Effective Assessment in EYFS: How Do We Know That They Know What They Know?
Revisiting the 80/20 ratio and understanding effective and authentic assessment judgements in the EYFS
The accuracy of data, and a collective confidence in this, is the single most defining factor of its purpose. Without the knowledge and belief that the information is reliable and truly reflects children’s attainment (and therefore achievement) such a data set is simply a random collection of numbers.
In considering assessment in the EYFS, this is acutely so, as this forms the ‘entry level’ of information that potentially plots a lifetime’s trajectory, so ascertaining its accuracy becomes vitally important.
How we truly understand and, confidently assert that a child ‘knows’ or ‘understands’ a concept or knowledge has always been contentious and vexed. Attempts to ‘test’ children in diluted versions of systems and approaches that are successfully provided for older children do not recognise the unpredictability of young children’s perceptions and responses nor the fact that they are (understandably) not usually encultured into the ethnography of testing as ‘finding the right answer’.
Such approaches with young children are highly problematic and unsuccessful, and, unless the test criteria is so narrow as to be severely limited, do not provide data that engenders confidence, accurately reflects children’s attainment or is usable to support progression.
The tradition in the EYFS, and its philosophical and statutory predecessors, is to use a form of ‘observation’ by which the child is carefully watched by the adult, usually in a self-led situation, and, from this the adult makes a judgement about the nature and level of the knowledge and understanding that has been demonstrated; typically against a criteria which is attained or not. The adult might prompt the child with a carefully judged question, or even set up a particular resource to tease out a specific aspect of learning, but the key principle is that this is achieved in a ‘naturalistic’ context with the child unaware that an assessment is taking place. The statutory EYFS Profile (and its previous iterations) is embedded in this principal.
While this approach is not without its challenges; subconscious bias, the ‘halo’ effect, and variables of how many observations constitute a judgement, it is generally accepted as an appropriate and trustworthy approach; although effective training and rigorous moderation are prerequisites for its success. However, even within this broad consensus the issue of ‘truly knowing’ a child’s attainment remains.
During my tenure as the National Lead for the EYFS Profile, I was responsible for creating and implementing the (now somewhat infamous) 80/20 ratio. As part of assessing for the statutory EYFS Profile, it stated that:
“When making a judgement for the EYFS Profile, practitioners should draw at least 80 per cent of evidence from knowledge of the child, observations and anecdotal assessments, and no more than 20 per cent of evidence from adult led and adult direct activity assessments” (EYFS Profile Handbook 2008)
Despite its subsequent misinterpretation and misuse, the rationale that underpinned this was clear and considered. If a child only demonstrated a knowledge, skill or understanding when prompted or instructed to do so by an adult in an adult led context, then this was not sufficiently embedded enough in the child’s achieved learning for them to use it confidently and independently. In short they might ‘possess’ the knowledge, but were unable – or unwilling – to apply and use it. However, if a child did use this in a self-led situation and was able to apply it to their everyday activity, then clearly this was significantly and sufficiently embedded to become part of their secure lexicon of knowledge skills and understanding. The purpose of this was to ensure and generate accurate and valid assessment, that, in this case, constituted a national data set for attainment at the end of the EYFS. This was clarified in documentation and formed an important ‘lens’ during moderation to ensure the judgements truly reflected a child’s attainment.
In the intervening years since this principal was boldly established within the culture of EYFS assessment, our understanding of how children represent their learning and how nuanced this can be has, as one might expect, grown, developed and evolved. Although the principle remains an important one – children need to be able to ‘use’ the knowledge and skills they have been taught, I do believe that the time has come to refine the model and accept some of the nuances and additional considerations that were not necessarily acknowledged in the original statutory requirement.
- Different aspects of the EYFS’s Learning and Development Requirements might need different treatment.
The Prime Areas of Learning and Development; Communication and Language, Personal, Social and Emotional Development and Physical Development are ‘constants’ that exist in most if not all activities that children undertake. Therefore they are eminently visible and therefore assessable. Equally so with the Characteristics of Effective Teaching and Learning; these critically important learning behaviours are also present in much of what children do, especially when it is self-led. However, the Specific Areas of Learning and Development; Literacy, Mathematics, Understanding the World and Expressive arts and design, are more concerned with ‘bodies of knowledge’ and particular skills that are not always part of the everyday. Therefore, seeing these consistently in a purely self-led context might prove challenging and provide minimal material with which to make a judgement. Therefore, do we need to accept that some aspects will invariably present themselves, while others will need more overt ‘construction’ and scaffolding in order for accurate information to be gleaned, and that a more responsive and professionally considered ‘hybrid’ approach might be required?
- Although self-led episodes of learning are the ideal way of demonstrating what is ‘known’, the binary principle implicit in the 80/20 ratio needs to take account of more nuance. There is a place in between, where a child may ‘know’ something but chooses not to demonstrate this independently. Although this may suggest that it is not yet sufficiently internalised to be used confidently, a knowledge and / or skill is nevertheless present and, for the sake of accuracy, needs to be acknowledged. Especially in terms of the Specific Areas of Learning and Development (as mentioned above) perhaps judgements need to include an additional strand to indicate this. Therefore we need to acknowledge and develop a three-way strata of judgement; an ‘entry level’ where the learning is not secure, a level where the learning is present following an adult prompt or direction, and a level where the learning is demonstrated independently.
Learning Ladders has launched a new EYFS module specifically for International Schools, designed in collaboration with Jan Dubiel and leading Independent International schools. Find out more and book a consultation here.
(Updated from an original blog written January 2021)