PhD research exploring Primary School assessment reform of the 2010 Coalition Government, showed preferential attitudes towards statutory test outcomes above teachers’ own assessment of their students. This blog article considers the place of teacher assessment in primary statutory assessment drawing on views of teachers themselves and members of the senior leadership team in two Primary Schools in the West Midlands. Both schools found the assessment reform difficult particularly the lack of guidance and support provided by the Government. Troublingly, the focus on statutory testing to monitor Primary schools has persisted, emphasising learning which can be tested objectively and in some cases erasing the notion than learning can occur if it cannot be measured.
History of SATs
For those who have young children in schools and those still young enough to remember it, Statutory Assessment Tasks (SATs) are a major event in Primary Schools. They happen at the end of Key Stage 1 (6-7 years old) and Key Stage 2 (10-11 years old).
In the early 1990s, SATs were teacher assessed tasks judged by the teacher, and not an externally imposed test. However, it was not long until SATs were adapted to resemble the formal tests we see today. This is still controversial, exposing children as young as 7 to formal testing, particularly when child mental health is in crisis. While KS1 SATs allows the inclusion of teacher assessments, KS2 SATs only requires teacher assessment for writing; while reading, maths and grammar are judged exclusively on formal timed tests. In 2018 the requirement to submit any teacher assessment in KS2 for tested subjects was removed. This raises an important question- what motivation is there for schools to invest time and resources in KS2 teacher assessment when it is no longer formally required and when statutory testing is the measure of school success?
Research and commissioned reports on assessment (Bew, 2011) strongly support the inclusion of teacher summative (systematic and planned collection of evidence of learning to judge what is learnt over a time) judgments in primary education (Black et al., 2011; Harlen, 2005). It enables wider sampling of the curriculum and opportunity to explore student learning in different formats outside of reading and writing.
"Teachers know their children, they know whether something is a one-off or whether the children understand things…whereas in the test it’s just what that child could do in an hour on that day"
"…there are certain skills that can’t be tested with SATs and that’s where teacher assessment comes in…"
Why has teacher assessment been side-tracked historically for statutory testing? This comes down to objectivity- that learning should be observable and measurable within defined parameters, independent of individual perspectives and interpretations. Teacher assessment by its nature is dependent on teacher interpretations and thus has been criticised for lacking objectivity (Johnson, 2013). Another perspective on this is that teacher assessment is bias, consciously or unconsciously because teachers know their students well. But this essentially is what makes teacher assessment effective in capturing learning testing cannot.
Lack of trust
Evidently, this research showed teachers shared these concerns and displayed a lack of confidence and trust in their own assessment despite believing the benefits outweighed the downsides of testing on learning. Any differences between teacher assessment and SATs was seen as a fault with the teacher assessment rather than the test.
"…standardised testing. It’s the only reliable measure and we know that because of the huge disparity between [it and] teacher assessment"
Tests were seen as reliable and objective, therefore, were felt to be better for measuring end of KS2 than teacher assessment. Teachers feeling this way may have resulted from not feeling trusted to assess.
"…I suppose, teachers are now not allowed to just know that they [students] can do this and this. You’ve got to prove they can…"
A need for training
It is important to note that while schools are ranked on league tables with KS2 SATs results, research does not necessarily support the use of teacher assessment in the place of SATs. Teacher assessment is already under resourced with a lack of training and experience for teachers to produce more trustworthy assessments (Collins et al., 2010). This is not a criticism of teachers but an analysis based on many years of statutory assessments de-skilling teachers. This de-skilling is the result of KS2 SATs assessments measuring not only student attainment, but also school quality, holding schools to account with these results. For this to be possible, student attainment data requires standardisation, so it can be analysed and interpreted on mass. This makes objectivity a priority above the scope of learning an assessment can capture. Due to the subjectivity of teacher summative assessment, it would be a huge task to do this, therefore, tests are better suited to this job. The results are interpreted to represent learning on all of the curriculum content based on a representative sample. But in practice what this research demonstrates is that results are skewed largely towards learning which is tested, and can be tested with KS2 SATs and not attainment from the curriculum as a whole.
"…for those children who we were worried about…in want of a better word, [we] narrowed the curriculum that we taught them because we knew that that was the basics basically that they needed to be able to cover in year six"
Reallocation of teaching
Using standardised tests does encourage useful practices such as - working harder, teaching more and working more efficiently (Koretz, 2008). However, it also encourages practices of test coaching, reallocation of teaching time to tested subjects and content, and, in the extreme cheating. Participants discussed reallocation of teaching time to tested content and targeted preparation of testing through the year.
"…some children at the very, very low ability were drilled on a certain number of objectives which meant that they could, by the skin of their teeth, pass the test"
One reason why teacher assessment is recommended as part of end of KS2 statutory assessment is to mitigate practices such as teaching to the test which undermines the interpretation of the test as sampling the whole curriculum.
The test itself is not at fault, rather the pressure placed on KS2 SATs as an objective measure of learning, and therefore an indicator of school success. The absence of any detailed non-statutory assessment guidelines exemplifying curriculum assessment standards, compounded pressure to teach to the SATs tests as there was no other yard stick for the schools to anticipate and predict student performance. Such pressure on the SATs as a way of demonstrating school quality essentially resulted, for these teachers, in undermining the very learning they were trying to achieve.
Predicting SAT scores
For one school in the study, there was an added conflict from an expectation that their own assessments should predict and match the SATs test outcomes. One teacher referred to it as luck that their assessment and the SATs test result matched. Another suggested that their teacher assessment was at fault for not anticipating that a student would struggle under test conditions.
"I think teacher assessment should have been more spot-on but it wasn’t. So, those children didn’t perform very well"
Comments from senior leadership echoed this expectation that teacher assessment should indicate the test result, and, where this differed greatly the teacher assessment was seen to be at fault.
"…[with] our year six teacher assessments this year...children were expected to be in the 90s…But in the test situation it was a very different picture."
"I still would say we’ve still got work to do with all our teachers. A lot of it is going back to confidence in making sure that things are independent…It’s getting that balance right and making sure the teachers are really clear about, how much evidence do we need to say that the children are secure"
The assumption here from the SLT was that SATs assess learning objectively compared to teacher assessment, therefore, to improve teacher assessment it must resemble the independence required in a test. Essentially this is standardising non-statutory teacher assessment against test expectations- an influence of objective views on assessment. Yet, educational assessment research disputes that teacher assessment should be compared to test results as they are both, as Harlen (2005) states ‘importantly different’. Both schools involved in the research were moving towards using termly privately purchased tests to inform teacher assessment, which mirrored SATs format and scaled marking (grade boundaries adjusted to reflect national cohort achievement each year). Teacher assessment in these schools would largely remain for formative purposes- to evaluate and inform future teaching. This is a valuable and essential part of learning, but would not maintain any standing of summative assess learning against the curriculum in KS2.
When A Level results were downgraded from previous predicted grades based on their teacher’s assessment, unfortunately I was not shocked. I had already seen in my own research that faith in teacher assessment in KS2 was low. Testing dominated the picture with both schools increasing their use of tests in the absence of non-statutory assessment frameworks, and less attention paid to teacher assessment by the DfE. This research hit upon two troubling findings related to this. The first that, in pursuit of an objective measure, preferential views of testing above teacher assessment has devalued and potentially de-skilled teachers in assessment of their students, creating an atmosphere of distrust, even from the teachers themselves. The second is that in continuing to only use statutory testing to judge KS2 pupil attainment, and thus, the quality of the school, as Biesta (2017) says, we need to consider whether we are ’…measuring what we value, or…valuing what is being measured.’ (p.316), as one teacher commented;
"…you’re coming down to test technique rather than actually teaching them"
These are two very different things with the latter continuing a worrying trend of teaching to the test, at the expense of the very learning tests are assumed to capture.
- Bew, P., 2011. Review of Key Stage 2 testing, assessment and accountability: Progress Report. London: Department for Education
- Biesta, G. (2017). Education, Measurement and the Professions: Reclaiming a space for democratic professionality in education. Educational Philosophy and Theory, 49(4), 315-330.
- Black, P; Harrison, C; Hodgen, J; Marshall, B; & Serret, N. (2011): Can teachers’ summative assessments produce dependable results and also enhance classroom learning? Assessment in Education: Principles, Policy & Practice, 18:4, 451-469
- Collins, S; Reiss, M & Stobart, G. (2010). What happens when high‐stakes testing stops? Teachers' perceptions of the impact of compulsory national testing in science of 11‐year‐olds in England and its abolition in Wales, Assessment in Education: Principles, Policy & Practice, 17:3, 273-286,
- Harlen, W. (2005) Trusting teachers’ judgement: research evidence of the reliability and validity of teachers’ assessment used for summative purposes, Research Papers in Education, 20:3, 245-270,
- Hurn, A,. (2020) Ofqual's A-level algorithm: why did it fail to make the grade? The Guardian [accessed 15/10/2020 https://www.theguardian.com/education/2020/aug/21/ofqual-exams-algorithm-why-did-it-fail-make-grade-a-levels
- Johnson, S. (2013) On the reliability of high-stakes teacher assessment, Research Papers in Education, 28:1, 91-105,
- Koretz, D.M., 2008. Measuring up. Harvard University Press.