CMMOD: Collective Memory Multi-source Observation Dataset (v1.0)

CMMOD is a multi-source observation dataset for the study of collective memory dynamics. It covers 122 curated public historical events observed across four independent observation systems: community-level discussion on Reddit (23–65 community grids, weekly and monthly panels), retrieval on Wikipedia (daily pageviews), press coverage in The New York Times (daily and weekly derived counts), and public search interest via Google Trends (monthly indices with official exports). The dataset comprises an event registry with aliases and cross-source mappings, 1.59 million rows of aggregated observation panels spanning 1941–2026 (core window 2015-07 to 2025-12), anonymised propagation-relation layers (crosspost, in-site reference and author-migration edges with HMAC-hashed identifiers), and a complete audit trail with per-file checksums, coverage summaries and construction lineage. No post text, plain author names or native platform identifiers are included; all second-level timestamps are reduced to day or week bins to prevent linkage re-identification.