When Capability Sounds Like Crisis: replication package (disclosure-based AI readiness instrument, S&P 500 IT firms, FY2023-2025)

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

Replication package for the manuscript "When Capability Sounds Like Crisis: Reliability Without Validity in Disclosure-Based Measures of Artificial Intelligence Readiness". Contains the fetch, scoring, panel-build and analysis code (Python), raw and derived data (CSV and SPSS .sav), codebooks and measurement decisions, results tables, SPSS output, and validation checks for an eleven-facet firm-level artificial intelligence readiness instrument built from SEC EDGAR 10-K text and XBRL financials, GitHub and HuggingFace artifacts, Google Patents counts and the Stanford HAI Global AI Vibrancy Index, for 73 S&P 500 information technology firms across fiscal years 2023 to 2025 (213 balanced firm-year observations). See 00_README/00_REPLICATION_README.md for the run order and the one documented reproducibility gap (manual Google Patents counts).