Early analysis shows wide disparities between 5-14 levels pupils are said to be at and their actual reading level. Two primary 7 children in the same class with an experienced teacher were at level C, yet were actually three years apart in reading ability.
The Stirling research will be finalised later this month and coincides with the launch of an Inspectorate investigation into 5-14 testing and assessment, announced last week by Helen Liddell, the Education Minister.
HMI has been ordered to report early in the new year and is expected to back standardised tests, which could be externally or centrally marked, on entry to primary and at two-year intervals through the 5-14 curriculum. A further option could be a record of achievement at the end of S2 instead of a test. The aim is to provide robust evidence of pupil progress following the national focus on baseline assessment and target-setting and the need to calculate value-added improvement.
Mrs Liddell insisted she would not be proposing one particular solution, a view that has gone down well with local authorities. She said it was difficult at present to reach an overall view on pupils' attainment and that tests which merely confirmed broad levels did not give parents a reliable guide to their children's performance.
Alternative approaches are now being trialled in Stirling and East Renfrewshire. East Renfrewshire is using standardised testing in English and maths throughout the 5-14 levels. Tests are carried out in the first fortnight in February by senior promoted teachers and are centrally marked to lessen class teachers' workload.
Ian Fraser, head of early and special education in East Renfrewshire, said 5-14 assessment was never meant to provide the rigorous analysis now required for target-setting.
Margaret Doran, head of the schools' service in Stirling, said evidence showed many children were not properly assessed. "It's not to do with the teachers' views but the 5-14 curriculum. We are just using another tool which helps the teacher ask questions about how the children are performing," Ms Doran said.
She said it was a "cause for concern" that flawed 5-14 levels were used as the basis for indicators of school performance in the Government's target-setting agenda.
Stirling employs the Edinburgh reading test to contrast reading levels with teachers' judgments and is collecting evidence about pupils' chronological age, reading age and level of attainment and studying the correlation.
Elsewhere, Fife has abandoned standardised testing at primary 3 and primary 5 over the past two years because of concerns about teacher workload but retains a compulsory Edinburgh reading test in primary 7. David Cameron, the council's head of performance review, confirmed wide disparities in pupils' levels when schools compared standardised test results with national tests.
"Clearly, schools are voting with their feet with assessment materials they are purchasing. National test materials are not precise enough," Mr Cameron said.
TIME FOR A FRESH LOOK
Fred Forrester, depute general secretary of the Educational Institute of Scotland, said the Inspectorate's private admission that 5-14 targets were inaccurate lay behind the latest review.
"National testing was given a bad name by Michael Forsyth but I would agree it is time for a fresh look at the matter," Mr Forrester said.
It would be helpful if externally moderated tests were used for internal purposes and as a basis for value-added measures.
But there were "amber flashing lights" about a specific focus on national testing in S2 such as the Conservatives proposed at the last election. The review should look at the whole of 5-14 assessment, Mr Forrester said.