My new paper asks, "theorizing after looking at the data is bad, right?" Actually, post-hoc theorizing might be OPTIMAL in the modern era of huge datasets and mature theories. 🧵

Other than the persistence of these arrows (down with the arrows!), this is quite something to behold.

Why do dictators target the socially isolated? 📢 New theory by Konstantin Sonin: Propaganda works best when people are either totally atomized or tightly unified rather than in between. 🎯 Optimal strategy? Don’t focus on influencers. Target the fringes of the network.

For many years, we knew that 1-on-1 tutoring was the most powerful educational intervention you could do. But we couldn't scale it. Now, with AI, we finally can. Great things are ahead.

Difference-in-differences for practitioners from the diff-in-diffs dream team!

Very excited to publicly share news about a new tool, Papercheck, that Lisa DeBruine and me started to develop more than a year ago! In an introductory blog post, we explain our philosophy to automatically check scientific papers for best practices. daniellakens.blogspot.com/2025/06/introd…

Thrilled to see our work out in Journal of Politics! We theorize how autocrats use redistributive messaging to claim democratic legitimacy—and show it can shape public opinion. Grateful for the journey with @eddySFYeung and the support we received throughout.

Currently in FirstView: In “Odd Profiles in Conjoint Experimental Designs: Effects on Survey-Taking Attention and Behavior,” Kirk Bansak and Libby Jenke consider how survey-takers respond to odd combinations of conjoint attributes.

🧵 NEW PAPER - "Quantifying Narrative Similarity Across Languages" is live at SMR. We combine NLP w pairwise LLM evals to track ideas across media ecosystems and languages. We also figured out a way to eval unsupervised probs on widely understood supervised performance metrics.