No one reading this article will be in any doubt as to the importance of high-quality formative assessment in teaching and learning. The term, often used interchangeably with “assessment for learning” and in contrast to “summative assessment”, refers to assessment activities undertaken by teachers – or students themselves – to obtain evidence which is then used to adapt teaching and learning methods to meet student needs and improve learning outcomes.
However, just because formative assessment doesn’t involve the high stakes with which its summative counterpart is often associated doesn’t mean it’s easy to do well. Indeed, previous research has highlighted not just that attempts to support teachers in developing their formative assessment skills at significant scale are difficult, but that they can backfire if done poorly.
As part of research funded by the Education Endowment Foundation, a team at the National Institute of Social and Economics Research and I worked with SSAT – the Schools, Students and Teachers Network – to carry out a randomised controlled trial of the Embedding Formative Assessment (EFA) programme. Embedding Formative Assessment is a professional development programme for schools, developed with educationalists Dylan Wiliam and Siobhan Leahy, that aims to support teachers to embed formative assessment strategies in their practice to improve pupil learning outcomes.
The programme itself consists of light-touch training and support, mostly focused on training and supporting school leaders and teachers who then go on to lead teaching and learning communities (TLCs) within schools. Most schools arrange these TLCs during time already dedicated to professional development, and each of their meetings involves teachers reporting on their use of techniques since the last meeting, sharing new formative assessment ideas to try, and personal action planning for the coming month. Teachers are also asked to pair themselves for monthly peer lesson observations between TLC meetings.
To look at the impact of schools implementing EFA, 140 secondary schools across England were recruited into the project, with 70 allocated to receive EFA, while 70 were allocated to receive a substitute payment, which they could put to any purpose (except EFA). This allocation was carried out at random so that the only systematic difference between the two groups was whether or not they were implementing EFA. After two years of implementation, we compared pupils’ GCSE test scores between the two groups of schools, providing us with a highly relevant outcome measure.
We were extremely pleased to find that schools that implemented EFA saw higher pupil attainment than those in the control group. This was particularly true for the group of schools where monitoring work suggested the biggest changes in practice had taken place.
The results are really encouraging for this approach to improving teachers’ implementation of formative assessment in the classroom and, hence, pupils’ academic attainment. The size of the difference between the EFA and control groups is equivalent to almost two months of pupil learning: especially worthwhile given the post-Covid catch-up context and the fact that this is a low-cost, scalable programme. The size of the effect was also slightly larger among pupils with lower levels of prior attainment, suggesting the potential for this work to narrow the distribution of test scores.
The Education Endowment Foundation are now supporting further work to scale up the delivery of Embedding Formative Assessment as part of their accelerator fund. This new project is focusing particularly on schools in the regional schools commissioner regions of the North, the East Midlands and Humber, and the West Midlands.
While we’re not directly involved, we’d nevertheless encourage schools to take part in this important research. It stands to add powerfully to our understanding of how we can scale up programmes and practice like this across schools throughout the country – not to mention offering immediate benefits for teachers and pupils.