The other day I saw (once again) the question, “How long does it take to develop an hour of elearning?”
My reply, not meant sarcastically: How long is a ladder?
I don’t really know what “an hour of elearning is.” Well, I do: it’s a chimera. The phrase has more assumptions built into it than Southwest Airlines has boarding passes.
Even if you could find this imaginary user who makes an average number of responses and takes an average number of branches–so what? If I’ve designed the course so it adapts to the learner’s choices and responses, then I’m pretty much guaranteed to have sections where a lot of developer work went into material that most learners won’t see.
It’s kind of like asking how long it takes to stock 150 feet of grocery shelves without considering whether you’re talking about quarter-ounce spice bottles or 50-pound bags of dog food.
I realize that an organization or a client wants some idea of how long things take. But I rarely see such concern about a related question: how long does it take to learn?
I’m currently working with a great client; they’re eager to avoid bullet-pointed talk-shop training. They’ve involved experienced professionals and line managers in defining useful skills that the target audience needs but typically lacks. We want people grappling with realistic problems, and we want them reflecting on how they did when they grappled.
But we’ve only got so much time.
I’m trying to figure out ways to maximize the formal (or should I say “focused?”) time that our planned half-day sessions permit. That means a number of things:
- “Worked examples” as handouts or supplements to activities. If the problem were for you to design a simple web page, the worked example might be a different page that embodied some of the coding or design features you’re supposed to apply
- Optional support for procedural information, like a cheat sheet with HTML codes or a CSS with call-out explanations. The point is that the facilitator probably isn’t going to go over this ahead of time. Instead, she’ll offer it: “While you’re working on the problem, you might find the cheat sheet useful for understanding the style sheet.”
- Enforced delay (or, as I like to call it, time to think).
“Enforced delay” is my shorthand for dealing with the retrieval process essential to learning. It takes time to recall information from long-term memory. A drawback to live discussion is that the first people to respond to a question can end up stopping the retrieval process for others.
One theory of how learning works is that as we bring information from long-term memory into short-term, we reprocess that information. In a sense, we learn it again. It’s not like punching a button and replaying a recording. (This is why some of our strongest memories can differ so much from what really happened: we’ve relived the event time and again, and each recalling of it has transformed it.)
So what’s this enforced delay? This kind of thing:
Present a problem or activity for individuals to consider and actively respond to.
I have a graphic that says:
On your own, read the sample document.
Identify at least three problems.
(“Identify” = “write down”)
Have individuals combine responses in groups.
Nothing all that surprising about group activity in organized training. I think it’s important for the group to attack problems that have value–and for there to be enough time for the group to think about what it’s doing.
Have the group articulate its results.
This is another round of “think and do.” The group’s been working on whatever the problem is. Now the group reorganizes that into a summary. It’s the difference between “write an elevator speech” and “deliver it.”
Don’t fall asleep, now. Nothing here is all that startling. Still, the facilitators for these sessions may be line managers or others who’ll benefit from a little guidance: allow people time to think. Use silence to invite further discussion. Instead of handling down principles, start by encouraging people to derive some of their own, and then compare those with those from other sources.
I see more things going on. For example, I sense in myself a conflict between the need for people to do their own learning, and my desire to organize or structure things so they can learn faster–perhaps more arrogantly, “more efficiently” [as in, because of my efforts).
Material for another post; I’ve got to get back to creating activities that give people time enough to think and do.
CC-licensed ladder photo by mahalie.
7 thoughts on “How to learn, or, think and do”
Kia ora e Dave!
I guess, there are many ways of measuring elearning.
It’s like other quantities to do with people. How big is a mouthful? How heavy is a handful? What’s the length of a step forward? How long is a good sleep? How many hours does it take to learn to touch-type?
I know, you could just as well measure elearning in terms of number of characters of text, number of questions asked or megabits of html (or whatever), as measure it in hours. Conceded.
But length of time taken to develop an hour of elearning is actually more tangible than measuring it in links, pages or megabits of download.
How long does it take to prepare and deliver an hour-long classroom lesson? We all know that for some who attend the lesson, it might seem to us like they’d been there a couple of minutes. And for other learners, it might seem like they’ve been there a whole morning. How long is a morning?
Does every learner get the same learning out of an hour reading? Of course not.
As a teacher in a day-school, I’ve been expected to squeeze all sorts of unbelievable tasks into 10 hours or more in one day. Never mind the weekend. So presumably 1/10 of that effort represents an hour’s teaching(etc) on my part. How do we measure that? In kilojoules?
On one of the posts you may have come across, I explained that an hour of elearning that I claimed to have developed was measured alongside its print-based equivalent. The same strange quantity relates to print-based learning, believe it or not. Hours of learning.
How else do we decide what (on average) represents what is to be taught to learners in a year? It’s just a length of time the same as an hour is. So it doesn’t take a PhD to work out what’s supposed to be taught in an hour.
So the developers have to do the almost-impossible in order to fill in their time-sheets – like teachers do when they set about preparing for an hour lesson.
I think it’s a bit unfair to criticise these people who, like me, have often been given the task to provide elearning according to measurements of time.
Please – don’t knock the teacher.
Ken, I wasn’t knocking the teacher, although I’m quite openly questioning the premise behind the “how long does it take” question.
To ask “how long does it take to develop an hour of elearning” leaves out a couple of key considerations, like:
…And: who’s asking?
Back in the era of computer-based training (CBT) delivered via mainframes, I ran the CBT program for Amtrak, the U.S. rail passenger system. I surveyed other companies that used the same software we did–organizations like Marriott Hotels, Holiday Inn, John Deere, USAir.
When I said “how long,” I explained I meant “from the time you decide ‘we need a course on X’ through ‘okay, we’ve piloted and revised; we can release the course.'”
Our numbers came in all over the place — from 40:1 to 400:1. What that meant, for example, was that at Amtrak, I could (and did) create in a week an online course that took the average person less than an hour to complete. That was for a straightforward, mainly procedural topic that the developer understood, well-suited to text-based interaction, intended mainly for people already experienced with our reservation system, with the course built by a person skilled in the authoring system and in instructional design for online learning.
And supported by a robust “training train” system allowing learners to practice risk-free in situations that they chose for themselves.
At the other end of the spectrum, the 400:1 ration (ten weeks to produce an average learner-hour), you had a high degree of simulation for a complex topic that was new for most learners.
Without that kind of qualification, I don’t see much value to the “how long” question, though I certainly understand why managers would like to have an answer.
Isn’t the answer always: it depends? Right now my org is trying to quantify this by having people track what they do when they develop a course. The idea is to get some real numbers. We do technical training, and here’s where I see the “it depends” come in: How new is the product, how experienced is the developer, is the product released, do we have the product, do we have all the components required to develop training for the product, do we have other software that may be affected by the product to create a realistic environment, are the developers skilled in these other products…
Well you get the idea. I understand the need for some number, but the minute you give the stakeholders a number, won’t they hold you to that number no matter the circumstances?
Hi, Gina; thanks for joining in.
I worked in corporate training / learning for decades, and I agree with you on several fronts.
One (maybe not obvious from the post) is that it can be very good to “get some real numbers.” It’s hard in the middle of a project to think about what you’re doing, how long it’s taking, and whether you’re getting the results. It’s valuable after the fact if you can analyze that data and use it to improve results.
Second, you’re talking about your own organization and its own work, not a theoretical average from all over the map.
As for stakeholders–yes, I do think some stakeholders are at high strength for turning an estimate or a rule of thumb into a law of nature. One way I can think of to deflect that is to look at the stakeholder’s area of expertise and try to find analogies–situations in which a simplistic average is, well, simplistic.
If I knew how to do that consistently, I’d probably have a book on that topic.
Kia ora Dave!
And thanks for replying to my comment. I agree with you that there are a lot of factors to be considered when estimating development times. But isn’t that always the case where coding is concerned? There’s the specifications, the planning and the coding to put a too simple summary for it. Development of elearning really is no different from much of that.
I have a lot of questions, and I don’t expect you to answer them. But, for instance, how does a software developer estimate the time taken to develop what they do? Obviously this has been done in the past, and Project Management will have a few pointers to offer us with this I’m sure.
For me, the most interesting factor is one you mentioned. “How well?” How that is quantified or estimated is in itself problematic. There are many parameters to decide the ‘wellness’ of any learning that is developed. My short answer to that, of course, is, “Does it teach effectively?” It’s amazing how effective simplicity can be when it comes to teaching and learning. It’s amazing how convoluted and intricate an elearning development plan can be when the developer wants to earn more cash.
Ken, I had in mind a lot more than coding. Coding is the easiest part of it, in fact. To the extent that it’s procedural, it’s pretty straightforward. The ease of so-called authoring tools, however, often gives a false sense of accomplishment.
“Oh, see, it’s easy to create a multiple-choice question.” “I can use this for a drag-and-drop.”
I remember a very serious debate at a client site between two “learning specialists.” The heart of the matter: what was the correct ratio of “lecture screens” to “testing screens?” In other words, they seemed to feel there was a golden heuristic — ten to one, eight to to — that was part of a recipe for learning.
They seemed surprised when I asked what the topic was.
Kia ora e Dave.
Yes but coding is still concerned for that’s the end of a process – not the only process. As we agreed, there are a lot of factors to be considered when estimating development times, and yes, pedagogy and well-design of learning modules are just two of them.