3. What’s the Goal?

(IN WHICH WE LEARN THAT BUILDINGS LEARN LIKE PEOPLE DO, AND THAT YOU SHOULD LET YOUR LEARNERS DRIVE THE CAR, NOT JUST RIDE ALONG)

If you don’t have a clear destination, you can’t plot a clear path and you certainly can’t communicate that path to your learner:

In determining the path for your learner, you want to do these things:

• Identify what problem you are trying to solve.

• Set a destination.

• Determine the gaps between the starting point and the destination.

• Decide how far you are going to be able to go.

CLIENT:                        The salespeople just need to understand the basics of insurance underwriting/cellular service/cloud computing. That’s the learning objective.

LEARNING DESIGNER: OK, sure, and why is it important they know that?

CLIENT:                        Well, they just need to be grounded in the basics.

LEARNING DESIGNER: Uh huh. And what will they actually do with that information?

CLIENT:                        They just need to know it.

LEARNING DESIGNER: And what bad thing could happen if they don’t know it?

CLIENT:                        Well, they’d look stupid in front of clients.

LEARNING DESIGNER: Ah! Great. So maybe the learning objective is something more like “Salespeople should be able to answer customer questions accurately.”

CLIENT:                        Yeah, I guess that makes sense.

Before you start designing a learning experience, you need to know what problem you are trying to solve.

A lot of learning projects start with the goal rather than the problem, but that puts you in the position of solving problems you don’t actually have, while failing to address the real issues.

For example, let’s look at this goal and some possible solutions:

All of those solutions could potentially address the stated goal. Let’s take a look at the same set of choices with the gap identified first.

The goal is still valid, but now some of the possible solutions are better than others—observing other supervisors and role-playing become better solutions than taking a knowledge-based course.

Here’s another example:

Let’s take a look at this same one with the gaps:

As with the previous example, when we use the gap definition, some of the possible solutions are better than others.

Some questions to ask to help identify the problem:

• “What bad thing will happen if they don’t know this?”

• “What are they actually going to do with this information?”

• “How will you know if they are doing it right?”

• “What does it look like if they get it wrong?”

• “So why is it important they know that? Uh huh, and why is that important?” (repeat as needed)

Students need to learn to be better managers.

That’s like saying, “Meet me in Africa.” It’s a destination, but it doesn’t help you book a flight. In those instances, you want to start breaking it down:

Students need to schedule restaurant employees so all shifts are adequately covered.

Students need to provide appropriate feedback to an employee who is chronically late for work.

Once you start breaking these down, you can formulate much more specific routes and destinations to get where you need to go.

I might take a class in film appreciation just for my own satisfaction, but I wouldn’t expect the learning objectives to make me a professional film critic. A community-education course on Taiwanese cooking, a painting class at a local art museum, or even a high school French class isn’t really set up in response to a “problem.”

What do the learners want or need to get out of the experience? If you are creating learning experiences that are primarily for enrichment, you will still create better experiences if you consider and define those wants or needs.

The more specific you can be about this, the better you can design the path to get there. For example, let’s say someone is teaching a Java programming class and the objective is:

Students will understand how to program in Java.

This is a learning objective pretty much guaranteed to make a formally trained instructional designer’s head explode:

There are actually a number of problems with this learning objective, but let’s start with the use of the word understand.

It’s a totally reasonable place to start—of course we want our learners to understand, but there’s no visible way to see if someone “understands.” It’s hard to design for something as fuzzy as “understand,” so we need to define it more.

One way some people get around this is by using “doing” words in their learning objectives:

Students will be able to explain the value of computation for modeling and simulation.

Students will be able to describe the proper use of method calls.

Students will be able to define and describe the use of core data structures such as arrays, linked lists, trees, and stacks.

They use words like define, describe, and explain because those are observable actions—you can witness someone describing, defining, or explaining.

Of course, this is a hedge; it doesn’t really get you around the problem, unless you are having people memorize and recite definitions. In fact, it’s almost as difficult to determine if they can define something as it is to know if they understand something. Besides, you don’t actually care if they can define it—you want to know if they can do it.

Using those kinds of approximations is basically saying, “I don’t know exactly how to explain what I want, but I’m pretty sure this gets them in the neighborhood.”

Ultimately, since we care not about what they know but about what they can do, the learning objectives should reflect that:

The student will be able to create a simple, fully functional user interface that collects customer data and transmits that data to the database.

So when you are creating learning objectives, ask yourself:

• Is this something the learner would actually do in the real world?

• Can I tell when they’ve done it?

If the answer to either of those questions is no, then you might want to reconsider your learning objective.

OK—now let’s take a look at each of these. While you look at the answers, think about how you could change the learning objective to make it meet both criteria.

So if we wanted to fix that last objective, how could we make it better?

Original objective:

Learner will understand the limitations of JavaScript as a programming tool.

A possible revision:

Learner will be able to identify the best programming tools for a specific task and be able to state the reasons why.

Or perhaps:

Learner will be able to state whether or not JavaScript is an appropriate programming tool for a specific task and give a correct rationale for the decision.

There may be times when it’s not feasible to have real-world tasks. For example, if you are teaching someone nuclear physics, there may be some conceptual material that just doesn’t have an analogous real-world task but that is necessary to understand in order to tackle later concepts. Still, even in those instances, there’s probably something the learner is supposed to do with the information, even if it’s just to use it to understand something else. Also, learning objectives related specifically to motivation gaps may include what learners know or believe.

These recommendations are guidelines, not unimpeachable rules. Use your judgment. If you feel like you need to do backflips to get a learning objective to work, then that’s usually a clue that you need to unpack it more—either break it down or keep asking why until you’ve uncovered the real purpose.

The first way is to think about how sophisticated or complex you want your learner’s understanding to be. One scale for this is Bloom’s Taxonomy (this is the later version, revised by Anderson & Krathwohl in 2001):

• Remember

• Understand

• Apply

• Analyze

• Evaluate

• Create

So, for example, if someone is reading The Non-Designer’s Design Book, an excellent book on the basics of graphic design, they would learn about the graphic design principles of Contrast, Repetition, Alignment, and Proximity (which adds up to a nicely memorable acronym).

If we looked at this from the perspective of the taxonomy, it might be like this:

The items are intended to become more cognitively demanding as you go (e.g., “remember” should be easier than “evaluate”), and some learning designers view the list as a progression (e.g., you have to understand before you can analyze).

Logically, the idea that this is a progression makes sense, but it’s not a recipe for a learning design. For example, analyzing a series of advertisements could be a good way to understand the principles, or guiding learners through the creation of a print advertisement could be a good way to learn an application.

In fact, you could create a great lesson by inverting the order completely:

• Familiarization

• Comprehension

• Conscious Effort

• Conscious Action

• Proficiency

• Unconscious Competence

If we apply this scale to our CRAP example, it might look something like this:

By the time you hit unconscious competence with something like the graphic design principles, you are doing things like seeing movie posters out of the corner of your eye and wincing at the misalignment of the text elements. You’ve fully automated the ideas, and no longer have to put a lot of conscious effort toward the tasks.

Generally it takes quite a bit of time and practice to develop unconscious competence. If you think back to learning how to drive, it probably was months and years before you no longer had to expend a lot of conscious effort on the task. Some of the unconscious driving competencies you might now have probably took years to develop.

It’s more unusual to invert this scale, but it can happen. Language learning is an example. You probably have unconscious competence in a number of grammar rules (you do them right without thinking about it) that you couldn’t consciously explain.

Another way to think of it is as an XY axis—how sophisticated or complex do you want their understanding to be, and what level of proficiency do you want to get them to?

Not surprisingly, the higher you go on either scale, the more time, practice, and skills development you’ll need to do. You absolutely cannot get past Conscious Action without a significant amount of practice distributed over time. You also can’t get to the higher levels of sophistication without multiple examples, and without the opportunity to interact with those examples and get feedback.

A single exposure to the material will likely only familiarize the learner, or take them just a little bit further than that. If you are being asked to get learners to those higher levels, but being given only one point of contact with the learners (one class session or one elearning course), it almost certainly can’t be done.

That frequently manifests itself as a slide like this at the beginning of training presentations:

Ever seen one of those slides at the beginning of a training class? If not, maybe you blocked it out (I’ve certainly done that a time or two). If so, I can only hope the learning objectives were a bit better written than these, but why do we announce the learning objectives to our learners in the first place?

There are a few different reasons you might want to tell your learners what the objectives or goals are:

• To focus the learner’s attention on the key points in the objectives

• To let learners know what to expect

• To let them know what level of performance they should be working toward

There are other reasons why you, as the learning designer, want to have clear learning objectives as well: You want a clear design direction so you know what you are trying to do (and just as importantly, what you are not trying to do), and you want to show a target or benchmark for assessing whether you’ve succeeded at the end of the project. But again, these learning objectives are more for you as the designer than they are for the learner.

Will Thalheimer, an instructional design expert, has a taxonomy of different types of learning objectives (Thalheimer 2006):

New instructional designers frequently struggle with writing good learning objectives, partly because they are often being asked to cram all four of these functions into a single statement. That’s like trying to pack all the gear for a family of four into a single carry-on bag—it’s somewhere between difficult and impossible. It also means that learners are subjected to horrible instructional design jargon. That jargon can serve a purpose for the design team in making sure the learning objectives are precise, but it’s just cruel to bludgeon your learners with it.

So how might you think about each of these categories?

And, for the record, just say no to learning-objective slides at the beginning of the course. If you want to communicate the objectives to the learner, use a challenge, a scenario, or a “your mission, should you choose to accept it” message. There are a multitude of ways that aren’t bullet points on a slide to accomplish the goals of focusing the learner’s attention and letting them know where they are headed.

• Knowledge gaps

• Skills gaps

• Motivation or attitude gaps

• Habit gaps

• Environment gaps

• Communication gaps

So which comes first, the learning objective or the gap? The answer is both, pretty much. It’s likely you are starting with some kind of need or challenge, and you are going to want to investigate both learning objectives and performance gaps. There are times when one will inform the other. For example, if you want to train people how to use the super-cool analysis feature in your software, you might have a learning objective about learners being able to run analysis reports. Subsequent gap analysis might tell you that learners do know the procedure for running the reports—they just don’t know they can access their own data in the analysis. That would lead you to a very different learning objective.

Years ago I had a fairly blechy job teaching GMAT prep classes. The class met for an entire weekend to prepare prospective MBA students to take the GMAT exam the following weekend.

With a typical student, we stood a decent chance of improving their Quantitative (math, logic, problem-solving) scores, but we usually couldn’t make much of a dent in their Verbal scores.

In the Quantitative section, we could teach them some quickie shortcuts for math problems, remind them of the geometry formulas they hadn’t seen since their sophomore year of high school, and get them used to the wacky “data sufficiency” format that shows up on the test.

These skills were information-based, based on activation of prior (albeit rusty) knowledge, or very quick to learn, and could therefore be brought to a reasonable level of mastery in a few hours (whether they retained those skills is another matter).

In the Verbal section, they needed skills like vocabulary, reading comprehension, complex analysis, and reasoning. As you might imagine, these are not skills you acquire in a weekend (try decades). There are very few quickie shortcuts that you can teach someone if the foundations of their language skills aren’t there.

Some knowledge or skills can be acquired quickly, but others are slow and take a long time to develop.

So how far down the path can a learner go?

I’ve had clients who told me they want to teach problem-solving skills in a half-hour elearning course.

Since problem-solving is a skill that takes a looooong time to develop, I usually sigh quietly to myself and then tell them a variant of the above story.

Basically, the idea is that some things change quickly (the actual contents of the room might change daily, the interior decorating might change in months to years), some things change more slowly (the space usage, the interior layout, and the actual structure might change in years), and some things change only very slowly (the foundation might change in years, decades, or centuries). (Brand 1994)

Here’s what Mr. Brand says (on the Long Now Foundation site) about the pace layering of cities and civilization:

The fast parts learn, propose, and absorb shocks; the slow parts remember, integrate, and constrain. The fast parts get all the attention. The slow parts have all the power.

This raises the question, What is the pace layering of learners? What can change quickly, and what changes more slowly?

In the GMAT course I taught, we could—at best—rearrange some furniture (and hope that it stayed rearranged until they took the test the following week). We weren’t going to really change anything like their verbal skills—those were part of the structure and foundation.

Take a look at this example. Todd is a brand-new restaurant manager. He just got promoted, and he needs to get his skills up to par as quickly as possible. Which of the skills that Todd needs to acquire are fast? Which are slow?

Todd’s New Job Requirements

Loosely speaking, approving timesheets, communicating changes, checking security and sprinkler systems, and verifying table settings are probably pretty fast skills. Creating weekly schedules, designing and implementing creative seasonal promotions, recognizing and preventing harassment, and training on safe alcohol sales are probably medium or mixed sets of skills, and things like ensuring a respectful environment, forecasting inventory requirements, minimizing waste, recognizing and rewarding employees, and resolving employee conflicts are slow skills.

Fast skills typically have more explicit rule sets. Faster skills tend to be things where you can make a list of the right answers. Slower skills tend to be things that have more tacit rule sets—it’s hard to say what “right” is, but you might know it when you see it. So, for example, table setting is pretty explicit—you could explain what the rules are—but ensuring a respectful environment is harder to define as explicitly. When a skill is based on tacit rules, it’s usually necessary for the learner to see lots of examples so that she or he can start to recognize patterns.

So what can you do with this? If you identify that something is a slow task, how do you approach it?

Find a few throw pillows. What are some easy, cheap ways to make an impact? It might be a model, a tool, a job aid, a checklist—something that’s easy for your learners to implement right away and that will have an immediate impact. It won’t change their world, but it might solve a small but pesky problem. Don’t try to solve big problems with a throw pillow, though. They may brighten the room and be a cheap way to have an impact (and there’s nothing wrong with that), but they aren’t a substitute for the heavy lifting involved in real behavioral change.

Provide some sturdier pieces. Give them some more concrete material, but recognize that this is going to take more time; they will need to set it up, move it into place, get rid of the old piece, arrange their existing stuff in it, and get used to how it changes their current patterns. Don’t try to do that all at once; keep in mind that there are several steps and that all of them need to be supported, unless you want the unassembled item sitting in its box in the storage area indefinitely.

Recognize that you aren’t going to change their structure. If they have some renovations already in place, you might move them along a little, or you can help them start some planning for future changes. This sounds easy, but actually it’s really hard, because it involves letting go of the deeply held belief that we can do major renovations in a short period of time. We can’t—and it’s a waste of resources to pretend we can. If we approach a learning design with the longer view in mind, acknowledging what we can and cannot accomplish, we can create better ways to help people and ensure that there is a long-term plan.

Respect the foundation. The foundation is based on a learner’s personal bedrock, which is composed of elements like culture and personality. If your structural changes aren’t going to sit well on the foundation, then you are better off changing your design, because it’s really unlikely that the foundation is going anywhere.

Let’s take a look at the management skill of hiring the right people. That can be a pretty slow skill to develop. People who have been managers for 20 years can still be learning new things and getting better at this.

What can we do to help Todd improve his skills in this area?

• Define the problem before coming up with solutions, to ensure you are actually solving the real problem and not a problem you don’t have.

• Use the two questions “Is this something the learner would actually do in the real world?” and “Can I tell when they’ve done it?” to make sure your learning objectives are useful and usable.

• Decide how sophisticated your learners’ understanding needs to be and how proficient they need to be, and then design accordingly.

• Recognize if you are teaching someone a fast or slow skill, and use strategies appropriate to developing that type of skill.

Bloom, Benjamin S. 1956. Taxonomy of Educational Objectives, Handbook I: The Cognitive Domain. New York: David McKay Co Inc.

Brand, Stewart. 1994. How Buildings Learn: What Happens After They’re Built. New York: Viking.

Gery, Gloria. 1991. Electronic Performance Support Systems: How and Why to Remake the Workplace Through the Strategic Application of Technology. Boston: Weingarten Publications.

Thalheimer, Will. 2006. “New Taxonomy for Learning Objectives,” Will At Work Learning Blog, June 1. www.willatworklearning.com/2006/06/new_taxonomy_fo.html.