Data and Reproducibility Code for "Temporal Persistence in Online Information Seeking: Long-Term Google Trends Evidence from the Sewol Ferry Disaster"

This package provides supporting data, reconstructed Python analysis code, numerical results, and validation reports for the manuscript “Temporal Persistence in Online Information Seeking: Long-Term Google Trends Evidence from the Sewol Ferry Disaster.” The analysis uses monthly Google Trends relative search-interest indices for three Korean-language queries in South Korea: 세월호 추모 (REM; remembrance), 세월호 기억 (MEM; memory), and 세월호 희생자 (VIC; victims). The analysis period is January 2014–November 2025 (143 months); post-event analyses cover January 2015–November 2025 (131 months). Values reported as “<1” are coded as 0.5, consistent with the author’s analysis workbook. Sensitivity analyses use alternative values of 0 and 1. The composite indicator is the equal-weight mean of the three search signals. The package includes the original comparison export, a documented Python script, monthly and annual analysis tables, lagged Pearson correlations, leave-one-out sensitivity results, source checksums, and validation reports. The code uses the Python standard library and can be run with “python analysis.py”. Twenty-two automated checks verify rounded numerical claims in the revised manuscript. The analysis code was reconstructed during manuscript preparation; it is not the original historical analysis script. VALIDATION_MIDPOINT.md is the final validation report and supersedes the earlier zero-coding report retained for transparency. Original retrieval settings not recorded in the export are identified as unavailable. Data source: Google Trends (https://trends.google.com/). These indices represent normalized relative search interest, not absolute search counts. Reuse of the original Google Trends data remains subject to Google’s applicable terms and attribution requirements. Documentation and generated results are licensed under Creative Commons Attribution 4.0 International (CC BY 4.0). The analysis code (analysis.py) is licensed under the MIT License. Original Google Trends data are excluded from these license grants and remain subject to Google’s applicable terms and attribution requirements. See LICENSES.md inside the archive for the file-by-file license scope.