How the R-Generation programme is inspiring young people to become recycling ambassadors in the UK Nestlé Waters UK
Hands on with history at The Roman Baths The Roman Baths
Learn about the importance of empathy in this hilarious space adventure! Hachette Childrens
Sky Arts launches a nationwide celebration of the arts in primary schools and everyone’s invited Sky Arts
Get your free download of the new issue of The Headteacher
Oxford University Press Courses
In education, there is a tendency to look for the ‘next big thing’ – which would make sense if we were already doing all the things known to improve student achievement.
But as long as there are things we know would improve student learning that are not routine practice in our classrooms, we should stop worrying about the next big thing and instead focus on doing the last big thing properly.
Jeffrey Pfeffer, a management guru, calls this the ‘knowing-doing gap’. He points out that typically, companies that fail do not go broke because they had bad plans. Most had good, or even great plans – but they just didn’t carry them out.
This is why ‘sharing good practice’ is such an attractive, but ultimately misleading idea in education. Most teachers have had enough good ideas to last them a lifetime. What they don’t get is support in putting these ideas into practice.
20 years ago, Paul Black and I embarked on a rather ambitious project. We wanted to know what kinds of changes teachers could make in their teaching that would have the biggest impact on how much children learn.
What we found was that using assessment to find out what children have learned, and using this information to adjust teaching to better meet their learning needs, produced more positive benefits than just about anything else that we looked at.
Of course, this idea was not new. 30 years earlier, David Ausubel, an American psychologist, had said that the most important determinant of effective teaching was what the learner already knew. Good teaching, he said, starts from where the learner is, rather than from where we would like the learner to be.
Moreover, since students do not always learn what we teach, and different students learn different things from the same teaching, good teaching involves finding out what students did learn before we try to teach them anything else.
This is why formative assessment, or assessment for learning, is so important. Ausubel’s intuition was correct, but now we had the evidence that it really was the most important thing in teaching.
Of course, some people ask, “How do we know that formative assessment isn’t just a passing fad?” The answer is simple. A focus on formative assessment requires teachers to relate two central issues in teaching – ‘What did I do as a teacher?’ and ‘What did my students learn?’
As long as teachers are focusing on the relationship between those two central issues, they will continue to improve their practice for as long as they stay in the job.
While the idea behind formative assessment is simple, it turns out that the practice is rather complex. Different people define formative assessment in different ways.
Some writers have suggested ‘assessment for learning’ focuses on the teacher’s role, whereas involving students in assessing their own learning should be called ‘assessment as learning’. Some believe that students assessing each other should be included; others think it should not.
In order to develop an inclusive framework, 10 years ago Marnie Thompson and I suggested that formative assessment could be thought of as involving three processes in learning – where the learner is going; where the learner is now; and how to get there.
If we then think of the role of the teacher, the learner and her or his peers, we get a 3 x 3 table, within which certain cells can be combined to produce the figure below:
In the years since Paul Black and I published our original research, the evidence in favour of formative assessment has grown even stronger.
In 2010 the Educational Endowment Foundation first published its Teaching and Learning Toolkit, which examined research from all over the world in terms of its quality and relevance for classroom practice, the cost of implementation, and the impact on student achievement.
It concluded that the three most cost effective interventions were feedback; helping learners become more active owners of their own learning; and collaborative or cooperative learning – which are, of course, three of the five strategies of formative assessment identified above.
The other two formative assessment strategies are necessary precursors to these three. You can’t give feedback until you find out what is going wrong (or indeed, right), so you have to elicit evidence of what it is that students have learned.
And you don’t know what evidence to elicit until you know what it is you want students to learn.
In this sense, the five strategies of formative assessment form a minimum set of requirements for improving student learning. Put simply, if you want to raise student achievement, right now there is nothing we know of that will have a bigger impact than developing the use of formative assessment.
Unfortunately, while the research evidence about the power of these strategies is clear, in many schools the strategies have not always been implemented in ways that are faithful to the research evidence.
Here, I’ll summarise some of the key issues that are involved in effective formative assessment in primary classrooms:
The idea that learners should know what they are meant to be learning is perhaps obvious, but in many schools this strategy has degenerated into a rigid requirement that every lesson should begin with a learning intention, copied by the teacher on to the board, and then copied by the students into their exercise books.
Students do need to know where they are going, but often it makes more sense to start the lesson with a ‘big question’ to engage the students in an idea, with the learning intention introduced later.
Helping students understand what good writing looks like is best done by giving them examples at different levels of quality, rather than written descriptions, since the words teachers use often do not mean to students what they mean to teachers.
It’s also important that the examples we give students are felt to be achievable. If the standard is too high, students may give up in the belief that such work is beyond them.
When teachers teach their students one-to-one, finding out what students have learned is relatively straightforward – although the questions used to elicit this information should be carefully planned.
However, when the teacher is teaching a group of students, monitoring learning becomes much more complex. Teachers ask questions to determine what their students have learned, but in most classrooms, the answers come from the confident students.
The problem with this is that hearing from the same few willing volunteers gives the teacher very little insight into what is happening in the minds of the other students.
A policy of selecting students at random to answer questions will avoid the problem of hearing only from ‘the usual suspects’ – but even then, the teacher will have little idea about what most of the students are thinking.
To get a better idea of the learning of the whole class, teachers need to regularly use an ‘all-student response system’. Schools can use electronic voting systems (sometimes called ‘clickers’) to collect a response from every student, but the expense of such systems may not be justified.
Electronic voting systems will allow teachers to record every student’s response, but whether this is a good thing is far from clear. Indeed, the idea that every single incorrect answer one has ever given in a class is recorded in a spreadsheet forever is rather intimidating to many students.
This is why simple systems such mini-whiteboards, or ‘finger voting’ (where students hold up one finger for A, two for B, three for C and four for D) may in fact be better.
Once the teacher has scanned the whiteboards or the finger-votes, she can decide what to do to best meet the students’ learning needs, and any students who responded incorrectly know that there is no permanent record of their failure.
Teachers are bombarded with advice on how to give feedback, and much of this seems sensible. The problem is, little of what’s suggested is supported by the research evidence.
Teachers are told that feedback should be descriptive rather than evaluative, and this may be good advice much of the time. But there are also times when it is helpful for students to know exactly how they are doing.
Teachers are told to be positive rather than critical, but sometimes, it is more important to point out things that are going wrong.
The real issue here is what the student does with the feedback. As every teacher knows, giving the same feedback to two similar children can produce different results. It may cause one child to strive harder, and the other to give up.
Teachers need to know their students – they need to know when to push, and when to back off. And students need to trust their teachers to have their best interests at heart and to know what they are talking about.
Ultimately, good feedback is any feedback that gets students doing more of what we want them to do – and the most important factor here is the relationship between the students and the teacher. Get that right, and everything else is relatively unimportant.
There is now a substantial body of research that shows collaborative or cooperative learning can considerably improve student learning. The problem here is that the same research shows that collaborative learning is rather hard to get right.
Two conditions are particularly important. The first is that students have to be working as a group, rather than just in a group. The second condition is that students have to be mutually interdependent.
In other words, the group work has to be set up in such a way that the best efforts of every student are necessary for the group to succeed.
Typically, teachers set up group work in such a way that two or three members of the group can do all the work, in which case the benefits of collaborative learning are lost.
Ensuring that both of these conditions (and especially the second) are met takes time, and considerable teacher skill – but once students realise that no-one in the group can be successful unless everyone in the group is successful, student motivation will be increased, lower achievers will get more support, and high achievers will be forced to deepen their thinking when helping others.
As with the other strategies, the idea that students should be owners of their learning is attractive. In many primary schools, students are asked to ‘traffic light’ their work by indicating how confident they are that they have learned what they are meant to be learning with a red, yellow or green circle.
The problem with this is that some students are confident but wrong, while others are diffident but right. For self-assessment to be useful, it needs to be anchored so that students know what it means to understand something, and this is not always easy.
Rather than anchoring against external criteria, it can be particularly helpful for students to compare their work with their own previous work in the same area.
By getting students to avoid comparison with other students and instead examine whether their current work represents a personal best, students will see that however they compare with others, they themselves are making progress.
They will see that, as Jeffrey Howard says, “Smart is not something you just are, smart is something you can get”.
Finally, it’s worth remembering that no matter what you want your children to learn, and no matter what you think happens when learning takes place, formative assessment is an essential ingredient of good teaching.
If you believe that it’s important for children to learn facts by rote, you’ll need formative assessment to find out how much reinforcement and repetition is needed for students to commit what they learn to long-term memory.
If you believe in mastery learning, you’ll need formative assessment to find out which students might need more support to gain the required level of mastery.
If you believe that students construct their own knowledge, you’ll need formative assessment to find out whether they have constructed the correct ideas.
Students do not always learn what we teach, and good teaching starts from where the learner is. Formative assessment really is the bridge between teaching and learning.
Dylan Wiliam is emeritus professor of educational assessment at the UCL Institute of Education; for more information, visit dylanwiliam.org or follow @dylanwiliam.
Everything you need for every subject across Key Stages 1 and 2.