Evaluation of pushing out of children from all English state schools: Administrative data cohort study of children receiving social care and their peers

Abstract

Background: Pushing out (off-rolling) occurs where pupils are illegally excluded from school. Those receiving children's social care (CSC) services (children in need (CiN), on child protection plans (CPPs) or looked after (CLA)) are thought to be at increased risk, but limited evidence inhibits understanding of this phenomenon. The extent of pushing out can be inferred from non-enrolment in administrative data.

Objective: To estimate proportions of children not enrolled across secondary school (aged 11-16, up to year 11) and to explore the association between CSC history and non-enrolment in year 10/11.

Participants and setting: >1 M pupils in year 7 (aged 11/12) in English state schools, 2011/12 and 2012/13.

Methods: We estimated the proportion of children not enrolled across years 8 to 11, disaggregated by CSC history. We assessed with regression modelling the association between CSC history and non-enrolment in years 10/11.

Results: Of children without CSC history, 3.8% had ≥1 year not enrolled by year 11. This was higher in those with a history CiN (8.1%), CPP (9.4%) or CLA (10.4%) status. The odds of non-enrolment in years 10/11 were higher among those with CLA history vs non-exposed peers (OR 4.76, 95% CI 4.49-5.05) as well as in those with CPP history (3.60, 3.39-3.81) and CiN history (2.53, 2.49-2.58). History of special educational needs further increased non-enrolment odds, including after confounder adjustment.

Conclusions: Findings imply that children with CSC history are more likely to be pushed out from school than children without, especially those with special educational needs.

Keywords: Administrative data; Children in need; Children's social care; Off-rolling; Pushing out; Special educational needs.

Fig. 1

Schematic overview of analyses. Data from the same cohort of children were used to address both aims, namely children aged 11/12 at the start of academic year 7 in 2011/12 or 2012/13 (or aged 11 at the start of those academic years if designated as following any year). * Coverage varies for each cohort. See Table 1 for details. FSM free school meals; IDACI income deprivation affecting children index.

Fig. 2

Annual and cumulative proportions of children not enrolled across years 8 to 11. Data in tabular form are available in Supplementary File 5. CiN children in need; CLA children looked after; CPP child protection plan. Annual and cumulative proportions of children not enrolled across years 8 to 11. Data in tabular form are available in Supplementary File 5. CiN children in need; CLA children looked after; CPP child protection plan.

Fig. 3

Predicted probabilities of non-enrolment in years 10/11 by social care status and special educational needs and disability status derived from Model 5 (Table 3). Dots represent the predicted probability of non-enrolment in years 10/11 (and sitting <5 GCSE exams) as calculated from the coefficients in Model 5 (Table 4 and Supplementary File 6). Error bars are 95% confidence intervals. Values for other variables in the model were set to their reference categories. CiN children in need; CLA children looked after; CPP child protection plan; GCSE General Certificate of Secondary Education; SEND special educational needs and disability; yr year. Predicted probabilities of non-enrolment in years 10/11 by social care status and special educational needs and disability status derived from Model 5 (Table 3). Dots represent the predicted probability of non-enrolment in years 10/11 (and sitting <5 GCSE exams) as calculated from the coefficients in Model 5 (Table 4 and Supplementary File 6). Error bars are 95% confidence intervals. Values for other variables in the model were set to their reference categories. CiN children in need; CLA children looked after; CPP child protection plan; GCSE General Certificate of Secondary Education; SEND special educational needs and disability; yr year.