I’m going to be delivering a mobile learning course for a university this spring. Consequently, I’m currently beginning the design. I need to practice what I preach in the sense of good learning design, so I’m working through the usual decisions. The real question I have is whether I can make it transformative. There are limitations, but one of my mantras is about design having many possible development forms. So…I’ve got to make a good stab at it. Here’s my preliminary thinking on bringing the course to life.

I’ve actually been working a lot on the design of university learning experiences. There have been several instances in the past year or two that have really pushed my thinking in this space. This naturally includes  application-based instruction, as well as meaningful (and minimal)  content, and good assessment design. It’s been handy for this!

Naturally,  Designing mLearning  will be the text (The Mobile Academy is focused on formal education, and these students are focused on the workplace).  As it turns out, the book isn’t written in the order I want to deliver the course. So while I’ll have specific readings each week, I’ll instead recommend that the students read it in one go (it’s not a  long book), and that two different cuts through the material will be a better learning experience.

I’m also thinking about the assignments: having them do meaningful things. E.g. designing solutions. Also, in the right order, to facilitate useful processing. I suppose I should worry about whether the workload will be too much However, as it’s a compressed course the expectations per week are higher. And I will argue that I’m having them  do more than  consume, so the workload’s ok.

The important thing, to me, is getting the emotional trajectory right. I reckon I need an aspirational goal up front, and then ensure I deliver.  It needs to work on a week-by-week basis  and overall.  I’m making sure they’re doing the right things, and then I’ll fine-tune. I like the chance to integrate my thinking and put a stake in the ground. It’ll be interesting to see how it goes.

As a side note, mobile seems (to me) to be resurrecting. I would think it’s now mainstream, but some folks are still getting started. Hey, it makes sense, so better late than never! Where are you in mobile?

A number of years ago, I wrote about activity-based learning.  And I firmly stand behind the model there. It’s not a major campaign, but quietly permeates the things that I do. However, I realize that there’re two misnomers in the label, and it’s time to rectify that. It’s about instruction, and it’s about application. I need to make those distinctions clear.

The original point was to find a way to make it hard for content  (read: info dump and knowledge test) to be the basis of curricula. So, instead of a series of content, or knowledge, to be a curriculum,    a curriculum is a series of activities.

There are, of course, specific constraints around the activities, and I realize it’s about knowledge  application. There are lots of activities that  aren’t going to lead to meaningful learning. The key is retrieval, but I think what’s important is retrieval  to do something. Hence application, applying knowledge to make decisions. It can be behavioral decisions or the decisions inherent in creating meaningful work product, but it’s about cognitive skills in context.

The focus on decisions is because I believe what will make the difference to organizations is making right, or better, decisions. Not, for instance, just knowing things. And part of the application core is about doing things.

I argue that learning is acton and reflection. And, consequently, instruction is designed action and guided reflection. Similarly, application is designed actions to require decisions. And this  is about instruction. It’s not learning, though that’s the goal. But instruction is a probabilistic game; as I’ve paraphrased Dorothy Parker, you can lead learners to learning, but you can’t make them think.

‘Application-based instruction’ it is. It’s like problem-based learning, or case-based, but it’s instruction not learning, and it’s an umbrella for those and other such initiatives. The real question is whether this labeling makes important distinctions. I think it does, but I’m biased. What do  you say?

I’m a fan of the notion of ‘learning experience design’ (not so sure about  platforms;  I need to investigate them more ;).  The idea of integrating effective education and engaging experiences is something I’ve been  on about for a  very long time. And I want to push it a little further. I want to talk about transformation.

What am I talking about? So, I’ve previously referred to Pine & Gilmore’s concept of the Transformation Economy. That is, going beyond experiences (e.g. themed restaurants) to ones that change us. And I argue that’s what we do; we create (or should) experiences that give us new skills, new abilities to  do.  But I want to push it further.

Here I’m talking about deliberately using the idea of transformation as a learning design goal. Not just change, but leveraging the emotions as well as cognition to have the learner not just feel empowered, but transformed!  This may sound like a lofty goal, fine for a TED Talk (just read the book; recommended), but is it practical for elearning?  Well, that’s an interesting question.

Let me spin it another way: I do  not think we should be shooting for an information dump and knowledge test. For two reasons: one is that it’s not inspiring. More importantly, however, it also isn’t effective. You end up with what cognitive scientists call ‘inert knowledge’. You’ll learn it and pass a test on it, but when it’s relevant in practice it won’t even get activated!  Because you’ve never used it in ways like you practice.

I think if we are actively thinking about transformation as a goal, we might do a better job of thinking about the necessary practice and the emotional engagement.  We can focus on thinking “what will lead to the transformation we want”, and “how do we make people want it and celebrate when they’ve made the breakthrough?”  And I think this is a useful perspective.

Even for things like compliance, I’d suggest that we should be having visceral reactions like “Ok, I get it <bad behavior> is pretty heinous”, and “safety  is important, and I commit to following these rules”.  For more important things, you’d like them to feel “yes, I see, this will change how I  do this!”

Yes, it’s ambitious. But why set ourselves limited goals? When I was teaching interface design, I maintained that if I accommodated the engineers lack of background in Psych, I’d get them only so far. If I pushed them, they’d end up farther than if I was conciliatory. Similarly, here, I think we’ll do a better job if we think ‘ambitious’, and end up not as far as we’d like. I’ll suggest that’s better than satisfactorily achieving mediocrity. Most importantly, I truly think we’ll do a better job of design if we strive for transformation.

And, if there’s nothing transformative about what we’re covering, should we really be using our resources?  Let me put it another way: why  shouldn’t we do this? Seriously, I’m asking.  So, what’s your answer?

 

For sins in my past, I’ve been thinking about assessments a bit lately. And one of the biggest problems comes from trying to find solutions that are meaningful yet easy to implement. You can ask learners  to  develop meaningful artifacts, but getting them assessed at scale is problematic. Mostly, auto-marked stuff is used to do trivial knowledge checks. Can we do better.

To be fair, there are more and more approaches (largely machine-learning powered), that can do a good job of assessing complex artifacts, e.g. writing. If you can create good examples, they can do a decent job of learning to evaluate how well a learner has approximated it. However, those tools aren’t ubiquitous. What is are the typical variations on multiple choice: drag and drop, image clicks, etc. The question is, can we use these to do good things?

I want to say yes. But you have to be thinking in a different way than typical. You can’t be thinking about testing knowledge recognition. That’s not as useful a task as knowledge retrieval. You don’t want learners to just have to discriminate a term, you want them to  use the knowledge to do something. How do we do that?

In  Engaging Learning, amongst other things I talked about ‘mini-scenarios’. These include a story setting and a required decision, but they’re singular, e.g. they don’t get tied to subsequent decisions. And this is just a better form of multiple choice!

So, for example, instead of asking whether an examination requires an initial screening, you might put the learner in the role of someone performing an examination, and have alternative choices of action like beginning the examination, conducting an initial screening, or reviewing case history. The point is that the learner is making choices  like the ones they’ll be making in real practice!

Note that the alternatives aren’t random; but instead represent ways in which learners reliably go wrong. You want to trap those mistakes in the learning situation, and address them  before they matter!  Thus, you’re not recognizing whether it’s right or not, you’re using that information to discriminate between actions that you’d take.  It may be a slight revision, but it’s important.

Further, you have the consequences of the choice play out: “your examination results were skewed because…and this caused X”.  Then you can give the principled feedback (based upon the model).

There are, also, the known obvious things to do. That is, don’t have any ‘none of the above’ or ‘all of the above’. Don’t make the alternatives obviously wrong. And, as Donald Clark summarizes, have two alternatives, not three. But the important thing, to me, is to have different choices based upon using the information to make decisions, not just recognizing the information amongst distractors. And capturing misconceptions.

These can be linked into ‘linear’ scenarios (where the consequences make everything right so you can continue in a narratively coherent progression) or branching, where decisions take you to different new decisions dependent on your choice.  Linear and branching scenarios are powerful learning. They’re just not always necessary or feasible.

And I certainly would agree that we’d like to do better: link decisions and complex work products together into series of narratively contextualized settings, combining the important types of decisions that naturally occur (ala Schank’s Goal Based Scenarios and Story-Centered Curriculum and other similar approaches).  And we’re getting tools that make this possible. But that requires some new thinking. This is an interim step that, if you get your mind around it, sets you up to start wanting more.

Note that the thinking here also covers a variety of interaction possibilities, again drag’n’drop, image links, etc. It’s a shift in thinking, but a valuable one. I encourage you to get your mind around it. Better practice, after all, is better learning.