Research Data Management Overview

Before you share your research data, you should consider what type of access constraints the data may have. While sharing all data openly would be the best option for reaching a broader audience in the research community and promoting transparency with the public, there are many reasons why data use may come with constraints. Often this includes working with personally identifiable information (PII), national security data, or proprietary datasets. Here are a few things to consider when choosing how your data should be shared:

Open Access

• Data is freely available to anyone without restrictions
• May use creative commons (CC) licenses to specify reuse conditions
• No extra security required for data storage; sensitive data must be de-identified
• Citation is encouraged for secondary use

Managed Access

• Access is granted after review or approval; may require application or account creation
• Suitable for data that is sensitive but can be shared under conditions
• Researchers must agree to terms before accessing data
• Citation is encouraged for secondary use

Controlled Access

• Data is highly restricted, often requiring a formal, legally binding data use agreement
• May involve IRB approval, encryption, or secure computing environments
• Long-term commitment of the data steward to handle permission requests
• Citation is encouraged for secondary use and may require report back of publications based on the data

Here are a couple of terms you should familiarize yourself with when considering how your data could and should be reused:

• Data Use Agreements (DUA) - establishes the permitted disclosure of a unique data set, allowable uses, and any privacy or security requirements necessary for receiving and handling the data. Check out Purdue's page on Data Transfer and Use Agreements
  • Example: This is a sample DUA from Harvard Dataverse for datasets that have de-identified human subject data
• Embargo Periods - a temporary restriction on public access to data after a research project is completed; the data may be stored in a repository but only the metadata is openly available to others
  • Example: A researcher sets a 12 month embargo to allow time to publish primary findings before others reuse the data

Raw research data is not eligible for copyright protection because it is viewed as a compilation of facts, which cannot be copyrighted. Nevertheless, you can apply licenses, like a Creative Commons license, to regulate how others may utilize your data. It is essential to include a license when distributing your data and code. Even if your funding agency requires you to make your data and code available, a license will specify what future users are allowed to do with it. This can increase the likelihood of other researchers choosing to reuse your data and code, which is advantageous for you.

Licensing data:

Creative Commons (CC) is one of the primary standard license groups for data. There are six different CC license types with a range of restrictions. Open Data Commons also offers a set of legal tools and licenses to help you publish, provide and use open data.

• Read about CC licenses and permissions offered for each CC license offered
  • Use Creative Commons' tool to help you choose a license for your work
• Read about Open Data Commons' license options here and how to apply for them

Licensing software and code:

When choosing a license for data and code it is important that the data license is compatible with the license for software and code. For example, a CC license, which cannot be applied to code, may integrate better with an MIT License. Below are resources for finding and learning more about open source software and code licensing.

• Open Source Initiative (OSI) provides a list of OSI approved licenses allowing software to be freely used, modified, and shared
• GitHub's Choose a License tool help users find a license that best fits their situation
  • They also provide a list of common open source licenses sorted by conditions for use

While raw research data itself is generally not protected by copyright because it consists of facts, there are still important intellectual property considerations when publishing and sharing research outputs.

Copyrights:

• Software elements, such as the specific code used in the program or elements in the user interface, can be protected by copyright.
• Learn more about copyright basics and procedures from the Purdue University Copyright Office

Patents:

• If your data relates to an invention, sharing it publicly may affect patentability.
• This Purdue Libraries research guide Patent Resources on the Web shares materials that will help you search for patent information

More Intellectual Property resources from Purdue:

• Office of Research page with Intellectual Property resources
• Purdue Policy Office on Intellectual Property institutional policy
• Office of Technology and Commercialization and licensing Purdue technology