Paper Info
Title
Developing reflection analytics for health professions education: A multi-dimensional framework to align critical concepts with data features
Abstract
Reflection is a key activity in self-regulated learning (SRL) and a critical part of health professions education that supports the development of effective lifelong-learning health professionals. Despite widespread use and plentiful theoretical models, empirical understanding of and support for reflection in health professions education remains limited due to simple manual assessment and rare feedback to students. Recent moves to digital reflection practices offer opportunities to computationally study and support reflection as a part of SRL. The critical task in such an endeavor, and the goal of this paper, is to align high-level reflection qualities that are valued conceptually with low-level features in the data that are possible to extract computationally. This paper approaches this goal by (a) developing a unified framework for conceptualizing reflection analytics in health professions education and (b) empirically examining potential data features through which these elements can be assessed. Synthesizing the prior literature yields a conceptual framework for health professions reflection comprised of six elements: Description, Analysis, Feelings, Perspective, Evaluation, and Outcome. These elements then serve as the conceptual grounding for the computational analysis in which 27 dental students’ reflections (in six reflective statement types) over the course of 4 years were examined using selected LIWC (Linguistic Inquiry and Word Count) indices. Variation in elements of reflection across students, years, and reflection-types supports use of the multi-dimensional analysis framework to (a) increase precision of research claims; (b) evaluate whether reflection activities are engaged in as intended; and (c) diagnose aspects of reflection in which specific students need support. Implications for the development of health professions reflection analytics that can contribute to SRL and promising areas for future research are discussed.
Year
DOI
Venue
2019
10.1016/j.chb.2019.02.019
Computers in Human Behavior
Keywords
DocType
Volume
Reflection,Learning analytics,Natural language processing,Professional education,Dental education,Health professions education
Journal
100
ISSN
Citations 
PageRank 
0747-5632
3
0.41
References 
Authors
0
3
Name
Order
Citations
PageRank
Yi Cui18712.46
Alyssa Wise221319.86
Kenneth L. Allen330.41