BBC NEWSAmericasAfricaEuropeMiddle EastSouth AsiaAsia PacificArabicSpanishRussianChineseWelsh
BBCiCATEGORIES  TV  RADIO  COMMUNICATE  WHERE I LIVE  INDEX   SEARCH 

BBC NEWS
 You are in:  Education
News image
Front Page 
World 
UK 
UK Politics 
Business 
Sci/Tech 
Health 
Education 
Hot Topics 
UK Systems 
League Tables 
Features 
Entertainment 
Talking Point 
In Depth 
AudioVideo 
News image


Commonwealth Games 2002

BBC Sport

BBC Weather

SERVICES 
Friday, 10 May, 2002, 15:45 GMT 16:45 UK
New doubts over test improvement
1996 test
Flashback: Have children really improved in five years?
Children's performance in standard reading tests has shown no improvement in recent years, researchers say - unlike the gains seen in the official national tests in England.

The finding casts new doubt on the validity of the national curriculum test results - which have been fiercely defended by the Department for Education.

The research was carried out by Durham University's Curriculum, Evaluation and Management Centre.

Each year it compiles performance indicators that track aspects of pupils' primary schooling.

The main aim is to give feedback to schools on the progress being made by individual pupils.

Maths improved

Data have been analysed from the same 122 schools, where each year since 1997 a little more than 5,000 pupils took the same tests.

The scores are standardised - adjusted to give the average for the group - with a mean score of 100.

They show that children's maths performance has improved over the five years from a score of 100 to 109.

But their reading has gone from 100 to 101 - regarded as too small a change to have any significance.

Their vocabulary score has gone from 100 to 102 - again, of no real significance.

'Not easy'

In the national tests over the same period, the percentage of children reaching the level expected for their age in English has gone from 63% to 75%.

Durham's Professor Peter Tymms said the basic problem was that the government was using a new test each year.

"They have to decide what the mark will be that corresponds to the same level from the previous year. That's not an easy thing to do.

"They spend a lot of effort doing it but they can't possibly get it right.

"If each year they get a slight error and it will add to the error from the year before."

Secret

This could add up to a drift upwards, which looked as though standards were rising when they were not.

He said the system was "hopeless".

The only way to check year on year was to use the same, secret test - as was done in the United States and used to be done in the UK.

This works because the children do no practice and the teachers do not "teach to the test" - because they are not facing the pressure of published league tables and targets which demand ever-higher performance.

'Hard work'

The Department for Education said primary school standards had seen the biggest rise in history since 1997.

"The evidence speaks for itself," a spokesperson said.

"Teachers, pupils and parents have worked very hard to achieve these results and people's time would be better spent celebrating that, not knocking it.

"The curriculum tests are very valid and reliable measures of pupils' achievement.

"To suggest that pupils haven't done as well as they have is to slip into the old habit of knocking success." He added: "This is old evidence that was disputed at the time that has been slightly re-heated.

"It was wrong then and it is wrong now."

See also:

09 May 02 | Education
School test papers not delivered
11 Apr 02 | Education
The tests are back
10 Apr 02 | Northern Ireland
Exam body wants to scrap tests
31 Mar 02 | Education
Tests boycott call by teachers
Internet links:


The BBC is not responsible for the content of external internet sites

Links to more Education stories are at the foot of the page.


E-mail this story to a friend

Links to more Education stories



News imageNews image