The process of selecting studies or topics for my blog is sometimes easy, sometimes difficult. I have to admit that it can be very tempting to go for large studies or striking results. I know from experience that these can generate more ‘clicks’. But I personally found last weekend’s study about the feelings scientists experience during research too fascinating not to share.
Today’s study concerns a debate in education that just won’t go away: explicit instruction versus inquiry-based learning. Should you explain something to pupils or students first and then let them get to work? Or is it better the other way around? You can probably already guess the answer, but the nice thing about this new study is that it neatly illustrates and confirms some existing nuances.
The study, by Jimmy Zambrano and colleagues, published in Learning and Instruction, involved 211 students in Ecuador studying pre-hospital care. It is worth noting that a significant amount of educational research takes place in medical education, often of high quality. The topic was clinical pharmacology, more specifically, the use of medication for high blood pressure. This is, not coincidentally, quite complex subject matter. Students have to connect several factors at once. Consider factors such as the type of medication, its mechanism of action, potential side effects, and patient characteristics. The researchers refer to this as high element interactivity: different elements must be processed and connected simultaneously.
The researchers randomly divided the students into two groups in what I think is a simple but elegant design. Both groups received exactly the same learning activities: twenty minutes of explicit instruction and twenty minutes of inquiry-based learning. Only the order differed.
One group received explicit instruction first and then moved on to inquiry. The other group started with inquiry and received the explicit instruction afterwards. The explicit instruction was also delivered through the same recorded video, ensuring that the content was identical. You can object to this, of course, because good instruction also involves interaction, but let’s call that a limitation of the study.
Two weeks before the experiment, the students did a knowledge test. This allowed the researchers not only to determine which sequence worked best on average, but, more importantly, whether the effect of that sequence differed depending on the students’ prior knowledge.
Now for the results. As you might have expected, the students who received explicit instruction first performed better on average. They scored 12.95 out of 19 after the intervention, compared with 11.66 for students who started with inquiry-based learning. But averages can hide important nuances. That was also the case here. There was a clear interaction between the order of instruction and students’ prior knowledge.
For students with little prior knowledge, receiving an explanation first was clearly more beneficial. At one standard deviation below the mean for prior knowledge, the difference was almost four points on a nineteen-point test. That is a more than substantial difference. The advantage became smaller as prior knowledge increased. Around the mean, there was still an advantage of almost two points for the group that received explicit instruction first.
But then the researchers, as they had expected, found the expertise reversal effect. For students with a lot of prior knowledge, the effect was reversed. At one standard deviation above the mean, students who started with inquiry scored approximately 1.6 points higher than students who received explicit instruction first.
The researchers were even able to identify a kind of intermediate zone. With little prior knowledge, explicit instruction first was clearly better. For a middle group, the order made no statistically significant difference. With sufficiently high prior knowledge, however, inquiry-based learning first was better. In this sample, approximately 64 per cent of the students belonged to the first group, 19 per cent fell within the intermediate zone, and 17 per cent belonged to the group with the most prior knowledge.
This is not a groundbreaking study, but it is nice to read one with such a clear design and equally clear results. The authors describe their study as a conceptual replication of earlier research by Zhang and Sweller (2024) with secondary school students. That study also found that, with complex subject matter, beginners benefited more from explicit instruction before problem-solving, whereas for students with more prior knowledge, the reverse order could be more beneficial. Sweller’s cognitive load theory is also an important explanation for the findings in the present study, although the researchers did not actually measure cognitive load.
But… what does this mean for everyday practice? Should we administer prior-knowledge tests to all pupils at all times and then differentiate accordingly? I fear that would be overdoing it. Fortunately, teachers have many other ways of gaining insight into prior knowledge, and you usually know what your pupils have already encountered in your own lessons. So all kinds of formative assessment can help. Simply ask questions at the beginning of and during a lesson, have pupils briefly explain something, present them with a problem, look at their answers on mini-whiteboards, or simply listen carefully to what they say.
But the second part is perhaps more difficult. What do you do with that information as a teacher? After all, not every pupil will reach the point at the same time where less explanation and more inquiry become useful. I do think this does not mean that you have to design a separate learning pathway for every pupil. Perhaps a different view of differentiation is needed here. Not necessarily by giving everyone different content or by creating different pathways, but, for example, by varying the point at which you reduce support and expect greater independence. And that particularity need not be true for each and every topic. The learner may already possess adequate prior knowledge regarding a particular topic and yet be a complete novice when it comes to the next.