Ssince the 1980s, we have seen a massive increase in testing. This emphasis on assessment of learning has deflected attention away from another important activity - assessment for learning. Put simply, we have been putting too much effort into target-setting and not enough into target-getting. Of course we need to set targets for schools, teachers and learners, but that is the easy part; the hard bit is achieving them.
Our 1998 review of 600 research studies worldwide, Inside the Black Box, showed quite clearly that the best way to achieve higher test standards appears to be to improve the use that teachers make of their day-to-day assessments of students. To see how these benefits might be achieved, we worked initially with 24 (later 36) teachers in six schools in England over 18 months. We found that improving formative assessment raised GCSE scores by more than half a grade per student per subject.
As they worked with the ideas from the research, the teachers came to change four aspects of their work: question-and-answer dialogues in class, marking homework, peer-and self-assessment by pupils, and involving pupils in setting and marking their tests. All four share two features that are fundamental to effective teaching and learning: they give students feedback they can use and they involve students in their own learning.
The most common form of feedback is to give students a mark, often with a comment. Research has shown that marking is better than giving no feedback at all, but giving comments produces substantial improvements. What is surprising is that giving marks and comments together produces no improvement. When students get both a mark and a comment, they first look at their own mark and then at their neighbour's. They hardly ever read the comments. So teachers who craft helpful comments are wasting their time if they also give a mark. Of course, because writing helpful comments that tell students what they need to do to improve takes time, it may follow that teachers may not be able to mark all of a student's work. What is important, however, is that when teachers do give feedback, they ensure that students act on it. At present, a teacher will typically spend more time marking a student's work than the student will spend following it up - which suggests that the teacher's time is less valuable! Of course, students do need to be given some sort of feedback about how well they are doing in terms of marks, grades or levels, but this need not be more than three or four times a year.
Teachers often ask us: "What will the parents think?" We have found that parents' reactions are usually positive. As one said: "When my child got four out of ten, I didn't know what to do, but with these comments, I know exactly how to help him."
Getting students to comment on each other's work can have a substantial impact on learning for two reasons. First, students are much better at spotting weaknesses in other people's work than in their own. Second, students are much tougher on each other than any teacher would dare to be. And they can take criticism from their classmates much more easily than they can from the teacher.
This leads naturally to students teaching each other. Students can often communicate far more effectively with classmates than the teacher can. More importantly, most children pretend they have understood an idea after the second explanation from a teacher, even if they have no idea what is going on. Some do this because they feel they should not monopolise a busy person's time, while others want to avoid looking stupid in front of the teacher. When they are explaining things to each other, however, they do not pretend. They interrupt when they have not grasped something, and ask for repeated explanations until they have understood. In many situations, therefore, peer-assessment and peer-teaching are not poor substitutes for improved pupil-teacher ratios, but a powerful way to promote learning.
If we are serious about raising standards then, instead of telling students that they are successful or unsuccessful, which is what happens when grades and marks are given out, we need to focus our feedback on issues that they can do something about. Just focusing on things like "effort" is not enough. Telling students that they need to "try harder" is no better than telling a bad comedian that he needs to be funnier. Our feedback must tell students not just what needs to be improved, but also how to go about it - one of the most common "pet hates" among students is feedback such as "Be more systematic". As one student commented: "If I knew how to be more systematic, I'd have done it first time round!"
We also need to involve students much more in their own learning. One unfortunate result of the drive to raise standards is that teachers have been working too hard in the classrooms and students have not been working hard enough. Teachers are trying to do the learning for the students, whereas research shows that no one can do your learning for you. To some teachers, involving students more in their own learning feels like a loss of control - our teachers said as much at the start of our project. It was scary. By the end, however, they described the same process as one of sharing responsibility. Perhaps even more important than the improved results was that teachers were enjoying their job more and the students were enjoying learning more. Teaching well is compatible with getting better results. Giving frequent marks is not.
Dylan Wiliam is assistant principal and professor of educational assessment at King's College, LondonPaul Black is professor emeritus of science education at King's 'Inside the black box' and 'Working inside the black box' cost pound;3, from: Department of Education and Professional Studies, King's College, Franklin-Wilkins Building, 150 Stamford Street, London SE1 9NN. Tel 020 7836 5454 ext 3189 or go to King's Formative Assessment Project; www.kcl.ac. ukdepstaeducation.
Click on "The research we do" and then go to the Assessment for Learning Group.