Selecting data for preservation

When it is time to preserve your data, you will need to carefully consider exactly which components of your research need to be preserved. Answering "yes" to any of the following questions with regard to a particular data file or set of data may indicate that those data should be preserved for the long-term.

  • Do the data support published research?
  • Are the data vulnerable?
  • Are the data required for your research but from another source (i.e. not your original research data)? If so, is the future availability of those data from the original source uncertain?
  • Do you wish, or are you required, to share your data?
  • Are the data historically significant?

In addition, you should also consider whether you will need to preserve multiple versions of a file or whether the most recent version will be sufficient for preservation. It may also be important to consider whether the project is still in progress or whether it is complete. Long-term projects, such as those that involve sampling of a single site repeatedly over months or years, may require periodic preservation of data before the project is actually considered "finished."

Once you have decided what data you want to preserve, you may want to submit it to a domain-specific repository or to the Stanford Digital Repository -- or both!

We have compiled some general guidelines (pdf) for selecting research and other works for preservation in the SDR. This document also includes specific examples and guidelines for various disciplines. Please refer to this guidance as you are selecting data for preservation. Do not hesitate to contact us if you have questions.