LMS Learning Analytics In 2019: An Update – MoodleNewsCMS
We’ve come a long way since conflating student clicks with engagement. But after an explosion of approaches, techniques, solutions and buzzwords, it can be a real challenge to identify Learning Analytics roadmaps worth pursuing.
Follow our Learning Analytics Competency Roadmap series.
Excited about the arms race for better, more intuitive and useful Learning Analytics from the top LMS providers? Keep it strong by the following recap of latest updates to core solutions or third-party providers.
Moodle Learning Analytics: Exciting 3.7 release coming up
Moodle Learning Analytics continues its steady progress towards a more friendly solution. Here are some of the issues on the works, currently fixed or ready for Moodle 3.7.
The Analytics model will send data to Moodle
How do Moodle users take advantage of the Analytics engine, or do not? To answer, the team is introducing a series of changes on the Moodle site registration. When Moodle checks your site, it counts the number of users, courses and actitivties made. From its latest check it was able to claim that by March, 1.5 billion quiz questions have been created.
After the latest update, your Moodle site will send Learning Analytics stats to the team. These include number of models installed and trained, samples, and some central tendencies. See the full list here. Moodle will not collect student-level information nor personal data.
Better model editing controls
Tensorflow is one of the most popular machine learning solutions today. Built by Google on the Python language, it is the basis of Moodle’s prediction engine. It is powerful, versatile and scalable. But its adoption continues to lag. The team works non-stop on a better experience for Model developers and tinkerers.
An upcoming update will allow a better experience building and collaborating on models. Users will be able to develop specific model configurations and share them. The models will not include any “learning” from specific data, only initial specifications. Target, indicators and the time-splitting methods are part of the import-export features.
New model targets available
Two new targets are now readily available for users. They can now choose “Course Completion” and “Competency Achievement” as targets. It means users will able to predict if students will complete a course or achieve a competency.
To be sure, in the Moodle predictive engine any variable can become a model target. This updates makes it possible to define a target without modifying the code of the model.
Sakai: Apereo Learning Analytics Initiative (LAI)
An initiative similar to Moodle’s Project Inspire (today’s Analytics API) and the Working Group, it currently works at a smaller scale despite serving several technologies beyond the Sakai LMS.
A series of projects under the umbrella of Apereo LAI involve the development of an analytics dashboard, increasing compatibility with xAPI and other standards, and other supporting and related developments.
A commercial bundle of Apereo’s Learning Analytics solutions are available under the Shuhari brand.
Although the main pages and documentation have not been updated since 2016, the public GitHub page for the project shows steady development on a weekly basis.
An interesting element of Apereo’s approach is the deep focus on modularity, vastly superior to Moodle’s. The Dashboard, for example, is build separate from the model, and in fact other organizations have built their own Dashboards, something which Apereo seems to encourage.
What is perhaps Apereo’s most noteworthy learning analytics development consists on a cards-based display of “single discrete visualizations.”
A newer release of what it was OpenLRS, Apereo’s take on the Learning Record Store, but with the word Warehouse instead. The latest upgrade, come last January, adds risk scores for class users (“MongoRisk”) and timezone offsets.
Blackboard Analytics (formerly X-Ray)
Thanks to the Blackboard Data Science team led by Dr. John Whitmer, the EdTech giant is one of the thought leaders in learning analytics. The acquisition of X-Ray analytics has morphed from an add-on to a bundle that today provides customers with solutions across many dimensions. Besides Blackboard Analytics, the company offers Blackboard Predict, and Blackboard Intelligence.
Unfortunately, this makes us unable to keep track of the new features and improvements on the technology. You can still find some info and schedule a demo at blackboard.com/education-analytics.
Zoola Analytics by Lambda Solutions
An analytics solution by a Totara and Moodle partner, Zoola is estimated to be the most widely adopted learning analytics system, to a large extent thanks to its adoption by large corporate clients. This speaks to its focus on robustness, as well as its seamless integration with Moodle or Totara LMS, in particularly those served by Lambda.
One of the most popular and well-regarded analytics platform, the only analytics-based Moodle Premium Integrators is readily available for Canvas LMS, Blackboard and Moodle. Certified Partners of the latter, such as eCreators, offer IntelliBoard baked into its Moodle solution, with the option to upgrade to its premium tier.
IntelliBoard is considered best-in-breed when it comes to real-time monitoring, customizable dashboard, and even xAPI compliance. Their latest dashing new feature is LISA, a natural languange processor that creates custom reporting, much like Google Analytics.
A free trial is available. For more, visit intelliboard.net.
Disclaimer: Lambda Solutions and IntelliBoard are MoodleNews Partner.
This Moodle Technology related post is made possible by: Moonami a company that provides a full range of Moodle services that combine the flexibility, scalability, and power of Amazon’s world-leading cloud platform (AWS) with fanatical Moodle support. Click here to learn more.