Reproducibility Materials for the CHARLS Empty-Nest Operationalisation and Depressive-Symptom Classification Study

This record contains reproducibility materials for the manuscript “Empty-Nest Operationalisation and Depressive Symptoms among Rural Older Chinese Adults: A CHARLS Study.” Version 2.3.0 includes the byte-identical executed final-analysis script retained for provenance, a functionally matched portable derivative, figure-generation scripts, aggregate derived results, multiple-imputation specifications and diagnostics, full-sample classification analyses, repeated cross-validation outputs, alternative CES-D-10 threshold analyses, survey-design robustness checks, household-composition analyses, and supplementary 2020 analyses. The archive also contains the final supplementary workbook, all main and supplementary figures, a README, release metadata, and a SHA-256 manifest. Restricted CHARLS microdata, participant identifiers, and row-level analytic intermediates are not included. Registered users must obtain the CHARLS data independently under the CHARLS project’s data-use terms. The downstream workflow begins from the restricted base_2018_household_linked.pkl intermediate. The required inputs, portable execution procedure, provenance record, and reproducibility boundary are documented in the README. The public CSV files contain only aggregate or resampling-level outputs. Version 2.3.0 supersedes version 2.2.0. It adds two binary-outcome survey-robustness rows to Supplementary Table S10 and updates Supplementary Table S15 accordingly. The corresponding estimates were already available in the archived aggregate output, S14_survey_weight_cluster_robustness.csv. This reporting update did not alter any statistical estimates or substantive conclusions.