Reproducibility package for the cross-repository audit of open polymer datasets (2020–2025)

This deposit accompanies the manuscript "Reuse quality of open polymer datasets: a reproducible cross-repository audit (2020–2025)" (W. Meyer, submitted to Patterns).

It contains the full reproducibility package for a polymer-domain audit of metadata quality across the eight major open generalist research-data repositories (Zenodo, Figshare, Mendeley Data, Harvard Dataverse, OSF, 4TU.ResearchData, Dryad, Fordatis) over 2020–2025. The corpus comprises 53,038 polymer-related records assembled via a 74-keyword controlled-vocabulary harvest, deduplicated, audited, and scored on an 11-criterion binary FAIR-aligned reuse score.

The deposit contains:- frozen raw harvest JSONs (polymermetalab_raw_harvest.tar.gz, ~250 MB)- processed corpus (polymermetalab_processed_corpus.csv, ~90 MB)- full source code — harvest, pipeline, audit, analysis (polymermetalab_code.tar.gz)- all 13 analysis tables and 8 figures (polymermetalab_results.tar.gz)- full audit output and reproducibility-test reference set (polymermetalab_supplementary_audit.tar.gz)- step-by-step reproduction recipe (REPRODUCE.md) and SHA-256 manifest (manifest.json)

Given the deposited raw JSONs as input, the published CSV tables and figures are regenerated byte-identically (33 of 33 outputs verified end-to-end on 2026-05-03; wall-time ~6 minutes on a 2023 Apple-silicon laptop).

Code is licensed under MIT; data and analysis output under CC-BY-4.0. See LICENSE.txt for full terms.