Learnlets

Secondary

Clark Quinn’s Learnings about Learning

ID Support Thyself

by Leave a Comment

Want to dig a bit deeper into improving design processes. Here, I look at tools,  asking IDs to ‘support thyself’.

As usual, the transcript:

One of the things I do is help organizations improve their design processes. Last week, I talked about when to team up in the process of learning design. Another component of good design, besides knowing when and how to draw in more minds, is baking learning science into your processes. That‘s where tools help. I expect that most orgs do have process support, but…baking in learning science seems not to be there. So here I‘m exhorting IDs to ‘Support Thyself’.  

As I discuss in my forthcoming book, there are nuances to each of the elements of learning design (as I also talked about for Learnnovators). That includes meaningful practice, useful models, motivating intros, and more. The question is how to help ensure that as you develop them, you make sure to address all the elements.

One approach, of course, is to use checklists. Atul Gawande has made the case for checklists in his The Checklist Manifesto.  In this great book, he talks about his own inspiring efforts in the context of other high-risk/high-value endeavors such as flight and construction.   There are clear benefits.

The point is that checklists externalize the important elements, supporting us in not forgetting them. It‘s easy when you do yet another task, to think you‘ve completed a component because you‘ve done it so many times before. Yet this can lead to errors. So having an external framework is useful. That‘s part of the rationale behind the Serious eLearning Manifesto!

I had originally been thinking about templates, and that‘s another way. And here, I‘m not talking about tarted-up quiz show templates. Instead, I mean a tool that leaves stubs for the important things that should be included. In examples, for instance, you could leave a placeholder for referencing the model, and for the underlying thinking. Really, these are checklists in another format.  All in all, these are ways that you can  Support Thyself!

What you don‘t want to do is make it too constraining. You want to create a minimum floor of quality, without enforcing a ceiling. At least other than the ones your own schedule and budget will import. But you want to be creative while also maintaining effectiveness.

And you can do this in your authoring tool. Just as you may have a template you reuse to maintain look and feel, you can have placeholders for the elements. You can also provide guidance for the elements, in a variety of ways.

There are lots of forms of performance support. And, just as we should be using them to assist our performers (even doing backwards design to design the tools first then any learning), we should be using them to overcome our own cognitive limitations. Our cognitive architecture is amazing, but it‘s prone to all sorts of limitations (there‘s no perfect answer). We can suffer from functional fixedness, set effects, confirmation bias, and more.  

I‘ll admit that I created an ID checklist. The only problem was it had 178 elements, which might be unwieldy (though it did go through the whole process). But you should make sure that whatever tools you do have cover the necessary elements you need. I did create a more reasonable one to accompany my ‘Make it Meaningful‘ initiative (coming soon to a theater or drive-in near you).  

Our brains have limitations that influence our ability to design. Fortunately, we can use technology as support to minimize the impact of those limitations and maximize the contributions of our outcomes. And we should. Thus, my encouragement for IDs to Support Thyself!

Separate content from description

by 3 Comments

Once again facing folks who aren’t using styles, I was triggered to think more deeply about the underlying principle. That is, to separate content from description. It’s a step forward in what we can do with systems to bring about a more powerful human-aligned system.

And, as always, here’s the text, in case you (like me) prefer to read ;).

I‘ve ranted before about styles, but I want to make a slightly different pitch today. It‘s not just about styles, it‘s about the thinking behind it. The point is to separate content from description.

So, the point about styles is that they‘re a definition of formatting. You have elements of documents like headings at various levels, and body text, and special paragraphs like quotes, and so on. Then you have features, like font size, bolding and italics, color, etc. And what you see, too often, is people hand-formatting documents, choosing to do headers by increasing the font size, bolding, etc. And, importantly, having to go through and change them all manually if there‘s a desire for a change in look.  

The point of styles is instead merely to say this is a heading 1, this is a figure, this is a caption, and so on. Then, you separately say: heading 1s will be font size 16 bold and left-justified. Figures and captions will be centered, in font size 12. And so on. Then, should someone want to change how the document‘s formatted, you just change the definition of heading 1, and all the heading ones change.  

It goes further. You can define that all heading ones have a page break before (e.g. a new chapter in a book). And you can define new styles, like for a callout box (e.g. colored background), etc. You can have different heading ones for a book than for a white paper. And some styles can be based on others. So your headings can use the same font as your body text, and if you want it all to change, you change the source and the rest will change.  

Which is wonderful for writing, but the concept behind this is what‘s really important to get your head around. That is, separating out role from description. That‘s what‘s led me to be keen on content systems. The notion of pulling up content by description instead of hardwiring together content into an experience is the dream.  

It‘s all about beginning to use semantics, that is the meaning of things, as a manipulable tool. Many years ago, I led a project creating an adaptive learning system. We were going to have content objects defined by topic, and learning role, and tagged in terms of media, difficulty, and more. So you could   say: “pull a video on an example of diversity set in a sales office”. Our goal, with a suite of rules about what when to move up or back in difficulty, was to specify what learning content the learner should see next, etc.  

This is how adaptive platforms work. When Amazon or Netflix make a recommendation for you, there‘s not someone watching your behavior, instead it‘s a set of rules matching your particular actions to content recommendations. If you‘ve ordered a lot of British mysteries, and you haven‘t seen a particular series that lots of other people like, it‘ll be likely to be offered to you.  

This is the opportunity of the future. We can start doing this with learning (and coaching)!   We can start pulling together your learning goals, job role, current progress, current location (in time and space), etc, and offer you particular things that are appropriate for you. And, like our learning system, it might be recommendations of content, but you can choose others, or ignore, or…As Wayne Hodgins used to say, present the ‘right stuff‘: the right content to the right person at the right time in the right way….

The point being, just like styles, if we stop hardwiring things together, hand-formatting learning experiences, we can start offering personalized and even adaptive learning. Yes, there are technical backend issues, and more rigor in development, but this is the direction we can, and should, go. At least, that‘s my proposal, what say you?

Personalized and adaptive learning

by 1 Comment

For reasons that are unclear even to me, I was thinking about personalized versus adaptive learning. They’re similar in some ways, but also different. And a way to distinguish them occurred to me. It’s kinda simplistic, but I think it may help to differentiate personalized and adaptive learning.

As background, I led a project to build an intelligently adaptive learning platform. We were going to profile learners, but then also track their ongoing behavior. And, on this basis, we’d serve up something appropriate for learner X versus learner Y. (We’d actually recommend something, and they could make other choices.)

It was quite the research endeavor, actually, as the CEO had been inspired by Guilford’s learning model. I dug into that and all the learning styles literature, and cognitive factor analysis, and content models around learning objectives, and revisited my interest in intelligent tutoring, and more. I was able to hire a stellar team, and create an approach that was scientifically scrutable (e.g. no learning ‘styles’ :). We got it up and running before, well, 2001 happened and the Internet bubble burst and…

In some sense, the system was really both, in the way I’m thinking about it. I’ve seen different definitions, and one has adaptive as a subset of personalized, but I’m going a different way. I think of personalized as pre-planned alternatives for different groups, whereas adaptive reacts to the learner’s behavior.

Our use of initial profiling, if we only used that, would be personalized. The ongoing adaptation is what made it adaptive. We had rules that would prioritize preferences, but we’d also use behavior to update the learner model. It’s something they’re doing now, but we had it a couple of decades ago.

So, my simple way of thinking about personalized versus adaptive is that personalized is based upon who you are: your role, largely. We’d swap out examples on marketing for people selling services versus those selling products, for instance. Or if we’re talking negotiation, a vendor might get a different model than a lawyer.

Adaptive, on the other hand, is based upon what you  do. So, for instance, if you did poorly on the last problem, we might not give you a more difficult one, but give you another at the same level. Twice in a row doing badly, we might bring you another example, or even revisit the concept. This is what intelligent tutoring systems do, they just tend to require a rigorous model of expertise.

Of course, you could get more complicated. Personalization might have a more and less supportive path, depending on your anxiety and confidence. Similarly, adaptive might throw in an encouraging remark while showing some remedial materials.

At any rate, that’s how I differentiate personalized and adaptive learning. Personalized is pre-set based upon some determined differences that suggest different learning paths. Adaptive calculates on the fly and changes what the learner sees next.  How do you see it?

The plusses and minuses of learning science research

by 1 Comment

A person who I find quite insightful (and occasionally inciteful ;) is Donald Clark. He built and sold Epic, an elearning company, and now he leads a learning AI company, Wildfire. He’s knowledgeable (for instance, having read up and summarized centuries of learning theorists), willing to call out bad learning, and he’s funny. And so, when he reported on a new study, I of course looked into it. And I find that it points out the plusses  and  minuses of learning science research.

To be clear, this is about his product, so there’s a vested interest. However, he’s got integrity; he’s not going to sully his reputation with a bad study. And, it’s a good study. It rightly demonstrates an important point. It’s just that it stops short of what we need for full  learning.

So, his product does something pretty amazing. You give it content, and it can not only answer questions about the content (as, for instance, some chat tools do), it can turn the tables and ask  you questions about the content. That is, it can serve as a sort of tutor. Which is all to the good.

What it can’t do, of course, is design meaningful practice. As Van Merriënboer’s Four Component Instructional Design (4C/ID) points out, you need to know the information, and you also need practice applying it. And I reckon we’re still far from that. So, while this is part of a whole solution (and Donald knows this), it’s not the full solution. He’s subsequently let me know it can do language tasks, which is impressive. I’m thinking more of contextualized scenarios, however.

The study demonstrates, as you might expect, that breaking up a video into reasonable chunks, and having system-generated questions asked in-between, led to 61% better retrieval, going from getting 8 to 14 questions right. That is a big improvement. it’s also impressive, since it’s generating those questions from video! That is, it parses the video, establishes a transcript, and then uses that to generate a knowledge base. Very cool.

And it’s a well-designed study. It’s got a control group, and a  reasonable number of subjects. It uses the same test material, for an AB comparison. Presumably, the video chunking was done by hand, into four pieces. The chunking and break might account for the difference, which wasn’t controlled for, but it’s still a big improvement. Granted, we know that watching a video alone isn’t necessarily going to improve retention (except, perhaps, over some other non-interactive way of dumping content). But still, this is good as it’s an improvement and a lot of work was saved.

What I quibble about, however, is the nature of the retrieval. The types of questions liable to be asked (and it’s not indicated), are knowledge questions. As suggested above, knowledge is a necessary component. But using that knowledge to make decisions in context is typically what our goals are. And to achieve such goals, you basically have to practice making decisions in context. (Interestingly, the topic here was equality and diversity, a topic he has complained about!)

Knowledge about a topic isn’t likely to impact your ability to apply it. What will  make a difference are actually doing things about it, like calling it out, having consequences, and actively working to remedy imbalances. And that requires separate practice. Which he’s acknowledged in the past, and rightly points out that his solution means you can devote more resources to that end.

Thus, the plusses of learning science research are we nibble away at the questions we need to answer, and find answers about the questions we ask. The minus, of course, is not necessarily asking the most important questions. It’d be easy to see this and say: “we’ve improved retention, and we’re done”. However, it won’t necessarily lead to reducing the behaviors being learned about, or building ability to deal with it.  There are plusses and minuses of learning science research, and we need to know the strengths, and limitations, of it when we hear it.

Top 10 Tools for Learning 2020

by Leave a Comment

It’s time, once again, for Jane Hart’s excellent Top 10 Tools for Learning survey. And, so, it’s time once again for my reflections. Here are my take on the top 10 tools that support my learning.

The first way I learn is to process what I’ve seen. That, toolwise, is largely about representing and communicating.

Processing Tools

1-2. Writing is arguably the top way I reflect. And, so that’d put Microsoft Word at the top of my list. That’s where I write books and articles first. And, of course WordPress is how I write my blog (e.g. here!).   Writing is a way to sort out how I think about things. As I say, things that end up in presentations and books tend to show up on blog first. Well, one of the main ways.

3-4. Besides writing, two ways I sort out my understandings are to diagram and to outline. I use OmniGraffle as a general purpose diagramming tool because, well, it largely works the way I want to think about it. Diagrams, mind maps, even recently as sort of posterboard. And I use OmniOutliner to do, well, outlines. Another way to map out structures. I’d use a less costly tool, but…the columns feature is really helpful for annotation. Both, unfortunately, are Mac only (and sadly quite dear).

5. Keynote is how I create presentations, another way I do, and then share, my thinking. Diagrams are a big part of my talks, punctuated with stock photos to represent concepts (from Pixabay and occasionally Unsplash). I believe (and don’t have evidence for) that using an image that relates to the concept but doesn’t exactly communicate it leaves open some curiosity that then gets connected. And that this leads to better comprehension (I avoid bullet points in live presos, and save them for handouts). Anyone got that data?

The second thing I do is see what other people are pointing to and have to say, and ask them questions   as well. So the second category is about interacting with others.

Social tools

6. Twitter is a regular feature of how I see what people are pointing to, as well as pointing to things I’ve found as well. Chats there are fun, too. Like Jane, Tweetdeck is my tool of course on my Mac. I have to use the Twitter client on iPad/iOS, since they’ve taken away Tweetdeck on the iPad (grr).

7. I like FeedBlitz as a way to sign up for blogs, as it brings them into my inbox, instead of me needing a separate app. Reading a select list of blogs is one of my tactics. That’s how people can sign up to get my blog in email, too.

8. Slack has also been a major component of getting things done, mostly with IBSTPI. It’s a handy way to get things done with others.

9-10. Social networks are a big part of my learning, which means that Facebook and LinkedIn also play big roles. Facebook’s more personal, ie less about work, but I learn about   many societal things there. And LinkedIn is a place for learning as well, professionally as opposed to personally.

And…

Honorable Mention: to round out the picture (10 is such an arbitrary number ;), sharing collaborative documents, e.g. Google Docs, is a major way to collaboratively process and learn together. Also socially, Zoom and BlueJeans (the latter’s almost the same, and what ISBTPI uses) are used a lot to discuss and negotiate understandings. And email, of course (using the Mac Mail client) is a major way I learn, e.g. blogs appear there, and it’s a major way I interact.

DuckDuckGo has become my goto search engine (and Brave as my browser,  at least on my Mac, awaiting cross-device sync), because I don’t need to spread my data any further than necessary. And searching is a big part of my learning.

As an aside, owing to the pandemic, like everyone else I’ve been doing much more with Zoom to interact with colleagues than I had in the past. And I find, interestingly, that the ways I reach out are more opportunistic: I’ll use FB Messenger, or a Twitter DM, or a LinkedIn message, or an email depending on who, why, and what tool I’m in at the time. There may be some method to the madness, but I’m not confident on that point ;).

So, there’re my Top 10 Tools for Learning. I hope you’ll post or send your list to Jane too, so we can continue to see what emerges.

 

Experimenting with conference design

by 2 Comments

As part of coping in this time of upheaval, I’m trying different things. Which isn’t new, but there seem to be more innovations to tap into. In addition to teaching a course on mobile learning, I’m one of the speakers at a new online event. And, what’s nice, is that they’re experimenting with conference design, not just moving straight online.

To be fair, the Learning Guild has had a continual practice of trying different things at their conferences, and it’s been good. And, so too, was the most recent TK by ATD.   But this is different. Two of my colleagues organized it as a response to our ‘new normal’, Will Thalheimer and Matt Richter. And their stated goal is changing the way we conference.

The key, of course, is to leverage what’s different, and possible, online. It’s running from June 22 – July 31. That’s not a typo, it’s all of July and the tail end of June. That’s a long time!   They’ve recruited a suite of experts from around the world (they’re really trying to do this across boundaries include time and geography). And, to let you know, I’m one (so take my comments with the appropriate caveats ;).

They’re also tossing out traditional ideas and open to new ones. Speakers are expected to build an experience that’s spread out over the time. Yet also designed so that you can come in late, or early, and drill into what you want when you want. They’re also planing on having synchronous events – debates, panels, socializing – again using technology.

Note that it’s not free. There are some free conferences being put on, mostly webinars. And those are good. This is different. It’s deeper. It’s a stab at looking afresh. And I’m not sure it could even have come from any existing framework.

And, we won’t know if it all will work. We’re designing this in the time between now and launch. There’re bound to be hiccups. Which, of course, means there’re bound to be learnings. I know I want to talk about Learning Science 101. And something else. Lots I could (I welcome suggestions). I’m inclined to think it might be Emotion and Learning. But it could also be LXD. (There are all linked, of course.)

But it’s a high quality group (er, mostly…they did let me in). AND, importantly, it’s focused on evidence-based content. There may be sponsors, or even an exhibit hall, but every presenter is honor-bound not to push anything that’s not legit. Most importantly, there’s enough quality that overall it’s bound to be worth it.

I’m excited, frankly. I have to come up with some different ideas. And I like that. I’m glad that they’re experimenting with conference design. We all win, regardless! It’s part of learning, challenging yourself. So, do yourself a favor. Check it out. It may not be for you, but keep an open mind!

 

 

How to be a world-class educational technologist

by 4 Comments

On LinkedIn, I was asked: “I would like to ask sir, how can I be a world-class educational technologist?” And I thought that was a very interesting question. (Of course, my immediate response should be “how should I know?” ;) But I thought I’d do a bit better. So here’s a recast of my response.

First, I get requests about how to get started as an instructional designer (particularly offers to come work for me). And, well, I’m an independent consultant, and just haven’t been a business builder. But I want to respond helpfully, and it’s one of those things that happen enough that I have this canned response:

If you want a bootstrap, working volunteer for a not-for-profit (NFP) foundation is a good step if you can.   Areas of specialization? Depends on what you like: kids – K12 or NFP, higher ed, adult – organizational L&D. They differ. As to skills, make sure you know the major authoring tools, e.g. Lectora, Captivate, and/or Storyline. And of course have some background in instructional design/learning science.   If you haven’t covered performance consulting, look into it so you don’t design a course when there’s a better/simpler solution. Make sure you have a portfolio of work. Good luck!

In this case, I also pointed him to a previous post, where I’d outlined some roles for learning experience design.

Then, thinking at the bigger scale of not just getting going as a new ID, but persisting, I added this:

Overall, you have to have the passion, it’s a long road. Have a good understanding of learning science, a fundamental grasp of technology, a mind for both design and process, and then put it to work doing real projects! Continue to read, reflect, and then as you start getting your mind around it, start sharing your thinking and get feedback (and listen to it!). Start local, work outward to sharing regionally, nationally, and internationally. If you learn, adapt and improve, and persist, you can get there.

I think that’s the path to improvement, regardless. In short. There’s more: I have just finished reading Eric Barker’s Barking Up the Wrong Tree  (affiliate link);  thanks to an ATD Sacramento event attendee, and found it having very interesting recommendations. Things like setting goals, giving, getting mentored, and more.

I think aiming to be a world-class educational technologist is a noble goal. Even if you don’t succeed, you’re liable to be better than if you just go through the motions. Now, I’m sure you’ve found things I’ve missed, so have at it!

Content systems not content packages

by 1 Comment

In a conversation last week (ok, an engagement), the topic of content systems came up. Now this is something I’ve argued for before, in several ways. For one, separate content from how it’s delivered. And, pull content together by rules, not hardwired. And it’s also about the right level of granularity. It’s time to revisit the message, because I thought it was too early, but I think the time is fast coming when we can look at this.

This is in opposition to the notion of pre-packaged content. MOOCs showed that folks want to drill in to what they need. Yet we still pull everything together and launch it as a final total solution. We are moving to smaller chunks (all for the better; even if it is burdened with a misleading label). But there’s more.

The first point is about content models. That we should start designing our content into smaller chunks. My heuristic is the smallest thing you’d give one person or another. My more general principle is that resolves to breaking content down by it’s learning role: a concept model is different than an example is different than a practice.

This approach emerged from an initiative on an adaptive learning system I led. It now has played out as a mechanism to support several initiatives delivering content appropriately. For one, it was supporting different business products from the same content repository. For another it was about delivering the right thing at the right time.

Which leads to the second point, about being able to pick and deliver the right thing  for the context.  This includes adaptive systems for learning, but also context-based performance support. With a model of the learner, the context, and the content, you can write rules that put these together to optimally identify the right thing to push.

You can go further. Think of two different representatives from the same company visiting a client. A sales person and a field engineer are going to want different things in the same location. So you can add a model of ‘role’ (though that can also be tied to the learner model).

There’s more, of course. To do this well requires content strategy, engineering, and management. Someone put it this way: strategy is what you want to deliver, engineering is how, and management is overseeing the content lifecycle.

Ultimately, it’s about moving from hardwired content to flexible delivery. And that’s possible and desirable. Moreover, it’s the future. As we see the movement from LMS to LXP, we realize that it’s about delivering just what’s needed when useful. Recognizing that LXPs are portals, not about creating experiences, we see the need for federated search.

There’s more: semantics means we can identify what things are (and are not), so we can respond to queries. With chatbot interfaces, we can make it easier to automate the search and offering to deliver the right thing to the right person at the right time.

The future is here; we see it in web interfaces all over the place. Why aren’t we seeing it yet in learning? There are strong cognitive reasons (performance support, workflow learning, self-directed and self-regulated learning).  And the technology is no longer the limitation. So let’s get on it. It’s time to think content systems, not content packages.

 

Sophia the Robot #DevLearn Keynote Mindmap

by 1 Comment

DevLearn opened with a keynote from Sophia the Robot. With an initially scripted presentation, and some scripted questions from host David Kelly, Sophia addresses the differences between AI and robots, with a bit of wit. The tech used to make the illusion was explored, but the technology was put to the test with some unscripted questions, and the responses were pretty good. An interesting start!

Tools for LXD?

by Leave a Comment

I’ve been thinking on LXD for a while now, not least because I’ve an upcoming workshop at DevLearn in Lost Wages in October. And one of the things I’ve been thinking about are the tools we use for LXD. I’ve created diagrams (such as the Education Engagement Alignment), and quips, but here I’m thinking something else. We know that job aids are helpful; things like checklists, and decision trees, and lookup tables. And I’ve created some aids for the Udemy course on deeper elearning I developed. But here I want to know what  you are using as tools for LXD? How do you use external resources to keep your design on track?

The simple rationale, of course, is that there are things our brains are good at, and things they’re not. We are pattern-matchers and meaning-makers, naturally making up explanations for things that happen. We’re also creative, finding solutions under constraints. Our cognitive architecture is designed to do this; to help us adapt to the first-level world we evolved in.

However, our brains aren’t particularly good at the second-level world we have created. Complex ideas require external representation. We’re bad at remembering rote and arbitrary steps and details. We’re also bad at complex calculations.  This makes the case for tools that help scaffold these gaps in our cognition.

And, in particular, for design. Design tends to involve complex responses, in this case in terms of an experience design. That maps out over content, time, and tools. Consequently, there are opportunities to go awry. Therefore, tools are a plausible adjunct.

You might be using templates for good design. Here, you’d have a draft storyboard, for instance, that insures you’re including a meaningful introduction, causal conceptual model, examples, etc. Or you might have a checklist that details the elements you should be including. You could have a model course that you use as a reference.

My question, to you, is what tools are you using to increase the likelihood of a quality design, and how are they working for you?  I’d like to know what you’ve found helpful as tools for LXD, as I look to create the best support I can. Please share!