Addendum to A Primary Assessment Primer

After a great discussion at tonight’s NUT Section general meeting on primary assessment, based upon the previous post, I thought I would explore in greater depth a few issues that arose, plus a few that I left out of the original piece.

Developmental differences in Reception

First of all, it was noted that, at the beginning of the academic year, summer-born children are almost a full year younger than those with birthdays in September. Proportionally, this means that children entering Reception may only have had 80% of the life experience of their peers (i.e. comparing a 4-year-old to a 5-year-old). This is one of the key factors affecting the reliability of Baseline tests: how can we meaningfully assess the development of children through a standardised test when many children have had significantly less time to develop? Quite apart from the fact that the child development is inherently uneven across children, this just does not make sense as a proposed metric.

Assumptions about the acquisition of knowledge

Much of standardised testing, particularly at the primary level, makes assumptions about the acquisition of knowledge that are not in line with reality. The kind of assessment we have been discussing assumes that knowledge is acquired linearly, and thus that children can be graded along a fixed path of knowledge which they should have acquired by a certain point in their schooling. Unfortunately, real life does not work in this way. The acquisition of knowledge is a complex and many-faceted process, affected by many more factors than simply what has been taught and what has been remembered, and thus assessment that is based on this kind of assumption is not useful as a tool with which to measure children’s development, and especially not as one with which to decide the fate of teachers and schools nationwide.

What is ‘progress’?

The previous section raises, then, the following question: what is ‘progress’? Assessment is entirely focussed on measuring progress, ensuring that children are making progress, and judging teaching and school effectiveness on the amount of progress being made. Sadly, this idea of ‘progress’ is not well-defined. Again, child development is a complex set of processes that is entirely unique to each child, and what may constitute an important educational step for one child may not for another. Hence, a system constructed on oversimplified numerical measures of ‘progress’ (that is to say, a value-added score based on performance in standardised tests) is meaningless as a means of actually charting pupils’ development. In fact, most, if not all, teachers and teaching assistants are very much aware of the real-terms progress that each child has made through their daily interactions with them, rather than through performance on a test, or against an arbitrary list of criteria. Can little Alex use a relative clause yet? Well, no, but in the last six months she’s learnt how to form all of her letters properly, can now write in sentences, has written some of her most imaginitive stories yet, and has overcome her separation anxiety issues in coming to school. But that, seemingly, is not ‘progress’, at least not in any manner that would look palatable to Ofsted.

Teacher assessment

From the teacher’s perspective, such a flawed system of assessment is also problematic as it exists to assess not only the pupils, but also the teachers (and, to a large extent, the school). With the advent of performance-related pay, many teachers are required to meet targets related to children’s performance at assessment, and ‘progress’ as evidenced by this assessment. Unfortunately, as discussed above, the entire model of assessment and ‘progress’ is based upon flawed premises, and thus the system becomes yet another stick with which to beat the teacher, or, in the case of judgments made by Ofsted, the school. In a climate where the Government actively favours forced academisation, we cannot afford to be subject to flawed measures of school effectiveness.

Assessment and social class

The class nature of assessment is also something that should not be ignored. Of course, socioeconomic background has a significant impact on performance in tests, but more insidiously than that, the way in which assessment data is used inherently classist. Assessment data is already used to construct league tables of schools, and this will only be exacerbated by the introduction of Baseline testing. League tables, as mentioned in the speech in the previous post, encourage ‘free-market’ competition between schools, with consumers (i.e. parents and carers) hypothetically able to take their business elsewhere should a school not meet their expectations. The idea is then that this will eliminate underperformance in schools in a somewhat Darwinian fashion. In reality, such a process is inherently discriminatory against the less well-off, as those can afford to move to better-performing schools leave behind those who can’t, concentrating less fortunate families in underperforming schools. This effectively leads to quality of provision being determined by socioeconomic background, or, to put it much more bluntly, those who need high-quality education provision the most become those who receive it the least. Rather than encouraging all schools to improve, it entails the ghettoisation of public education, and is, as such, inherently classist.


Finally, a very short note to say that the trend towards a neoliberal, business-oriented model of education is very much part of the Global Education Reform Movement (or GERM) which has plagued public education systems across the world for the last several decades. I won’t go into depth about it here, as an explanation of the GERM and its effects will form the basis of an upcoming blog post, so stay tuned!