What Not to Keep: Not All Data Have Future Research Value

The rise of academic library involvement in research data management has presented numerous challenges for academic libraries. Although libraries and archives have always had collection development policies that defined what they would or would not collect, policies for selecting research data for preservation are in their infancy. This study surveyed and interviewed health sciences academic researchers. From this research an initial list of eight types of health research data were identified as data that should not be preserved and made public. These include research data that are: sensitive or confidential; proprietary; easily replicable; do not have good metadata; test, pilot, or intermediate data; bad or junk data; data that cannot be used by others for a variety of reasons; and older data that are not used and have no obvious cultural or historical value. Conclusions drawn from the study will help librarians and archivists make informed decisions about which types of research data are worth keeping.