ELearning is Still Broken

Student taking online class

Read Part II: Welcome to the Age of Learning

While there are many examples where eLearning is being used successfully in industry today, there is still much room for improvement. 

The two most important deliverables for any learning system should be that it increases the learner’s skills (provably improve those skills as measured by tests) and be as painless as possible (ideally learning should make the student want to come back the next day). But the majority of eLearning solutions are neither. Often they are just traditional courses being delivering digitally. Key problems of completeness, consistency, validation, reinforcement, adaptation to the learner and lack of focus have yet to be adequately addressed.

These are really important problems to solve because we are relying on eLearning to keep us relevant and appropriately skilled with the ever increasing pace of change in technology-based industries (or just to keep us competent in using the next-generation smartphone). And this is just the technology that humans are creating – just wait until we have to keep up with what an AI creates.

But there is hope. The new paradigm of ‘lightning learning’ provides criteria to help measure the problems and shows how to fix them. The principles of lightning learning provide a faster, deeper path to mastery that is more enjoyable for the learner.

Examples of where eLearning isn’t working

Here are just a few of the examples of where eLearning isn’t working:

  • Elearning tools not actually used. A recent study of K-12 school districts found that 67% of the eLearning software goes unused. It is probably as bad or worse in the industry. For instance, I personally have a Lynda subscription languishing on LinkedIn right now that is costing me $32 per month that I'd better start using or cancel.
  • Students not showing up. During the COVID crisis, many middle school teachers are unable to get even 50% of their students to attend their required weekly zoom sessions and have almost given up on assigning homework as it is often not completed.
  • MOOC students not showing up. Elearing is not fun or engaging. In the Science article “The MOOC Pivot”, MIT researchers recently looked at seven years of data from the EdX MOOC (Massive Open Online Course) with courses from MIT and Harvard. It covered 5.63 million learners and 12.67 million course registrations. They found that only a very few students finish the courses that they start (around 3%) and it’s been getting worse (dropping by 1 percentile point per year). And once the students try the MOOC they don’t come back: only 10% of students return the next year after taking a course the previous year.

What’s broken with eLearning?

Here are the main brokennesses (yes, I just made up that word) of eLearning systems from the learner’s point of view:

  • They waste the learner’s time.
  • They don’t follow up to ensure that the learner has really mastered the material.
  • They don’t validate that the learner actually learned something.
  • They frustrate the learner with content that is too difficult or bore the learner with content that is too easy.

What a good eLearning system should have

If you don’t mind wasting your time in order to get that certificate of completion then don’t worry about which eLearning system you choose. But, if you really want to learn and achieve mastery of something new, you need an eLearning solution with the features of lightning learning. Here are the key criteria for an awesome lightning learning platform:

  1. Focused – A good lightning learning system will not waste your time repeating things you already know or teach you things that are outside the bounds of what you currently need to know. It should hold sacred the lightning learning goals of ‘just-in-time learning’ but also ‘just-enough learning’.
  2. Adaptive – A good system won’t throw you into the deep end for content that is too challenging. It will have the ability to rapidly assess your current level of skills and understanding and start you at the right level of expertise. It will also have high-quality definitions of required prerequisite knowledge for any topic. Best case it will be constantly recalibrating the challenge level to keep you in the ‘Goldilocks zone’ … not so hard that you are frustrated and not so easy that you get bored.
  3. Complete – You need to be confident that your eLearning system doesn’t have any holes in the knowledge it is presenting to you. Since you don’t know the material, you will also not know if something is missing, so you need to rely on the reputation of the system you are using.  Top eLearning systems often will create a body of knowledge (BOK) that captures everything that one needs to know about a topic. This is the best way to ensure that learning is complete.
  4. Consistent – As you use a particular learning system, you will begin to become accustomed to the idiosyncrasies of how it presents knowledge. You may also become familiar with the style of instruction. Ideally, a different lesson, on a different topic, delivered by a different teacher will have to conform to a familiar style of presentation and have consistent levels of quality and production value. K-12 learning systems like Khan Academy have done this in the past by having just one instructor (though this is changing as even the polymath Sal Khan can’t teach everything).
  5. Reinforced – How many times have you gone to a weeklong conference or training, leave ready to apply your newly acquired knowledge, but then get distracted when you return to the office?  The knowledge quickly starts leaking out of your memory before you get a chance to apply it. This happens when knowledge is presented to you but you never really ‘own’ it well enough to use it day to day. This problem can be remedied with two techniques known as ‘recall practice’ and ‘spaced repetition’ where concepts are reviewed at regular intervals so that they stick with the learner.
  6. Validated – A good eLearning system should validate whether the student really knows the material and has put in the required time and effort. This can only be done through testing (or as the educators like to say ‘assessment’). The good news is that research has proven, beyond a doubt, that testing is not wasted time but, in fact, is a very efficient learning time. Sadly most learning systems don’t embed assessment. Probably because learners are not aware of how important testing is as a part of learning. It may also be that educators and trainers generally would prefer not to be measured except by qualitative measures like surveys or whether or not the students completed the course.  Without validating that the learner has achieved mastery there really isn’t any reason to do the training in the first place. We can extend the old business adage “what isn’t measured isn’t managed” with “those who aren’t measured can’t improve”.

The exciting thing is that we really do know how to do all these things. It’s like NASA in the early 1960s. All physics and science existed to put a man on the moon. It just needed to be applied.

In my next blog post, I’ll talk about some great eLearning solutions that are already here and improvements that are coming soon!

Related articles: 

Welcome to the Age of Learning

What is Lightning Learning?

Commodity AI and the Next Best Experiment

References: 

 “Why MOOCs Don’t Work”, Doug Lederman, www.insidehighered.com

Stephen J. Smith

Stephen Smith is a well-respected expert in the fields of data science, predictive analytics and their application in the education, pharmaceutical, healthcare, telecom and finance...

More About Stephen J. Smith

Books by Our Experts