Sohum Bhatt is a PhD researcher at itec, the interdisciplinary research group of KU Leuven and imec. He is researching learning analytics and recommendation systems in the context of the i-Learn project.
This is a personal blog post by Sohum Bhatt, who is a researcher at KU Leuven.
The use of technology in the classroom is constantly increasing. At the same time, the possibilities to optimise and personalise the learning experience of students are growing. But in order to be able to personalise efficiently, a good analysis must be made of the behaviour and results of the individual pupil. In a digital educational environment, this analysis is done on the basis of the data generated within the environment. This data analysis is called learning analytics. Learning analytics focuses on learning and aims to gain insight into the learning process through the analysis of data. Data which, due to increasing digitalisation, is abundant. It is therefore no surprise that scientific interest in learning analytics has also risen sharply in recent years.
Thanks to its broad focus, the field of learning analytics has the potential to have a huge impact on education. However, we are currently seeing the greatest effects mainly within the university context. The i-Learn team wants to change this by also introducing primary and secondary education to the power of learning analytics.
In this blog post, I will look at the usefulness of learning analytics from three perspectives: the learner, the teacher and the care coordinator and/or educational supervisor.
The usefulness of learning analytics for learners should not be underestimated. Its implementation can lead to more personalised support for learners in need, as teachers (and digital tools) gain insight into the learner’s learning process. Thanks to learning analytics, a teacher can be made aware of the exact aspect with which a learner is struggling.
Furthermore, learning analytics also offers the possibility of providing customised feedback. For example, an online learning environment that focuses on English as a new language may automatically provide different feedback to learners who repeatedly misuse verb tenses than to learners who repeatedly misuse prepositions.
Moreover, when learning analytics elements are incorporated into a dashboard (which visually displays individual progress or errors), learners can also more accurately assess their own skills. Through this better estimation of their achieved level, learning analytics can encourage students to deepen a topic that they master less well, or to look for challenging additions to a subject in which they excel.
By integrating learning analytics into their lessons, teachers are given the opportunity to make better estimates of their students’ level of achievement. For example, let’s look at the topic of computational thinking. With the help of learning analytics, a teacher is able to track how a student is acquiring this skill. This gives teachers a very valuable insight into the learning process of each individual pupil. For example, when a class is assigned to develop a game, the teacher – via the learning analytics dashboard – can monitor how a particular student is learning Boolean logic or approaching the programming process. This can help the teacher to get a clearer idea of what aspects should be emphasised more or less in future computational thinking lessons.
In addition, learning analytics can help teachers identify who is experiencing difficulties in acquiring the material. This help is not limited to the identification of a problem, but can also provide insights into the cause of the problem. Perhaps the learner in question does not like the material being used, or has found the instruction confusing, or has simply made a mistake. All of these situations may result in the same outcome – an assignment that was not completed correctly – but the causes may be very different. Learning analytics can help teachers gain a better understanding of why certain errors occur. In addition, it can serve as a useful tool in the future – with the help of a recommendation system – to use other material that is more attuned to the living environment of the pupils. In this way, the teacher’s task of teaching becomes easier and better matches the learning style of each pupil.
Finally, learning analytics can paint a picture of the impact of certain teaching techniques and materials on certain class groups. This makes it possible to personalise the learning process on the basis of the class as a social and human unit.
For care coordinators and educational supervisors
A more administrative use of learning analytics can help identify early and automatically students who need more support than others. This can help care coordinators provide students with the appropriate support more quickly, with a clear focus on the type of help needed for this particular student. For example, if a student’s performance is noticeably worse than normal over an extended period of time, a computer system can alert the care coordinator to the difference in performance. The care coordinator can then have a conversation with the pupil to find out what is going on and how the school can help to get the pupil’s results back up.
Finally, by implementing learning analytics, pedagogical supervisors can also gain more insight into the impact of used teaching techniques and materials. Certain teaching strategies may have specific effects, such as increasing students’ motivation or decreasing the feeling that certain subjects are difficult. This useful information can then be used by pedagogical supervisors during their coaching of teachers.
This brief overview clearly shows that the implementation of learning analytics is useful on several levels. That is why the i-Learn team is also conducting scientific research into learning analytics, based in part on the data obtained from the i-Learn prototype. Where possible, we put the insights we gain from this into practice.