Best Practices for Storing, Archiving and Preserving Data
Research data and related files require reliable and trustworthy storage at all phases of the research process. Best practices include documenting the information below either in a Data Management Plan or as part of project protocols. For a more detailed guide to storage and archiving best practices, see the attached PDFs.
At a minimum ensure you can document:
- Ownership and responsibility for the data at all phases (active, archived) of it's life-cycle.
- For collaborative projects, Memoranda of Understanding (MOUs) should be developed that detail roles and responsibilities towards the data.
- Who has access to the data and how it is restricted.
- If data includes PII (personally identifiable information) or other sensitive data, access must be limited and data must be stored in a secure location.
- Process and procedures for creating and verifying backup copies during the project.
- Follow the 3-2-1 rule: There should ideally be 3 copies of the data, stored on 2 different media, with at least 1 stored off-site or in the "cloud".
- Length of time data must be maintained and why, e.g., raw sensor data must be kept indefinitely, analyzed final data should be kept for 10 years or until raw data can be re-analyzed.
- Data that can be re-created, such as OCR text files, may only need to be kept until the software that created it is superceded by better technology.
- Where data will be archived permanently.
In order to preserve data so it is accessible and useable in the future it must:
- Have adequate descriptive metadata for correct interpretation and use by future researchers (see Describing Your Data : Data Dictionaries)
- Be available in an open, non-proprietary, commonly used format (see attached PDF Best Practices for Choosing File Formats below).