Data integrity

Purpose

Data integrity is essential to research, regardless of data classification, since unauthorized modifications to your data can disrupt a project, produce incorrect results, undermine trust in academic research, and affect your reputation. It is important that throughout the research workflow and lifecycle that the integrity of the data is verifiable.

Audience

RESEARCHERS IT STAFF

On this page

Initial considerations

Establish a resilient backup strategy and back up your data.

• Should unauthorized modifications take place, it is important that you have a “clean” (unaffected) backup to revert back to.

  • Back up data

Departmental and divisional resources and recommendations.

• Contact your local IT group supported or recommended file verification and data integrity tools that are available to you.

  • https://uoft-infosec-cf.atlassian.net/wiki/spaces/ISH/pages/4948958/Additional+help#%F0%9F%96%A5%EF%B8%8F-Information-Technology-(IT)

What can I do?

Compute and check checksums on your integral data files.

• Hashing algorithms provide a unique output sequence for every unique input. This means that even if a single bit of a file has been altered, a radically different output sequence will be produced, even if the input file might initially appear unmodified.

Windows

• File Checksum Integrity Verifier

  • https://en.wikibooks.org/wiki/File_Checksum_Integrity_Verifier_(FCIV)_Examples

MacOS

• SHA Checksum command

  • https://ss64.com/osx/shasum.html

Linux

• SHA256 Checksum command

  • https://www.man7.org/linux/man-pages/man1/sha256sum.1.html

Maintain version control.

• Practicing manual version control or leveraging a version control system allows you to properly manage and retain the numerous revisions to a file that might take place over the course of a task or project.

  • https://onesearch.library.utoronto.ca/researchdata/organizing-your-data#Version%20control