There is no doubt that Sats are deeply pernicious, but it is not fair to identify testing as the problem
If the key stage 2 tests are cruel and damaging for children, then the key stage 1 tests for those as young as six must be bordering on abuse. The condemnation of the KS2 Sats has become synonymous with the condemnation of testing.
A high proportion of schools boycotted this year’s tests because of the harmful effect they have had on schools. And while there is no doubt that Sats are deeply pernicious, identifying testing as the problem is not a fair assessment. Think of the tests you set your class - spelling and times tables, for example. They aren’t arduous, the kids don’t lose sleep worrying about them, and you don’t have to skip PE lessons to fit in mock exams in preparation. You might say that that is entirely different because those are internal tests for your own informal assessment. And it is different - the crucial difference being the purpose of the tests.
And herein lies the problem. Your tests are aiming to find out where your class is and to motivate them to try their hardest. You are also trying to gauge how successful your teaching has been. In this informal scenario, you, the teacher, feels no pressure to tailor your teaching to ensure that pupils perform to a certain level. If for example, you told the class that you were only testing them on words beginning with “t”, you know that they would practise only words beginning with “t” - a completely undesirable outcome, because it would be much less informative for you.
And you wouldn’t coach the kids with a pre-test spelling crammer because the results would indicate how well you had prepped them rather than what they had absorbed in lessons. But, as obvious as that is, as we all know it is precisely this type of narrowing and coaching which has accompanied Sats.
Teaching to the test, not testing per se, has made Sats so problematic. Both results and the teaching and learning experience have become distorted. A Civitas survey of secondary school teachers revealed that 79 per cent had found that up to a third of their Year 7 cohort’s abilities were lower than their KS2 Sats results. This encapsulates the pressure which has taken Sats hostage. The reality is that the stress which pupils experience in Sats is teacher, not test, induced.
The response to a decline in this year’s KS2 science results was that an end to testing had put an end to artificially inflated results. However, what has turned the situation around is not the end of testing but the end to the pressure of hitting the correct benchmark. Because it is the pressure to prove children have achieved the “correct” level, rather than the pressure of testing which is causing problems, teacher assessment will not solve the problem. As long as teachers have to demonstrate the proficiency of pupils in their assessments, rather than assess them, it doesn’t matter what gauge is used, it will always be distorted.
This has been showcased by the fact that the latest teacher assessment results were so similar to those in the externally marked tests. In light of the robust evidence showing externally marked Sats results to be inflated, the similarity between the two sets of results implies that the focus on reaching the expected level makes teacher assessment as vulnerable to distortions as the externally marked Sats tests.
This underlying pressure is also one of the main reasons value-added measures have not been the hoped-for panacea in primary testing. Despite diversified benchmarks, these still standardise the child and still require them to reach a certain level of improvement.
In short, it is the misuse, not use, of testing that is causing problems in primaries. Used appropriately, testing can be an invaluable tool for educators rather than a stick to beat them with. Assessing what a child knows in a subject and how they are able to marry content knowledge and skills provides useful information on many levels - for teachers, for parents and for those designing the curriculum (ideally, also teachers). And standardised testing can easily find room under the effective-testing umbrella.
A compelling argument for standardised testing, put forward by New York University educationalist Diane Ravitch, is that it can be a democratising tool. While we tend to regard testing as undemocratic in light of the achievement gap, Ravitch argues that standardised test scores can be seen as a more easily accessible way of understanding how your child is doing at school - without having to understand the education system. In line with this, survey evidence from both sides of the Atlantic tends to show that parents are in favour of primary school testing.
But for primary school-level standardised testing to be a valuable tool, there are two prerequisites. First, it is imperative to recognise its limitations. In doing so, there must be parallel indicators of “performance”. One of the greatest causes of the pressures associated with today’s Sats has been an over-reliance - and therefore an over-emphasis - on test data. Schools have come to be judged almost solely on their Sats results, and instead of using broader gauges of school effectiveness Ofsted has increasingly narrowed in on results.
Second, “expected levels” need to be replaced with a more nuanced, broader scale of attainment. Fairly arbitrary expected levels have, completely inappropriately, turned Sats benchmarks into “qualifications”. Yet what qualifies pupils is their time spent in primary school - not the scores on their Sats papers.
Anastasia de Waal is director of Family and Education, Civitas.