Harvard Dataverse
Institute for Quantitative Social ScienceHarvard Dataverse is a free research data repository, open to researchers anywhere, run by Harvard's Institute for Quantitative Social Science on the open-source Dataverse software. Depositors publish datasets that are versioned under DOIs and downloadable through a web interface and an API. Tabular files from Stata, SPSS, R, Excel and CSV are ingested with variable-level metadata, and dataset metadata exports as Croissant, DDI and DataCite.
Scored as the Harvard installation. The Dataverse software also runs many other institutions' repositories, which are not entries here.
Openness
5 high confidence- license
- Apache-2.0(OSI)
- source
- public(github.com/IQSS/dataverse
- core features withheld
- no — academic institute at Harvard
Harvard Dataverse runs a current release of the Dataverse software, which Harvard's IQSS develops in the open under Apache-2.0. Deposit is free, and no feature of the software is held back for a paid edition.
- https://datamanagement.hms.harvard.edu/share/data-repositories recorded 2026-10-07
"Harvard Dataverse Repository is free for all researchers worldwide (up to 1 TB)."
- https://dataverse.harvard.edu/api/info/version recorded 2026-10-07
{"version":"6.10.1","build":"iqss-4"}: the Harvard installation reports the Dataverse release it runs.
- https://raw.githubusercontent.com/IQSS/dataverse/HEAD/LICENSE.md recorded 2026-10-07
"The Dataverse software is licensed under the Apache License, Version 2.0".
- https://raw.githubusercontent.com/IQSS/dataverse/HEAD/README.md recorded 2026-10-07
"Dataverse®, the open source software platform designed for sharing, finding, citing, and preserving research data. Developed by the Dataverse team at the Institute for Quantitative Social Science".
Adoption
4 medium confidenceThe installation's own metrics API counts every file downloaded from it, read here over the trailing month. That is file traffic rather than a count of researchers, and one dataset download can fetch many files.
- https://dataverse.harvard.edu/api/info/metrics/downloads recorded 2026-10-07
{"status":"OK","data":{"count":177441883}} downloads in total, read 2026-10-07.
- https://dataverse.harvard.edu/api/info/metrics/downloads/pastDays/30 recorded 2026-10-07
{"status":"OK","data":{"count":9682680}}, read 2026-10-07.
Capability
5 medium confidenceResearchers anywhere can deposit data for free, and each dataset is versioned under a DOI that the API serves for download. Tabular files are ingested into a standard form with variable-level metadata and can be described in Croissant for machine learning tools, which puts it beside OpenML rather than Zenodo.
- https://datamanagement.hms.harvard.edu/share/data-repositories recorded 2026-10-07
Harvard Dataverse: "Assignment of dataset DOIs Yes"; "free for all researchers worldwide (up to 1 TB)".
- https://guides.dataverse.org/en/latest/api/native-api.html recorded 2026-10-07
Native API reference: create and publish datasets, list versions, and download files by dataset persistent identifier.
- https://guides.dataverse.org/en/latest/user/dataset-management.html recorded 2026-10-07
"Files in certain formats - Stata, SPSS, R, Excel (xlsx), CSV and TSV - may be ingested as tabular data"; exports "Croissant ... Dublin Core DDI ... DataCite 4 ... Schema.org JSON-LD".
Verified 2026-10-07