Abstract Introduction
School exclusion is an increasingly important public health issue. Excluded children are more likely to experience poor educational outcomes, unemployment, criminal justice interventions and develop severe mental health problems. In Walsall, children as young as 5 years old have been excluded from school. Identification of the local factors that increase risk of exclusion would enable early and targeted interventions to potentially break the cycle of disadvantage. Methods: We linked 3 datasets (school census, school exclusions logs and social care records) from the academic years 2015/16 - 2017/18 inclusive, using a sequence of methods. Initially, school census and exclusions records were linked using unique pupil number, which were then matched to social care records using a concatenation of surname, first name and date of birth. This rendered a linked dataset of 62,536 Walsall children. Data was interrogated using PowerBI and the CRISP-DM methodology.
Results
244 pupils were permanently excluded and 2136 had fixed term exclusions. In the non-excluded group, 50.5% were male, median age is 10 years, 17.2% are known to social care, 17.2% benefit from free school meals (FSM), 12.3% have a special educational need/disability (SEND). In the excluded group, 76.9% are boys, median age is 14 years, 44.5% are known to social care, 39% benefit from FSM, 35% have a SEND need.
Conclusions
Data linkage and mining has enabled identification of significant differences in demographic, socioeconomic and educational indicators in children excluded from school. Work is ongoing to develop a robust predictive model to minimise exclusions in Walsall. External funding details