Framing the News: India 2026 — coded dataset and methods (v1, first public release)

575 homepage stories from 10 Indian English-language digital news outlets, coded for narrative frame, topic, trigger and underlying message using PEJ Framing the News (1998) definitions plus an India-specific Institutional Critique frame. Constructed week within one fortnight, 18 June – 2 July 2026 (seeded, reproducible via the shipped sampling.py). Every story carries a resolvable Wayback Machine archive URL. 62 outlet-days; 326 unique events; 44 of 575 codes (7.7%) self-flagged low-confidence with a published sensitivity check; a seeded 20% human-validation subsample (115 stories) ships blind for Cohen’s κ. Findings are PRELIMINARY pending that validation. Package: v1 report · METHODS_v1.md (full methods, deviations log, audit trail) · dataset_v1.json (575 records) · sampling.py · rosen_scores.json · exclusion-evidence appendices for The Wire and Times of India · dropped_cells.json · validation_sample.csv / validation_key.csv. This version supersedes the previous deposit (a 170-story instrument) in full: that draft failed an adversarial audit on data provenance, exclusion-narrative accuracy, sampling reproducibility and unpublished Rosen Index inputs. The audit findings and their dispositions are logged in METHODS_v1.md §8. Interactive instrument: https://framing.onlinejournalism.in/