Formal school exclusions over the educational lifecourse in Wales

Abstract

Objectives
For the first time, this paper takes a dynamic approach to studying formal school exclusions over a pupil’s education lifecourse (from end of Primary school to end of Secondary school). Using this approach we identify five school exclusion trajectories and then examine their association with educational attainment.

Method
We use a single population cohort of 27,085 pupils in maintained schools in Wales who were aged 15 years old in 2016/17. Using pupil-level administrative data we construct a time series of exclusion data for each pupil. We created school exclusion trajectories using latent class growth analysis before performing regression analysis using pupil-level, school-level and area-level characteristics. This allows us look at the different characteristics of each school exclusion trajectory and then their association with educational attainment at GCSE.

Results
We show that the frequency of formal school exclusions varies over pupils’ lifecourse and identify five main exclusion trajectories, reflecting these variations over time, as follows: no exclusions, low level exclusions, early rise in exclusions, high-level Year 11 exclusions and late rise in exclusions. We find that there are not many differences in the characteristics of the pupils by trajectory. But we do see that the pupils most likely to be formally excluded from school are increasingly homogenous among the groups most frequently excluded over time, specifically male, with special educational needs, consistently eligible for free school meals and persistent absence. We also show that when a pupil gets excluded during their lifecourse does not affect their attainment, but frequency of exclusions does.

Conclusions
These general trajectories may suggest differences in the underlying reason for school exclusions that are not necessarily evident in pupils’ official records or as reported by individual schools. We also show the differential impact of these exclusion trajectories on later educational outcomes. Highlighting these different trajectories may be useful to policymakers and practitioners.