Test then show: a learning design pattern
There is a formula, known as the quadratic formula, that enables you to calculate if, and where, a parabola curve cuts the horizontal axis in a graph.
Most students learn this formula in early high school. A common method of teaching it is more or less as follows:
- Provide the formula
- Demonstrate how to use the formula to solve the equation for a parabola
- Get the student to practice: give them the equation for a different parabola, and ask them to use the formula to solve that one
In other words, you “show then test”. You show the student how to do something, and then you test them (or get them to practice) on it later.
Sounds pretty intuitive, yeah? In fact, why wouldn’t you do it this way? In fact, when you break it down, why would you ever teach anything any other way?
Well, there can be a number of situations where it makes sense to reverse it, to “test then show”.
The idea here is to put the learner into a position where they are first asked to contribute to the learning experience in some way, to demonstrate their current understanding or capability. (This is the “test”, though it does not and should not have anything to do with actual assessment!)
Then, once the learner has had a go, the learning moves on to “show”, where the teacher (or other students!) provide content, context and guidance around the topic at hand.
I’ll go through three main reasons for using this approach and the sorts of settings where these reasons apply, and I’ll give some examples for each one.
Reason #1: Get the learner to reflect on their own assumptions, biases, or blind spots
A father and son are in a car accident. The son is badly wounded and rushed to the ER while the father stays at the scene to get treated for minor injuries.
The head doctor walks into the operating room and takes one look at the patient and states: “I can’t operate on this boy. He is my son.”
How can this be?
Sometimes, part of the learning outcome is for the learner to reflect on their own preconceptions. More specifically, there are times where the point is to experience these preconceptions, to better address them and learn from them.
In many of these cases, blatantly pointing out the preconception can defeat the purpose of the learning. Once you tell someone about a bias they may have, they’re more likely to be on the lookout for it… at that moment. But it can deprive them of the aha! moment that goes with actually reflecting on the bias or assumption.
One answer to the above riddle is that the head doctor is the boy’s mother. If you struggled to discover this answer, now’s probably a good time for you to reflect on your own unconscious gender bias. Which is more effective than me telling you that unconscious gender bias exists and is a real problem.
It doesn’t have to be as blatant as calling out “blind spots”, either. At its heart, “question your assumptions” applies whenever you want learners to reflect on their understanding prior to the lesson.
For example, I’ve designed a number of courses around harassment and bullying in the workplace. Often, telling/showing that “bullying is bad” is not as effective a learning experience as, for example, asking the learner to:
- First consider and identify where the bullying is happening in subtle scenarios, in which the harassment or bullying might not be so obvious. They may even claim that harassment isn’t occurring in the given scenario, which is an opportunity for an even stronger learning outcome.
- Or ask them to consider their level of responsibility in such a situation. A lot of people will tell you that bullying is “obviously bad”, but what should you do if you happen to witness it? In this situation, having someone genuinely think about it first will be more instructive than simply throwing the rule book at them.
Reason #2: Reinforce common sense
Then there are learning settings that are the inverse to the above. Instead of getting the learner to “check their assumptions at the door”, you may want to get them to apply common sense and reinforce that they’ve already “got this”.
This is particularly effective in the case of adult learners who come to learning with lots of life experience and existing mental models that you can leverage.
For example, one of my colleagues is currently working on a suite of courses on ethics, business practices, and so on, targeted at people working in the financial services sector. Many learners that will complete the courses are already working in the field.
Suppose you want to “teach” this audience about the do’s and don’ts around accepting a bribe. It is likely that many of the learners already:
- Know what bribery is
- Can spot it when they see it
- And they know that it is wrong
That’s not to say that there aren’t things to learn — bribery can be subtle to navigate in some situations. But the target audience comes to the learning experience with some common knowledge.
My colleague starts each course with a “Thought bubble” — a hypothetical scenario to get the learner to nominate “what they would do” in a given situation. Sometimes this is to encourage the learner to “question their assumptions”. But in many cases, they provide the learner with confidence that what’s required here is simply common sense — which they have!
This form of test-then-show can also be useful anywhere communication skills come up. We all have experience talking, listening, showing empathy, and so on. Asking a learner how they would deal with a delicate situation with a friend or colleague will give them the opportunity to prove to you (and themselves) that they already know how to do this. And if they don’t — there’s a motivated opportunity to learn! More on that later…
Reason #3: Productive failure
A young high school student who has no familiarity with the quadratic formula, and who barely knows what a parabola is, probably isn’t going to learn much without you showing them first.
But, there are lots of situations where failure can be productive learning. In fact, in many cases, a learner will achieve much deeper learning by trial and error, instead of following a rule book.
For example, suppose your target audience is learning about fire safety equipment. Specifically, how to properly choose and position fire detection and prevention equipment in a building. There are a tonne of rules and regulations around how to do this. But why don’t you give the learner some building plans, and ask them to have a go?
Now, just to be clear, this is not the same as Reason #2 (common sense). The learner will probably come up with a bad design for the safety equipment at first. But while they are doing this, they are thinking it through, trying to make sense of why to put this smoke detector here and that hydrant there. They may need to be given guidance later on — but that’s OK!
Or take coding, as another example. I can say from personal experience, as a very amateur coder, that 95% of the fun in programming is trying, failing, trying again, to get your code to achieve a (frustratingly simple) outcome.
That might sound obvious, but there’s a key lesson here for learning design.
The more you hold a coder’s hand — that is, the more you show them how to do it before you test them — the more they are just going to try and follow whatever rule or rubric you’ve given them. This is what we call extrinsic motivation, and it leads to surface learning.
The opposite to extrinsic motivation is, of course, intrinsic motivation, meaning that the coder (or more generally the learner) is motivated from within — to tinker with the parts, bang this part with that to see what happens, to work out what is going on. This leads to deep learning.
How do you test then show?
The above reasons and examples should give you an indication of ways you can go about using test-then-show in your learning design. The method will depend on the situation. (For example, which of the above reasons apply?)
But a general flow is as follows:
Step 1: Provide context and reassurance
Many learners get anxious about being “assessed”, and the idea of being tested before being told how to do something is especially daunting. So, it is important that you do provide context. Often you’ll want to get across the core message that this is a safe space for them to make mistakes. The nature of this “test” is not to assess them, but to provide an opportunity for learning.
For example, you should assure the fire safety learners that we aren’t going to be installing any equipment based on their first stab at it!
Step 2: Provide the test
Give the learner the setting, and ask for their input, whatever that may be. It may be to make a decision, to spot what’s wrong, to try and build something, or to simply give their thoughts on a matter.
You will often want to conclude this step by teasing out the immediate point of the “test”. If it was to get the learner to question their bias, what bias was it, and how did this come out in the test? If it was to confirm a piece of common sense, did the learner successfully demonstrate this common sense? If not, what do they need to learn more about?
Step 3: Show
Using the test and its immediate repercussions as a spring board, launch into any further context or guidance you aim to provide.
For example, if your learning outcome was about unconscious bias, the outcome of the “test” may be to see how this bias can manifest. Now’s the time for you to show your learners more about unconscious bias, and point them towards any other experiences to teach them about that.
Or, if your learning outcome was about fire safety equipment design, the “test” was your learners’ opportunity to try and work out the elements of such a design. Now’s the time to dig deeper and look at those rules and patterns that govern good fire safety design — now with a cohort of learners that have thought about it, and are as ready as they’re going to be to learn more.
