GillbornsubtoSewell Notes on: Gillborn, D., Bhopla, K., Crawford, C., demack, S., Gholami, R. Kitching, K., Kiwan, D, Warmnington, P.   Evidence for the Commission on Race and Ethnic Disparities.B irmingham: University of
Birmingham CRRE. DOI: 10.25500/epapers.bham.00003389

Dave Harris

They focused especially on research concerned with the experiences and attainments of black British children and young people, especially 'students of Black Caribbean and dual heritage (mixed: White/Black Caribbean) ethnic origin' (2)

They answered four questions provided by Sewell, with summaries first

Question one: what are the main causes of racial and ethnic disparities in the UK and why

Racism means unfairly discriminating against minority ethnic groups including 'unrecognised bias' and 'stereotyping, neglect and/or omission''. Research shows the multiple ways in which it is a key factor with the education system affecting policy and practice at all levels. However racism has never been seen as the only relevant factor in understanding disparities. Statistics can be useful in mapping broad trends but there are always questions about reliability and validity of data . Quantitative research is shaped by assumptions theories and interests of statisticians which can introduce unintended bias data [same with subjective data too, of course].

The Timpson Report [cited in Sewell] shows the problems by confusing association with causation, and by adopting a 'garbage can' approach with too many factors, often tenuously linked to the problem, which lessened the 'apparent significance of each'[even after factor analysis?]. Racism is often seen as a residual category after other more dominant factors, but it 'threads through and influences these other issues', such as income, class and poverty.

Education statistics have been widely misunderstood, for example in presenting the white working class [in general] as the lowest attainment group — 'this is factually incorrect and socially divisive' (2) [white working class boys eligible for free school meals, FSM, is one of the lowest attaining groups, but they are a minority of white working class students] . Eligibility for FSM is a crude proxy measure of poverty which excludes most people who consider themselves working class — 60% of adults consider themselves working class [a notorious overestimate of course] but only 11% of white British school students are FSM. In every ethnic group FSM students achieve lower average results than those not eligible. 'White British FSM students are not [always, they argue]  the least likely to succeed in any of the main measures of achievement' Looking at the 87% who are not FSM, 'the lowest attaining groups are consistently Gypsy/Roma, Black Caribbean, dual heritage and Pakistani students', but an obsessive focus on white British FSM students has diverted attention from these inequities [white categories might include gypsy/Roma?].

There are frequent deficit analyses that stereotype black communities and divert attention from schools. Qualitative research 'Drawing on interviews and observations inside schools' has long shown that black students experience more negative teacher expectations allowing for gender and social class and that these are institutionalised through streaming, setting and tiering. Colourblind school-based policies such as zero tolerance approaches to discipline 'actually discriminate in systematic ways' and produce more frequent and harsher sanctions for black students.

Question two: what could be done to improve representation retention and progression for people of different ethnic backgrounds in public sector workforces?

The teaching force is disproportionately white and 'underprepared for multi-ethnic classrooms'. School leadership is often such and 'replicates patterns of institutional rates inequity'. 'More than 1/3 of minority teachers report having experienced discrimination at work in the last 12 months'. Those who do reach leadership positions report feeling unsupported and over scrutinised and judged more harshly. Meaningful training is urgently required. Current leadership training is mostly silent on race and racism. Ethnic monitoring is needed, but it is often 'an empty gesture' (3)

Question three: how could educational performance across different ethnic and socio-economic status groups be improved?

OFSTED needs to treat race equality as more than an optional extra. Trainee teachers should undertake serious work in relation to patterns of discrimination but there is no current formal requirement. Equality Impact Assessments should be used in a more serious and constructive way. More data should be collected, not just restricted to quantitative material, building on the Ethnicity Facts and Figures Website, which needs to be improved.

Question four how should the school curriculum adopt in response to ethnic diversity?

Key facts about black history are 'inaccurately conveyed or ignored entirely' erasing the vital role played by minority peoples. The English literature curriculum is 'almost entirely devoid of ethnic diversity. The main characters in children's books are almost 8 times more likely to be animals than people of colour'. There is a public appetite for greater diversity and more teaching about racial injustice according to a 2020 survey, and for each major ethnic group.

In more detail:

On racial disparities and its causes, they cannot adequately summarise all the research so they focus on the research based on black British children [bit disingenuous -- these are important to them politically?] especially black Caribbean dual heritage, who are often separated in official statistics even though they are demographically interconnected. Black Caribbean communities are long established and have often been at the forefront of campaigns for racial justice, although they continue to experience persistent and significant race inequity in education 'regardless of their social class status' [references to Gillborn's own work]. Inequalities are not limited to education. They are extensive and long lasting showing there are no easy fixes, [and so they seem to be?] structural, rooted in the organisational skill the training of teachers and the priorities in policies regardless of which party is in power.

The Commission on Race and Ethnic Disparities must be aware of the persistent and complex operation of racism, actions and processes that lead to unfair discrimination. Of course racism is not the only factor, but it is a key factor, and one which has been difficult to address, for example the disappearance of the issue from policy even after the impetus of the MacPherson report – 'many stakeholders working the field believed that much of the initial progress had been wiped away' (4).

Problems with statistics include ignoring children in private schools, thereby ignoring 'a group of disproportionately white and highly advantaged students'. Often surveys use a small group of children to estimate the national picture, but that has problems in sampling and interpretation [a particular problem with testimony too of course!] There are unrecognised assumptions, and the more researchers manipulate the data the more they are likely to introduce unrecognised biases. This is especially so with claims of finding hidden patterns and explanations revealed only by complex manipulations such as regression modelling. One US report 'included a long shopping list of issues (many with no obvious link to the question)… What critics call 'garbage can modelling: "simply the signal is overwhelmed by the noise"' [rather obscure reference to an American study -- introduced because they want to criticise Timpson below?]. There's the problem of overfitting a model, asking too much of the available data, and this can have implications.

The example is 'recent discussions about the significance of racism and discrimination as a factor in the disproportionate impact of the covid 19 pandemic on people of minority ethnic heritage'. The problem seems to be that many of the factors involved were taken to be discrete drivers of inequality without realising that they were shaped by discrimination and unfairness including poverty, home ownership, parental education and pre-existing health issues. Racism was taken to account for discrepancies after all the other conceivable issues had been tested, but it  cross contaminates other factors. It appears to operate as  something which is just left over but it is a more complex set of processes 'working through each aspect of society'.

The Timpson study on exclusion failed to mention racism at all, although it was considered by Sewell. It confused association with causal mechanisms and it failed to understand how racism works and thus downplayed the extent of race inequity. It is agreed that we cannot infer that one thing causes another, but did argue that it had found what drives higher rates of exclusion. It provided detail about its calculations but not enough to rework them. As a result it 'has the characteristics of '"garbage can" modelling'. The regression analysis assumes ethnicity is only relevant when everything else has been accounted for, but it does not consider for example that black students are more likely to live in economically disadvantaged households, more likely to be labelled as SEN, more likely to attend low attaining schools. Overall, these calculations 'reduced the supposed level of disproportionate exclusion experienced by black British students' and removed it altogether for Gypsy Roma and Traveller students [a nice point here saying that Roma children are actually excluded from school at five times the white rate which means, for Timpson 'they are not excluded as much as they should be all other things considered' (5)]. Timpson concluded that ethnicity is not a main factor and there are more complex causes, and this 'belittles the historical experiences of black and other minoritized communities'. It has ignored earlier criticisms of similar statistics and claims superior insight.

Headlines frequently proclaim that white working class students are the least attaining group, which fits in with the left behind narrative [several newspaper references are provided,and there is good analysis in Gillborn 2010]. Since 2010 the Education Select committee has conducted 90 enquiries, but only two were focused on a single ethnic group, in both cases white students. The whole focus is problematic:

First, these figures exclude most white people who think of themselves as working class, around 60% of British adults. Focusing on those eligible for FSM means we are talking about 11% of white British students in state schools [still a significant minority though]. Overall, white students are still least likely to be in the FSM group, with only Chinese and Indian students less likely. More likely to be FSM are 'Gypsy/Roma (39%), Bangladeshi (25%), mixed: white/black Caribbean (23%), black Caribbean (22%) and black African (20%' ) (7). 'White British FSM students are among the lowest attainers but they are not the least likely to succeed', according to the government's Ethnicity Facts and Figures website:

The measures are those achieving the Ebacc, those achieving a strong pass, grade 5 above in English and maths, and the average score in relation to attainment eight, (performance in eight GCSEs). The findings are:

Overall, FSM students do less well than their peers in the same ethnic group and this is the case in each ethnic group. Gypsy/Roma students are the lowest attaining group for each measure regardless of FSM. For ethnic group,  the lowest among FSM 'although not always in the same order; Gypsy/Roma, White British, black Caribbean and dual heritage' (8). Among non-FSM students the bottom four places are again occupied by Gypsy/Roma, black Caribbean, dual heritage and Pakistani students, and this time the order is always the same and white British students do not appear in this group at all

Focusing on white FSM kids is a misrepresentation and it can 'inflame racial hostility'. It does not describe the experience of 60% of white people or identifiers as working class and can eclipse wider questions of unequal attainments between ethnic groups. Those who are not eligible for FSM rarely feature in policy speeches at all and among these, 'four minority ethnic groups consistently experience significant inequities of attainment'.

In-school factors have been neglected, despite the Wanless Report that drew on research saying that black students were treated differently in schools. Often this work was greeted by attempts to deploy 'the deficit perspective' that tried to focus on mine outside communities instead. Qualitative research has shown for a long time that 'black students experience systematically more negative teacher expectations than their white peers of the same gender and social class background' [lots of his own research cited here].

Question two on representation retention and progression opportunities.

Quality of teachers and teaching skills is an important factor, but 'despite the best of intentions and sometimes without conscious awareness… Why teachers tend to view black students as more likely to cause trouble than to excel academically'. This tendency is currently described as '"unconscious bias"' and it can become part of the fabric of the school and become reinforced by processes. If so 'it is a textbook example of institutional racism' (10). Factors that are important include the teaching force that is disproportionately white and underprepared for multi-ethnic classrooms, school leadership, unawareness of the experiences of minoritized teachers, including '55% report being described as "oversensitive", "paranoid" or "aggressive" when they challenged racially unacceptable behaviour' (11). Meaningful training for gatekeepers is crucial, including an awareness of racism and institutionalised barriers, reviewing of key educational leadership programs, ethnic monitoring and so on.

Question three on improving education performance across different ethnic and socio-economic groups.

Attainment has risen dramatically since the late 1980s so young people are capable of succeeding. OFSTED should drive more change, and race equality should be a mandatory aspect of Ofsted inspections again. Initial teacher education should include discrimination related to race and ethnic origin as a formal requirement. This should be meaningful equality impact analysis, as in the aftermath of the Stephen Lawrence enquiry, with proper scrutiny rather than attempts to justify policy. There should be better collection and use of data, but not the garbage can style. We should also rethink the categories: 'mixed' is not a useful category, 'Asian' is not useful category, 'Chinese' is misleading — Chinese students account for 0.4% of state school students compared with 5.7% black and 11.4% Asian, so it is misleading to list them as a major ethnic group.

Question four on the school curriculum.

This is now a highly contested area, with wide acknowledgement that the curriculum needs to be updated, made more diverse and representative, with evidence that this can have direct positive impact on attendance and achievement. If we remember that 'around one in three school students identifies as of minority ethnic heritage, it must surely be a cause for concern' [but there are marked regional differences]. Specifically:

The official UK citizenship has an inaccurate history section that erases minorities as people; the English literature curriculum is almost entirely devoid of ethnic diversity; children's books are dominated by white characters; there is a public appetite for greater diversity and a demand for teaching about race and racial equality.

[Loads of references to follow-up]