The Society for Learning Analytics Research (SOLAR) defines learning analytics as
the measurement, collection, analysis and reporting of data about learners and their contexts for purposes of understanding and optimizing learning and the environments in which learning occurs.
Learning analytics has the potential to:
- Empower students,
- Offer instructors formative feedback,
- Identify, earlier, students in need of support,
- Illuminate curriculum connectivity,
- Improve curriculum alignment,
- Improve assessment of learning,
- Improve evaluation of teaching.
In the field of learning analytics we ask questions such as:
- What can clickstream data from learning technologies tell us about how students learn?
- What does ‘learner engagement’ look like in virtual learning environments and how can we promote and support it through learning design choices?
- What are the most common themes in student feedback comments?
- What are the most common enrollment choices our learners make to complete their degrees? Do ‘high achievers’ make different choices than ‘low achievers’?
- Which students are most at risk of failure? (How early can we identify them and provide better support?)
- Who are our students? Where do they come from and how has that changed over time?
In practice, learning analytics involves the extraction of large volumes of data from institutional databases and learning technology systems, and the application of various statistical and analytic techniques. Although most educational institutions gather large amounts of data about their learners, courses, instructors and processes, much of this data has remained unused because until recently, analysis has been extremely laborious. Now, new analytic tools and platforms are available that allow us to mine and examine large amounts of data, with the goal of identifying trends, patterns, and correlations that can inform us about teaching and learning.
- eLearning Analytics
Does learner behaviour in online learning contexts contribute to their eventual learning outcomes? This project seeks to discover which learner practices may correlate meaningfully with success by applying statistical analysis and social network analysis to data from an institutional LMS and/or other learning technologies. The goal is to inform development of interpretive tools that will act as an “early warning system” of students at risk of failure.
-
Visualization of course enrollment patterns and pathways Analysis of temporal data
‘Time’ is an important dimension of many educational data sets, whether we we are exploring patterns of student and instructor activity in a classroom, learner problem-solving strategies online, or student course enrollment choices. Important information can be lost if we only explore averages and aggregates. Projects in this area seek to investigate and illuminate temporal information that may be significant for teaching or learning.
- Text analysis
What are the most common topics of discussion in an online forum? Can we analyze student writing for evidence of critical thinking? Which themes arise most frequently in student feedback commons? Work in this area is underway to explore and test analytic approaches to text.
Leah P. Macfadyen 