Visit HKU DataHub to explore research data and scholarly outputs published by HKU researchers, and the DataHub Guide for how to use the repository.
Visit the online guide to Open Data
It is common to see that along with the growth of open science in the recent decades, the global research community is increasingly sharing or publishing research data, making it available to others. Many academic research funders and academic journals are also imposing data policies that require data accessibility.
Widely adopted by the academic communities, the FAIR data principles first introduced in 2016 is the set of guidelines that help researchers make better use of, and engage with a broader audience with, their research data. Research data shared are required to be findable, accessible, interoperable, and reusable.
When you are planning to share your research data, you are suggested to consider the below to maximize reusability of your data:
Share your data and codes in open trusted repositories
Get a persistent identifier (e.g. DOI) for your data and use it in the associated publication for others to cite
Document your data, code, workflows, software required to open the data in separate file(s), and share alongside your data
Select appropriate data format and tools for higher interoperability
Use open license for your shared data and code
Read our guide on Open Data for more details on the benefits of data sharing, the FAIR data principles, and selecting repositories for sharing.
Research data can be shared in a variety of locations. Several researchers may opt to share directly via emails and personal websites or upload their data files directly on the journal publisher website in their journal article. They may seem convenient, yet it might not be the best option often due to limited accessibility, long-term preservation not guaranteed, versioning not supported, etc.
In compliance with the FAIR data principles, a long-term repository that provides a permanent identifier is the most recommended option for data sharing.
When you are considering which data repository would best suits your data, consider the followings:
Does your funding sponsor (or journal publisher) require or recommend a specific data repository?
Is there a domain-specific repository that is widely-used in your research field?
Does your organization/institution offer a data repository?
Do you think the tools offered by the repository for data discovery and distribution are suitable for your data?
Does the repository provide open data access and support the FAIR data principles (e.g. offer persistent identifiers like DOI, data licensing)?
According to the Open Science Training Handbook, as recommended by OpenAIRE, researchers may consider the order of preference as follows:
Use a disciplinary repository established for your research domain with recognized standards in your discipline
Use an institutional research data repository
Use other general repositories that are designed to accommodate multi-disciplinary research data
Search for other data repositories in a global registry such as re3data or FAIRsharing.
An external disciplinary or data-type specific repository often follows discipline-specific metadata standards and data curation practices. Primary considerations should be given to these data repositories as they could better facilitate discoverability, understanding, and reusability of the shared datasets in a specific area of study. Researchers and research students are recommended to seek advice from their colleagues or supervisors for locating suitable disciplinary specific repositories in their subject fields.
For instance, the Trans-NIH BioMedical Informatics Coordinating Committee (BMIC) maintains a list of National Institutes of Health (NIH) supported domain-specific repositories.
Researchers can also identify appropriate discipline-specific repositories by referring to a data repository registry like re3data.org or exploring the catalogue of databases available in the FAIRsharing collection.
The University of Hong Kong has an institutional data repository, HKU DataHub, powered by Figshare. HKU researchers and students could cite, store, publish and share their research data and other digital materials on this repository. It serves as a persistent ‘home’ for research data generated by HKU community members, providing long-term storage, global open access, identifier generation, and secure protection, etc.
Visit HKU DataHub to explore research data and other digital scholarly outputs published by HKU researchers and HKU DataHub: The Guide for guidelines on how to use the repository.
To enhance the visibility of the research within HKU community, HKU researchers who deposit their data primarily at disciplinary repository may create an item record at DataHub with the link directing others to where the data is stored. Read the guide for uploading linked files to DataHub.
Several generalist repositories offer cost-free services and accounts for researchers from multiple disciplines to deposit their research data. They usually accept all types of data and are most commonly used for data that cannot go into a domain- or discipline-specific repository.
The Generalist Repository Ecosystem Initiative (GREI)
The GREI is an initiative of the NIH to bring together seven generalist repositories in a collaborative working group. The repositories participated in the GREI are highlighted in the below table:
See also a comparison chart of the GREI-participated generalist repositories.
Research data is intellectual property that could be under the ownership of researchers, the supporting institution, or the funder. HKU researchers may refer to the university’s policy on Intellectual Property Rights and the Research Data and Records Management for relevant regulations.
When sharing your research data, using an open license is one of the recommended best practices to increase the reusability of your data. An open license specifies what can and cannot be done with an original work regardless of its form. It grants permissions and states restrictions. According to the definition by Opendefinition.org, an open license is one which grants permission to access, re-use and redistribute a work with few or no restrictions.
Understanding the licensing terms of a dataset before reusing it is crucial to prevent copyright infringement and other intellectual property concerns.
The most common open license used for academic work and datasets is the Creative Commons (CC) licenses. See the textbox below for more details on each CC license with different combinations, and more information on our guiding page.
For open license frequently used specifically for open-source software or codes, read our guide on Open-Source Software and Codes.
There are six Creative Commons license options. The Creative Commons license on a copyrighted work answers the question: What can a user do with this work?
| Elements | Copy and distribute the material |
Attribute the creator | Distribute, remix, adapt, and build upon the material |
Share the modified material under identical terms |
Commercial use |
|---|---|---|---|---|---|
|
CC BY
|
Allowed | Required | Allowed | Not required | Allowed |
|
CC BY-SA
|
Allowed | Required | Allowed | Required | Allowed |
|
CC BY-NC
|
Allowed | Required | Allowed | Not required | Prohibited |
|
CC BY-NC-SA
|
Allowed | Required | Allowed | Required | Prohibited |
|
CC BY-ND
|
Allowed | Required | Prohibited | (Modification prohibited) | Allowed |
|
CC BY-NC-ND
|
Allowed | Required | Prohibited | (Modification prohibited) | Prohibited |
|
CC0
|
Allowed | Not required | Allowed | Not required | Allowed |
| Elements | Copy and distribute the material |
Attribute the creator | Distribute, remix, adapt, and build upon the material |
Share the modified material under identical terms |
Commercial use |
Note:
Different publishers may have different license to publish (LTP) agreements.
As an author, when you choose a license, you will read through the license terms and consider which license suits you best.
For example, do you prefer CC-BY-NC-ND license, if you need to grant only to the journal publisher (but not to other users) the right to sell or rent your article?
Learn more and license chooser:
Disclaimer: The information and materials provided on the website are for general informational purposes only and do not constitute legal advice.
If you are utilizing third-party data published by others in your research or publications, you must provide a citation for the dataset to give credit where credit is due (original author/producer) and to help other researchers to locate the materials.
Acknowledgements and citations contribute towards fostering a culture of sharing data without fear of ideas or recognition being stolen. Data citations also aid in the transparency of how data is being used. By citing data, original authors and new researchers can easily track how the data are being used to answer different questions.
In general, a citation for dataset often includes the following components:
Authors and their affiliated institutions/organizations
Title
Version
DOI (or URL if a unique identifier is not available)
Creation date
Additional fields may also be specified or required by individual repository/journal
The Australian Resource Data Commons (ARDC) provides standard data and software citation templates and examples:
Standard data citation
Template
Creator (Publication Year): Title. Publisher. (resourceTypeGeneral). Identifier
Example
Hanigan, Ivan (2012): Monthly drought data for Australia 1890-2008 using the Hutchinson Drought Index. The Australian National University Australian Data Archive. (Dataset) http://doi.org/10.4225/13/50BBFD7E6727A
Standard software citation
Template
Creator (Publication Year): Title. Version No. Publisher. [resourceTypeGeneral]. Identifier.
Example
Xu, C., & Christoffersen, B. (2017). The Functionally-Assembled Terrestrial Ecosystem Simulator Version 1. Los Alamos National Laboratory (LANL), Los Alamos, NM (United States). [Software]. https://doi.org/10.11578/dc.20171025.1962
More resources:
Digital Curation Centre, How to Cite Datasets and Link to Publications
