We wrote up our findings on Trump's endorsements and the outlook for 2022 for The Conversation U.S.. w/ @hjghassell Andy Ballard

For the sake of my mentions, I would like to stress that I am in fact not Michael Heselschwerdt, the man responsible for incorrectly conducting the EUFA Champions League draw... #ChampionsLeague

New paper at LSQ with Andy Ballard and co-authors. We examine polarizing rhetoric in congressional tweets (2009-2020), finding that electoral and ideological factors affect rhetorical strategy. Members also receive more engagement and donations. onlinelibrary.wiley.com/doi/10.1111/ls…

more ideologically extreme members of Congress, those from safer districts, and those who are not in the president’s party are more likely to send polarizing tweets; polarizing tweets garner more engagement, increasing campaign funding #polisciresearch onlinelibrary.wiley.com/doi/abs/10.111…

My new article examining the use of incivility by candidates on Twitter in the 2020 congressional elections. Republicans and non-incumbents tended to be more uncivil. Notably, candidates were also more uncivil when their opponents were also more uncivil. journals.sagepub.com/eprint/WRVMVGI…

During a crisis leaders often encourage unity. At the same time, polarization is increasing - and a crisis can be used to shed further negative light on opponents. How did these competing incentives shape elite rhetoric during the pandemic?, asks Michael Heseltine at #APSA2022 1/5

"Dynamics of Polarizing Rhetoric in Congressional Tweets" Andy Ballard, Ryan DeTamble, Spencer Dorsey, Michael Heseltine, and Marcus Johnson at American University is also in Volume 48, Issue 1. Read it here: onlinelibrary.wiley.com/doi/10.1111/ls…

Write-up of my latest paper by Philip Bump in the Washington Post. Trump really focused his endorsements on the primaries in 2022 which is where his involvement has the potential to make a positive impact. In the generals, not so much... sciencedirect.com/science/articl…

🚨New working paper! With Bernhard Clemm, across tasks and languages, we test the viablility of LLMs like GPT-4 to replace human coders for machine learning training data, with mostly strong results! We also offer a suggested pipeline for researchers. osf.io/cx752

🚨New paper w Bernhard Clemm: Testing GPT4 on a range of common political science coding tasks in 4 languages. We find generally strong results at a fraction of the time and cost. Also introduce an effective double-code and validate approach for LLM coding journals.sagepub.com/doi/10.1177/20…

This paper tests the performance of GPT4 in text coding across a range of common political science coding tasks in 4 languages. Michael Heseltine & Bernhard Clemm find that GPT4 coding is highly accurate, especially for shorter texts. Read more here: doi.org/10.1177/205316…

Thanks to Michael Heseltine for the vision & leadership on this project! LLMs obviously come with many issues – but there is increasing evidence that they can facilitate social-science research. A (non-exhaustive) 🧵 with examples I have come across:

New Preprint 🚨: Read our new paper: arxiv.org/abs/2403.13362 where we try to answer the question: Can we leverage LLMs and realistic bot accounts to incentivize users to consume more news content? More below👇[1/6]

🚨 New at AJPS w/ Hennes Barnehl and Magdalena Wojcieszak. Do partisans change news consumption based on daily sentiment about their team? Dems read more on bad days for Republicans and less on bad days for Dems. Reps read more conservative news on negative days for Rs onlinelibrary.wiley.com/doi/10.1111/aj…