Our situation: Educators often find themselves in possession of large amounts of text-based materials, such as student reflections, narrative feedback, and assignments. While these materials can provide critical insight into topics of interest, they also require a substantial amount of time to read, interpret, and use. The purpose of this article is to describe and provide recommendations for text analytics.
Methodological literature review: An overview of text analytics is provided, including a brief history, common types of contemporary techniques, and the basic phases of text analytics. Several examples of common text analytics techniques are used to illustrate this approach.
Our recommendations and their applications: Practical recommendations are provided to support the use of text analytics in pharmacy education. These recommendations include: (1) clarify the purpose of the text analytics; (2) ensure the research questions are relevant and grounded in the literature; (3) develop a processing strategy and create a dictionary; (4) explore various tools for analysis and visualization; (5) establish tolerance for error; (6) train, calibrate, and validate the analytic strategy; and (7) collaborate and equip yourself.
Potential impact: Text analytics provide a systematic approach to generating information from text-based materials. Several benefits to this approach are apparent, such as improving the efficiency of analyzing text and elucidating new knowledge. Despite recent developments in text analytics techniques, limitations to this approach remain. Efforts to improve usability and accessibility of text analytics remain ongoing, and pharmacy educators should position their work within the context of these limitations.
Keywords: Machine learning; Sentiment analysis; Text analytics; Text mining; Topic modelling.
Copyright © 2022 Elsevier Inc. All rights reserved.