Data, Information, Evidence, and Knowledge:: A Proposal for Health Informatics and Data Science

Area of Science:

• Information Science
• Knowledge Management
• Philosophy of Science

Background:

• The widely cited data-information-knowledge-wisdom (DIKW) hierarchy provides a framework for understanding the transformation of data into wisdom.
• While influential, the DIKW hierarchy has faced critiques regarding the definition and transition points between its stages, particularly the ambiguous role of wisdom.

Purpose of the Study:

• To revisit and modify the existing DIKW hierarchy.
• To propose a refined framework, termed data-evidence-information-knowledge (DIEK), by de-emphasizing wisdom and inserting evidence.
• To define the distinct stages and transitions within the proposed DIEK model.

Main Methods:

• Conceptual analysis and modification of the DIKW hierarchy.
• Introduction of 'evidence' as a distinct stage between information and knowledge.
• Definition of data, information, evidence, and knowledge within the proposed DIEK framework.

Main Results:

• A modified hierarchy, DIEK, is proposed, shifting focus from wisdom to a more empirically grounded progression.
• Data is defined as raw symbols, contextualized into information.
• Information becomes evidence when compared to standards, which is then transformed into knowledge through hypothesis testing, success, and consensus.

Conclusions:

• The proposed DIEK model offers a more nuanced and empirically testable progression from data to knowledge.
• Key checkpoints for transitioning evidence to knowledge include relevance, robustness, repeatability, and reproducibility.
• De-emphasizing wisdom allows for a clearer focus on the foundational steps of knowledge creation.