[an error occurred while processing this directive]
BBC News
watch One-Minute World News
Last Updated: Saturday, 6 January 2007, 00:27 GMT
Testing times for school assessment
By Mike Baker
Education correspondent, BBC News

Mike Baker
Change is in the air for testing, targets and league tables.

It seems these central pillars of the education scene in England for the past 15 years are about to be transformed.

But before teachers and pupils cry "hooray" - let's be clear: there will still be testing, it will just be different.

We will hear more details from the government next week, but it is already quite clear from what has been said in public that some big changes are on the way.

At the heart of the changes is the new agenda of "personalised learning".

Last week, the Gilbert report on this stressed the importance of each child progressing at his own pace, and using his own style of learning.

This does not fit well with a system that puts children through formal testing at the same age and at the same moment of the year. Indeed the review suggested a new notion of testing pupils "when ready".

The current formal tests are not much good for informing teachers how their pupils are progressing

We will have to wait for the education secretary's announcement on Monday to see exactly how the government plans to marry the concept of "testing when ready" with its desire to keep tests as an accountability check on schools.

One option would be for schools to switch to more informal teacher assessments. These can be more easily carried out, during normal teaching time, as and when pupils are ready.

But this government, like those before it, is very wary of using teacher assessments as the basis for monitoring school performance.

Those with long memories will recall that a debate over the relative roles of teacher assessment and formal tests raged when national testing first loomed in the late 1980s as part of the national curriculum.

The original idea then was for informal assessments made by teachers to sit alongside the formal, against-the-clock tests. Equal weight would be given to both types of assessment.

But, very soon, the teacher assessments took second place to the more traditional tests whose results became the sole basis of targets and league tables.


However, just after Christmas, the government's favourite think-tank, the Institute of Public Policy Research, recommended a return to teacher assessment.

The IPPR report (Assessment and Testing: Making space for teaching and learning) got rather buried in the end of year festivities. But it is worth a closer look.

It attempts to resolve a long-standing conundrum - namely how to devise a system of testing which both helps improve pupils' learning and holds schools to account for their performance.

The current formal tests are not much good for informing teachers how their pupils are progressing and how they can be helped to improve further. That is partly because they cover a very narrow range of the skills and knowledge that children need. It is also because, rather shockingly, they are so inaccurate.


Research by Professor Dylan Wiliam, of the Institute of Education, suggests that around 32% of test results at 11, and 43 % at 14, are wrong by at least one level.

To make them more reliable, the tests would have to cover a much broader range of skills and last much longer. This is clearly not practical.

Yet the sort of testing that really helps a teacher assess how a child is doing, and helps them devise a pupil's teaching needs, can really make a difference to raising standards.

Indeed, according to research cited by the IPPR report, this sort of assessment (known as "formative assessment") can approximately double a pupil's rate of progress.

So the IPPR paper proposes that schools should go back to teacher assessment for the purposes of checking pupils' progress.

However, it does not abandon the idea of testing for the purposes of holding schools to account.

There are some big issues here

For this, though, it argues it is only necessary to test some of the pupils in some of the subjects.

Providing the sample is large enough for each school, this would give a statistically reliable measure of a school's performance.

Incidentally, while the inaccuracy of the current tests is a problem when assessing individual pupils, it says this is less of a problem when results are aggregated across a whole school.

It is an interesting idea. However, there are some practical problems to overcome; how to ensure, for example, that schools do not select the most able pupils in each subject to be tested.

It would reduce the time taken up by formal testing, and release more time for teaching or for teacher assessments.

Will the government take such a bold step and tilt the pendulum back towards teacher assessments? We will just have to wait and see.

We do know, however, from Alan Johnson's previous public comments that he wants a new way of measuring all pupils' progress that takes account of the improvement (or otherwise) made by all pupils, including the least and the most able.

One problem with the current system is that it only records those who pass a particular threshold, such as reaching Level 4 or getting five good GCSE passes.

In other words, a school could get 80% of pupils achieving five good GCSEs whilst doing nothing to help the other 20%. Yet it would still appear to be doing well in league tables.

There are some big issues here. Finding a way to deliver "testing when ready" whilst also providing a more useful and reliable system of formative assessment and measuring the progression made by all pupils will certainly not be easy.

Johnson signals targets shake-up
16 Nov 06 |  Education
School test 'should be optional'
05 Jan 07 |  Education
Change on way in tests and tables
04 Jan 07 |  Education

Has China's housing bubble burst?
How the world's oldest clove tree defied an empire
Why Royal Ballet principal Sergei Polunin quit


Americas Africa Europe Middle East South Asia Asia Pacific