ indifference- in-indifference

I recently upgraded my MacBook to a much more powerful spec but it's still dying trying to run a linear model with 3000 fixed effects. So that's probably the wake up call I need to make my code more efficient :(

I've been replicating some data visualizations from different papers for a homework assignment and am really proud of this one! Also really useful to use nonparametric estimation to characterize treatment effect heterogeneity.