JMIR Publications

ABSTRACT

Background:

Digital education has expanded greatly since the COVID-19 pandemic began. New substantial amounts of data about how students learn have become available for Learning Analytics (LA).

Objective:

The aim of this scoping review was to examine the use of LA in healthcare professions education and propose a framework for the LA life cycle.

Methods:

We performed a comprehensive literature search of 10 databases: MEDLINE, EM- BASE, Web of Science, ERIC, Cochrane Library, PsycINFO, CINAHL, ICTP, Scopus, and IEEE Explore. We screened papers and included them if they met the following criteria: papers on healthcare professions education, digital education, and collected LA data from any type of digital education platform.

Results:

We retrieved 1,238 papers, of which 65 met inclusion criteria. From those papers, we extracted some typical characteristics of the whole LA process and subsequently we proposed a framework for the whole LA life cycle as: digital education content creation, data collection, data analytics and purposes of LA. Assignment materials were the most popular type of digital education content, used in 47 papers. The most commonly collected data types were the number of connections to learning materials, identified in 53 papers, followed by test scores in 19 papers, and time spent on learning materials in 15 papers. Forum interactions and textual data were collected in 11 studies each. Descriptive statistics was the most commonly used data analytics in 58 studies, followed by inferential statistics used in 40 papers, data visualization in 20, machine learning in 18, and social network analysis in nine papers. Finally, among the purposes for LA, understanding learners’ interactions with the digital education platform was cited most often in 56 papers, and understanding the relationship between interactions and student performance, in 41. Far less common were the purposes of optimizing learning: the provision of at-risk intervention, feedback, and adaptive learning was found in 11, five, and three papers, respectively.

Conclusions:

We identified gaps for each of the four components of the LA life cycle, with lack of an iterative approach while designing courses for healthcare professions being the most prevalent. We identified only one instance in which the authors used knowledge from a previous course to improve the next course. Only two studies reported that LA was used to detect at-risk students during the course’s run, compared to the overwhelming majority of other studies in which data analysis was performed only after the course was completed.