Resources for repositories and data platforms
Users engage with datasets in many ways, so understanding data reach requires insights into this diversity of interactions. Make Data Count provides recommendations and tools to help repositories and data platforms collect, standardize, and share data-usage indicators.
Why implement data usage tracking
- Demonstrate value: Usage indicators show how a repository’s data collections are used, offering valuable evidence to report to host institutions and funders about the platform’s impact within research communities.
- Improve services: Usage indicators reveal trends in data usage to inform strategic decisions into the development of products and services to address evolving community needs.
- Drive visibility and recognition: Usage indicators help researchers discover relevant datasets, fostering visibility and reuse. They also signal trust and support more holistic research assessment beyond traditional metrics.
By adopting Make Data Count recommendations, repositories and data platforms can collect data views, downloads, and citations to demonstrate the value and impact of their data collections and advance community understanding of data usage.
Data views & downloads
Make Data Count enables repositories and data platforms to collect and normalize usage counts, so they can be compared and interpreted across platforms. This ensures transparency and builds trust, supporting the inclusion of data-usage measures in reporting and evaluation.
The COUNTER Code of Practice for Research Data provides a standard for the normalization of usage counts. Repositories and data platforms may implement this standard for data-usage tracking via two paths:
the DataCite usage tracker or log processing.
Usage tracking implementation options
| Feature | DataCite usage tracker | Log processing |
| Counts of views and/or downloads | Collected through landing pages | Server-wide collection |
| Normalization | Usage tracker normalizes counts in alignment with COUNTER Code of Practice for Research Data | Repository processes usage reports per rules outlined in the COUNTER Code of Practice for Research Data |
| Usage reports | Generated by usage tracker, monthly | Generated by the repository or platform, at any time |
| Submission of reports to DataCite | The usage tracker directly submits reports to the Usage Reports API | The repository or platform completes submission of the reports |
Usage counts collected via the implementations above are reported to DataCite and made available via the DataCite Usage Reports API and DataCite Commons.
We recommend that repositories display the usage counts on the dataset landing pages. Repositories may also display repository-level statistics for usage across their collection of datasets; for such statistics, we recommend a monthly distribution over single aggregate usage.
Example of usage counts displayed on DataCite Commons, for a dataset at the Dryad repository: https://commons.datacite.org/doi.org/10.5061/dryad.7d7wm37wr.
Insights from normalized usage counts
Make Data Count recommendations have been implemented by a number of repositories. The seven repositories in the NIH-funded Generalist Repository Ecosystem Initiative (GREI) have implemented these workflows to consistently collect, standardize and share usage counts for the datasets they host. This enables the repositories to showcase the value of their datasets and provide NIH insights into the use and impact of NIH-funded data.
Implementation of Make Data Count recommendations for data usage tracking at Figshare and Vivli.
| Implement Make Data Count recommendations to monitor dataset usage and impact. Contact us to learn how the usage tracker enables usage tracking to demonstrate your repository’s value and help researchers get recognition for their datasets. |
Data citations
Data citations are links between datasets and another output. Examples of data citations include a journal or a preprint that cites the dataset underlying the results, a software that cites the training dataset used during development, or a derived dataset that cites the original dataset on which it builds. Such citations signal the use of data as part of scholarly activities and provide valuable information into the reach of datasets.
Collecting data citations
Data repositories and platforms can collect and report data citations by registering the citation in the dataset metadata. This relationship to the other object is registered via the relatedIdentifier field of the dataset DOI metadata, and designating one of the following relationTypes:
- IsCitedBy
- IsReferencedBy
- IsSupplementTo
DataCite provides documentation on how to contribute citations. Make Data Count and the repositories in the GREI initiative have developed best practices for data citations: https://doi.org/10.5281/zenodo.10562429.
Finding data citations
Data citations registered via DOI metadata can be found and consumed via DataCite Event Data, Data Cite APIs and DataCite Commons.
Example of data citations displayed on DataCite Commons, for a dataset at the ICPSR repository: https://commons.datacite.org/doi.org/10.3886/icpsr29621.
In addition to metadata, links between the dataset and other outputs may be established via different methodologies. The Data Citation Corpus aggregates data citations collected via DOI metadata, as well as other sources that identify data-article links via curation or text mining. The Corpus includes citations for datasets with both DOIs and accession numbers.
Data Citation Corpus dashboard: https://corpus.datacite.org/dashboard
Repositories and data platforms can check for citations to datasets in their repositories by exploring the Data Citation Corpus data file: https://doi.org/10.5281/zenodo.11196858.
| Do you have data citations you would like to contribute to the Data Citation Corpus? Do contact us on info@makedatacount.org. |
