Children are suffering stress-related vomiting and sleeplessness as some teachers drill them for months prior to the National Assessment Program - Literacy and Numeracy (NAPLAN), according to the first national study into the impact of the high-stakes testing regime.
The University of Melbourne survey of 8353 teachers and principals raises significant concerns about the ''unintended side-effects'' of NAPLAN, including teaching to the test, a reduction in time devoted to other subjects and a negative impact on student health and staff morale.
Almost half of teachers said they held practice NAPLAN tests at least once a week for five months before the tests every May.
About 90 per cent of those surveyed said some students felt stressed before NAPLAN tests, with symptoms including crying, sleeplessness, vomiting and absenteeism.
The tests, which assess the literacy and numeracy skills of students in years 3, 5, 7 and 9, have been conducted nationwide since 2008.
The study's researchers have called for a national debate into whether there are other ways the data could be collected without the negative impacts revealed in their findings.
''We are narrowing the curriculum in order to test children,'' said lead researcher Nicky Dulfer. ''There are ways we can support numeracy and literacy learning without limiting children's access to other subjects like music, languages and art.''
However the School Education Minister, Peter Garrett, said the results of the survey did not at all reflect the feedback he consistently received.
''Instead, principals and teachers tell me that NAPLAN has proven a really valuable tool to help track student performance and direct attention and resources where they are needed,'' he said.
Mr Garrett said there was nothing in any of the tests which students needed to learn above and beyond what was already in the curriculum.
''There is no reason that the teaching of other subjects should suffer or that students should feel stressed.''
The Australian Curriculum Assessment and Reporting Authority says ''excessive test preparation is not useful'' and ''NAPLAN tests are not tests students can 'prepare' for''.
But the national survey found 39 per cent of teachers held weekly practice tests - and 7 per cent held daily tests - in the five months before NAPLAN. ''Unsurprisingly, teachers also reported that many students became very bored with this,'' according to report authors Ms Dulfer, John Polesel and Suzanne Rice.
More than 70 per cent of educators surveyed said they taught to the test and 69 per cent said NAPLAN had led to a reduction in the time they spent teaching subjects that were not tested.
There were also concerns about the effect of the tests on students' self-esteem. ''As one teacher put it … some students have a belief that they are viewed as dumb by the rest of the community,'' states the report, The Experience of Education: The impacts of high stakes testing on school students and their families.
However, 46 per cent of teachers and more than two-thirds of principals believed NAPLAN information was useful.
Although the federal government tries to ensure NAPLAN results are not used to create school league tables, the majority of teachers believed a purpose of the tests was to rank schools and police their performance.
Only 42 per cent saw NAPLAN as a diagnostic tool. ''As one teacher posited, 'results come out too late in the year to make a significant impact during that year'.''
Ninety per cent believed lower than expected NAPLAN results would mean that a school would have trouble attracting and retaining students.
More than 2280 teachers knew of students who had changed schools as a result of poor results. Other research indicated middle-class parents were more adept at using NAPLAN results to help them choose schools, ''whereas parents with less social and economic capital have less capacity to use this information to their advantage''.