By Gary Eason BBC News Online education editor

It's not only on the touchline that competitiveness shows

From next year, testing at the end of Year 2 is being made more flexible, with greater reliance on teachers' judgements of children's progress.

Researchers at the University of Leeds, who evaluated this year's trial in 5,000 schools, said the move away from formal tests at the age of six or seven was generally welcomed by parents, who regarded the issue of stress as important.

But their report, prepared for the Qualifications and Curriculum Authority (QCA), found "tensions" in parents' views.

Right choice?

The common desire to have a good educational experience for their children was often combined with a desire to have full information about the school.

They wanted reassurance that the school was "doing its job" and that their choice of it for their child had been correct.

So they did not want children tested - but they did want test results.

Most did not mind losing test results for their own children.

"However, they seemed to want to know the test results for the school, in order to make comparisons with other schools and the national picture.

"This is not possible," the researchers pointed out, "so there may be a perception issue that will need to be addressed."

One of the research team, John Threlfall of the Leeds University School of Education, said: "That's a case of wanting to have your cake and eat it."

He said parents had a dual role: immediate concern for their child's state of mind, and trying to make the best choices for their child's future, so wanting them to be in as good an environment as possible.

"Perhaps they have been told, or believe from some other source, that testing is a completely reliable source of information and you can use that to tell whether one school is better than another."

If the perceptions of the professionals were right, however, the accuracy with which children's attainments were reported improved with the mixture of testing and teacher assessment.

"A test is a one-off snapshot of a particular day and everyone can see that some children are stressed and don't perform well, some just get lucky - it's something they have been doing recently and doesn't represent their overall achievement level."

Trust

Moreover the Leeds researchers found "an almost universal belief" among parents that their own child's teacher and school was good, and could be trusted to be fair and accurate in assessing their child's progress.

But there was a widespread feeling that teachers in other schools could not be trusted.

So they were happy to rely on their child's teacher's judgement but wanted others to use an objective test - so that their child's achievements and those of the school - would not be "undermined by unfair comparisons".

A questionnaire comment from a parent at a small school was:

"We are fortunate to have an excellent teacher who knows the children very well.

"In larger schools with higher staff turnover I would be concerned that the teacher assessment would be unfair."

The evaluation found, incidentally, "no evidence" that schools had used the new arrangements to inflate their results.

Tables

Contrary to what some people, including teachers, seem to think, test results at this stage are not used directly for the published performance tables.

But they do go into the annual comparative performance reports that schools and education authorities get from the inspectorate, Ofsted.

And they form the basis of the new "value added" measures, indicating how much progress children have made by the time they take the national tests at the end of primary school - which do form the "league tables".

Parents also wanted to be told more about their child's progress during the year, so as to be able to help.

The Leeds research report said this might become more feasible under the new assessment arrangements.