Construction of a Theoretical Framework for Scientific Data Governance

Area of Science:

• Data Science
• Artificial Intelligence
• Information Governance

Background:

• Data-intensive and AI-driven sciences face governance challenges with multi-source heterogeneous data.
• Complex interactions of stakeholders, processes, and content complicate scientific data governance.

Purpose of the Study:

• To propose a theoretical framework for scientific data governance.
• To elucidate the complex dynamics of scientific data governance.
• To inform and improve scientific data governance practices.

Main Methods:

• A non-systematic literature review was conducted to classify data stakeholders and data lifecycle.
• Bibliometric analysis was employed to identify elements of scientific data governance.

Main Results:

• The proposed framework includes three core dimensions: data stakeholders, data lifecycle, and data governance elements.
• Five distinct governance systems were identified: organizational operation, technical support, risk prevention and control, value realization, and regulatory systems.

Conclusions:

• The theoretical framework provides a structured approach to understanding and managing scientific data governance.
• The identified governance systems offer practical components for implementing robust data governance strategies.