Hannah Waight

the USA built a system of higher education so good that smart/rich people from across the world came here, spending billions to learn here, subsidizing education for Americans while spending money to live in our cities and towns. our government arbitrarily decided we should stop doing that

Thank you @thomasdavidson.bsky.social and Danny Karell for all your hard work on this! Really thankful to have been a part of this special issue!

A screenshot of the Sociological Methods & Research website showing the special issue title

I’m delighted to share that the August 2025 special issue of Sociological Methods & Research on Generative AI is out now. Along with my co-editor, Daniel Karell, we put together this issue to build on the conference we organized last year.

Here's a thread on each of the ten papers:

Thank you for fighting for our public land @wyden.senate.gov !

✨New✨ postdoc opportunity to collaborate with @mollycopeland.bsky.social and myself on an exciting project on geography, community, and mental health at @ndsociology.bsky.social. Happy to talk to anyone interested. Please resky (or whatever retweeting is called here)!
apply.interfolio.com/169206

our sources included Russian, English, and Ukrainian

Excited to share “Quantifying Narrative Similarity Across Languages” was just published! Amazing work by my wonderful coauthors — read more below!

and by you!! @jasong.bsky.social really enjoyed working with you on this paper

Our replication package is available at this link: doi.org/10.7910/DVN/...

This article will appear in print in the August SMR special issue on generative AI. Special thanks to Daniel Karell and @thomasdavidson.bsky.social for organizing this special issue! All the articles in the issue are fantastic, please read them all.

Our paper shows that large language models can be used for complex labeling tasks unattainable by previous measures. We furthermore find fine tuning to hold particular promise for these types of tasks. We also provide a framework for out-of-sample validation of our rare event estimand.

The column recall holdout estimates the percent of 47 holdout pairs recovered by each estimator. The Precision score is the percent of predicted (out of sample) positive cases which were labeled as TPs by human coders. The F1 score is the harmonic mean of two scores. The column total predicted pairs includes the total number of pairs each estimator predicted to be “same claim, same subject” pairs. The bolded estimates indicate the highest performing estimators across all types by metric.

We benchmarked the performance of our approach against a range of existing measures of related estimands. Our measure outperformed these relevant alternatives. We show the performance of our large language model estimators (“SBERT-LLM”) versus considered alternatives in the table below.

Among U.S. media outlets, the percent articles in our bioweapons corpus that share narratives with Russian state media, Ukranian media outlets, and other U.S. sources. Low quality U.S. media websites (left) are more likely than mainstream popular U.S. news sources (right) to print stories that contain the same narratives as Russian state media articles.

We use this method in a case study of U.S. news website coverage of the war in Ukraine. We show that low quality U.S. news sites were more likely than mainstream U.S. news sites to have overlapping claims (narrative similarity) with Russian newspapers.

We leverage recent advances in NLP to measure whether two newspaper articles are making the same claims about the same underlying subjects. We use document embeddings to reduce the number of comparisons to a tractable number and large language models for pair annotation.

Researchers are often interested in tracking the flow of ideas and claims across texts. This is a very challenging target to estimate, however, as due to copyright and journalistic norms authors will often reuse information and ideas without using the same words, phrases or even language.

Quantifying Narrative Similarity Across Languages - Hannah Waight, Solomon Messing, Anton Shirikov, Margaret E. Roberts, Jonathan Nagler, Jason Greenfield, Megan A. Brown, Kevin Aslett, Joshua A. Tuck... How can one understand the spread of ideas across text data? This is a key measurement problem in sociological inquiry, from the study of how interest groups sh...

I am thrilled to share a new article in Sociological Methods & Research, “Quantifying Narrative Similarity Across Languages”. My co-first author Sol Messing and our collaborators developed a new approach to measuring “narrative similarity” between texts: journals.sagepub.com/doi/10.1177/...

Thanks Erik!

Excellent new work by one of our graduate students

Read the first article from the ASA Sex & Sexualities Section's new journal, in which editors Krystale E. Littlejohn @drklittlej.bsky.social, U of Oregon, & Amy L. Stone @amylstone1.bsky.social, Trinity University, consider the importance of nurturing a sociology of sex & sexualities. bit.ly/3EyV56r

Life course illegality: how the life course and aging shape the experience of illegality Abstract. Scholars have studied in detail how immigrants experience illegality in the US. Many have focused on immigrants’ fear of deportation as central t

📣🚨 NEW ARTICLE: Many #immigrants are aging & remain undocumented. What is their experience of illegality as they age–specifically their fear of deportation? New publication @sfjournal.bsky.social! Please read and share widely. 🔗 here: bit.ly/SF25-LCI (summary below) /1

Hannah Waight and Adam Goldstein show that inequality perceptions have become increasingly polarized by partisanship.This gap has been driven by Republicans, whose increasing disavowal of growing inequality contributed to an overall decline in Americans’ perceptions in the new gilded age.

Elite political polarization on distributional matters proceeded mass partisan polarization. We show this by examining data on Democratic and Republican party platforms from the Comparative Manifestos Project.

Given the cross sectional nature of our data it is beyond the scope of our data to directly identify the cause of Republicans' asymmetrical downturn. We discuss in the conclusion, however, that one likely possibility was elite signaling.

Thank you @jonathanmijs.com we were very inspired by your work on how people perceive social structure!

9/ Our replication package is currently under review, but will be available at this link: doi.org/10.7910/DVN/.... Please contact me at hwaight@uoregon.edu if you would like to access the data before then.

8/ We are very grateful to the editorial team at AJS and our fantastic reviewers, whose anonymous comments greatly improved the manuscript and our analysis.

7/ Beginning in the early 1990s, before the takeoff of cable news viewership during the early 2000s, Republicans began to evince an increasingly divergent conception of macro distributional reality. This divergence has attenuated alongside the rise of conservative economic populism since 2008.

6/ Our results shed new light on the perceptual connections between the dual economic and political polarizations of American society during the late-20th and early-21st centuries.

Observed and Counterfactual Effects Trend Lines for Inequality Perceptions, 1992-2013. Aggregate estimates averaged from model-based simulations under observed and counterfactual scenarios.

5/ Using a counterfactual simulation, we estimate that if the difference between Republicans and Democrats had remained at its 1992 size, the aggregate subsequent downturn in inequality perceptions would have been halved.

This plot shows the estimated percent of individuals perceiving the “rich are getting richer” by five-year bin and party identification.

This plot shows the estimated percent of individuals perceiving the “rich are getting richer” by five-year bin, party identification, and socio-economic status.

4/ These figures show that the downturn was asymmetrically concentrated among identified Republicans, especially Republicans with middle and high socio-economic status. Patterns of inequality perceptions reflect a growing class divide within the Republican party.