Social inequalities do not begin in school, but gaps between children widen during school. As cognitive demands on children become more complex, the gaps widen further.
The OECD report on Scottish schools does not claim that social inequalities are primarily a phenomenon of late primary and (especially) secondary education. Rather, it is during these years that the gap widens.
Our report recognises that early efforts must be made to prepare children for the cognitive and cultural demands of school. That is why it praises the high level of pre-school participation, argues for higher proportions of qualified and specially trained staff in pre-school, and commends Scottish efforts to tackle poverty on a broad front.
Our conclusion is that Scottish primaries are very effective and inclusive. Researchers who have highlighted the importance of good-quality pre-school and smaller class sizes in the early primary years would predict that Scottish efforts on these fronts should be rewarded by a lessening of achievement differences. They would not expect these differences to disappear, but expect the gap to widen more slowly than in the past - and that appears to be happening.
But researchers would argue that, unless schools serving the most disadvantaged communities are resourced to provide sustained support throughout the later primary years, the social gap will widen and inequalities will grow apace in secondary.
It is how resources are used that matters, not simply their quantity. This is why the OECD examiners stress the importance of cultural and pedagogical change, and why they see the nature of the curriculum as so important.
Vocational studies, internationally, represent a most important avenue of reform. Scotland has been cautious. There is a risk that vocational education will be used as a relegation stream and undermine working-class access to the academic curriculum.
Weaker learners - disproportionately from working-class homes - already experience relegation: they leave school with few or no qualifications, and enter a labour market which offers few options and little future to unskilled and poorly schooled workers.
Curriculum reform is not a magic bullet, and it will not be enough to raise levels of achievement without effective action to tackle the environment of social deprivation and problems in the ways good schools work.
To tackle the effects of poverty, Scotland needs to allocate resources between local authorities in the best manner, fund national priorities and link those to agreed goals, and assess the performance of its resource strategy against those goals. There is a danger of hiding behind poverty as the source of all educational (and other) ills.
National tests, despite the suspicion attached to them, should be used to measure progress against national standards, taking into account family attributes and the social context of the school. What can we expect, what can we ask and are we making progress?
How do we ensure that all children are well prepared to manage the demands of the curriculum if we do not have reliable test data over the different stages of schooling to alert us to weaknesses, help set goals, and frame strategies for improvement which are capable of evaluation?
Not all the good work of schools can, or should, be measured by tests. But are we to wait until national exams at the end of compulsory schooling before we know how well our children fare in national terms?
National tests should also be used to assess how well educational policies work and how consistently in the 32 local authorities. The concordat between local and national government will not lessen the need for information but increase it, especially in the context of sharp regional variations in achievement related to social deprivation.
Richard Teese was the rapporteur of the OECD review on Scottish schools. This is a personal view which first appeared in the `Scottish Educational Review'.