Your Source for Learning
Technology, Strategy, and News
ARTICLES       RSS feed RSS feed

Share:

email print

One Size Won't Fit All: The Future of Personalized Digital Learning

by Jeff Batt

July 18, 2016

Feature

by Jeff Batt

July 18, 2016

“This does call for creating courses in a different way and for becoming more user-centric. It forces thinking beyond static page turners. It requires that we create courses that can flex and adapt based on user data. This may seem hard, and we may need to engage with web development teams and other talents outside the normal learning teams, but this approach forces us in the right direction. It is the path we need to take for digital learning.”

In June, I attended The eLearning Guild’s new FocusOn Learning Conference & Expo. There was one particular takeaway from the conference that I think we need to embrace more as an industry, and I want to share that thought with you. The thought is this: What if your digital learning could adapt, adjust, read, and assess the needs of each of your “digital students,” then serve up a personalized course specific to each of them based on their personal learning history? How cool would it be if we could create a personalized learning experience for each student who takes our courses?

Personalized learning? What is that?

What do I mean by personalized learning? Imagine, if you will, that we are in a classroom together. I am the teacher and you are the student. You came to my class to learn how to use a certain software program. Before the class, you decided that you wanted to review some tutorials and documents that would help you feel ready to dive right into the content. But others in the class did different things to prepare, and some did not prepare at all, so each person is on a different level.

As the teacher, I prepared a lesson, but I wanted to make sure I would be ready for any type of question that might come up. So I almost over-prepared to make sure I was knowledgeable enough about the tool and ready to answer any question about the software. I then organized a plan based on where I thought the class would go, but I also prepared for other situations that might happen. So as a teacher, I am able to adapt and adjust my lesson depending on the learners’ needs and the questions that may come up as I am teaching the class. But suppose that, during the class, you have questions that are different than what I had planned.

Now, I can take your questions and respond in one of two different ways. I can ignore your questions and continue with my plan, or I can change my plan and spend some time helping you understand some of the basics, and then adapt the rest of the content as I go. If I am a good teacher, I will choose the second option. I would see the students’ needs or identify their knowledge gaps and then do what I could to help fill those knowledge gaps. In that way, when I get back on schedule it will go more smoothly than if I had tried to ignore those knowledge gaps and just push the content I prepared.

For the most part, as “digital teachers” we tend to just push our content as the solution, a “one-size-fits-all” course, ignoring where users may be in their current knowledge. In person, I am able to adapt, change, adjust, read, and assess the student needs in real time and change my approach based on what I learn about the individual learners. Because of that, I can personalize my teaching to better help everyone in the class.

So how do we do that digitally? How do we adapt, adjust, read, and assess our digital students’ needs to create a personalized experience in a digital format? Most current eLearning tools only allow you to push content, creating a presentation without really allowing for the customization and personalization that the users need. Yes, I know some allow you to collect the user’s name and then plug it in throughout the course. That is a good start, but I am talking about creating a completely different course for each user.

Getting to know the user

So how do we get around that? Do we have to stop teaching digitally and move back to in- person courses? No, of course not. How do we create digital learning that can adapt, adjust, process, and assess our learners’ needs? How do we create training that can get to know the users, identify their knowledge gaps, figure out what they have done and what they haven’t done, and then serve up instantly a personalized digital course that best suits their needs?

One solution could be as simple as adding “get-to-know-you questions” at the very beginning of a course. The responses would allow you to create logic on the fly for the course: which sections to serve up and which sections not to deliver. You could even modify the content based on user responses.

If you have not heard of Consensus (www.consens.us), you should check it out. This application targets sales organizations, but I think the concepts behind the approach could easily be applied to the learning sector. Instead of just asking prospects to click a button and watch a one-size-fits-all sales demo video, the technology first asks questions of the individual. Knowing what the person already knows and what they don’t know, what their needs are, and which features are of most interest to them, the application creates a custom sales demo for each prospect. This demo shows longer sections of the video that target the prospect’s interest and shorter video snippets for features the prospect cares about less.

Holy smokes! Can you imagine doing that for learning? What if you gathered some user info before a course to set up the logic and then programmatically adapted the course for the user? It is not as personalized as a one-on-one discussion, but it gets closer than static one-size-fits-all learning.

Knowing how to write some basic code could accomplish this. You could prompt the users with questions, store their responses in some variables, and then programmatically take them down different paths based on those responses.

The xAPI—personal learning history

We have been hearing about the xAPI (Experience API) for years now. We know that it allows us to track data from any location and that we can get more detailed data than we ever could with SCORM. With the xAPI, we can assemble what are called “statements.” This allows us to track various learning activities, such as: Jeff Batt attended FocusOn Learning Conference. This statement has three important elements—the actor (who it is: “Jeff Batt”), the verb (the action: “attended”), and then the object (name of the thing the actor acted upon: “FocusOn Learning Conference”). Sending over these statements allows us to start tracking learning history. Every time we send a statement, that event is now stored in an LRS (learning record store), which gives us greater insights into what our learners are doing and how our content is performing.

But probably the most fascinating part is that it also gives us the ability to programmatically go into the LRS and, on the fly, see what a learner has done and hasn’t done, and use that data in whatever way we need to use it. Imagine: With a learner’s history built over time in the LRS, we could create a personalized course that first goes into the LRS and asks the question, “What has the learner already learned or experienced?” With the information that the LRS gives us, we could adapt, change, adjust, read, and assess the student’s needs, and deliver a different course based on their personal learning history. This is just like I explained in the first example (“Personalized learning? What is that?”), but now we don’t need to ask the users questions or ask them to fill anything out first—the course can obtain the information automatically, and in fact, it would have more data to work with than it could get from a form.

Figure 1 is a schematic of the way this could work.

Figure 1: Schematic of xAPI support for personalized learning

This may seem “Big Brotherish,” but it is not far from what is already being done on the web. How does Amazon know what I want? Facebook uses what it knows about you and your behaviors to deliver customized, personalized marketing ads for things you like. Target has also been doing this for years.

I have done some initial tests of this concept, and it looks promising. I am excited to explore this use of the xAPI in more depth. This does call for creating courses in a different way and for becoming more user-centric. It forces thinking beyond static page-turners. It requires that we create courses that can flex and adapt based on user data. This may seem hard, and we may need to engage with web development teams and other talents outside the normal learning teams, but this approach forces us in the right direction. It is the path we need to take for digital learning.

So where can you start? The eLearning Guild offers xAPI Camp pre-conference workshops at all of its live conferences (details of the next xAPI Camp are available on the DevLearn 2016 website). I think that would a great place to start, to dive in, and to really get to know the xAPI. I see this as a way to stop just pushing our content in a one-size-fits-all format and start creating personal experiences. What are your thoughts? Do you see eLearning going in this direction? What would it take to do this? Please share your thoughts in the comments on this article.

Topics Covered

(35)
Appreciate this!
Comments

Login or subscribe to comment

Could you share the technical environment that you used to test this concept? Which LRS did you use? How did you design the courses so that 'statements' could be sent over to the LRS?
@diptendu Great questions! We build our tests in HTML, CSS and JavaScript. Reason being we had more control on how we sent over statements and how we were able to get statements and adapt our content based on what we got back from the LRS. Standard rapid development tools do not really allow for this yet. The LRS we have done our tests on is Watershed. Great LRS, highly recommend it. As far as the design of the statements since it is mostly initial concepts for now we just decided on some basic statements to start tracking and now we are in the strategy phase of really planning how to use and track each verb. So the verbs we are using are experienced, completed, replied to and verbs like that so we could see what they have done and adjust accordingly. Hope that helps. Let me know if you need any more clarification.
Related Articles