Data management is an ongoing process throughout any research project. It is not only important to plan how you intend to manage data from the very beginning, but it is also important to update this plan every step of the way from grant writing, to data collection, to analysis and project termination. Your data management plan should be considered a living document that can be altered as required over the course of the project in order to best suit a project's evolving needs.
In this short guide, we will give an overview of key concepts in data management, and introduce you to the basics on data lifecycles and what a good data management plan should include. For a list of terminology and what the terms mean in the context of this guide, please see the
Data Management: Why Does it Matter?
Development of good data management practices may:
- Help you and other scientists who will use your dataset in the future by supplying adequate metadata to comprehend its features.
- Save time by requiring less support for a dataset's use from the source, be it within-lab or at the data sharing stage.
- Prepare data for sharing.
- Save time if deployed from the start of a project, by ensuring data is organized in a similar version to its end state, thus eliminating the need to organize it later.
- Improve overall efficiency.
- Support easier reuse of data for multiple publications.
- Provide more control over human errors that occur throughout the project. Help meet requirements of ethical review boards which often require at least a basis data management plan to be submitted with clearance applications
- Help you meet requirements of funders, many of whom now demand that data management plans accompany grant applications. Some also demand data sharing as a condition for funding.
- Some major journals also require data sharing upon publication.
- Expand your future employment possibilities. Working with big data requires the skills to manage it, and this is unlikely to change any time soon!
There are many reasons to practice good data stewardship. The above are just a few of the most common motivations!