Blog

Learning Analytics: an essential learning tool for teachers

When we e-shop or read an article online, every click, like or swipe on a website or mobile application represents events that are sent to servers and stored in databases. This raw data is then filtered, studied, and classified to allow meaningful information to be extracted. Information that might otherwise be hidden due to the sheer volume of data. The resulting “clean” data is called analytical data. It is used in industry, marketing, but can also be applied to human-centered activities such as learning and education.

1. Definition of Learning Analytics and data processing

Les Learning Analytics correspondent à la collection, à l’analyse et au reporting des données d’apprentissage. C’est une pratique de plus en plus utilisée dans l’éducation notamment grâce aux avancées dans le domaine mais aussi l’utilisation grandissante des outils digitaux dans les écoles. Il existe plusieurs Learning Analytics is the collection, analysis and reporting of learning data. It is a practice that is increasingly used in education due to advances in the field and the growing use of digital tools in schools. There are various stages in the implementation of Learning Analytics.

Collection

During the learner’s learning path, several interaction data are collected:

• Firstly, interaction data between the learner and their educational environment. This includes the learner’s enrolment in learning modules, the time they spend and their feedback.
• Then, there is the interaction data between the learner and the learning content which usually represents most of the data collected. This can be test results, time spent on questions or reading learning content.

Analysis

This raw data will then be used to set up more advanced metrics and indicators. For example, it will be possible to define progress and commitment indicators that will allow for better monitoring and understanding of the learner, their use of the tools and their progress.

Visualization

An important part of Learning Analytics is to make the analyzed data understandable in a very visual way. Dashboards containing relevant indicators and metrics are often set up and highlight this data with the help of graphs. The actors concerned will therefore be able to use them to better understand the situation of learners and the potential barriers they encounter in their learning path.

2. Why use Learning Analytics? Who benefits?

By taking benefit of the vast amounts of data available and as data infrastructures improve (from data collection and analysis to visualization and recommendation) learning data analysis offers several significant benefits to learners, teachers and learning organizations.

Learners benefit from better monitoring of their learning and a more personalized experience that allows them to progress at their own pace. 

For teachers, Learning Analytics will enable them to monitor each learner individually, without taking up too much time on their work. On the contrary, this data, intelligently presented in the form of a dashboard, will help teachers and instructors to put training plans for each learner in place.

Learning analytics will also benefit learning organizations. They offer quantifiable feedback on the effectiveness of the educational programs implemented, which will allow them to be challenged and to constantly seek to improve the processes implemented. 

Learning analytics are therefore of growing interest to all the players in the educational ecosystem, from learners to instructors and the learning organizations themselves.

3. Use cases for Learning Analytics

Learner modelling & profile detection

The different interactions of learners, also called learning traces, are very interesting for characterizing and modelling the learner. Metrics such as learner motivation and engagement can be defined from the time spent on the learning path, the number of learning contents viewed, etc… 

Based on these models, it is possible to detect specific learner profiles such as dropouts, cheaters, or talented learners. Teachers will be able to use this information to gain a better understanding of the learners and to implement appropriate educational actions.

Better learner monitoring & remediation

Students can receive more meaningful feedback using learning data. Remedial support for a learner on a topic is provided as soon as the teacher is motivated by the perceived needs of the learners. Traditionally, these needs are only perceived by what a teacher observes in his or her classroom or what can be reflected in the assignments handed in. This constraint not only limits the amount of data that a teacher can act on, but also introduces a delay between the moment when help is needed and the moment when a teacher is finally able to perceive this need and intervene. 

With learning data, the teacher tracks in real time the learner’s activities in and out of class (on online assignments), which are then aggregated and presented to the teacher in the form of an activity report for each learner. The teacher will then only have to implement remedial or encouraging actions for the learner. Thus, better follow-up is guaranteed for each learner.

4. Cautions to be taken 

Although the benefits of learning analytics are great for most educators, there are also concerns about the ownership of learner data, how the data is used, how possible errors and biases are handled, and the validity and implications of teachers’ interpretations.

Educational institutions, as guarantors of the protection of users, often minors, must therefore be clear and transparent about the data they collect and the purpose of the collection and use of the data.

The case of Domoscio

Domoscio Spark’s Learning & Predictive Analytics module allows to adapt pedagogical actions in a predictive way thanks to a better understanding of the learners’ profiles.

Some metrics and learner indicators are monitored and shared via dashboards throughout the learning process of each learner. They are used to detect as early as possible profiles that require teacher intervention to avoid dropping out of school, for example.

Conclusion

Learning Analytics become an essential pedagogical tool for teachers, allowing them to have a better understanding of the learner and thus adapt their pedagogical actions. Underlying these indicators are data and artificial intelligence models. It is therefore important that personal data rights are respected, and that the teacher remains at the center of decision-making to avoid any possible bias against the learner.

Sources

[1] Solar Society for Learning Analytics Research https://www.solaresearch.org/

[2] Kelsey Miller, Northeastern University Graduate Programs (18 février 2020) What is Learning Analytics & How Can it be used? https://www.northeastern.edu/graduate/blog/learning-analytics/

[3] Cerfi, Learning Analytics: à quoi cela sert-il? https://www.cerfi.ch/fr/Actualites/Learning-Analytics-a-quoi-cela-sert-il.html

Illustration by Storyset

---