Catalogue of testing disasters uncovered
Many primary schools protested to the teacher unions and the School Curriculum and Assessment Authority this summer after receiving key stage 2 results which were widely different from teachers' assessments. Some complained that children with special needs were given higher grades than their more academic classmates and many pointed out that the science test was considerably easier than the English or maths papers.
At the time, SCAA officials argued that they had received relatively few protests or appeals against grades. But a survey undertaken by a group of researchers from King's College and the Institute of Education, University of London, reveals that there was remarkable consistency in the complaints lodged by a nationally representative sample of 30 schools. The schools in the survey, which was funded by the Economic and Social Research Council, were situated in a northern shire county, a Midlands metropolitan area, a prosperous southern Home County and an inner London education authority.
The European Conference on Educational Research heard that teachers in all of the schools complained that insufficient time had been allowed for the maths tests. Twenty-seven of the 30 teachers questioned also said that the maths test required an unacceptably high reading level.
"The overwhelming feeling was that the overall level of the papers was so hard that it would have caused underattainment due to feelings of panic and inadequacy," the researchers reported. "Maths A was the only test in which a team member observed a child crying at the end, among a general 'stunned silence'. Teachers said that it was 'very difficult', 'formidable', and that some pupils had 'looked at it aghast' and needed reassurance to encourage them to continue.
"Teachers made particular mention of the problems of Level 3 pupils. One said it 'knocked them back completely', another said that 'the children who are borderline 3s had a thoroughly horrible 35 minutes'."
Most teachers felt that the science test was generally fair to pupils (although seven of the 22 schools that said the scores were unexpectedly high did question the validity of the test). They also thought the spelling and handwriting tasks were reasonably fair, but they made several complaints about the English papers.
As with the maths test, the need to recognise achievement across Levels 3-6 meant that lower-ability children lost out. "Teachers were pleased that there were easy questions at the beginning, but many thought it unfair to present weaker and less mature pupils with both a story in which the main point was too subtle for them to understand, and a difficult poem."
The tight time limit for reading comprehension was also thought to disadvantage slower pupils. Teachers were unhappy that many pupils did not finish and were therefore unable to do themselves justice. One commented: "There was nobody who had real reflection time at the end ... even the brightest child couldn't look back at any answers and I don't think that's very fair."
The researchers also noted that the amount of preparation for the tests varied widely. Some children in schools where there had been no revision or test practice were therefore likely to have been disadvantaged. "This difference also affects the validity of the results," the researchers said, "since they may well be measuring training rather than more fundamental competencies. "
However, despite all these drawbacks - and the fact that the tests were leading to more setting and ability grouping - 24 of the 30 teachers and 25 of the 28 heads saw some advantages in the national tests. Furthermore, the majority of heads (16) had no objections to them.
"It seems that the current model is seen as a broadly viable one, providing that the tests could be improved in line with earlier recommendations so as to give more valid results," the researchers concluded. "Teachers would prefer, however, rather than two separate results, some combination of teacher assessment with national test results to give a single more valid result for each subject. In the longer term, moderated teacher assessment without national tests seems likely to receive the bestsupport."
Validity and impact of national tests in the primary school: the teacher's view, by Margaret Brown, Bet McCallum, Brenda Taggart, Jackie Branson and Caroline Gipps.