New Tools for Studying Bias and Belief: Institute for Policy Research - Northwestern University

Grateful to my coauthors Michalis Mamakos, Tessa Charlesworth, and @sylviapperry.bsky.social for the collaboration, and thanks to @ipratnu.bsky.social for highlighting the work.

New Tools for Studying Bias and Belief: Institute for Policy Research - Northwestern University

Honored to see our UML paper featured by the Institute for Policy Research at Northwestern University.

The article highlights how unsupervised machine learning can help social scientists uncover hidden structure in beliefs and attitudes.

www.ipr.northwestern.edu/news/2026/ne...

Thanks for hosting me! It was wonderful chatting with such a great group of folks!

🚨New WP "@Grok is this true?"
We analyze 1.6M factcheck requests on X (grok & Perplexity)
📌Usage is polarized, Grok users more likely to be Reps
📌BUT Rep posts rated as false more often—even by Grok
📌Bot agreement with factchecks is OK but not great; APIs match fact-checkers
osf.io/preprints/ps...

Research in Action: How Dr. Samuel Sommers Exposed Unconscious Bias in Jury Selection

As part of an effort to celebrate scholars who apply their research in service of social change, SPSP is highlighting the late Dr. Samuel Sommers - who served as an expert in several cases on different aspects of the criminal justice system.

Learn more: https://ow.ly/wr7V50Y9sKA

New work led by @jonds7.bsky.social on responses to police brutality (justification and outrage), with analyses of YouTube video comments and a behavioral experiment.

Very grateful to have gotten to work on this with the team! 🙏

These young people are, in their own way, teaching seminars on identity every time they post. And it reminds me that the best social science takes what folks already sense about the world and treats it with the theoretical and empirical seriousness it deserves.

www.instagram.com/reel/DQ-1vvq...

Another reminder that if you got students who wanna do a social psych PhD focusing on identities (specifically Asian, Latine, MENA Americans), I’m recruiting! I’m hustling hard here so pls send people my way 😅
see website for details www.pbandjlab.com

This project has been years in the making — and it’s been an absolute joy working with @avamadesousa.bsky.social, Lauryn Hoard, Wendi Gardner, @williambrady.bsky.social, and @sylviapperry.bsky.social.

So grateful for this team. 🙏
Full preprint here: osf.io/preprints/ps...

Across both studies, justification consistently aligned with less support for police reform, while outrage aligned with more.

That link between outrage and reform was strongest for the Black victim — raising questions about how racialized stereotypes might blunt the very outrage needed for change.

Study 2: Participants watched two matched videos—one with a Black victim, one White.

Those who endorsed more justification also reported less outrage and lower support for reform.
When the victim was Black, participants more often described him as “superhuman.”

Comments justifying police conduct were ~80× less likely to mention accountability.
Outrage comments were ~11× more likely.

In public conversation, justification and outrage emerge as opposing ways of interpreting the same harm.

Study 1: We analyzed 250k YouTube comments from 57 videos of police brutality using GPT-3.5 + Google Jigsaw.

Two main reactions appeared:
➡️ Justifying the officer’s actions
➡️ Expressing moral outrage

They rarely overlap — each offers a different lens. ⚖️

New preprint out 📄
“Why Reform Stalls: Justifications of Force Are Linked to Lower Outrage and Reform Support.”

Why do some cases of police violence spark reform while others fade? We look at how people explain them—through justification or outrage.

osf.io/preprints/ps...

✨New preprint! Why do people express outrage online? In 4 studies we develop a taxonomy of online outrage motives, test what motives people report, what they infer for in- vs. out-partisans, and how motive inferences shape downstream intergroup consequences. Led by @felix-chenwei.bsky.social 🧵👇

Thank you so much Abby ❤️🥹

Thank you so much family! Hope you’re doing well!!

🎓 Paper: osf.io/preprints/os...
📺 YouTube Tutorials: youtube.com/playlist?lis...
📦 GitHub: github.com/jedoriscar/U...

#RStats #MachineLearning #SocialPsych #AcademicTwitter #DataScience

This paper was a labor of love and I’m so grateful to my brilliant co-authors Michalis Mamakos, @sylviapperry.bsky.social , and Tessa Charlesworth for their support, feedback, and generosity throughout the process.

We also include extensive annotated R code, visual walkthroughs, and a free YouTube tutorial series to make this genuinely hands-on. If you’re curious about how UML can help you uncover new patterns in your data—or just want to brush up on clustering and dimension reduction—check it out!

Each section walks through:

- what the method is

- how to implement it in R

- how to interpret the results

- real examples using Project Implicit data

We focus on 4 key UML techniques:

- K-means clustering

- DBSCAN

- Principal Component Analysis (PCA)

- Market Basket Analysis (MBA)

Social cognition research is drowning in big, complex data—but we often use the same familiar methods. Unsupervised Machine Learning (UML) gives us tools to explore patterns without predefined outcomes. We wrote this paper to demystify UML and help researchers use it meaningfully.

🧵 A brief thread on what this tutorial paper is about & why I’m excited about it:

Thrilled to share that my first first-authored publication is officially out (soon to be in press at Social Cognition)! 🚀

"From Data to Discovery: Unsupervised Machine Learning in Social Cognition"
📄 OSF preprint: osf.io/preprints/os...

Let’s gooooo! Dr. Vallabha has a nice ring to it!

Photo of Dr. Vallabha in front of her power point slides

Despite all the upheavals, some bright spots!

Yesterday, @shreevallabha.bsky.social successfully defended her dissertation; approved with no revisions! 🎉

Dr. Vallabha’s dissertation work is the most comprehensive study of moral humility to date 👀

4. With her consent, the University can confirm that Ms. Oztürk is a third-year doctoral student in good academic and administrative standing. Her research focuses on how young adults can use social media in positive, prosocial ways and she is described by her faculty as a hard-working student dedicated to her studies and the Tufts community. The University has no information to support the allegations that she was engaged in activities at Tufts that warrant her arrest and detention. The University has seen an outpouring of support for Ms. Öztürk over the last week from Tufts students, faculty and staff. These individuals have described Ms. Öztürk as a valued member of the community, dedicated to her academic pursuits and committed to her colleagues.

NEW: Tufts files a declaration in support of Rümeysa Öztürk: "The University has no information to support the allegations that she was engaged in activities at Tufts that warrant her arrest and detention." www.tufts.edu/president/sp...

It’s been a devastating 24 hours with 3 NIH grants terminated…all focused on the health and wellbeing of trans or intersex youth. If folks have any ideas of foundations that may be interested in supporting the completion of these projects so years of work aren’t wasted, please let me know. 🙏

I would like to share that the family requests donations be made to the Innocence Project (innocenceproject.org/donate/) or the American Civil Liberties Union (action.aclu.org/give/now) in lieu of flowers. 💖💐