Decision-making scenarios work best when they require realistic decisions and avoid preaching. Let’s look at some examples.

Not a real on-the-job decision

Carla, a sales person, is meeting with Amit, a new customer. She shows him a megawidget.

“You’ll love this megawidget,” Carla says.

“I don’t want a megawidget,” Amit says. “I came in here for a microwidget.”

What is this an example of?

1. Product Boundary Issues
2. Customer Misvetting
3. Courageous Upselling

What’s wrong with this scenario?

We’re not asking the learner to make a challenging decision like the ones they make on the job. We’re checking the learner’s short-term memory: Can they still recognize “Customer Misvetting,” which we defined three screens ago?

We’ve disguised a quiz question as a scenario. It’s better than a generic quiz question, but it doesn’t require the kind of thinking that learners need to do on the job.

Also, the question tests only whether the learner can apply the right label to a problem. It doesn’t test whether the learner can correct the problem.

A better question would ask what Carla should do, with the correct choice being the type of action that will correct a case of “Customer Misvetting.” Then we’d be testing the learner’s ability to recognize the problem and their ability to solve it.

A real decision

How is the following scenario different?

Carla, a sales person, is meeting with Amit, a new customer.

“I might be interested in your J-12 microwidget,” Amit says. “At 79 wigabits, it has enough power for paramatizing. But I’ve read that it runs hot.”

What should Carla say?

1. “Are you referring to the Widget World review?”
2. “Actually, the J-12 has only 60 wigabits. I think you’d need the K-77.”
3. “Our studies have never shown any heat issues with the J-12. Would you like to see the test results?”

Why is this question better?

It more closely mirrors what salespeople actually have to do, which is listen and respond to customers’ needs and concerns. We’re not asking learners to regurgitate facts but to make choices in nuanced situations.

Feedback

Let’s say the learner chooses option C above—they think Carla should tell Amit, “Our studies have never shown any heat issues with the J-12. Would you like to see the test results?”

What’s the difference between the following types of feedback for that choice?

“Telling” feedback

While this response won’t derail the sales conversation, it could make Carla seem defensive and possibly increase Amit’s skepticism. It would be better for Carla to show that she’s familiar with what the industry is saying about our widgets. Try again.

“Showing” feedback

“I’m not surprised that your studies don’t show any problems,” Amit says, sounding a little annoyed. “But Widget World does rigorous, independent testing, and they found heat issues. What can you say about their results?”

“Telling” feedback destroys the sense of being in a story and, worse, it relieves the learner of any need to think. “Showing” feedback emulates real life: something happens and the learner draws a conclusion from it, using a lot more brain. It’s also more concrete and therefore more likely to be remembered.

If you’re worried that your learners won’t be able to extrapolate from “showing” feedback, you might be tempted to provide additional “telling” feedback, like “Carla needs to show that she’s familiar with what the industry is saying about our widgets.” But fight this temptation, because:

• Your learners are adults and have been learning from experience for decades. Trust them to extrapolate. If you’re truly concerned about their ability to draw conclusions, test your material on some typical learners.
• If these mini-scenarios form a longer scenario, adding “telling” feedback would interrupt the story. You’d become like an annoying parent who interrupts a movie to tell the children to never do what the protagonist just did.

Development

I wrote the sample scenarios as text because that was easiest given my limited time. Obviously, interactions like these could be developed in any number of ways, from photos with dialog bubbles to video.

More at Learning Solutions in Orlando

Some of this post is taken from the handout for my talk on scenarios scheduled for March 24 at the Orlando Learning Solutions conference. I hope to see you there!

I’ve been tweeting a daily idea to spark your instructional design creativity. To get your daily ID idea, follow me on Twitter. (I don’t say much!)

Some recent tweets in the series:

• Ask your SME, “What are the 3 most common mistakes people make?” Turn them into branching scenarios.
• Using a story? What challenge is your character facing? Make them suffer!
• Are you solving a performance problem, or are you just turning information into a course?
• What are you creating to support your course? Job aids? Guidance for managers? Follow-up discussion? Additional tips in emails?
• Teaching a complex procedure? Course: How to use the job aids. Job aids: Everything else.
• What compelling scenario can you use to start your course? P.S. “A new employee is wondering…” isn’t compelling.
• Have your learners finish an almost-completed problem, then take away support in stages until they’re flying solo.
• What do you really want? “Describe the Heimlich maneuver” or “Save lives?” Don’t stop at declarative knowledge.
• How can you turn a dos/don’ts list into something more memorable? Try short scenarios that show the results; have learners draw the conclusion.

I use the #IDideas hashtag, and other people sometimes chime in with their own ideas. You can see past ideas here.

Image © iStockPhoto: mattjeacock

Do stakeholders want to add text to your materials? Here’s one study you can use to show how wordiness can hurt learning.

The study compared three lessons about the same weather process. All lessons used the same illustrations but varied in the number of words.

The lesson with the fewest words resulted in the most learning.

Read the original publication (PDF) from the Journal of Educational Psychology, or see the summary on pp. 109-115 of Efficiency in Learning by Ruth Clark, Frank Nguyen, and John Sweller.

“We shouldn’t use contractions because then people won’t take the content seriously.” Sound familiar?

Or maybe you’ve heard this: “We shouldn’t use contractions because they’re confusing for people who speak English as a second language.”

The result of these beliefs can be robotic chanting like the paragraph that you are reading now. I will not use contractions as I say that sometimes we become obsessed with details of grammar that are not actually useful, and as a result of this obsession we do not see the big picture. We are too busy enforcing small rules that do not help the learner, so we do not realize that our learner is thinking, “I will leave this course now because this text I am reading did not come from a human being.”

“They won’t take it seriously!”

Here’s what Ruth Clark and Richard Mayer have to say about “conversational” style in e-Learning and the Science of Instruction:

“Research on discourse processing shows that people work harder to understand material when they feel they are in a conversation with a partner rather than simply receiving information.”

Clark and Mayer dispute claims that the content must sound stiff and serious for learners to take it seriously. They go on to cite studies that showed that saying “you” improved learning, as did using a human narrator instead of a machine-simulated one and adding a friendly helper in the form of Herman the Bug.

I like to think that it’s safe to extend Clark and Mayer’s “personalization principle” to contractions, because people and bugs named Herman use contractions, though I’m not aware of research specifically on contractions in elearning. But if our goal is to sound like a human being, contractions definitely help, especially in an audio script.

Even the US Federal Plain Language Guidelines tell us to use contractions when appropriate: “Contractions make your writing more accessible to the reader. Research shows that that they also enhance readability.”

“Contractions confuse people who speak English as a second language.”

Some people worry that learners who speak English as a second language have trouble understanding contractions. I’ve tutored many new English speakers, and contractions are one of the first things they learn to understand. They might not use contractions themselves for awhile, but they understand them.

The trick is to know your learners. If they rarely use English at work, then yes, they would benefit from extremely simple language (or, better, materials in their language).

However, if you expect your learners to know enough English to read your advanced text about cross-border pharmaceutical regulations, then there’s no reason to worry about contractions. If a simple “don’t” stumps your learner, they won’t understand the rest of the course, either.

“Contractions interfere with automatic translation.”

Some people also say that contractions mess up machine translation. I’m not sure where this belief comes from. Even the free, feeble Google translator understood and translated this: “If you don’t know how to read, you might have trouble using an online course.” It was also fine with sentences using “doesn’t,” “couldn’t,” “hasn’t,” “we’re,” and “we’ll,” and then my short attention span kicked in, but I’m confident that other common contractions survive translation as well.

What do you think? Have you heard a convincing argument against contractions? If you speak English as a second language, is it hard for you to understand text that uses contractions?

Image ©iStockPhoto | hundreddays

Probably everyone on your team agrees that elearning should be concise and lively. But does everyone agree on what “concise and lively” looks like? Here’s one way to get everyone on the same stylistic page.

Quantify, quantify

When we talk about writing style, we can get bogged down in personal preferences that are hard to communicate. But if we use readability statistics to quantify style, it’s easier to guide writers.

I’m not talking about the nearly useless “ninth-grade reading level” requirement in your corporate style guide. Instead, let’s look at the Reading Ease measurement that’s part of Word’s readability check. It’s a much more practical guide, especially if you compare your score with that of familiar publications.

What does this chart tell us?

Want to be popular? Aim for a high score.

The highest-circulation magazines tend to have the highest readability scores. Coincidence? I think not!

Instructions can be short and lively

I included Better Homes and Gardens and Family Handyman because they cover a lot of the same territory that elearning does: they motivate you to make a change and tell you how to do it. They also manage to get a high readability score while using terms like “oakleaf hydrangea” and “personalized wrench.”

What score should you aim for?

Many plain-English advocates suggest aiming for a score in the 60s, and that’s my preference, too (this blog post gets a 63). I’ll settle for the 50s if necessary.

Unfortunately, a lot of elearning ends up in the 40-something “Suits” category thanks to corporate drone.

De-drone to improve your score and motivate learners

The reading ease formula considers sentence length and the number of syllables in words, so short sentences with short words score better. But changing your style to get a higher score can also have a profound effect on how the reader feels about you. Here’s an example.

It is expected that all employees will strive to achieve the highest standards of customer service, as service excellence is a competitive differentiator in the market and improving customer service is key to the Firm’s strength as a business. To that end, this course demonstrates the six-step Customer Delightification process which…

Our competition does a pretty good job of customer service. But soon they’ll find out that “pretty good” isn’t good enough, because we’re going to do better. This course will give you …

We stopped talking around the issue and stated it directly, the way our CEO might say it. And by using “we” and “you,” we made clear that we’re human beings in a conversation, not robots issuing edicts. These changes also improved our reading ease score by a bajillion percent.

Quick ways to increase your score and sound like a human being

• Say “you” and “we.”
• Cut 98% of adjectives and adverbs.
• Write active sentences that make clear who does what.
• Use strong verbs instead of wimpy “is.”
• Look for tacked-on clauses (“blah blah, which…” “blah blah, because…”). Turn them into standalone sentences.

How to check your score in Word

The readability check is part of Word’s spelling and grammar check. So, check your spelling. If you don’t see a window with readability statistics, you need to turn on the feature:

1. Open Preferences.
2. Choose Spelling and Grammar.
3. Check the box next to Show readability statistics.
4. Check your spelling. You should see the readability results.

Be sure to check a big chunk of text–500 words or more. Short snippets give unreliable results.

Check both on-screen text and narration scripts

All the text associated with your material should be concise, easy to understand, and direct. A lot of narration sounds dull and de-motivating because it’s coming from the “Suits” category.

Why not use grade level?

1. Grade-level statistics have too much baggage. People worry about offending their audience by writing “below” their educational level. For example, a stakeholder could say, “Our learners all finished college. Therefore, we should write at grade 16. Writing lower than that dumbs down the material.” Using the reading ease score and keeping the conversation focused on magazines read by adults avoids these issues.
2. Grade levels aren’t global. “Seventh grade” means different things in different cultures, while the reading ease score isn’t tied to the US educational system. You can really localize the process by determining the reading ease scores of local magazines and comparing your materials to them.

For way more about this topic, including research and how-to guides, see Writing for the Web.

You’re standing in the Daniel Boone National Forest wearing 97 chiggers and a banjo. You’re surrounded by old-time musicians, and they’re playing this tune:

Download audio file (natchez.mp3)
(“Natchez Under the Hill” as played by The Fatted Calf String Band)

You want to play along. But you’ve never heard the tune in your life. What do you do?

If you answered, “Pull out my copy of 10,273 American Old-Time Tunes and read from the book,” you would be very, very wrong.

To join the jam, you need to be able to play by ear–you need to be able to wing it. You need to adapt the rules you’ve learned from other old-time tunes to this new situation.

To transfer their learning to their jobs, your learners need to be able to wing it, too. They need to apply the rules they learn from your courses to new situations that you could never foresee.

Unfortunately, it’s hard to teach people to wing it. It’s easiest to just tell them what to do, but what we need to do is to teach them how to think.

What’s wrong with telling them what to do?

Let’s say you’re writing a short course on having difficult conversations. The course is supposed to give everyone the same basic knowledge before the face-to-face training. You’re doing the section about using “I” language.

A common approach would be to write this:

Be sure to use “I” language. For example, say, “I was disappointed when you were late to the client meeting.”

This might be accompanied by a photo of someone saying, “I was worried when you didn’t respond to my emails.” And later on there’s a multiple-choice question about “I” language in which the learner chooses an “I” statement from statements that clearly aren’t “I” statements.

What’s wrong with this? It patronizes the learner. Or perhaps I should say, “I feel patronized when someone just tells me what to do and then gives me a simple quiz about it.”

I also don’t see any reason to believe that the rule is useful, plus I know that it’s never that simple in the real world. So why pay attention at all?

What’s the alternative?

Show, don’t tell. Present a scenario that has grey areas and let the learner draw a conclusion about it. Only then make your point, if necessary. Here’s one approach:

1. Don’t say anything about “I” language. Instead, show a manager speaking with a team member. The manager says in a dialog bubble, “I feel frustrated when you’re so argumentative.”
2. On the next screen, we see the team member saying, “Argumentative?! I’m not argumentative!”
3. The manager thinks in frustration, “Why is he so mad? I used “I” language!”

So far you’ve done two things: You’ve introduced the concept of “I” language to the 3 people who have never heard of it, and you’ve demonstrated a misuse of it that makes the learner think.

Next, you could ask the learner, “What just happened?” The correct option is the point you want to make: “The manager used judgmental language instead of focusing on a specific example.”

You could then have the learner choose a different statement for the manager to say. Instead of immediately identifying a choice as “correct” or “incorrect,” you could use a short, one-screen branch to show the effect of the chosen statement on the listener. Of course, the correct option is a well phrased “I” statement that keeps the listener calm and receptive. Feedback points out why that statement worked.

And then, if you must, you could summarize what was just demonstrated about “I” language and its correct use. But by now, your learners have already drawn the right conclusions about it, and all you’re doing is reinforcing their own thoughts.

But that sounds like a lot of work!

It’s easy to list some rules on the screen and require learners to “know” them. But does simply “knowing” a rule change people’s behavior?

Instead, if we use a few more screens and a couple of photos to show the results of decisions, we’re encouraging the learners think for themselves. And conclusions that people reach on their own are far more persuasive than rules that other people give them.

Instead of passively reading a simplistic rule that rarely works simply in the real world, our learners are learning how to think about a type of situation—so they’ll be able to wing it.

What do you do to help learners learn to think? Do your stakeholders like grey areas and scenarios? Let us know in the comments.

Whew. Long post. I’m sorry I don’t have time to make it shorter. I’m very busy with client work (thanks, clients!) and travel.

How good is your ear for dialog? Find out with Dude or Droid, a simple drag-and-drop activity I created to try out Dragster.

As you decide who said each blurb, notice the cues that you’re responding to. What makes dialog sound natural, and what makes it sound stiff?

Click the image to start the activity, and pretend the “TRIAL” watermark isn’t there. Then come back here for some dialog tips and a mini-review of Dragster.

What did you notice about the dialog?

In the droid’s lines, you probably saw these symptoms of unnatural dialog:

• Lack of contractions (“you are wearing”)
• Obsessively grammatical sentences
• Formal wording (“wish” instead of “want”)
• Introductory -ing phrases (“Upon examining the data…”)
• Legalistic weaseling (“and/or”)
• No idioms or slang
• Buzzwords that no human should say (“key value-add”)

We can make the droid sound more like a dude with a few changes:

• Use contractions: “She is our best chainsaw juggler” becomes “She’s our best…”
• Break sentences into fragments of different types: “If you wish to play the banjo, please go outside” becomes “You want to play the banjo? Then go outside” –and it has more personality, too
• Choose informal words: “wish” becomes “want”
• Replace -ing introductions with past tense: “Upon examining the data” becomes “When we examined the data”

What about “Our biggest deal just fell through?” Why did I make that a dude line? Mostly because it uses an idiom (“fell through”). Most droids don’t know idioms and would say something like “Our biggest opportunity is no longer viable.”

Dragster mini-review

To create the activity, I used version 3 of Dragster, an easy-to-use web tool that creates Flash drag-and-drop activities that are SCORM compliant. A basic version of the tool is available free; the version I tested costs £45 a year.

The workflow is intuitive and fast. Once my graphic was ready, it took less than 15 minutes to create and publish the activity in Dragster.

I got confused once, when I thought I could save some draggable labels and edit them later. Apparently, once you’ve saved the labels, you can’t change them, though you can add more labels.

The technique for defining target areas will be familiar to PowerPoint and Keynote users: you draw an invisible shape over the target area. Because your shape can have an almost infinite number of points, you can accurately define complex “correct” areas. You can also set up “close, but not quite” areas. (I just used super-easy rectangular targets.)

The draggable items can be images or text labels. You can save learners’ scores, and learners can also collaboratively work on an activity and add their own labels. One feature I would recommend for future versions is individual feedback for each dragged item, such as a hint that appears when the learner drags something to the wrong spot.

Your customization choices are limited, but that’s a fair price to pay for speed and ease of use. Learn more at the Dragster site.

Click the green arrow to help determine the next celebrity verb!

Can’t afford a full-fledged simulation? You can still recreate the learner’s world in your materials, even if your only tool is the lowly multiple-choice question.

Let’s say you’re writing materials for people who create custom pet hedgehogs using genetic engineering. You might be tempted to write a question like the one below.

It’s a good idea to include parrot genes in a custom hedgehog.

1. True
2. False (correct)

How could you make this question more realistically reflect the learner’s world?

First ask yourself, “Why does the learner need to know this fact?”

Then write a question that tests both the learner’s knowledge of the fact and their ability to apply it in the real world. “What if…” questions come in handy for this.

What would happen if you included parrot genes in a custom hedgehog?

1. The hedgehog would become depressed by its inability to fly.
2. The hedgehog would charm customers with its bright plumage.
3. The hedgehog would develop an offensive vocabulary.

The “before” question tests only basic knowledge. The “after” question tests the same basic knowledge and at the same time tests whether learners understand why the fact is important.

Of course, make sure your materials answer the question.

Obviously, the “after” question has more shades of meaning and suggests that the materials contain more information than simply, “It’s a bad idea to include parrot genes.”

I’m a big advocate of writing questions before you write the main materials. So if you want your low-tech materials to simulate real life, you might:

1. Write questions that learners must be able to answer in order to fully understand (not just perform) their jobs. These probably aren’t fact-based questions but instead are “What if” and “Why” questions.
2. Then write the materials that will precede the questions, assuming you’re doing the standard “tell, then test” approach. Also consider putting the questions first, then explaining why the correct answer is correct.

What are some other ways to give basic questions more depth? Are simple fact checks ever appropriate?