'Tests for four-year-olds as soon as they start school are harmful'

The government is ploughing on with its attempts to introduce baseline assessment – ignoring the fact that it's impossible to develop a fair test for children of that age, writes one former primary teacher and inspector

Colin Richards

News article image

Let’s be clear: a test is a test is a test – even if it is called a teacher-mediated baseline assessment by the Department for Education. 

And it is to be administered to four-and-five-year-olds in the very first half-term of their entry into school when some of them will only have just turned 4.

The government believes that such testing is necessary, not for the good of the individual children in the here and now, but to provide a means of evaluating their primary schools around seven years on through the establishment of so-called "progress measures".

Do we have the know-how to devise a reliable and valid test for children at such a young age? Many of us in education don’t think that we do.

The fate of the recent ill-judged attempt to provide three comparable approaches to such assessment should have given the DfE, at least, a warning and, at best, cold feet. Yet, it is ploughing on – but with only one test developer, so there's no need to worry about comparability...   

If tests for very young children can be devised, and it’s a very big "if", they will take years to develop – yet the proposed testing is to be piloted in 2019-20 and introduced from Autumn 2020. 

A major part of the problem lies with the children themselves.

How can tests be fair for all pupils?

They learn in very different ways and at different rates, so developing tests that are fair to all is difficult, almost certainly impossible. At that age children are particularly volatile: not only do they find it difficult to sit still, but they also change from day to day, almost minute to minute at times. How can testing capture that changeability?

Children come to school with a wide range of achievements, but the proposed testing only looks at a fraction of these. Many of the most important, such as self-confidence, wanting to learn, willingness to cooperate with others and a degree of personal independence, cannot be measured or tested at all.

Early literacy and numeracy are important, of course, but not all-important and not as important as emotional security, which is the foundation of all school learning.

Which brings us to the effects of the testing on the children themselves.

They may not fully realise the importance of the tests but they will soon pick up signals if their parents or teachers are anxious about them, as many will be. The result will be many worried children whose worry will get in the way of their early learning and will threaten their enjoyment of the challenge of a new school.

Some of the most anxious children are likely to be the youngest, who could be almost a year younger than their classmates taking the same tests. Would you like to take tests just weeks into a new job and tests that could well label you "good","OK" or "poor"?  Presumably not, but that’s how many children will feel, consciously or unconsciously, just weeks into school.

Lastly, the government assumes that children’s performance at 5 can be compared meaningfully with their performance almost seven years later. But how reasonable is that?

Not only are the children taking the test different people, but also the tests themselves are different and their results are not comparable.

All this leads most early years experts and very many teachers, let alone retired inspectors like me, to believe that the tests are likely to be harmful.

Of course, we do believe that children’s achievements on entry to school need to be recognised and built upon, but not through testing conducted so early in their school lives. Class teachers are best placed to find out where young children are and what they need to learn next, on the basis of observing and working closely with them during their first few months in school. Such assessments are likely to be much more sensitive to individual children’s needs than any tests provided by the government. 

But there is a strong case for believing that the worst won’t happen. The testing will not become established. Why? Because as every grandparent knows, four-year-olds are notoriously and gloriously mercurial and unpredictable. With these qualities, they will undoubtedly sabotage efforts to establish a reliable and valid baseline test and with it the government’s vain attempt to “measure” progress from age 4 to 11.

Colin Richards is a grandparent, a former primary teacher and inspector and an emeritus professor of education

Want to keep up with the latest education news and opinion? Follow Tes on Twitter and like Tes on Facebook

Register to continue reading for free

It only takes a moment and you'll get access to more news, plus courses, jobs and teaching resources tailored to you

Latest stories


Coronavirus and schools: LIVE 12/8

A one-stop shop for teachers who want to know what impact the outbreak of the virus will have on their working lives
Tes Reporter 12 Aug 2020