The link between pupils' race and their likelihood of succeeding in a British school has been a matter of concern since the early 1960s.
The African-Caribbean immigrant parents of that era had high expectations of schools in England, but were disturbed to discover their children tended to perform worse than their classmates.
In response to parents' complaints, the Conservative government finally set up a committee of inquiry in 1979. Six years later, the Swann report called for a change in behaviour and attitudes throughout Britain and said schools were uniquely placed to take a lead role.
Since then, education in England has been transformed by Ofsted, the national curriculum, national tests, targets and league tables.
But the final report of the Commission for Racial Equality before it was absorbed into the Equality and Human Rights Commission last year said that children's ethnic origin continue to play a part in determining their future, and that this was apparent from a young age.
So, are schools today better placed to break that cycle of low expectation and consequent low achievement since Swann in 1985?
Happily, the answer appears to be yes. The sheer wealth of data now available has enabled academics to pinpoint with real accuracy exactly what is going on, and allowed the Government to take action.
It has also shown that "colour blindness", where school staff treat everyone the same regardless of ethnic origin, and which is still practised in some schools today, is not the best answer.
Last autumn, the Government issued guidance to local authorities telling them to set a target for 2009 to narrow the achievement gap for each ethnic group with 30 pupils or more. In the frequently asked questions section of the guidance, it explicitly states that this does not mean simply setting the same target for all groups - something at least one authority did in 2006.
At first glance, target-setting seems a seductively simple tool for raising results. But targets throw up unintended consequences, not least the risk of reinforcing what George W. Bush once referred to as "the soft bigotry of low expectations".
The Secondary National Strategy has also published a guide for school managers, to help raise minority ethnic achievement. Include in the guide are examples of how to use data to improve attainment (see story overleaf).
While there was widespread concern about the underachievement of African-Caribbean pupils in the 1980s, the Swann inquiry was hampered by a lack of ethnically based exam statistics that might have revealed how different groups performed. It was not until 2002 that such data started to be collected.
Those statistics are now being used to positive effect, and while overall attainment levels remain similar to those in the 1980s, there is now a real hope that they will begin to improve.
The figures revealed that Chinese and Indian children outstripped their white counterparts, and confirmed that other ethnic minorities, particularly black Caribbean children, did proportionally worse.
The Government responded with Aiming High, an initiative that placed a legal duty on schools to promote race equality and highlighted the importance of using data to monitor change. Later that year, a pilot project was launched in 30 schools to raise the achievement of African-Caribbean pupils.
- Steve Strand, a reader in education at Warwick University, carried out a longitudinal study of ethnic minority achievement for the Department for Children, Schools and Families, which was published in March this year.
He found that, on raw results, Indian and "other" pupils do better in their GCSEs than white British pupils. On average, black African, Bangladeshi and mixed-race pupils have the same attainment as white British pupils. Pakistani students do slightly worse and black Caribbean students do substantially worse.
Dr Strand's analysis did not just consider race and achievement, but how race, class and gender interact.
Of the three factors, social economic status has the greatest impact on pupil achievement. The class gap - the difference between the highest-class children's attainment and the lowest - is roughly three times as large as the ethnic gap.
Once class is factored out, the gap between many of the races narrows - but clear differences remain. Pakistani and Bangladeshi pupils join Indians in doing better than their white counterparts. Black Caribbean children are the only group to underachieve.
This led Dr Strand to conclude that the two groups that should be cause for greatest concern are middle-class black Caribbean children and the white working class.
The three factors that appear to explain much of the difference in key stage 4 results appear to be pupil and parental aspirations, how children see themselves as learners, and how much homework they do.
Parental aspirations may explain why Indian children do so well, as their parents were found to be more likely to pay for tutors and most likely to have a home computer.
It may also explain why white working-class pupils, with low educational aspirations, do badly.
But it does not explain why, despite high aspirations, middle-class black Caribbean children and black African boys underachieve.
Dr Strand concluded that teacher expectations played a part. He found that, proportionate to the population, for every three white pupils entered for higher-tier KS3 science and maths papers, two black Caribbean pupils with the same KS2 scores were entered. They were the only ethnic group under-represented in this way.
"These are complex social problems," he says. "There is evidence that suggests there are low expectations in schools. But rather than a slanging match, we've got to see that it's a mutually negative process - low expectations cause kids to play up, which then reinforce those low expectations.
"It's no good then asking 'Is it schools failing or kids failing?' We have to convince teachers that they need to change without blaming them or putting all the responsibility on schools.
"Schools are one part of society, but only one part. Schools, however, are a good lever for change, because it is much easier to change what schools do, than it is to change attitudes throughout society."
The final evaluation of Aiming High, led by Leon Tikly, professor in education at Bristol University, found that there had been impressive changes where schools routinely gathered data that was then analysed and used for professional development and targeted programmes of support.
What did not work was "colour blindness". The Aiming High evaluation found progress muted where "schools argued that they treat everyone the same regardless of their ethnic origin".
Such attitudes may stem from an admirable ambition. But 60 years after the Empire Windrush brought Caribbean migrants to Tilbury docks, 40 years after Enoch Powell spoke of "rivers of blood", and 15 years after the murder of Stephen Lawrence, race remains an issue that cannot be ignored.
- "Minority Ethnic Pupils in the Longitudinal Study of Young People in England" by Steve Strand
How pupils perform at school can be heavily influenced by their gender, race and social class. Each factor has been the subject of intense debate.
The gaps between the performance of different groups has also led to public concern and a multitude of national and local initiatives.
In this special series, The TES examines each factor in turn to see how it affects teachers and pupils. In the fourth and final part, we will explore how the factors overlap and the challenges that this creates for schools.