Preserve your data

Data preservation is a series of managed activities necessary to ensure continued access to your research data as long as necessary. These activities are undertaken after the data is no longer in active use. Digital data especially needs to be preserved to mitigate the risk of software obsolescence, hardware failure, etc.

Best practices in preserving your data*

• Select data to retain: It's best to err on the side of keeping more data than less but it's unrealistic to preserve all of your research data (due to storage limitations). Always keep data that is used in publications, supports intellectual property claims, or can't be easily reproduced.
• Convert your data to non-proprietary formats: Proprietary file formats and software become obsolete quickly. Choosing non-proprietary file formats (CSV, PDF, MP3, etc.) will help to provide continuedl access to your data. Stanford Libraries has an extensive list of suggested non-proprietary file formats.
• Retain your data for 5-10 years: Your funding agency may have a data retention policy. Research data from federally funded projects needs to be retained at least three years. But err on the side of caution by keeping your data for longer.

How to preserve your data*

• Preserving the data yourself: You can preserve your data using your own storage hardware (such as an external hard drive). If you choose this option, your preserved data should be backed up. The major disadvantage with self-preservation of data is the significant time investment for maintenance and upkeep.
• Using a data repository: A data repository makes your data publicly available online and some (not all) commit to long-term preservation of your data. Before depositing your data, carefully read the repository’s online documentation to determine if they provide long-term preservation.

*Confidential data needs to be preserved more carefully than other data to ensure the privacy and confidentiality of human subjects.