Ang Yu

Sage Journals: Discover world-class research Subscription and open access journals from Sage, the world's leading independent academic publisher.

Someone else’s perspective on working in tech (and in fact at Meta) that I just came across: journals.sagepub.com/doi/10.1177/...

Maybe there's something in McDonald's coffee today, but I find myself totally amazed by the basic premise of statistics again on a Sunday morning. It's nothing short of a miracle that we have a way to know the world with mathematically guaranteed unbiasedness and efficiency.

Causal Mediation Analysis Cambridge Core - Research Methods In Sociology and Criminology - Causal Mediation Analysis

Something to look forward to 🤩: www.cambridge.org/core/books/c...

I just realized the two ASAs (sociology and statistics) will have their annual meetings in the same place (Chicago) and at the same time in 2027 👀

I had a wonderful time visiting the casual causal group at Berkeley today!

Thank you so much Julia for the wonderful summary (and advertisement 😉) of our paper!!!😁

I’ve been elected as the student representative on the ASA (the soc one, not the stats one) Methods Section council (after two failed runs in previous years😂)! I'm looking forward to contributing to the methodology community over the next 2 years in this role!

New PB in half marathon!

Thanks! And yes I know! 😄

Belated announcement: I joined Meta as a research scientist in May! Happy to meet up or catch up if you are in the Bay Area! 😁

Yu and Elwert Develop Approach to Better Understand Causes Behind Group Disparities IDS-supported research by sociologists Ang Yu and Felix Elwert has identified a statistical approach that allows social scientists to measure causes of group disparities.

IDS affiliate Felix Elwert of @uwsoc.bsky.social and @ang-yu.bsky.social have uncovered a new approach to measuring the causes of inequalities between groups. Read about it at ids.wisc.edu/2025/07/08/y...

What would you replace pumpkins with?

cdgd: Causal Decomposition of Group Disparities The framework of causal decomposition of group disparities developed by Yu and Elwert (2025) &lt;<a href="https://doi.org/10.1214%2F24-AOAS1990" target="_top">doi:10.1214/24-AOAS1990</a>&gt;. This pac...

The cdgd package now accommodates sampling weights! As always, let me know if you run into any issues using the package.
cran.r-project.org/web/packages...

Groundbreaking paper in JASA

These estimands quantify the counterfactual association between an outcome and a continuous covariate under interventions on an intermediate variable. In social stratification, they offer a principled framework for reassessing the great equalizer thesis and Mare’s school transition thesis.

Sage Journals: Discover world-class research Subscription and open access journals from Sage, the world's leading independent academic publisher.

My paper with Jiwei Zhao is now online in Sociological Methodology! In this article, we propose a new class of counterfactual slope estimands along with corresponding estimation techniques. journals.sagepub.com/doi/10.1177/...

I just defended my dissertation! 😌

www.annualreviews.org/content/jour...

Thrilled to receive the Best Student Paper Award from the CIES East Asia SIG for our paper published in Social Science Research with @ang-yu.bsky.social! It’s my first time at the conference, and I’m loving the vibrant community.

David you are too kind😆😆

Nonparametric causal decomposition of group disparities We introduce a new nonparametric causal decomposition approach that identifies the mechanisms by which a treatment variable contributes to a group-based outcome disparity. Our approach distinguishes three mechanisms: group differences in: (1) treatment prevalence, (2) average treatment effects, and (3) selection into treatment based on individual-level treatment effects. Our approach reformulates classic Kitagawa–Blinder–Oaxaca decompositions in causal and nonparametric terms, complements causal mediation analysis by explaining group disparities instead of group effects, and isolates conceptually distinct mechanisms conflated in recent random equalization decompositions. In contrast to all prior approaches, our framework uniquely identifies differential selection into treatment as a novel disparity-generating mechanism. Our approach can be used for both the retrospective causal explanation of disparities and the prospective planning of interventions to change disparities. We present both an unconditional and a conditional decomposition, where the latter quantifies the contributions of the treatment within levels of certain covariates. We develop nonparametric estimators that are n-consistent, asymptotically normal, semiparametrically efficient, and multiply robust. We apply our approach to analyze the mechanisms by which college graduation causally contributes to intergenerational income persistence (the disparity in adult income between the children of high- vs. low-income parents). Empirically, we demonstrate a previously undiscovered role played by the new selection component in intergenerational income persistence.

My paper on causal decomposition of group disparities is out in the Annals of Applied Statistics! If you are looking to explain group differences, this is likely the methodological framework for you! doi.org/10.1214/24-A...

No sir, it’s logit outcome group 😉

And social science is the Titanic? 😆

Disparity analysis: a tale of two approaches Abstract. To understand patterns of social inequality, social science research has typically relied on statistical models linking the conditional mean of a

academic.oup.com/jrsssa/advan...

Dr. Felix Elwert, Professor of Sociology and Biostatistics at the University of Wisconsin (@uwsoc.bsky.social), presents "Three Paths to Equality" as this year's Schuessler Lecture. Join us next Friday (2/14) in the SSRC at 2pm. Reception to follow.

A Calibrated Sensitivity Analysis for Weighted Causal Decompositions Disparities in health or well-being experienced by minority groups can be difficult to study using the traditional exposure-outcome paradigm in causal inference, since potential outcomes in variables...

onlinelibrary.wiley.com/doi/10.1002/...

Vilas awards for Felix Elwert and Michael Massoglia Two Sociology faculty have recently received Vilas awards. Congratulations to Felix Elwert for his Vilas Distinguished Achievement Professorship and Michael Massoglia for his Vilas Mid-career award! S...

Two Sociology faculty have recently received Vilas awards. Congratulations to Felix Elwert for his Vilas Distinguished Achievement Professorship and Michael Massoglia for his Vilas Mid-career award! sociology.wisc.edu/2025/01/07/v...

Got it. I assumed it was about how to present *all* coefficients of a regression. Also, in observational settings, if an estimand can be directly captured by a coefficient in a regression, the regression is likely too restrictive (to capture an ATE by a reg coef, one needs to assume constant effect)