What should primary assessment look like?
Dame Alison Peacock, executive headteacher at The Wroxham School, TES columnist and member of a number of government panels for education
‘Assessment should support, rather than drive, learning’
A ssessment currently serves too many masters. In English primary schools, it is used mainly to assess teacher and school performance via endless tracking. Pedagogy and curriculum should regain prominence, with assessment providing crucial information that supports, rather than drives, learning.
We do not need rigid baseline or key stage tests. Electronic platforms, including video and scanned work examples, would involve the child with a means of celebrating their achievement, thereby inspiring pride and ambition. These files would also enable teachers to gauge the quality of their students’ work through professional collaboration within and between groups of schools. If a child finds aspects of learning difficult, more detailed records will need to be kept as a means of ensuring that finessed teaching is provided.
Accountability achieved through self-regulation and peer-review would replace performance-related pay informed by data sets. National comparative indicators of young children’s attainment could be achieved via random anonymised sampling at any point in the school year and from any year group.
Shared success and collaborative achievement should be part of assessment, too, as optimum learning doesn’t always take place in isolation. Human qualities, such as kindness evidenced through empathic behaviour, also deserve celebration.
A formal presentation by each child to their family and teacher could take place each year as a means of illustrating and evidencing individual successes and challenges. Examples of work, alongside video clips chosen by the child and teacher, could be used to inform a dialogue with families about their child’s learning. This could include examples of scores achieved within adaptive online assessments accessed by the child as a means of practicing and improving skills.
Daisy Christodoulou, head of education research at Ark Schools
‘Comparative judgement is the way forward’
E ven though national curriculum levels have been abolished, sadly, too many of their flaws have crept back in because of the new interim national frameworks. These require teachers to evaluate their pupils’ progress against a list of different statements. This sounds simple, but in practice it is very difficult to do reliably, and it also creates huge amounts of work as teachers are forced to collect reams of evidence showing that their pupils have met each statement.
This method of assessment also has negative implications for classroom practice, particularly in writing. Teachers are forced into a “tick-box” approach to teaching and assessment, where pupils are rewarded for using certain features whether or not those features are appropriate or not, and where heavily coached or scaffolded pieces of work are preferred to more original ones. For example, an odd sentence like ‘forgettably, he crept through the darkness’ gets rewarded because it uses an adverbial opener, even though it doesn’t really make sense.
There is a way to solve this problem. Comparative judgment is a new form of assessment that allows teachers to mark complex tasks like extended writing in much less time, with much greater reliability, and without distorting teaching and learning. Instead of asking teachers to mark each essay against a set of statements, comparative judgment presents teachers with a series of pairs of essays and asks for a judgement. An example given by the organisation No More Marking (nomoremarking.com) is the following: the question answered by the students might be “What is an equation? Give examples of how equations can be useful” and the judgement teachers would be asked for would be “Who has a better understanding of equations?”.
For each pair, the teacher simply has to decide which one is best. If the teacher and her colleagues make enough judgments, the comparative judgment algorithm works out the order and scores all the essays in the mix.
No More Marking provides a free online system that allows you to upload essays, judge them, and download the results. Using this process, a group of about 10 teachers can get near-perfect agreement on the scores of 100 scripts in just half an hour. Traditional moderation can take 10 times as long and is nothing like as reliable.
Comparative judgment could potentially replace the current moderation process for writing at key stages 1 and 2. Of course, any big change like this would have to be undertaken with care, and with the full support of the teaching profession. At Ark Schools, we’ve been piloting this approach with our primary teachers, and we’re working with Dr Chris Wheadon at No More Marking to see if we can use comparative judgment to link standards across schools.
Roisin Ellison, Royal Society for the encouragement of Arts, Manufactures and Commerce (RSA) academies programme coordinator
‘We need a middle years baccalaureate with a broad focus’
W e would recommend the introduction of a Middle Years Modern Baccalaureate (MidBac), bridging the gap between primary and secondary by introducing a curriculum for 9- to 14-year-olds.
The MidBac accredits a whole education experience within a broad and balanced curriculum. It celebrates the wider achievements of young people within and beyond the classroom, encouraging a balance between knowledge, skills and experiences to prepare them for the 21st century.
It would improve the tricky transition phase between primary and secondary by promoting constructive cross-phase discussions between senior leaders, and lead to a truly cross-phase approach to pupil development. It would also recognise achievement from vocational, practical and creative parts of the curriculum. This is done by accrediting three areas of learning: core progress, honours (experiences) and skills.
The core progress reflects academic progress, reported by teacher assessors from KS2 onwards. The honours aspect encourages breadth and flexibility, though learning opportunities and experiences such as working on an extended project. The skills are those competencies needed in and beyond school, such as teamwork and problem-solving.
Dylan Wiliam, emeritus professor of educational assessment at UCL Institute of Education and author of books including Inside the Black Box and Assessment for Learning: why, what and how?
‘Tests must be focussed, frequent and not catch-all in nature’
Assessment is at the heart of effective teaching. If we do not find out what children have learned as a result of the teaching they have received, it is impossible to plan the next steps in their learning effectively.
The difficulty arises when we try to make the same assessment tasks satisfy a range of different objectives. Diagnostic assessments – such as a test of phonemic awareness – can be useful indications of what students can and cannot do. But when the same assessment is used to draw conclusions about the quality of education children have received up to that point, then there is pressure to make sure the students pass the test, robbing the test of its usefulness as a guide to the child’s learning needs.
Unfortunately, because assessing students takes time, there is pressure to make sure tests serve a variety of purposes. No assessment system will be perfect, but there are a number of principles that are useful to bear in mind when designing an assessment system for a school. As a starting point, the following may be useful.
Any assessment system that collects all the information at the end of the year, for example through an examination, will, because of the limited time available for testing, be unreliable. Evidence must be accumulated throughout the year.
An assessment system must cover all aspects of work that are important for the child’s development. Progress in reading and writing is relatively easy to record and report, but we must also record and report on the child’s progress in speaking and listening.
An assessment system is useful only if it is a simplification of reality, rather than an attempt to record every single thing that a child can do. Recording a child’s developing understanding of place value is a good idea. If you’re also recording whether the child can use Roman numerals, you’re probably spending too much time testing and not enough teaching.
If teachers are spending more time recording student achievement than they are planning teaching, the assessment system is out of control and needs reining back.
Pippa Morgan, head of education and skills at the CBI
‘A better balance of formative and summative assessment is required’
The CBI [Confederation of British Industry] has always advocated a holistic education, one that goes beyond the academic into the behaviours and attitudes schools should foster in everything they do. Given the importance of literacy and numeracy, progress tests will always have a key role to play. But if education is to genuinely encourage a love of learning, assessment should be designed to check understanding, not shape the entire approach of educators and force schools to focus on teaching to tests.
A new assessment model should better balance formative and summative assessment than the current system. Owing to Ofsted’s assessment criteria, schools have to be primarily focused on the subjects that are summatively assessed. Yet a meaningful education is about much more than this.
We would advocate for a genuine balance between testing and complementary feedback and assessment that gives pupils the chance to develop, learn and improve over time – shaping them for success in school and the lives ahead of them.”
Fiona Hughes, freelance writer based in Devon and mother to three children, two of whom are in primary school
‘Assessment should be informal and
done in a mixture of ways across all subjects’
I would like assessment to be handled internally by teachers and for it to take place throughout the year on an informal basis – integrated into normal school life. I’d hope this would be a way for teachers to know when a child hasn’t quite grasped something and needs extra support to get the hang of it, rather than to stack children or schools up against each other.
I’d like assessment to be done in a mixture of ways and across all subjects – through the written work that children are doing anyway, as well as through small group and one-to-one sessions (verbally) and more traditional assessment where suitable (for things like spelling and times tables, where you really do just have to memorise things).
The thought of my seven-year-old sitting down to do a formal exam makes me feel anxious on his behalf – it would help if the papers were presented in a way that more closely resembled normal classroom worksheets. And, if “big assessments” like Sats do have to happen, I’d prefer these to happen yearly and without fuss or practice papers and with a clear objective.
Sir Daniel Moynihan, chief executive of the Harris Federation
‘Secondary schools need more accurate information’
Every child should start secondary school with clarity about their knowledge so that their transition can be well-planned and the dip in Year 7 is minimised.
Currently, nearly all secondary schools retest children because they don’t feel primary Sats paint an accurate picture.
This wastes time and enables some secondary schools to make excuses about poor progress.
Instead, we should design simple and accurate computer-based tests for Year 6 students and require secondary schools to use these to inform their planning. These tests could be taken online, so they would be easy to administer with teachers getting the results back quickly. Children who don’t pass in May could undergo an intensive intervention programme and try again before starting secondary school.
We want good teaching throughout primary school, so let’s devise simple tests for each year group to sit alongside Sats. The results would be for internal use, but would also be useful for inspection and to establish a common understanding about what children can do. This would help to ensure schools focus as much on how well children do in other years as they do in Year 6.
And let’s test science, history and geography as well as maths and English. Valid tests are straightforward to construct for them, and they are important subjects that deserve time to be spent on them.
Hilary Leevers, head of education and learning at the Wellcome Trust
‘We don’t want to return to Science Sats, we need robust teacher assessment instead’
Like many other science organisations, we were pleased when science Sats were removed and teachers felt freer to let children explore science in richer, more experimental ways. However, despite still being a core subject, science lost priority in schools and was given less resourcing and teaching time.
Talking with colleagues across a range of organisations, none wish a return to science Sats, but we do need greater scrutiny of science at the school level – a key role that Ofsted should play.
We can assess science knowledge in written tests, and, to some extent, conceptual understanding, but inquiry and experimental skills are harder to capture.
So we need to be able to trust in robust teacher assessment, including moderation. But we know that many teachers are anxious about teaching science and they must be supported to participate in relevant professional development – much of which is freely available through the National STEM Learning Centre.
Michael Tidd, deputy headteacher at Edgewood Primary School in Nottingham and TES columnist
‘We need to separate accountability and assessment’
The challenge for me is how we separate the necessary burden of accountability from the powerful tool of assessment.
There are those who would argue that we should strip out the high-stakes accountability entirely. Tempted as I am, I just can’t hold with that view. The government has a duty to hold schools to account for the use of taxpayers’ money and pupils deserve to have their quality of education ensured.
My view is that tests offer one of the least biased ways in which we can measure such information on a large scale. However, the data should be treated with care.
We need to remove the arbitrary thresholds: scaled scores, or whatever form the results take, are more than enough to produce data that shows the progress individual schools are making; an ‘expected standard’ score is a unnecessary stick with which to beat schools in more challenging areas. By removing these, we won’t prevent the media producing league tables, but it might soften their edge.
We should separate the other elements of assessment from accountability. Teacher assessment in the classroom should be about children’s specific learning; information transferred to the next key stage should be about what a child can and can’t do, and parents ought to hear about where their child is succeeding and where they need more help.
Christina Hinton, neuroscientist and faculty member at the Harvard Graduate School of Education
‘Crucial skills are currently not prioritised as they should be’
We have to be mindful about the limitations of assessments. In some countries, schools measure whatever is easily measurable, and disregard the rest as unimportant. This can lead us to assess whether students are able to memorise facts and spit them back out, then make decisions based on that limited picture. Crucial skills, such as social emotional skills, are not prioritised as they should be because they are not easy to measure.
At this point, I do not recommend trying to measure social emotional skills and then make decisions based on the results. Some people are quick to try to assess skills like “grit” and evaluate schools based on the results. We don’t yet have a measure of grit that enables us to do that responsibly.
We need to look at learning holistically, and consider assessment as one piece of information, rather than the whole picture.
Keith Topping, professor of educational and social research at the University of Dundee
‘Current assessments are a waste of time’
Standardised tests have some use, but only up to a point and only with children old enough to adapt to them. I would use them on two occasions in primary: at Year 3 and Year 5. The tests would be of English (essentially reading, since speaking, listening and writing are too hard to assess by such tests), mathematics and thinking. The three subjects would take place in consecutive weeks. The test would be scored by an agency outside the schools and the results fed back to teachers, children and parents.
In addition, children should take computer-based adaptive item-banked tests of reading and mathematics. The minimum requirement would be that they were taken once in each year, but being item-banked they could be taken more often if anyone wished.
Teachers should be required to subjectively assess the performance of each child on two occasions during each year. The requirement to do this on two occasions is an onerous one, so the nature of assessment should be brief on each occasion.
The first assessment should be done around mid-term in the first term, and focus on what the teacher hopes to achieve with that child during the rest of the year. Specific objectives should be set. The second assessment should be done at the end of the year and assess to what extent the objectives were met and whether there were any other growth or decline points that were unexpected.
I would consider adding forms of peer assessment in Years 3-6, but this would be for immediate gain, rather than any summative purpose.
I would completely do away with current national curriculum assessments, which are antiquated, unreliable and a waste of time.