Using learning analytics to boost eLearning
Analytics vs reporting in Moodle. One of the great benefits of eLearning courses is that they contain lots of juicy digital data. This means a tutor can access data to analyse factors like student engagement, enabling them to monitor individual student behaviours and the strengths and weaknesses of their courses.
For example, tutors can see how much of your course a student has completed. They can also see which options the student has taken and even how long they have spent on any topic. What’s more, educators can monitor any assessments the student has submitted, their scores over time, and their interactions with them.
Ultimately, learning analytics help tutors evaluate student performance and their programs’ effectiveness. This data can also be used when designing future courses to optimise content in line with past successes.
Analytics vs Reporting
However, many analytics tools are simply descriptive in nature. That is to say; they report information only on what has happened already.
Some of the more sophisticated analytics tools include predictive analytics. That’s when data on student behaviours and achievements within an eLearning course can be used to predict future learning needs and development.
The AI model in Moodle can process data in the system and make predictions. It also provides insights based on simple rules. These insights are then emailed to relevant parties as notifications to act on.
The machine learning model analyses previous student interactions with activities and resources on the site (essentially training itself using sample data) and uses that data to make predictions for current students (insights).
Analytics Example
One example of this is our ‘Students at risk of dropping out‘ model. Our analytics software scans the data on student interactions within a course. It flags up behaviours – In summary, such as when a student disengages – that suggest the student is showing signs of dropping out. Yet, the tutor can then take appropriate steps to prevent this.
Interestingly, the tool uses a highly sophisticated system of analysis based on the Community of Inquiry model of student engagement. It monitors behaviours based on cognitive, social and teacher presence throughout a course.
All this offers invaluable insights to the tutor. However, students also benefit because interventions based on notifications can have a positive impact.
In addition to predictive analytics, the best tools include diagnostic and prescriptive insights. This means they can suggest why something happened and offer suggestions to help improve the situation, perhaps based on data from similar problems in the past.
Moodle Analytics Model
Moodle supports two types of analytics models, as per below:
- Machine-learning-based models, including predictive models trained using site history to detect or predict hidden aspects of the learning process
- “Static” models to detect situations of concern using a more straightforward, rule-based system of detecting circumstances on the Moodle site and notifying selected users.
Whichever system you use, learning analytics is an essential tool for eLearning providers. Without them, you risk flying blind and making similar mistakes in the future. Unfortunately, you will risk losing students when you could quickly have taken steps to retain them.
If you would like to know more about how to integrate learning analytics into your LMS, contact Pukunui to discuss the options.
