AfL and PMLD

The distinction between formative and summative assessment is long-standing but use of the term ‘assessment for learning’ (AfL) to describe the former derives from Black and Wiliam, 1990. They argue that changing what happens within the classroom is the most effective approach to improvement and that “assessment in education must, first and foremost, serve the purpose of supporting learning” (Black and Wiliam, 2006).

Amongst other assessment for learning strategies, they advocate the use of questioning to elicit evidence of learning. Unsurprisingly, Black and Wiliam are not focused on learners whose communication is pre-verbal and possibly pre-intentional and who may not even be cognitively aware yet that their actions can have specific effects. It would be easy therefore to dismiss AfL as not really applicable to learners with profound and multiple learning difficulties and those concerned with their learning.

In re-interpreting assessment for learning for the very specialised context of learners with PMLD, a talk given by Wiliam at a Cambridge Assessment Network Conference is perhaps more instructive. Here he makes it clear that assessment for learning is fundamentally about using “evidence of student learning to adapt teaching and learning in order to meet student needs”. He points out that there are “different timescales” for this and concludes that:

“If you’re not using information to make a difference to your teaching within a day or two then it’s unlikely to make a difference to student achievement. It’s the short cycle formative assessment that really matters, minute by minute, and day by day” (Wiliam 2006).

This seems much closer to the world of learners with PMLD and the practitioners who work with them, because acting ‘minute by minute’ or even ‘second by second’ is fundamental to the human interaction which lies at the heart of early learning and teaching of learners with PMLD.

AfL1As William points out, “learners create learning. Teachers create the conditions in which students learn”.  So, if they are to elicit evidence of learning teachers need to use everything they already know about the conditions that learners with PMLD in general, and any learner in particular, might require in order to access learning in the first place. They will use this evidence to structure conditions – to create and then maintain what Ware (2003) refers to as ‘a responsive environment’.


AfL provides the means to refine conditions continuously, by incorporating evidence as it emerges, including about what motivates the learner.

Simple but effective recording ensures that evidence is not lost and can have an impact in the future.



ZPDAfL is transacted within Vygotsky’s ‘zone of proximal development’. The teacher works to elicit, maintain and enhance the learner’s responses, continuously monitoring and adjusting the conditions so as to optimise support and recognising when such support can be reduced or withdrawn.






AfL means being alert to learner responses that are new or interesting, even if they are tentative or puzzling.  When a response appears for the first time, seeking to elicit its repetition (by re-presenting the stimulus) is a way to test the evidence. Leaving a gap between re-presentations (a ‘burst-pause’ pattern) will strengthen the learner’s ability to register what is happening  These strategies are close analogues to ‘questioning to elicit evidence of learning’ albeit within a pre-verbal context.



AfL involves professional reflection both within and after the moment.  What does the learner’s behaviour appear to indicate?   (How) can it be elicited again?   What were the conditions which supported it? Would changing anything enhance it further?  Does it say anything of more general significance?  All of these are relevant thoughts to ponder for everyone in the teaching team, especially the teacher.


Burst-pause, stimulus-response – these are the simplest patterns associated with very early learning. Using AfL, teachers can seek to extend and build these ‘contingencies’ into routines whose pattern learners come to recognise and through which they might begin to develop anticipation.

Routines are fundamentally ‘social’.  When learning is involved both partners share in the interaction, working together to take it forward. Sometimes, it is primarily a social ‘game’ – with no specific goal other than being human with another human being.  At other times, the interaction may be more specifically focused on making something happen out in the environment (like obtaining that motivating toy).

In the beginning, the focus may not necessarily be clear to a learner with PMLD, but in re-visiting the routine the teacher can ‘big up’ the aspects that matter and downgrade those which are distractors in that context.  We could say this is about ‘clarifying learning intentions’ – one of Wiliam’s ‘five key strategies’ for AfL.

Observing, reflecting upon and interpreting the (potential) meaning of learner responses is fundamental to AfL.  At the earliest developmental stages, the ‘Affective Communication Assessment’ (Coupe O’ Kane & Goldbart, 1998) is a tool which can support this process by refining the way in which observations are recorded and meanings arrived at.  AfL is about ‘adapting teaching and learning’. So when, for instance, the teaching team believe a learner’s response might mean ‘I like that’, they could decide to respond in accordance with that interpretation when it re-appears in the future and use the routine in which it arose to teach that specific communicative response.

Interpreting the meaning of learner responses requires that we see how they fit relative to some framework which represents progression. During the short history of education for learners with PMLD (ie. since 1971), the framework of choice has typically been ‘the curriculum’ conceived as a hierarchy of sub-skills.  This was based in the early days on a behaviourist interpretation of social competence and was more recently supposed to lead towards a top-down subject-based set of national expectations.  This approach does not ultimately assist assessment for learning since learning is valued only to the extent that it conforms with these narrow views of ‘curriculum’ and the framework is used like a ruler – placed alongside the response, we focus on measuring how many units (or ‘Levels’) it equates to. In other words, Assessment for learning (in the learner’s own terms) gives way to assessment of learning (measuring what value that learner’s learning has in relation to other learners).

An appropriate framework for learners with PMLD is one which disentangles progression from prescribed curriculum content, is centred on the learning and development of individuals and accommodates diversity of need and outcome. In this approach, assessment for learning interprets evidence in relation to developmental landmarks which are understood to lie along multiple learning pathways and whose potential can be explored through a broad range of curricular opportunities. Here the analogy is to trying to find where the learner is by looking for those landmarks on a map.  On the map, a general direction of travel (e.g. top to bottom) indicates progression, but the landmarks are not equidistant (hence I have avoided referring to them as ‘milestones’). Whilst it is possible to consider the distance the individual travels over time and the landmarks their learning has encompassed, the importance of the framework (the map) lies in its ability to enlighten, inform and give direction to learning.

According to Black and Wiliam:

“The distinction between assessment of learning and assessment for learning is basically about the intention behind the assessment.  So, if you’re assessing in order to help you teach better, that’s assessment for learning, and if you’re assessing in order to grade students, to rank them or to give them a score on a test, then that’s assessment of learning”.

Within special education, we seem all too often to confuse the two.


Back: Fair Assessment: principles                                                                           Next: Evidence