“Teacher-led observational assessment, properly supported and moderated, provides accurate and reliable information”
The current tensions around the nature, purpose and efficacy of assessment in the EYFS are nothing new. Given that what is assessed is an indicator of what we consider important and relevant, coupled with how we assess as an expression of our belief in how children demonstrate what they know and understand, this is hardly surprising. What is perplexing is that policy-makers stubbornly refuse to acknowledge the reality of young children, and how this relates to the process and content of assessment.
Part of this reality is, inconveniently, that there are generally significant cognitive, social, emotional and developmental differences between young children (up to the age of six) and children aged 11, 14 and 18.
Scientific evidence indicates that birth to six or seven is a specific and unique stage of development. It is also clear that this early childhood stage requires specialist knowledge and approaches to support and – in terms of summative assessment – represent this. Unfortunately, it is more convenient to ignore and subsume this reality into approaches and methodologies suitable for much older children.
This quest for convenience is precisely where the conflict resides.
Even the most cursory glance over recent history in this area provides ample supporting evidence of this. The transition from the original Baseline to the then Foundation Stage Profile in 2002 indicated the need for a single, teacher-led observational approach that accounted for all aspects of development and knowledge.
The reason for this was primarily the need for accurate and reliable data rather than statistical convenience. The previous range of available Baselines had included test-based models whose accuracy, regardless of its pedagogical usefulness or otherwise, was questionable.
The more recent Baseline debacle effectively re-ran the conflict, with the DfE accrediting six different choices for schools. Five of them consisted of narrow test- / task-based assessments and one – our own EExBA – provided a non-test / task-based, moderated teacher-derived assessment. Over 80% of schools that chose to use any of the accredited Baselines selected EExBA, prompting the DfE to revoke and ‘reconsider’ the policy…
The DfE’s enthusiastic (and unilateral) adoption of ‘Baby PISA’– a tablet-based test of specific skills – and its apparent intentions for the ‘reconsidered’ Baseline to consist of a similar online test of narrow aspects of learning – reinforces the disturbing tradition of ignoring reality for the sake of convenience.
There are deeper issues here about strategic understanding of early childhood education (ECE) and a denial that it is different, how this impacts on policy and expectations of practice. Much of this surfaced in our ‘Teaching Four & Five Year-olds: The Hundred Review of the Reception Year in England’ report (bit.ly/2fNAPwK), with one of its conclusions stating that
“1.2. The understanding of pedagogy and practice in YR and its uniqueness within a school environment is not always fully acknowledged at either national or local strategic levels. There is strong feeling amongst YR Teachers and Practitioners that pressures and tensions emanating from this lack of understanding can compromise effective YR practice and have a negative impact on outcomes for children.”
Inevitably this then has had the habit of affecting the understanding and approach to early years assessment, with the intention of ignoring what is known about young children, and in particular their understanding of the process, and the content of what is considered to be relevant.
In essence this returns to two of the basic principles of all assessment:
Considering the evidence provided by research into the impact of ECE on later outcomes leads to inconvenient but fairly clear conclusions. Studies cited in the research study to the Hundred Review indicated that Heckman, Wikart, Sylva et al (EPPSE) and Callanan et al (SEED) all strongly suggest that the critical factors in ECE relating to long-term development and success – and therefore what should constitute the content of assessment – are identifiable, although appear awkwardly within a wider view of assessment.
The awkward inconvenience of this reality is that what clearly matters to young children’s successful development, and is most likely to support and impact in their long-term development, is not purely the acquisition of specific knowledge but the all-round development of learning behaviours, dispositions and understandings – summarised in the EYFS’s ‘Characteristics of Effective Learning’ (CoEL).
Additionally, the areas of language and communication, physical development and personal, social and emotional development are also critical to long-term trajectories of success.
The reason this is awkward is that these key aspects can be difficult to fully assess, are possibly impossible to measure and don’t fit in a simple linear way with later measurable outcomes. Taking the central accountability of primary school outcomes in literacy and mathematics, it would be much more convenient if this were traceable to literacy and mathematical knowledge in the EYFS – especially with a Baseline policy.
Unfortunately this isn’t the case, and later outcomes in these two vitally important curriculum areas are, in ECE, equally dependant on the demonstration of securely embedded CoEL, language, physical development and personal/social/emotional skills in EYFS classes. Therefore any attempted assessment that avoids these areas, or focuses solely on literacy and mathematical knowledge, will be an unhelpfully partial view.
The idea of a ‘simple task-based assessment’ from which progress can be measured is a seductive solution. However, it belies our knowledge of the reality of children at this age, their perception and understanding of the world and how this manifests itself.
Testing – and by that I mean an assessment that relies on a response to a preset question to which there is a right answer – may or may not be an effective methodology for older children. It is quite clearly not for children in the early childhood age bracket.
When older children are subjected to a test scenario they are aware of how the process works. Most importantly they know there is a ‘right answer’ and the purpose of the test is to get as many right answers as possible. We learn the rituals of testing: what the ‘tester’ or ‘examiner’ wants us to do in order to get that right answer, how to ‘second guess’ what they are looking for and provide it for them.
Young children do not have the knowledge of the rituals of the test process, generally aren’t aware that there is a ‘right’ answer, and often respond to the questions with their own unconventional or creatively uninformed perspective, thus creating data that doesn’t necessarily demonstrate what they really know.
As long as policy-makers deflect reality because the nature of young children is untidy, unpredictable and takes time, skill, patience and reflection to ascertain, then these tensions and conflicts will continue. As long as they place statistical convenience above accurate, albeit challenging processes of assessment then there will always be a nagging antagonism between the necessary creation of accountability data and EYFS specialists and practitioners.
Although not convenient, effective assessment for accountability must include all the aspects that contribute to developmental trajectories and likely outcomes.
Although by no means a perfect process, teacher-led observational assessment, properly supported and moderated, does provide accurate and reliable information that can be used to effectively establish starting points from which accountability can be judged.