In October, I facilitated a Morning Buzz session about personalized learning at DevLearn 2017 Conference & Expo in Las Vegas. I have written about the topic before, and it has been on my mind lately. In eLearning, we tend to make one-size-fits-all courses. We don’t tailor the content to give individual learners options to explore topics they are unfamiliar with, and to advance if they already know a topic. We need to get past this one-size-fits-all eLearning and look at ways to create tailored digital experiences for our learners. We need to focus on personalized learning.
I speak from personal experience. I struggled with math in college, often feeling like the teacher was five steps ahead of me. This was not the fault of the teacher, who was trying to deliver the same experience to all 30 students—but not all of us were at the same level. Contrast that with a tutor, who assesses a student’s knowledge and then tailors the lessons to that learner’s needs. This is where digital learning could (and honestly should) be.
I focused my DevLearn Morning Buzz session around three basic questions:
- What is personalized learning?
- Why should we care?
- How can we start doing it?
The informal conversation yielded lively discussion. I will share some of the highlights with you.
What is personalized learning?
Custom learning paths: Allow learners to create their own career learning paths. What topics do they want to learn about? What role(s) in the company do they want to strive for? Can we help curate—or have others already in those positions curate—paths that would help them get there?
Starting from a learner’s current knowledge base: Instead of making assumptions about what learners know (or should know), let’s instead focus on creating training options that allow users to dive deep into areas they are foggy on.
Assessments: Part of personalized learning could include some kind of opt-out quiz or assessment that evaluates where the learner’s knowledge is currently, and tailors an experience to that.
Diagnosis leads to dynamic learning: I loved the simplicity of this concept. There are two important elements to personalizing learning. First, see where your learner is at—then, dynamically create content based on that diagnosis. This reminds me of a doctor prescribing a solution based on symptoms.
Curated content: David Kelly, executive vice president of The eLearning Guild, talks about this a lot. I think there is great power in having an expert (defined as someone who has already traveled the path) suggest to others how to travel the same path.
User’s learning history: xAPI can track a user’s learning history, providing greater detail than SCORM ever could. It allows us to customize a course based on what the user has done or not done.
Invisible: Not all learning happens on an LMS. Personalized learning must be where the learner is at, when he or she needs it. So we need to get into the places where our learners are and where learning happens in order to truly tailor the experience.
Why should we care?
Increase learner engagement: If users can understand the why, they will be more engaged in the content. Seeing how the content applies to them will help them retain it.
Positive experience: We all know the course that everyone complains about—the one that is hours long and contains non-relevant content. Such experiences give eLearning a bad name. If content is relevant and useful, the user will have a positive experience and word will spread.
Relevant content: Users care more when content is relevant. Concepts like just-in-time learning and better performance support apply here.
Competency-based: I love the idea of not only testing for competency, but adjusting the training to the competency of the user. Meet the learner at his or her current level of knowledge. (If I had something like this for math in college, I wouldn’t have struggled so much.)
Buy-in and feeling of belonging in the company: One of the best examples I saw of this at DevLearn was with Visa University. They have a tool that allows any user to create a learning path. Others can enroll in the learning path and even vote on the usefulness of it. As the path is shared across the organization, that individual gets rewarded with points and exposure to upper management. It is a way for someone to stand out and really own learning.
How can we start doing it?
xAPI: A lot of what we need to personalize learning can be done with xAPI; we just need to embrace and start using it. With xAPI, you can get specific details about learning behavior and query (or ask via code) what the learner has done. Based on the results, you can adapt or change your content. Think of this like Amazon recommendations. Amazon tracks your browsing and purchase history in order to personalize recommendations for you.
Boosters: Art Kohn talks about boosters. These are short, simple questions sent out at certain intervals to help the learner remember what they learned. The repetition helps with retention. The book Make it Stick (by Peter C. Brown, Henry L. Roediger III, and Mark A. McDaniel) talks about similar concepts.
Large, medium, and small-sectioned courses: Before you click Play, one company asks a couple of questions about what you are looking for. Based on your answers, it shows you the longer versions of the topics you are interested in and shorter versions of the topics you are not interested in. Although this is a marketing tool, it can also be used to personalize learning. What if you built small, medium, and large sections for each topic? Then, based on the quiz results, you can adapt the course to navigate to each relevant section. This way, the learner can dive deeper into the section(s) relevant to them.
Explain the why: Often, we don’t explain why something is relevant or how it could benefit the user. I think of my daughter when she was younger. If we sat her down and explained our decision, she would buy into it and be OK. But if we just said, “Because I said so,” she would throw a fit. If learners know the why, the how makes more sense.
Component-based reusable learning: Melissa Milloway often talks about this, and I think it is something worth exploring. A popular concept in web development is to make items component-based. This means you make parts of a website reusable so that the same components can be used in several locations. When you update it in one location, it updates everywhere.
Marketing analytics can also be used to customize a web page with personalized components relevant to that user. For example: If I have searched in the past for superhero merchandise, it may pull in a superhero merchandise component showing me some products I may be interested in, while another person with different interests may see something else. If we section off our learning into components, we may be able to piece together pages in different ways for each learner.