31 October 2008
The UK's strict school testing regime has helped put British children at the bottom of international learning and wellbeing league tables. So ministers' U-turn on Sats is a welcome change of direction
Few aspects of education policy since 1997 have been more controversial than the government's approach to testing and assessment. Was a high-stakes testing regime essential to raise standards or has it increased the inequality gap between different schools and their pupils?
Two major international studies have given a new context to the debate in the past two years. First was a 2006 report on the performance of 15-year-olds by the Organisation for Economic Co-operation and Development's Programme for International Student Assessments. This showed a big fall in the UK's comparative position. Secondly, Unicef's 2007 report on young people's wellbeing placed UK children firmly at the bottom of the OECD league table.
If our national testing regime is so good, why are our children falling behind their peers in other OECD countries? If our children are so unhappy at school, is this related to our unique national system of tests and tables? And do we always have to accept a trade-off between high levels of achievement for the most able and a long tail of under-achievement for the hardest to teach?
All these issues were examined during an investigation into testing and assessment carried out earlier this year by the Commons children, schools and families select committee, which called for a substantial change of direction. We found a lack of clarity over the main purpose of the testing regime and that the validity of using the same test results for multiple purposes was questionable.
While we accepted the value of some form of national testing, we believed that a light sampling technique was a statistically more valid way of tracking achievement at national level. We were deeply concerned about the evidence (most recently provided by the Ofsted inspectorate) on the extent of 'teaching to the test' and its impact on the broader curriculum.
Above all, we rejected the concept of high-stakes testing in which Sats, GCSE and A-level results are presented as the main factor in school performance. We called for a new accountability framework built on performance data that reflect a wide range of a school's achievements and not just raw scores.
We recognised the continuing problem of young people who are disaffected from mainstream learning. These are the youngsters most likely to cause low-level disruption in the classroom and antisocial behaviour on the streets, and most at risk of leaving school with few qualifications and chronically low self-esteem.
A major issue, particularly for a government committed to reducing social inequality and increasing social mobility, is to what extent our testing regime — in which almost a quarter of pupils are regularly and very publicly labelled as failures throughout much of their school career — is a major factor in the growing problem of disaffected and alienated youth?
The government's remarkable statement two weeks ago, announcing the end of compulsory Key Stage Three Sats and a new framework for performance assessment of schools, reflected much of the committee's report. It marked a welcome shift of emphasis in testing and assessment policy.
The timing was, of course, driven by contractual deadlines over the 2009 tests and the issue was forced by this year's termination of the ETS contract. Nevertheless, Children, Schools and Families Secretary Ed Balls has once again shown his willingness to respond to a broader range of expert opinion than his predecessors.
The door is now open for further analysis of the long-term value of full cohort external testing at Key Stage Two, the complexities of the single level tests being piloted and the potential for reintroducing sampling techniques to assess standards. There will also be lively debates on the structure of the new 'School Report Card', which could revolutionise the traditional league tables.
Other areas that still need to be examined include the balance between external and internal assessment; the optimal frequency of external testing; and what factors other than raw scores need to be evaluated in assessing quality.
Above all, we need to think of the purpose of testing. By focusing more on assessment for learning and less on assessment for ranking, we might find that all pupils improve. Rating a school's progress in boosting children's motivation levels, inspiring further achievement and raising self-esteem is preferable to using tests simply to create league tables.
If schools prioritise assessment for learning, and make use of the new flexibilities in the curriculum, many of the hardest to teach children might just start to enjoy learning, stay learning for longer — and ultimately achieve more. There is some way to go before British children reach Dutch or Scandinavian levels of wellbeing or Canadian levels of achievement. However, these welcome reforms to the testing and assessment system should help us go much further down that road.
l David Chaytor is Labour MP for Bury North and a member of the Commons children, schools and families select committee
PFoct2008
