When I first started teaching A-Level English language, I was heavily dependent on textbooks because I needed to be confident that what I was teaching aligned with the syllabus. But this led to lessons that resembled lectures.

This was a problem. We know that, even in the lecture’s natural home, a university, this style of delivery alone does not constitute learning. No one figures out how to ride a bicycle or swim merely by reading about it or having it explained on a chalkboard. There has to be some element of interaction between the learner and the content.

So, to solve the problem, I turned to the obvious fix: group work.

Consider this example. Students were given several non-fiction texts and a sheet with two adjacent word lists. These came in pairs: “written” and “spoken”, “formal” and “informal”, “personal” and “impersonal”, and so on. A horizontal line ran between each pair of words. The students were supposed to mark a position on the lines showing how far each text leaned in a particular direction.

An extract from a legal document, for example, would get a mark at the far left of the written-spoken line and a mark at the far right of the personal-impersonal line because it does not sound at all like spoken English and comes across as highly impersonal.

The idea was that each text had a unique linguistic fingerprint. All I had to do was give the instruction: “Work in groups to identify the characteristics of each text” and off they would go to learn that for themselves.

Except that this never happened. Instead, one student would immediately begin to read the texts. Another would speculate about how formal the legal text was. A third would say nothing but nod sagely as if in agreement. The fourth would switch off entirely.

And this is the trouble with group work in schools. Too often, the instruction to work in groups gives students no information about how to work. It’s like saying, “Work in groups to build this house” instead of assigning clear roles and responsibilities. What we have here is unstructured interaction: the task is clearly defined but the organisation of the team isn’t.

From hope to certainty

Let’s think about why we want students to work in teams in the first place. After all, why not have them complete the task individually?

Presumably, it’s because we know we need to scaffold the learning. The students are not yet sufficiently skilled to solve a problem individually, so we want to offer some guided practice with others first. By getting them to work in groups, we hope they will benefit from having to articulate their thinking, from hearing others explain the problem in different ways, and from having to justify their various points of view. But unfortunately, that’s all it is - a hope. The instruction “Work in groups” does nothing to make these outcomes a certainty.

So what can we do? To find out, I looked to the research. As the Education Endowment Foundation puts it: “Effective collaborative learning requires much more than just sitting students together and asking them to work in a group.” The foundation has found that collaborative learning can add five months to students’ progress over the course of a year. And its research is rigorous: “Over 40 years a number of systematic reviews and meta-analyses have provided consistent evidence about the benefits of collaborative learning.”

My own introduction to the research behind collaborative learning came after attending an introductory Kagan Cooperative Learning session while at a conference (I should disclose at this point that I have been a Kagan trainer for more than a decade now, although in this article I am discussing collaborative learning generally rather than focusing on a particular approach).

This led me to the book Classroom Instruction that Works by Robert Marzano, Debra Pickering and Jane Pollock, which again gave compelling evidence for the benefits of this approach. A meta-analysis looking at 938 separate studies found “an effect size of .78 when compared with instructional strategies in which students work on tasks individually”. As the authors put it: “Cooperative learning … has a powerful effect on learning.”

But what does all this mean in terms of actually managing a task?

Let’s return to the text-sorting task outlined above. We want to hold all students individually accountable throughout. So, first, we get them to complete it alone. Once they are in groups of four, we want them all to take part in the discussion, so we give them two minutes each to explain how they have categorised the texts.

This equalises the discussion and puts them on the spot: none must escape!

If a teacher calls on students one at a time in a class of 30 and gives them only 30 seconds each, it will take 15 minutes. This can be a long and disengaging process. But by using the collaborative learning method described, everyone can respond in eight minutes, because each team is working simultaneously. Following this, students may return to individual work, altering their initial answers as necessary. This systematic approach means that everyone will hear three other explanations before returning to their own work. Contrast this with group work where engagement is left to chance.

Training team players

Of course, there are challenges. Some students, often the most able, prefer to complete work by themselves. Others lack the necessary social skills. It’s important, not to overlook these aspects, so explain the rationale behind the approach.

Employers want people who can work in a team, so it’s important that developing these skills is a regular classroom experience. What if students lack the necessary social skills? Build that into your collaborative learning tasks with a “thank your partner” element. If students can’t politely disagree, give them some language to help: “I understand your point of view, but I feel … ” We do this with our own kids (“What’s the magic word?” ), so why not do it in the classroom?

The research demonstrates the impact on progress. But good collaborative learning also affects students’ attitudes (an Ofsted inspection at our school noted how it improved their confidence), their metacognitive thinking and teachers’ ability to assess their understanding, since it is so often vocalised.

Perhaps most importantly, it also means the students are spending more time engaging with lesson content than with the teacher.

Jon Eaton is an English teacher and research lead at Kingsbridge Research School. He is also a trainer in Kagan Cooperative Learning for TeacherToTeacherUK (Kagan UK)