PA · 2026
When AI models look at a political campaign photo, do they make assumptions about the candidate's ideology based on race or gender — even when the ad itself is completely neutral? …
Working Paper · 2026
Introduces the Big Five Adaptive Inventory (BF-AI), a freely available computerized adaptive test built on the 100-item IPIP-NEO via the catSurv R package. In simulation studies …
Working Paper · 2026
Proposes GD-GPIRT, a generalized dynamic Gaussian process item response theory model for longitudinal and ordinal observations. The model combines Bayesian nonparametric IRT — …
PSRM · 2025
Proposes a framework for evaluating the validity and trustworthiness of measures created using supervised machine learning. The paper argues that standard validation practices from …
NeurIPS · 2024
Personality psychology has long been caught between two competing visions: models that describe everyone the same way, and models tailored uniquely to each individual. This paper …
PA · 2023
Standard political measurement tools assume that people who give the same answer do so for the same reason — but what if a liberal and a conservative both vote "no" for completely …
R package for crowdsourced validation of topic models via Amazon Mechanical Turk. Creates and manages tasks that assess both topic coherence and label accuracy, integrating …
PA · 2022
Topic models have become one of the most popular tools for analyzing political text at scale — but are researchers actually checking whether the topics they find measure what they …
Cambridge University Press · 2022
Survey research has a fundamental tension: asking enough questions to measure something well versus not exhausting or losing respondents. This book introduces a smarter solution — …
R package for Bayesian estimation of the generalized graded unfolding model (GGUM) via Metropolis-coupled MCMC. Recovers ideal points in settings where respondents with opposing …