Research School Network: Making worked examples work – part 1 Amarbeer Singh Gill shares some of the theory underpinning worked examples & three learnings to help make worked examples work.

As a trainee I came across lots of advice about ​‘getting students to solve it for themselves’. I think, on some level, it makes sense why this is appealing: there will come a point where we’re no longer able to teach our students, so when that time comes we want to know they’ll be able to teach themselves.

It was only when I came across Clark, Kirschner, & Sweller’s paper, Putting Students on the Path to Learning, that I realised the mistakes in that approach. It’s not to say that giving students minimal support is ​‘bad’, just as the opposite isn’t ​‘good’. So, how can we make worked examples work for us?

Constantly using worked examples can be unhelpful, and (as we’ll see) can end up doing more harm than good. But, when done well, they’re absolutely fantastic for novice learners. Let’s start by first defining ​‘novice’:

Novices aren’t just ​‘new’ students or low-prior attaining students, they are any student for whom the knowledge being taught to them is new.

This includes students who may have been exposed to the learning previously, but have forgotten so much of it that it can be effectively considered as new learning.

From the above, it’s fairly clear most of our students are ​‘novices’ most of the time! So why are worked examples so useful for these students? Let’s consider an analogy: we’re about to start our first driving lesson. We have no prior experience of driving, but we get in the car ready to learn. The instructor sat next to us simply says, ​“Show me how you’d change gears when driving”. For those of us who do drive I’m sure you can appreciate how daunting a task this could be on your first ever lesson, and for those of us who don’t it’s even easier to imagine the feelings we might experience in that situation! The point is this is what it feels like to be a novice confronted with a new situation: we have absolutely no idea what all the pedals and buttons do let alone demonstrate changing gear!

In an alternate reality our instructor models exactly what we need to do: pressing the clutch, moving the gear stick, and releasing the clutch whilst narrating the process. We practice doing this a few times with the car turned off. Here we have the right scaffolds in place to make sure we know exactly what we need to do. It may look like we haven’t really ​‘learnt how to change gear’ and we’re just following a procedure and there is indeed some truth to it. But this is exactly the point!

Takeaway: Telling students exactly what they need to do is helpful when starting out.

As outlined by Daniel Willingham, ​“thinking is slow, effortful, and uncertain”¹. When faced with problems that are too hard students will often just opt-out but, more importantly, will probably learn very little. In our first scenario it’s

a) unlikely that we’ll figure out what to do by luck and 

b) on the off-chance that we do figure it out, our brains are so occupied with trying to ​‘guess’ our way through the problem that there’s a high chance we wouldn’t actually retain how we did it in order to do it again! 

In other words the limited capacity of our working memory (where we do our thinking) is overwhelmed².

By providing scaffolds, we ease that burden on working memory. Instead of thinking capacity being used up on figuring out what we need to do, we can instead devote the resources to making sense of the procedure that’s already been shown to us. This is why ​“just following a procedure” is such a great starting point when working with novices.

Takeaway: Worked examples allow thinking resources to be devoted to making sense of a given procedure rather than also having to figure it out, reducing the burden placed on our limited working memory capacity.

Does this mean we continually just keep giving students worked examples? Definitely not! As mentioned, the discussion is far more nuanced than this. Firstly we want students to achieve some level of independence and that won’t happen without carefully removing the scaffolding we’ve put in place. Secondly, we want to avoid the ​‘expertise reversal effect’: whilst worked examples can prove incredibly helpful for novices, the reverse is true when expertise gets developed³. Going back to our driving analogy: imagine if we get to a point where we can regularly change gears with ease but our instructor still insists on exemplifying it to us each lesson, we’d likely get frustrated because we feel like we’re being held back.

A reason this can happen is as we develop expertise we start relying on our own thinking to unpick the problem. In a situation where our thinking is developed and we’re still being given worked examples our attention is being split between trying to decipher the problem using our own thinking and focusing on the information from the worked example⁴: we’re either having to drown out the explanation so we can focus on our own thinking, or consciously hold back our own thinking so we can pay attention to the explanation. In either situation we’re placing more demand on our mental resources without it being relevant to the problem at hand.

Takeaway: Worked examples can become an unnecessary burden on mental resources if they are not removed appropriately.

In this part 1 we’ve looked at some of the theory underpinning worked examples and in part 2 we go on to look at some ways we can apply that theory in our classrooms.

Author:

Amarbeer Singh Gill
Lead Practitioner of Maths
@InspiredLearn_

References

^{1 Willingham, D. T. (2021). Why don’t students like school?: A cognitive scientist answers questions about how the mind works and what it means for the classroom. John Wiley & Sons. p.10}

^{2 Willingham, D. T. (2021).}

^{3 Clark, Kirschner, & Sweller. (2012).}

^{4 Kalyuga, S., Ayres, P., Chandler, P., & Sweller, J. (2003). The Expertise Reversal Effect. Educational Psychologist, 38(1), 23 – 31.}