Methods for Large Scale Text Data
Some of my latest work on frontier methods for Political Methodology.
Republicans perceive scientists are Democrats, but they can be persuaded to be more trustful of scientific findings
Americans perceive latent partisanship of scientists, and this influences their trust in scientific findings.
If a Statistical Model Predicts That Common Events Should Occur Only Once in 10,000 Elections, Maybe it’s the Wrong Model
Statistical description is not anything goes. Important quantities of interest arise from generatively accurate models— which inform our understanding of American democracy.
Abortion rights and issues-based frameworks for elections
We provide a new data-driven foundation for understanding the structure of influential stakeholders' online conversations in the climate and sustainability space.
How Social Media Attacks on Election Officials in 2020 Undermined American Election Institutions
Insights from Using Machine Learning and Natural Language Processing with Twitter Data
We examine the multivariate correlates of trust in university research and opinions about climate change using high-quality survey data.