Find a data repository
Repositories can help you:
- manage your data
- cite your data by supplying a persistent identifier
- facilitate discovery of your data
- preserve your data over time
Selecting your repository
In choosing a repository for your data, first consider:
- Does your funder or a publisher require or suggest a specific repository?
See our Research Funder Requirements page for information on some of the common funders. - Is there a repository specific for your discipline or type of data?
The Registry of Research Data Repositories can be helpful in browsing options.
MIT Libraries’ supported disciplinary repositories:- ICPSR: Social and behavioral sciences
- Qualitative Data Repository (QDR): Qualitative and multi-method research in the social sciences and related disciplines
- What are the desirable characteristics for a data repository?
See the following resources for specific guidelines:- Comparison chart for OSTP, NIH, USRN
- OSTP/NSTC, Desirable Characteristics of Data Repositories for Federally Funded Research
- NIH
- The TRUST Principles for digital repositories
While there are many options to choose from, we’ve highlighted a small set of recommended or supported repositories below to help researchers quickly review key features in making their decisions. Please refer to each repository’s documentation for the most current information. To access MIT-associated benefits:
- Harvard Dataverse: contact data-management@mit.edu
- Dryad at MIT: Account creation/linking
- OSF at MIT: Log in
- QDR: contact data-management@mit.edu
- ICPSR: Use libproxy.mit.edu address to create account or login
If you would like to discuss what might work for you and your data, or if you see an error in the information below, contact data-management@mit.edu.
| Dataverse | Zenodo | Dryad | OSF | |
|---|---|---|---|---|
| File management | ||||
| File & dataset size limits |
2.5GB/file; 1TB per researcher | 50GB/dataset (contact to discuss larger datasets) |
300GB/dataset (contact to discuss larger datasets) |
Projects currently have no storage limit. There is a 5GB/file upload limit for native OSF Storage. No limit for the amount of storage used across add-ons |
| Useful integrations | Open Science Framework (OSF); Dropbox |
Github (archive a Github repo in Zenodo) |
Zenodo (for software publication) Frictionless Data |
Many integrations via add-ons |
| Versioning support? | Yes | Yes | Yes | Yes, for OSF storage |
| Persistent, unique identifier support | DOI | DOI for each version with a “Concept” and DOI to represent “all versions” |
DOI | DOI |
| Permissions & access | ||||
| Allows multiple administrators? | Yes | Yes for Communities, unknown for individual items | Yes | Yes |
| User guestbook? | Yes | Yes | No | No |
| Data licensing options |
CC0 default; custom Terms of Use optional |
CC BY default and options of several Creative Commons licenses and open source software licenses | CC0 required | 14 licenses are available, or custom license creation |
| Allows embargo? | Yes | Yes | Yes, if allowed by journal | Yes, for registrations |
| Provides private URLs for peer review? |
Yes | Yes for software through Dryad integration; and by sharing secret link | Yes | View-only link with ability to anonymize contributor list |
| Other access restrictions available? | Yes, allow access to specific accounts, and allow users to request access | Yes, allow access through secret link, and allow users to request access | No | Yes, with request access and private sharing setting |
| Computational access |
Search API, Data Access API, and Dataverse Package for R on rOpenSci Project | OAI-PMH and REST API | REST API and Dryad Package for R on rOpenSci Project | The OSF API generally conforms to the JSON-API v1.0 spec |
| Administrative considerations | ||||
| Costs | Free to researcher up to 1TB; MITL partnership | Free to researcher, up to 50GB/dataset | Free to MIT research community due to MITL’s institutional membership. | Free to researcher; MITL membership |
| Curation services | Yes. Available to MIT users at Harvard fee levels. | No | Yes | No |
| Usage analytics | Downloads
Follows Counter Code Of Practice for Research Data Usage Metrics (Make Data Count) |
Views (unique), Downloads (unique), Data volume, Unique views, Unique downloads
Follows Counter Code Of Practice for Research Data Usage Metrics (Make Data Count) |
Views, Downloads, Citations
Follows Counter Code Of Practice for Research Data Usage Metrics (Make Data Count) |
Downloads (per version), Links, Forks Working towards Counter Code Of Practice for Research Data Usage Metrics (Make Data Count) |
| Trusted repository? * | Yes | Yes | Yes | Yes |
* Trusted repositories are those that commit to providing “reliable, long-term access to managed digital resources” (RLG-OCLC, 2002). “The TRUST Principles for digital repositories” offers a useful framework – Transparency, Responsibility, User focus, Sustainability and Technology – for evaluating the alignment of a repository to this mission.
Using data from a repository?
Cite the data to give credit to the data producer, enable others to use the data, and meet journal requirements.
We are grateful to the creators of the Generalist Repository Comparison Chart for their efforts on developing a concise comparison chart from which we took much inspiration for the content of this page.
Stall, Shelley, Martone, Maryann E., Chandramouliswaran, Ishwar, Crosas, Mercè, Federer, Lisa, Gautier, Julian, Hahnel, Mark, Larkin, Jennie, Lowenberg, Daniella, Pfeiffer, Nicole, Sim, Ida, Smith, Tim, Van Gulick, Ana E., Walker, Erin, Wood, Julie, Zaringhalam, Maryam, & Zigoni, Alberto. (2020). Generalist Repository Comparison Chart. Zenodo. https://doi.org/10.5281/zenodo.3946720