Three-quarters of the teachers questioned by researchers from Exeter University pronounced themselves “satisfied” with the 1996 science test at key stage 3, while the remaining quarter had “mixed feelings”.

“The papers are steering us in the way we want to go,” said one enthusiastic teacher. “They are acting as a signpost to what is expected, and standards are rising as a result, especially for children attaining the middle levels. ”

In maths, two-thirds of teachers were satisfied, 25 per cent had mixed feelings and 6 per cent were “dissatisfied”.

“At first I thought they were a waste of space but I have changed my mind, ” one maths teacher said. “They are especially good for pupils working at the lower levels.”

Modified rapture greeted the English tests, with only half the teachers expressing satisfaction with the Shakespeare paper. But two-thirds were happy with Paper 1.

The survey covered a representative sample of 294 schools. Teachers were not asked about the workload imposed by KS3 tests because that has not been a serious problem at this stage.

Other independent research conducted for SCAA also shows growing satisfaction with the testing of 7- and 11-year olds (key stages 1 and 2). In surveys carried out by the National Foundation for Educational Research (KS1) and Bath University (KS2), the majority of teachers at both stages said they found the tests and tasks manageable and educationally valid, and the results useful.

These findings are in marked contrast to those of a survey published last week by the Association of Teachers and Lecturers. Nearly half the teachers in the ATL survey thought the workload arising from the national curriculum tests was still excessive and more than half thought the tests “not at all worthwhile”. Discontent was most marked at key stage 1.

According to SCAA, the reason for the contrasting findings is that the ATL’s sample was less representative of the majority view. “This is not the fault of the ATL,” David Hawker, assistant chief executive of SCAA, said. “It’s simply that any survey conducted by a union will tend to draw responses predominantly from the dissatisfied.”

In the key stage 1 survey conducted for SCAA, 61 per cent of classroom teachers and 63 per cent of heads considered the testing workload manageable. The proportion of classroom teachers who rated the manageability of individual tests satisfactory or better ranged from 79 per cent for the Level 2 read aloud task to 98 per cent for the spelling test.

Whereas more than two-thirds of teachers in the ATL survey did not believe the tasks to be educationally worthwhile, the NFER survey found the proportion of teachers who rated them satisfactory or better in terms of educational appropriateness averaged 89 per cent. And 80 per cent found them quite or very useful in planning the curriculum, reporting to parents and governors, and for other purposes.

The Bath survey found more than nine out of ten teachers felt the workload was manageable, and confidence in the maths and science tests had risen sharply since 1995. In maths, the proportion of teachers who found all the questions valid had leapt from 48 per cent to over 80 per cent and, in science, it had risen from 58 per cent to over 70 per cent. The introduction of a calculator-free paper was welcomed by 94 per cent of teachers.

The popularity of the tests among parents and children had also shot up. The proportion of parents who support national testing at 11 had risen from two-thirds to three-quarters between 1995 and 1996, and the proportion saying their child had enjoyed the tests had jumped from 36 to 67 per cent.

However, only half of the children surveyed said they had enjoyed the tests but 87 per cent had not been at all upset by them.

The key stage 1 survey was based on returns from a representative sample of 200 schools and the key stage 2 survey on returns from 319 schools. These are preliminary findings; the full evaluations of the 1996 tests will be published by SCAA later this year.