Chenyue Jiao's Dissertation Defense

Research data sharing and reuse promise substantial benefits to science. Nevertheless, many valuable datasets remain inadequately curated to support meaningful reuse. Limited resources are available for curation, meaning that not all datasets can be curated to the same extent. Scientists and data management professionals instead face the complex task determining which datasets to prioritize for curation. These decisions are further complicated by differing perspectives among scientists and data management professionals on how data value should be evaluated. 

This dissertation argues that data value is not intrinsic; instead, it is a relational and emergent judgment shaped by stakeholders’ positions within knowledge infrastructures. Scientists and data management professionals assess value differently because they occupy different lifecycle, temporal, epistemic, and institutional positions. Drawing on qualitative research in the Earth Sciences, the study combines: semi-structured interviews (n=32) with Earth scientists at the National Center for Atmospheric Research (NCAR); interviews (n=20) with data management professionals working across universities, research centers, national laboratories, and data repositories; and qualitative analysis of institutional policies, repository documentation, data management guidelines, and related archival documents. The findings show that while both groups often express similar criteria (e.g., reuse, quality, uniqueness, documentation, and acceptability), they interpret and prioritize these criteria differently. Scientists emphasize immediate research use and disciplinary relevance, whereas data management professionals place greater emphasis on long-term usability, documentation, and the future reuse. These differences help explain uneven curation, persistent tensions in stewardship, and the limits of existing interventions that assume data value is self-evident or shared.

This dissertation contributes to Library and Information Science by providing a systematic multi-stakeholder analysis of how research data value is defined and assessed in practice. It shows that differences in data valuation emerge from stakeholders’ positions within knowledge infrastructures and the distinct forms of work they perform across the research lifecycle. By explaining how scientists and data management professionals assign value to research data, the study advances understanding of the social, institutional, and infrastructural processes through which data become prioritized for curation, preservation, and reuse. These insights can support more effective data stewardship, stronger collaboration across stakeholder groups, and more realistic approaches to promoting the long-term reuse of research data.

Questions? Contact Chenyue Jiao.