Concerns about unequal achievement in relation to race, sex and class are long standing in England, as they are in many other European countries (OECD, 2023; Eurostat, 2024). However, few studies have explicitly compared the relative size of race, sex and class achievement gaps, and even fewer have looked at the data from an intersectional perspective. The concept of intersectionality has been most closely associated with feminist analyses and more widely deployed in qualitative than quantitative educational research (Codiroli Mcmaster & Cook, 2019). However, some quantitative researchers in England have long considered interactions, particularly between ethnicity, sex and socio-economic background, in the analysis of educational achievement (for example, Strand, 1999). Rather than assuming simple additive effects, an intersectional analysis is achieved through the inclusion of statistical interaction terms which allow the ‘effect’ of one independent variable to vary at different levels of another independent variable. For example, Strand (2014) showed the educational achievement of White British, working class boys was the lowest of all groups in England at age 16. This better reflects how social categories, such as race, sex and class, permeate each other and combine to create ‘complex inequality’ (McCall, 2005). In this paper, we focus on recent educational achievement at age 16, because it is the end of statutory full-time education in England, and qualifications achieved then are key to young people’s future educational, economic, health and wellbeing outcomes (see, for example, DFE, 2018).

The specific research questions addressed were:

• What are the relationships between race, sex, socio-economic status (SES) and educational achievement at age 16 in England? How do these achievement gaps compare in size?
• Are additive models sufficient to account for the data, or are there insights to gain from explicitly modelling interactions between race, sex and SES?
• Which groups of pupils (defined by combinations of race, sex and SES) have the lowest and highest achievement?
• What factors might account for the range of achievement gaps?

The dataset is the Second Longitudinal Study of Young People in England (LSYPE2), which is the most recent nationally representative dataset in England with comprehensive measures of race, sex, class and educational achievement at age 16 (Strand, 2021). A nationally representative sample of 13,000 young people were recruited at age 13/14, and detailed 45 minute face-to-face interviews were conducted with them and their parents, as well as drawing from linked administrative sources such as the National Pupil Database. Sample boosts were included for each of the main ethnic minority groups, so that the samples are large enough to support robust national estimates for different ethnic minority groups. A total of 9,704 students completed GCSE examinations in summer of 2015, and we calculate each students’ total score across the best eight GCSE examination results they achieved. For ease of interpretation, we applied a normal score transformation so that the outcome is expressed in standard deviation (SD) units with a mean of zero and SD of 1. Socio-economic status (SES): We create a comprehensive measure of socio-economic status that combines parental occupation, parental educational qualifications and family household income, by taking the loading on the first factor of a principal components analysis of the three measures. The SES score is standardised to a mean of zero and SD of 1. Ethnic group: In 2015, 29% of the school population in England were from ethnic minority groups. We present an analysis in relation to the nine main ethnic groups in England (White British, White other groups, Indian, Pakistani, Bangladeshi, Other Asian, Black Caribbean, Black African and Any Other ethnic group). Because of similarities in their achievement profiles, we analyse students of mixed heritage together with the relevant ethnic minority group e.g. we combine Black Caribbean and Mixed White & Black Caribbean (MWBC) students. Sex: The students’ sex is recorded as either Male or Female. The most important thing though is to look at this data intersectionally, since everyone has a race, sex and class background, we don’t hold any of these characteristics in isolation. We employ hierarchical linear regression to model the achievement score, allowing for a significant three-way interaction and several significant two-way interactions (see Strand, 2021). The combination of nine major ethnic groups, two sexes and three levels of SES (low, mean and high), produces estimates for 54 unique combinations of race, sex and SES.

The largest achievement gaps are those associated with family socio-economic status. For example, the achievement gap between students from the 20% of homes with the highest household income and the 20% of homes with the lowest household income is extremely large (0.93 SD), over three times larger than the gap between boys and girls, which is small (0.29 SD), and over eight times larger than the gap between Black and White students, which is very small (0.11 SD) . Looked at intersectionally, the groups with the lowest achievement at age 16 are White British and Black Caribbean students from low SES backgrounds, who have mean scores well below the average for all students. This is most pronounced for boys (-0.68 SD and -0.77 SD, respectively), but low SES White British and Black Caribbean girls are also the lowest scoring groups of girls (-0.39 SD and -0.54 SD, respectively). These groups of young people are most at risk of poor long-term outcomes. The mean achievement score for most ethnic minority groups is substantially higher than the mean score for White British students of the same sex and SES, particularly at low SES. An important intersectional question is what factors make minority ethnic groups so resilient, particularly in the face of socio-economic disadvantage? There is a specific instance of ethnic under-achievement, with lower mean scores for Black Caribbean and Black African boys from high SES homes relative to their White British peers. Understanding this outcome means considering the intersectional nature, asking why only boys and not girls, and why exclusively among students from high SES or ‘middle-class’ homes? The implications for educational policy and practice will be discussed.

Codiroli Mcmaster, N., & Cook, R. (2019). The contribution of intersectionality to quantitative research into educational inequalities. Review of Education, 7(2), 271–292. https://doi.org/10.1002/rev3.3116 Cohen, J. (1988). Statistical power analysis for the behavioural sciences (2nd Edition). Hillsdale, NJ: Erlbaum. Department for Education [DFE] (2018). Post-16 education: Highest level of achievement by age 25 (DFE-RR815). Eurostat (2024). https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Educational_attainment_statistics McCall, L. (2005). The complexity of intersectionality. Signs: Journal of Women in Culture and Society, 30(3), 1771–1800. https://doi.org/10.1086/426800 OECD (2023). PISA 2022 Results (Volume I): The State of Learning and Equity in Education. PISA, OECD Publishing, Paris. https://doi.org/10.1787/53f23881-en. Strand, S. (1999). Ethnic group, sex and economic disadvantage: Associations with pupils' educational progress from Baseline to the end of Key Stage 1. British Educational Research Journal, 25(2), 179–202. https://doi.org/10.1080/0141192990250204 Strand, S. (2014). Ethnicity, gender, social class and achievement gaps at age 16: Intersectionality and ‘getting it’ for the white working class. Research Papers in Education, 29(2), 131–171. https://doi.org/10.1080/02671522.2013.767370 Strand, S. (2021). Ethnic, socio-economic and sex inequalities in educational achievement at age 16: An analysis of the Second Longitudinal Study of Young People in England (LSYPE2). https://www.education.ox.ac.uk/wp-content/uploads/2021/05/Strand_2021_Report-to-CRED.pdf