Supplementary Materials: How Large Language Models Source Brand Reputation Knowledge — A Cross-Industry Attribution Analysis (CRR Anonymous Submission)

This package contains analysis from 1,440 AI responses across 24 companies in 8 industries with 41,250 coded source attributions. Includes: master analysis dataframe, coded sources, sentiment analysis, similarity analysis, statistical tests, Jupyter notebook for full replication, and all manuscript figures. Data collection date: 19 March 2026. For full information read README.me file.