Taken to task

2nd May 2003 at 01:00
This month pupils aged 7, 11 and 14 will be taking tests many teachers believe to be fundamentally flawed. Tom Deveson and Peter Thomas voice their protests while Sue Horner presents the case for the QCA.


Franz Kafka's short story, Prometheus, is a miniature masterpiece, but it might not achieve level 3 were it submitted in this year's key stage 1 writing task. The changes to the English tasks are such that many experienced teachers are seized with doubt about their ability to match their assessments with those of colleagues or with their own previous judgments.

We can at least be thankful that the reading comprehension tests and the KS1 reading task remain the same. But the writing task has undergone a complete change. At KS1, teachers used to be able to suggest a type of writing that would allow children to demonstrate their strengths. This meant assessment of the task was usually close to teachers' summative assessment covering the entire key stage.

Now there are two set tasks whose apparent flexibility conceals a prescriptive urge. The first has to be a story; the second has to be a set of instructions. Children will spend a total of about 75 minutes on the two tasks, half an hour longer than in previous years. The fact that they have only one attempt to produce their best work means that at the age of seven they are being inducted into the culture of exams. This task is really a test. The mark scheme is not only new; it is specific to this pair of tasks only. Teachers will need to familiarise themselves with a new mark scheme each year. There is, however, no Standards Fund support for training or administration and marking.

Children in Year 6 are no luckier. They used to be allowed to pick from one of four options which gave them the opportunity to write in a genre that suited them. Now there will be two tests based on set prompts. The shorter lasts 20 minutes - 10 minutes less than the time allowed for seven-year-olds. As handwriting is to be assessed within the long writing test, children will have to sustain their best efforts during a timed activity. We all know children who will write inhibitedly and lose marks through fear that their work will not be "neat enough". Those who lose confidence under pressure will also suffer. If they don't respond to the prompts, they will score nothing under "composition and effect", however capable their writing may be.

The thresholds within the new KS2 mark scheme will be determined by the end of June, but no one outside the QCA will know how. Previously, teachers used to be able to sit with colleagues and "level" individual pieces of written work, drawing on a rich practice of making assessment against an agreed pattern of levels. This will no longer be possible; the final level is assigned after totalling marks for the long test, the short test, spelling and handwriting. No one, therefore, knows what "level 4" writing will look like.

Judging from my own experience and that of colleagues in providing training in assessment to hundreds of Year 2 and Year 6 teachers all over England this year, the greatest anxiety surrounds the mark schemes themselves. We have used marked and unmarked pieces provided by the QCA and taken great pains to demonstrate how to apply the schemes. Yet conscientious and experienced markers have produced disturbing discrepancies, coming up with a range of marks at KS1 that give levels 2a, 2b and 2c to the same work and a scattered range of marks at KS2 - the associated levels are undisclosed.

The greatest disparities come in bands C and E, where composition and effect is assessed. Headteachers, middle-aged classroom teachers and enthusiastic 25-year-olds have been equally frustrated. The marking of this year's writing tasks and tests seems to show that those who know the most about their children's work are to be trusted the least.

Tom Deveson works as a teacher,trainer and freelance writer.


The revamped English key stage 3 tests create an unpleasant sense of dej... vu. Once again we are told that the tests are rigorous, piloted and of value to teachers and pupils. Once again English teachers are expressing frustration and dismay, not only at the late notice of change, but at the quality of the material.

The new tests are based on KS2 and national literacy strategy testing models, even though the year cohort taking them (unless in NLS pilot schools) has not been exposed to the strategy. The only pupils who will be familiar with this kind of test are those who have practised on the so-called "optional" Year 7 and Year 8 tests.

The much-criticised, "dumbed down" Shakespeare paper has become the Shakespeare and More Writing paper. The new model requires pupils to respond to two Shakespeare passages, not one as before, but for only 18 marks, not 38 as before. The rest of the paper awards 20 marks for a short writing task. Devaluing Shakespeare is bad enough, but the least of what's wrong.

These new tests are fundamentally flawed. They are based on an inappropriate and Philistine model of reading and writing. Yet the sample papers are endorsed by the QCA - official logo: "Guarding standards".

The sample papers exemplify two writing tasks, one long and one short. Both require imagining a talk to an audience. There is a lack of parity between these tasks. One 30-minute task is allocated 20 marks (four of them for spelling) when the other 30-minute task is allocated 30 marks.

In the long writing task, a separate mark will be awarded for: a) sentence structure and punctuation; b) text structure and organisation; and c)composition and effect. However, in the short writing task, separate marks will be awarded for: a) sentence structure, punctuation and text organisation; b) composition and effect; and c)spelling.

If that's perplexing, there's more. The long writing descriptors are divided into six mark bands and the short writing descriptors are divided into five mark bands. Departing from established practice at KS3 and GCSE, these bands do not relate to notional levels of attainment. Markers will not know what notional levels they are working with, so their previous marking experience will be redundant. Numbers will be converted into levels in June, after the papers are marked.

One of the descriptors for writing which gets no marks at all is simple connectives - for instance, "and", "but" and "so". The example given to support this notion of nil award for the use of qualifying and causative connectives ("but" and "so") is: "Young people will find it more interesting and be able to relate to it more easily." This sentence shows some sophistication in the use of modal auxiliary and main verbs as well as the adverbial comparative. It is a remarkable example of writing that gets no marks at all - not even a compensatory or diagnostic level 3.

Other zero-points band descriptors specify numerous writing characteristics which will not merit one mark out of the six available. If you add together all the zero-score band descriptors of the six strands in the writing mark schemes, you find an extraordinary profile of zilch-quality writing. How this squares with Government targets for raising the number of level fives is a mystery. The only strand which has no zero marks band is spelling. It is difficult to know if this is an act of kindness or a blunder.

On the sample paper "Making it Real", one question requires simple retrieval of information about when the Frankenstein experiment took place.

Two bullet points indicate a two-part answer for the one mark available.

Then pupils are asked: "In the first paragraph, how does the way the final sentence is written build up tension?" This is a vastly more cognitively demanding question, yet only one mark is available. This is a failure to distinguish or value different attainment in a hierarchy of reading skills.

The model answers create more concern. One mark is available for stating the month and the exact time of the event as "November" and "one in the morning", but no mark for "November" and "at one am".

A more demanding question on how tone is achieved in a final paragraph offers the following model answers meriting one mark: "There is a humorous, flippant, ironic tone making the virtual human seem a less important achievement than before"; or "It starts with a conjunction, 'and', which sets an informal tone and suggests a less important or serious side to the virtual human." What is missing in these mark schemes is a proportionate valuing of pupils' attainment at both the lower end and the upper end of the ability range.

English teachers have been demoralised over recent years, yet they have continued, despite initiatives and targets, to develop pupils' self-esteem as well as their language and pleasure in reading. Teachers and pupils deserve better than this.

Peter Thomas is a lecturer in education at the University of Hull.


Change is often challenging and rarely comfortable. But the old ways are not always the best. New ideas and knowledge deserve attention and may demand action. What then matters is that change is based on evidence and experience. Changes to the tests this year are underpinned by a clear rationale for the assessment of both reading and writing. They offer continuity across the key stages, they are based on what we now know about writing and they offer a richer source of information about pupils' abilities.

Since the mid-1990s the QCA English team, together with academics and teachers, has looked long and hard at pupils' writing in a number of projects. We now know much more about the qualities of writing across the key stages and about the choices writers make. The tests have a greater range of writing tasks, varied in terms of purpose, style and form. One task, a longer one, attracts 60 per cent of the marks for writing and the other, shorter, task attracts 40 per cent, reflecting a balance of demand.

Teachers have increasingly focused on different skills at word, sentence and text level. The new mark schemes identify these ingredients and the marks indicate their relative significance. The main judgment on overall effectiveness attracts most marks, but we need also to recognise strengths in sentence and text construction and variety, especially in developing writers.

Evidence so far suggests that marking will be more accurate, as the scope for wayward judgment is less. Of course, a detailed picture of pupils' writing is, in itself, important, but the changes have more to offer. We know that the potential of these tests for analysis of pupils' strengths and weaknesses is much greater and we will distribute software in the autumn to help schools make the best use of these possibilities. Tracking progress will now be easier because of the continuity across the key stage and optional tests.

It is important that the facts of the changes do not get lost in random and misleading points focusing on details. It is pointless to pick out one question and suggest it is harder than another - of course questions vary in demand. We have pre-testing to ensure that, overall, a test is accessible to those at the lower levels and challenges the more able pupils too.

The QCA has also provided new tasks for pupils working outside the targeted range of the tests and these can be found on the QCA website at www.qca.org.ukcatestsabove_level.asp As for Shakespeare, the reading questions ask pupils to comment on scenes in terms of language, character, themes and performance, as they always have. The writing tasks ask pupils to use their study of the play as a springboard for their writing. This is scarcely dumbing down. In reading at KS3, the majority of the marks go to close reading of a variety of unseen texts and nearly 40 per cent to a prepared reading of a Shakespeare play - an appropriate balance.

The QCA tests are the result of extensive research and advice from teachers and other experts so pupils can show their best and have their achievements recognised.

Sue Horner is principal manager for English at the QCA

Log-in as an existing print or digital subscriber

Forgotten your subscriber ID?


To access this content and the full TES archive, subscribe now.

View subscriber offers


Get TES online and delivered to your door – for less than the price of a coffee

Save 33% off the cover price with this great subscription offer. Every copy delivered to your door by first-class post, plus full access to TES online and the TES app for just £1.90 per week.
Subscribers also enjoy a range of fantastic offers and benefits worth over £270:

  • Discounts off TES Institute courses
  • Access over 200,000 articles in the TES online archive
  • Free Tastecard membership worth £79.99
  • Discounts with Zipcar, Buyagift.com, Virgin Wines and other partners
Order your low-cost subscription today