|Create Date||May 8, 2017|
|Last Updated||April 20, 2017|
By Elliot Gowans, VP of sales EMEA, D2L.
We swim in a sea of data. It pours in from wearable technology, GPS location tracking, our smartphones, and even our appliances and vehicles. In fact, with all the recommendations, curations, and suggestions available to us now, even our data has data.
Many organisations have learned that data collection and analysis are important, but what they should really be asking themselves is, ‘how smart is our data?’
For educators, the data challenge is no different. Educators don`t necessarily need to access more data, but rather we need to learn how to progress from ‘more data’ to ‘smarter data.’ How do we navigate through the sea of information to give it meaning and context that helps student learning outcomes? One way educators can make practical use of their data is through predictive analytics.
Predicting future behaviour
To paraphrase Dr Phil: “Past behaviour is the best predictor of future behaviour.” That’s why we ask so many questions in a job interview about a candidate’s previous experience; we’re creating a data set about that person’s past conduct. But what if instead of a brief conversation, you had 20 million data points and the ability to analyse and identify patterns in an instant that let you know ahead of time which candidates are best suited to the job?
That’s what predictive analytics is all about. Even if you’re not familiar with the concept, you’re probably familiar with popular tools that use predictive analytics, such as the auto-complete feature in Google or personalised recommendations from Netflix. When we bring ‘big data’ together with machine learning, we have the power to change the field of education by giving instructors a new way to improve the learning experience, and learning outcomes.
We can make predictions about people who might be ‘at risk’ in a learning environment by comparing their current activity within a course to large sets of modelled data of past performance in similar courses. This can work across many types of data, from course access, to reading content, to assessments, to analysing their social interactions.
But that’s only the first stem. Some of the most progressive educators are combining predictive analytics with machine learning. Technologies like course recommendation and adaptive learning are helping instructors make a real difference in the classroom by zeroing in on the students who need help before it’s too late.
Tailor future learning activities
Adaptive learning uses past student behaviour to tailor future learning activities, from connecting the student to external resources, to recommending something as simple as talking with a peer or taking a break. Built on the foundation of pedagogical expertise of the instructor, adaptive learning lets the learning experience adapt to the student, instead of asking the student to adapt to the learning experience. After all, everyone is unique, and we all have our own ways of learning.
Enabling smarter decisions about selecting courses also has a tremendous impact on personalising learning for students and educators. Course recommendations provide new support for course selection, a high stakes decision that’s critical for student success. Selecting courses normally happens just six to eight times over an entire degree, and possibly fewer times for professional development. When we don’t make the most of these opportunities, we risk education costing more energy, time, and money. We also risk lower retention and completion.
Moving from term-based analysis to individual moments, adaptive learning provides the opportunity for greater personalisation of the learning experience in real time. This is when the student can benefit most from feedback on their learning activities. Smarter data coupled with a shift from mass production to mass personalisation by delivering the right data, at the right time, in the right place.
Predictive analytics enables better informed, evidence-based decisions that can deliver better outcomes for student retention, completion, and ultimately student success. Through the new capabilities for data-driven and evidence-based decision making, predictive analytics helps instructors and students become stronger swimmers in a sea of data, and brings new levels of rigour and consistency to drive student success.
D2L is a software provider that makes learning experiences better. The company’s cloud-based platform, Brightspace, is easy to use, flexible, and smart.
Download was expired on April 5, 2017 12:00 AM