Using a topic model to map and analyze a large curriculum

Area of Science:

• Medical Education
• Curriculum Development
• Educational Data Mining

Background:

• Understanding medical curriculum content is vital for assessing learning objectives.
• Medical education curricula face challenges due to vast content, topic diversity, and numerous faculty.
• Existing methods for curriculum analysis are often insufficient for complex medical programs.

Purpose of the Study:

• To develop a manageable representation of content in the pre-clerkship medical curriculum at Yale School of Medicine.
• To quantitatively map curriculum content to school-wide competencies.
• To enable tracking of specific content areas and measure inter-course integration.

Main Methods:

• A topic model was generated using all educational documents provided to students during the pre-clerkship period.
• The topic model was employed to quantitatively assess the coverage of each topic within the curriculum.
• Content mapping was performed against established school-wide competencies.

Main Results:

• The topic model successfully provided a quantitative overview of curriculum content.
• A previously under-identified content area, gender identity, was highlighted and its coverage tracked over four years.
• The model enabled quantitative measurement of content integration both within and between courses.

Conclusions:

• Topic modeling offers a robust method for analyzing and understanding complex medical curricula.
• This approach facilitates the quantitative assessment of curriculum content alignment with learning objectives and competencies.
• The methodology is adaptable to other educational settings where textual data from curriculum materials can be extracted.