This project checklist will help you set up and maintain robust data management practices and systems through the full life of a project.
Starting a Project
Determine your data management needs & responsibilities
- ❏ Determine who on the project team is responsible for managing your data.
Make sure you know who is responsible for what. Who collects the data? Who has access? Who is responsible for implementing a data management plan?
- ❏ Define what data, files, etc. you need to manage.
- ❏ Create a quick inventory of your data & related files.
- ❏ Ask what type of files & file formats? What are the estimated file sizes?
- ❏ Develop & document/share your file organization and naming system.
File naming is important for short and long term success. Create a standard way to name and organize your files to avoid access and version issues. Your system should be useable by everyone in your team.
More on File Naming:
These plans are also a great way for you to anticipate your data management needs and establish a shared understanding of the resources available to you.
- ❏ Be aware of funder and publisher requirements for Data Management Plans.
More on Data Management Plans:
Storing and Sharing Your Work
- ❏ Determine your active storage & sharing system.
During the active phase of your work, where do you store your data & files? This is different than your backups; these are the copies you are actively accessing and editing. Consider the amount of space you’ll need for the life of the project, who needs access, what tool integrations are needed, etc.
More on Storage:
- ❏ Set up a backup system.
Follow the “Here-Near-Far” backup practice of three copies in three different locations. Set up a ‘Crashplan‘ on your computers as one of your backups. Also see backup recommendations from UC Davis Information Technology.
❏ Establish access & security guidelines for your data.
Do you have sensitive data? Determine what access limits need to be in place for your different data files. Who can access what, when? Some external datasets have explicit access rules that need to be followed.
See: DataONE Best Practices Working Group, DataONE (July 01, 2010) “Best Practice: Identify data sensitivity.”
Document Your Project
- ❏ Determine what you need to document about your data (metadata) & how to capture it.
Check what is standard in your discipline or what a long-term repository may capture about similar data. Think about what you or someone else would need to know to use the data in the future. Start with a simple README (plain text file) for this information.
How to select the appropriate metadata for your field:
Evaluate Your System
- ❏ Test and revise your system as needed over the life of the project.
No system is ever perfect. Every system needs to evolve as projects change and the users of it identify better ways. Fight the urge to abandon an imperfect data-management system. Instead, revise it to better meet your needs.
- ❏ Think ahead to your longer-term data management & sharing goals and prepare for long-term data management in advance.
- ❏ Refer to the “Ending a Project” checklist below to start establishing your practices for longer-term storage and sharing.
Ending a Project
Determine Post-Project Data Sharing and Restrictions
- ❏ Determine what data should be openly shared and how any confidential data will be handled.
- ❏ Make sure any funder or intellectual property restrictions on the data are documented and followed.
- ❏ Determine publisher requirements for data sharing.
- ❏ Evaluate long-term sharing and storage systems. Store and share your work long term.
- ❏ Ensure those responsible for the project long term have appropriate access to your files and data.
- ❏ Make sure any needed software is accessible and properly licensed.
- ❏ Make sure ReadMe files are up to date.
Publish, Share and Preserve Your Data
- ❏ Adhere to FAIR Principles for Sharing Data.
FAIR data are:
- Findable: Data are identified through unique identifiers and clearly cited.
- Accessible: Data are publicly available, for example, in an Open Access repository.
- Interoperable: Data are in made actionable by being available in non-proprietary formats, for instance CSV rather than PDF or Excel files.
- Reusable: Data are properly documented through readme files, file naming protocols, and codebooks.
Unique identifiers disambiguate information. They are available for datasets in the form of DOI’s (Digital Object Identifiers).
Unique identifiers are important for authors too, and can be acquired through ORCID identifiers.
❏ Document the published & unpublished work (manuscripts, published papers, figures) that are based on or related to your data.
This checklist is modified from “Managing Your Data– Project Start & End Checklists” (PDF). Created by Phoebe Ayers & Christine Malinowski at MIT Libraries Data Management Services. Last Updated: 2020-04-17