The future of Assessment for Learning under Michael Gove


As we move towards more terminal examination at GCSE and a system which prizes knowledge over skills, Tom Middlehurst argues that Assessment for Learning has become more essential to classroom practice than ever.

Assessment for Learning (AfL) has for a long time been seen as the foundation of good classroom practice; indeed as the very business of teaching and learning. 

That teachers should provide their students with excellent formative comments has become a given. However, with the government’s proposals to increase terminal assessment at GCSE, truly embedded formative assessment is needed now more than ever.

There is nothing particularly new or innovative about AfL, rather AfL or formative assessment should be seen as a fundamental principle of good teaching.

Essentially, formative assessment is concerned with the constant and persistent monitoring of student progress by analysing data, using this as a vehicle by which to open up lines of dialogue with learners about where they are and – crucially – what they need to do next in order to progress. In practice, this process manifests itself in any number of ways within the classroom, including questioning, marking and feedback, and use of success criteria.

Again, none of this is new, and surely forms the basis of any conversation about learning between learner and teacher. However, when Professors Paul Black and Dylan Wiliam wrote their seminal work on AfL in 1998, Inside the Black Box: Raising standards through classroom assessment, they compared these different manifestations of AfL happening inside the classroom to things happening within a “black box”, where certain demands were thrown into the classroom from the outside and teachers were left to meet these demands on their own. While they highlighted this as a national problem 15 years ago, their observations reflect my experience of working in an academy in 2012.

Profs Black and Wiliam comment that certain inputs are fed into the classroom when management, students, and parents require test scores and grades, and that hopefully some outputs will be returned, but that no-one really knows what goes on within the classroom. 

So too, in my experience, there was a strong academy-wide drive for data collection, which required bi-termly inputting of grades and levels onto the intranet. However, there was little if no clarity about the ways I should have been preparing students for this summative assessment, marking these summative assessments, or indeed giving feedback to the students about their learning.

The focus on AfL which followed the previous government’s Assessment for Learning Strategy in 2008 highlighted the real need for continual dialogue between learners and other stakeholders. Despite this, within my own academy last year they had yet to formalise a policy towards this.

The AfL Strategy was a “significant part of the government’s commitment to developing personalised learning and to improving rates of progression”. Indeed, the very quintessence of what the government said in 2008 suggests an absence of summative assessment in favour of a formative approach, or rather a “bi-directional process between teacher and student to enhance, recognise and respond to the learning”, as is suggested in A Model of Formative Assessment in Science Education (Cowie & Bell, 1999).

How classroom teachers are to continue to embed this good practice in the new educational climate is a challenge which must be faced head-on. Arguably, practitioners should continually remind themselves of the principles of formative assessment, which need not be sacrificed regardless of government policy.

In Developing Formative Assessment in the Classroom (2001), Torrance and Pryor recognise two approaches – divergent and convergent. The first understands assessment as an “interactive and social component of teaching and learning”, the second as a “traditional” view of assessment which assesses a child’s knowledge.

This latter form of, usually summative, assessment is seen as potentially being a “judgemental evaluation which runs the risk of making weaker pupils feel stigmatised”. The authors suggest that convergent assessment includes closed questioning and setting closed tasks, taking a linear view of the curriculum and making quantitative or judgemental evaluations.

On the other hand, divergent assessment focuses on what the learner already knows and understands, which supports the understanding of those theorists who had the greatest influence on my own practice as a teacher.

Moreover, Profs Black and Wiliam do not see formative assessment as merely coming from what the learner already understands, but it is also “appropriate in all situations and helps to identify the next steps to build on the successes and strengths as well as to correct weaknesses”. In other words, assessment should not just be of learning (regardless of whether that is teacher or learner-led), but should empower students to progress further.

This is something which rang true in my classroom, particularly with regards to peer-assessment. I noticed that students were unable to apply the marking criteria to their partner’s work, despite heavily scaffolded tasks designed to reword the criteria in students’ own language. The peer-set targets for progression included several comments such as “write neater” and “include more information” which were ultimately irrelevant and vague respectively. 

I realised that my students were so used to being “spoon-fed” any information they received, including areas of progression, that by key stage 4 they were left without the cognitive tools which would allow them to do this for themselves.

In terms of cognitive AfL, this presented me with a problem given Profs Black and Wiliam advocate a Vygotskian theory of “zone of proximal development” in which learners move from a zone of high cognitive support to a zone of low cognitive support, where they are able to achieve tasks with limited intervention. My experience showed that my students, despite having had the task scaffolded through the lesson, were unable to make appropriate peer-assessment independently.

However, it has also been suggested that cognitive scaffolding does not work; Professor Paul Kirschner’s view suggests that if one accepts that learning is defined as a change in long-term memory and that the human cognitive architecture is formed of both long-term and short-term memory “where the long-term incorporates a massive knowledge base that is central to all of our cognitively based activities”, and thus independent learning is ineffective. 

Due to our human cognitive architecture, the notion of independent learning is irreconcilable; my key stage 4 students did not have the “cognitive architecture” to peer and self-assess.

While I agree with Prof Kirschner’s assertion to an extent, the success of previous enquiry-based activities based around constructivist theory (from my wider teaching experience) encouraged me to continue with a constructivist approach towards assessment.

I realised that the difference between other independent tasks and this assessment was that students were coming to this from a completely blank slate, never having been exposed to marking criteria previously. This is a crucial caveat to Vygotsky’s theory of constructivism. He argues that the “complex and genuine act of thought” (since known more commonly as “active learning”) occurs only when learners are at an “appropriate stage”.

With an education system that is in danger of moving further away from constructivist ideals to one that favours knowledge over experiential and skills-based learning, it is more important than ever to embed AfL at every level of school life.

  • Tom Middlehurst previously worked as an English teacher at an academy in Essex. He is now programme co-ordinator of student impact at SSAT.

Further reading
Inside the Black Box: Raising Standards Through Classroom Assessment:


Please view our Terms and Conditions before leaving a comment.

Change the CAPTCHA codeSpeak the CAPTCHA code
Sign up SecEd Bulletin