With the help of various APIs dedicated to capturing the behavioural data generated by your learners, it’s possible to create a system which customises learning pathways for individual students. In this article, we will explore how Tin Can API can be used to achieve an adaptive learning system.
In Tin Can, all personal data is sent to and stored in the learning record store. Tin Can also retains collective data which can be leveraged. Personal data can contribute to personalised learning experiences, and collective data can be used to improve UX, and create student behavioural predictions. By utilising this data, it is possible to create personalised learning and adaptive learning systems for your students.
Draw more accurate profiles of individual students by utilising personal data. What kind of learner is the student? Which source do they most often consult when seeking answers, YouTube or Wikipedia? Learning and development (L&D) experts’ opinions are especially valuable in determining which app data is most useful in identifying learner traits. Analyse your students’ attention spans by monitoring their web browser activity. How long does the student study for before they open up their email, or start watching music videos? These observations are possible through the Timestamp trait provided by the API.
If you have enough data to build models of your students’ behaviours, data mining is useful in allowing you to adapt to changes. You can use the data to predict a student’s learning rhythm, and detect abnormalities in their learning progress. For example, if a student is taking longer than usual to answer a quiz question, you can adapt by lowering the difficulty level, and providing more auxiliary materials.
Collective data can be utilised to make improvements in your UX. Ask questions about your students’ behaviours: do students seek extra sources to understand particular concepts? What apps are being opened and used when dealing with specific material? Maybe incorporate them into the learning material or the webpage. UI designers can also use collective data to find unique angles and leverage better UI.
Again, this is where L&D experts should be consulted to help identify what attributes of student behaviour are related to the score. A model can then be built which predicts learning ability (assuming the score indicates learning ability) with whatever the student is presenting, and then creates an adaptive learning path. With further observation of how students react to the process, this model can be adapted and optimised in return.
By utilising APIs like Tin Can, it is possible for RTOs to create adaptive and customised learning systems for their students. These systems can be optimised as behaviours change over time, and help you create personalised pathways for your learners.
VET moves fast. Stay informed, with blogs straight to your inbox.