Ethnicity, disadvantage and other variables in the analysis of Birmingham longitudinal school attainment datasets

Abstract

Explaining and responding to inequalities in attainment are significant educational policy challenges in England as elsewhere. Data on four cohorts of Birmingham Local Education Authority (LEA) pupils, each approximately 13,000, were analysed by ethnicity, deprivation, gender and other relevant individual pupil variables. For the four successive cohorts of children, aged five in 1997–2001, analysis shows the attainment trajectory of each ethnic group from Baseline/Foundation Stage Profile (age 5) to GCSE (age 16). The relative constancy over time, the changes from one key stage to the next and the differences within broad ethnic categories argue against simplistic explanations. The ethnicity variable accounts for a relatively small amount of variance in pupil achievement, with the same ethnic subgroups recurrently low attainers. Considering explanatory perspectives on educational inequalities and ethnicity in the light of these data, we conclude that a structuralist perspective offers the best explanation recognising economic exploitation, dominance and oppression at the national and local levels. Notions of institutional racism and Critical Race Theory (CRT) are considered to be inadequate and counter-productive, in part shown by their inability to accommodate the range of attainment levels and educational experience of different ethnic groups. More tellingly, they lack causal explanations relevant to the United Kingdom and deflect attention from the need for sustained effort to reduce poverty and disadvantage as it affects children.

Keywords:

• Ethnicity
• disadvantage
• gender
• attainment inequalities

Acknowledgement

The authors thank the Birmingham Assessment Unit and the Children, Young People and Families Information Management Team who provided the data for five cohorts matched across key stage assessment points. This was a large, unique body of data with which we were privileged to work. Without the work of the Birmingham teams, this analysis would not have been possible and attempts at understanding the complexities of the education system would have been harder.

The authors would like to thank Stephen Steadman for his critical inputs to the development of this article. Ray Godfrey was part of the original formulation of the project at Canterbury Christ Church University and has advised on analytical approaches. Gill Fairbanks and David Ewens were fierce in demanding clarity and syntactical propriety; the authors tried hard to meet their high standards.

Notes

1. Birmingham supplied five cohorts but the 1997 cohort was omitted as we considered four to be a sufficient set of repeated measures for the case to be made about constancy and change. The 1997 cohort is included in Table 12.

2. Chinese pupils, of whom there are few, achieve even higher levels especially at KS3 and KS4.

3. Changes across time for a group can only really be used to suggest an increase/decrease relative to the other groups. Because scores were necessarily standardised to make the different assessments numerically comparable, it is impossible to assess whether scores were actually going up or down across time in an absolute sense.

4. The exception is the enduring and shocking disparity with Irish Traveller and Gypsy/Roma children, continuing high on exclusions and very low on attainment.

5. “Institutional racism”, so central to the Macpherson report on the murder of Stephen Lawrence, is not a term that appears in two landmark equalities Acts of Parliament – Race Relations (Amendment) Act 2000 or the Equality Act 2010. It is likely that legal advice was that the term was too imprecise to bring cases of compliance or breach.

6. The Baseline protocol rested heavily on a combination of parental judgement and early teacher assessment at the start of schooling and was a “snapshot”; the FSP protocol guided teachers’ judgement on a range of criteria over a longer period. The analysis (Figures 1–6 and 9 and Table 7) show that these changes have not been great and have not affected trajectories of attainment during later stages of schooling.

7. Cole makes the point that “such language as a descriptor of everyday racism is useless as unifier and counter-productive as a political rallying point” (Cole 2016, 20). Though flawed theoretically, CRT may well serve these social and quasi-political purposes.