Replication package — Large Language Models and Conversational Counter-Arguments to Anti-Public Sector Bias

Data, survey instruments, and analysis code to reproduce the figures in the paper "Large Language Models and Conversational Counter-Arguments to Anti-Public Sector Bias" (Journal of Public Administration Research and Theory, 2026). The archive contains the complete survey datasets for three pre-registered experiments (not trimmed replication extracts), the survey instruments as fielded, a full variable codebook, and the Stata and R scripts that generate the paper's data-driven figures.