The concealed middle? An exploration of ordinary young people and school GCSE subject area attainment

In Britain school examination results are now an annual newsworthy item. This recurrent event illustrates, and reinforces, the importance of school level qualifications. The General Certificate of Secondary Education (GCSE) is the standard qualification undertaken by pupils at the end of year 11 (age 15-16). GCSEs continue to play an important and central role in young people’s educational and employment pathways. Within the sociology of youth there has been recent interest in documenting the lives and educational experiences of ‘ordinary’ young people.

There are many analyses of agglomerate (i.e. overall) school GCSE attainment. More recently attention has been focused on individual GCSE subjects. In this paper we analyse school GCSE attainment at the subject area level. This is an innovative approach and our motivation is to explore substantively interesting patterns of attainment that might be concealed in analyses of overall attainment, or attainment within individual subjects.

We analyse data from the Youth Cohort Study of England and Wales using a latent
variable approach. The modelling process uncovered four distinctive latent educational groups. One latent group is characterised by high levels of overall attainment, whereas another latent group is characterised by poor GCSE performance. There are two latent groups with moderate or ‘middle’ levels of GCSE attainment. These two latent groups have similar levels of agglomerate attainment, but one group performs better in science and the other performs better in arts GCSEs.

Pupils study for multiple GCSEs which are drawn from a wide menu of choices. There is a large array of possible GCSE subject combinations, and results in individual GCSE subjects are highly correlated. The adoption of a latent variable approach is attractive because it handles the messy nature of the data whilst not trivialising its complexity. The paper demonstrates that a latent variable approach is practicable with large-scale social survey data, and is appealing for the analysis of more contemporaneous cohorts.