As defined for the First Conference on Learning Analytics and Knowledge 2011, “learning analytics is 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 it occurs.”
As outlined in the Educause report, Building Blocks for College Completion: Learning Analytics: “In practice, learning analytics offers a robust way to use data to help assess student risk, predict student success, and improve tailoring of instruction and student services to meet student needs. When conducted within the boundaries of how an institution approaches the privacy of student data and in alignment with an institution’s philosophy about academic interventions and student self-direction, learning analytics can help pinpoint how a student might be veering off track in academic progress in ways that suggest interventions to improve academic performance. Using information gleaned from learning analytics, the institution can improve its early detection and remediation of risk factors for student failure or attrition in order to improve student persistence toward course and program completion. ... More broadly, institutions can apply lessons drawn from learning analytics to deepen student learning and engagement overall by fine tuning pedagogy, enriching student services, and generally honing how they deliver educational content.”
Powerful analytics integrated into online learning enable students to fully understand their strengths and weaknesses in terms of the learning objectives of their courses and programs.
Learning analytics also permit:
- In depth knowledge of what concepts and skills students find difficult and struggle with.
- Learning materials to be individualized around the strengths and weaknesses of the learner. Using adaptive learning systems (e.g. Knewton) as students demonstrate competency or lack of competency, the available learning materials are automatically adjusted to take learner performance into account.
- Identifying students requiring additional support and remediation for their learning.
- Providing feedback to course developers on the efficacy of their design.
In its Horizon Report > 2013 Higher Education Edition, New Media Consortium listed Learning Analytics with a two to three-year time-to-adoption horizon, seeing the promise as actionable data relevant to every tier of the education system. The article provides links to applications and sources.
There are many examples of effective learning analytics in use, especially now that most learning management systems have improved analytics embedded within them.
The examples include:
- Looking at Patterns of Interaction – between students, between students and their instructors, and students and the learning materials. This helps identify levels of student engagement and those students who are not fully engaged in their course work.
- Predicting Success and Failure – mining data from thousands of student interactions and activities permits the development, by subject, of predictive models of student success, drop-out and failure. A great deal of attention is being paid to reducing drop-out in colleges and universities. Being able to predict which students are “vulnerable” and the development of appropriate intervention strategies can have a significant impact.
- Adaptive Learning – using one or more of the available software tools (e.g. Apila, KnowRe, DreamBox, Knewton), instructors can create formative assessments, which lead students to the knowledge they require to be successful.
- For further information, some sites to consult include:
- Edutech Wiki
- Open Universiteit, Netherlands
- Third Conference on Learning Analytics and Knowledge, 2013