Educational data mining in medical education: A five-level approach.

Area of Science:

• Medical Education
• Data Science
• Healthcare Informatics

Background:

• Healthcare education systems generate substantial daily data.
• Analyzing this data can reveal insights into knowledge, skills, and attitudes.
• Such analysis has the potential to revolutionize medical education.

Purpose of the Study:

• To review the application of data mining in medical education.
• To provide a comprehensive overview across various educational dimensions.
• To offer recommendations for enhancing data mining techniques.

Main Methods:

• Literature review conducted from 2010 to 2022.
• Searches across major academic databases (IEEE, SSCI, Elsevier, CIVILICA, Science Direct).
• Inclusion of 34 selected articles after initial identification of 250, with quality assessed using Carnwell and Daly's method.

Main Results:

• A five-level taxonomy was developed: educational policy and management, instructional designing and planning, educational technologies, learning content, and learning outcomes.
• Recommendations for improving data mining efficiency at each level were presented.
• Educational data mining (EDM) facilitates better policy-making, planning, and decision-making.

Conclusions:

• EDM is a valuable methodology for medical education.
• It enables the description and prediction of educational trends.
• EDM enhances the potential for success in medical education through data-driven insights.