The Use of Natural Language Processing to Visually Summarize Characteristics of Narrative Feedback to Medical Students

• In Event: Assessment and Feedback in Health Professions Education

Abstract

This study describes how natural language processing (NLP) techniques are used to visually summarize large amounts of narrative feedback provided to medical students in an assessment portfolio. Narrative comments are an effective way to provide feedback on complex behaviors such as communication, professionalism and teamwork, but it is time intensive to read through numerous comments to judge performance. Logistic regression methods identified textual features associated with students in need of additional support based on five years of faculty reviews. Summary reports were created to visualize the presence of these features for individual students. To our knowledge, this study represents the first time NLP techniques have been used to summarize and visualize longitudinal narrative feedback on medical student performance.

Authors

• Christina Maimone

• Brigid M. Dolan, Northwestern University

• Marianne Green, Northwestern University

• Celia Laird O'Brien, Northwestern University