Is Adaptive Learning Technology Truly Transformational?
Published Sat Feb 21 00:00:00 EST 2015
How personalized learning technology will change the future of teaching.
Can personalized learning technology ever truly replace teachers? At McGraw-Hill Education, we have hundreds of data scientists and analysts working to understand the science of learning, and for us the answer is simple. Teachers are here to stay, but their role may change.
One of the most important impacts of personalized and adaptive learning is through facilitating and accelerating that change. A change where the unique characteristics of a teacher are critical for the most complex aspects of the educational system: The social learning environment and the creative struggle when students are trying to learn something challenging.
Here's a look at what adaptive and personalized learning can do to help with this:
Learn efficiently. There is a 'mountain of stuff' that students are supposed to learn. Stuff that in isolation is quite easy to understand. The challenge occurs when it is accumulated during a day, a week or a semester. It becomes overwhelming and thus hard to learn fast enough and retain – without help. How students get through this overwhelming pile of stuff varies from student to student. We have indications that it could be significantly more varied than anybody has ever imagined before. Helping the student monitor what is learned well, and what has been learned less well, is one of the key advantages of an adaptive system. It should be quite obvious that some level of adaptivity alone is not enough. Rather, HOW adaptive and personalized the system is able to perform is critical in determining its value.
Our SmartBook adaptive reading platform and ALEKS artificial intelligence-driven learning system have the world's most advanced capabilities in this respect. The level of granularity and sophistication that they offer in terms of understanding the learner's interaction is unprecedented. They are uniquely designed to help students more efficiently learn the basics and build academic confidence. The result of this we hear about often from teachers is that students show up to class better prepared to learn the more complex things – and the teachers spend less time remediating the gaps in learning 'the mountain of stuff'. That time recouped is a gift to both the student and the teacher that can be spent on better things.
Balance workload over time – preventing cramming. From middle school and up, cramming most likely plays a critically important role in inefficient and ineffective learning. We now have 'recharge' capabilities in our SmartBook platform that help students avoid this by allowing them to focus on what really helps them individually over the long-haul. The concept behind this technology is simple, but it is extremely hard to execute: Recharge selectively – based on a learner's individual needs – filters learning objectives from previous weeks of learning that should be 'recharged':
This is a monumental challenge to execute, dependent upon the sophistication and granularity of our adaptive engines and content. When you think about it, this is actually something that is impossible for a teacher to offer: To be able everyday to make a personalized repetition plan for each student taking into account everything that has transpired in the past. But then think again: Which teacher would not like at the end of a semester to see students show up in class prepared to integrate it all?
Accommodating a larger variety of individual ways of thinking and problem solving.
By the end of 2014, we realized that even our wildest expectations on how differently students learn were too conservative. The graph below illustrates the widely varied correct ways that 100+ college students solved the exact same problem in an adaptive math learning system.
This shows the near impossibility of always accurately predicting how students get from the problem at the top to the solution at the bottom. We have adaptive learning technology that can gracefully handle this complexity and variance among individual students. That is vital to be able to accommodate the students at the fringes of the graph. Those students would otherwise be penalized for preceding in a correct, unique manner that works best for them, only because the software engineers did not imagine their path. The consequence of penalizing a student like this is the most poisonous thing in learning: Demotivation!
How do we do it instead? We have made a breakthrough in computer science that allows students to show how they get from A to B allowing them – fully and freely – to formulate their intermediate steps. This technology is critically important to analyze individual differences in how students learn to achieve the best possible outcomes.
This may have a transformational impact on how students study when they don’t have access to a teacher – and how subsequent interactions with teachers occur. We develop cutting-edge products and technologies that address the full spectrum of subjects. Our technologies are already sophisticated enough to cover computational based learning adaptively, and more exciting developments are well underway.
What makes McGraw-Hill Education a learning science company is that we spend 100% of our attention studying student and teacher learning and behavior. We are pioneering and maturing educational technology based on a deep understanding of how people learn and the importance and critical role teachers play in this process. We are obsessed with the design of the next generation of education specific technologies and the most advanced educational content in the world.