Secondaries appear to be less influential, accounting for up to 7 per cent of variation in attainment in 1997 after pupils' previous performance and background was taken into account. This could make the difference between getting six Standard grades at level 2 and six at level 3 for a similar pupil attending the most and least effective of the 36 secondaries surveyed.
Differential effects have also undermined assumptions about the extent to which girls always do better than boys in exams. The study found that while boys began performing less well than girls as early as the late primary stages, girls made less progress in maths between primary 4 and primary 6.
The strongest predictor of S4 achievement, the study states, is pupils' S2 performance in reading and maths. Early intervention in secondary schools should be recognised as the crucial factor, as it is in primaries.
The report makes clear that all schools, not just the most advantaged, can make a difference. The secondary school with the highest free meal entitlement, for example, performed above expectation in two of the three Standard grade measures.
Professor MacBeath said: "Challenging schools' limiting beliefs about what they can achieve is at the heart of what school effectiveness and school improvement are all about."