Digital Object Identifiers (DOIs) Prove Highly Effective for Long-Term Data Availability in PLOS ONE
Bibliographic record
Abstract
A Review of: Federer, L. M. (2022). Long-term availability of data associated with articles in PLOS ONE. PLOS ONE 17(8), Article e0272845. https://doi.org/10.1371/journal.pone.0272845 Objective – To retrieve a range of PLOS ONE data availability statements and quantify their ability to point to the study data efficiently and accurately. Research questions focused on availability over time, availability of URLs versus DOIs, the ability to locate resources using the data availability statement and availability based on data sharing method. Design – Observational study. Setting – PLOS ONE archive. Subjects – A corpus of 47,593 data availability statements from research articles in PLOS ONE between March 1, 2014, and May 31, 2016. Methods – Use of custom R scripts to retrieve 47,593 data availability statements; of these, 6,912 (14.5%) contained at least one URL or DOI. Once these links were extracted, R scripts were run to fetch the resources and record HTTP status codes to determine if the resource was discoverable. To address the potential for the DOI or URL to fetch but not actually contain the appropriate data, the researchers selected at random and manually retrieved the data for 350 URLs and 350 DOIs. Main Results – Of the unique URLs, 75% were able to be automatically retrieved by custom R scripts. In the manual sample of 350 URLs, which was used to test for accuracy of the URLs in containing the data, there was a 78% retrieval rate. Of the unique DOIs, 90% were able to be automatically retrieved by custom R scripts. The manual sample of 350 DOIs had a 98% retrieval rate. Conclusion – DOIs, especially those linked with a repository, had the highest rate of success in retrieving the data attached to the article. While URLs were better than no link at all, URLs are susceptible to content drift and need more management for long-term data availability.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.192 | 0.672 |
| Meta-epidemiology (narrow) | 0.002 | 0.003 |
| Meta-epidemiology (broad) | 0.003 | 0.003 |
| Bibliometrics | 0.026 | 0.028 |
| Science and technology studies | 0.003 | 0.005 |
| Scholarly communication | 0.014 | 0.026 |
| Open science | 0.004 | 0.011 |
| Research integrity | 0.005 | 0.005 |
| Insufficient payload (model declined to judge) | 0.072 | 0.035 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".