account arrow-down-linearrow-down-small arrow-downarrow-download arrow-left-small arrow-leftarrow-link arrow-rightarrow-upaudio-less-volume audio-not-playing audio-plus-volume audio awarded books calendar close-modal closedate document education emailevent Facebookhamburger impact instagramjustice linkedin location-outline location opinion page phonepinterestplay pluspost preview project reports search-bigsearch-old search share startime twitterwelfare youtube zoom-in zoom-out

Four in ten children will be identified as SEND at some stage during their time at school

By Nuffield Foundation

The Nuffield Foundation is funding the Education Policy Institute to undertake an investigation into how the system for identifying children with Special Education Needs and Disabilities (SEND) works in England.

To launch the project, the research team has produced some preliminary analysis which shows for the first time how many children have ever had SEND over the course of their compulsory schooling. Although official statistics are available for the age profile of children with SEND, they are produced by taking a snapshot of the number of children with SEND at each age, meaning they do not capture the dynamic nature of the system by which schools continue to periodically assess and review SEND status, resulting in some children ceasing to have SEND, while others are newly identified. The official statistics, while valid and important, only show part of the picture, and this affects how we think about policy.


There is a striking difference between the percentage of children who have SEND in a given year group (at one point in time) and the percentage of children who have ever had SEND in that year group or previously. While a maximum of 23 percent of children in the 2016 cohort had SEND at any one time (the peak was in Year 5 when they were aged ten), a full 39 percent were recorded with SEND at some point between Reception (age five) and Year 11 (age sixteen). This makes SEND directly relevant to four in ten children, or twelve per class of thirty on average.

Between 2014 and 2016 there has been a slight reduction in the percentage of children who have ever had SEND by Year 11, from 40% to 39%. This results from a slightly lower percentage of new SEND identifications during secondary school. There has also been a small increase in the percentage of children ceasing to have SEND during secondary school.


The team used administrative data from the National Pupil Database collected via the School Census. These data are recorded by schools to reflect their decisions about when to provide support that is different from or additional to that normally available to other children of the same age. The coverage of the data is all pupils in England who were at the end of Key Stage 4 including those in both mainstream and special schools. SEND status is only recorded for those in state-funded schools.

In this preliminary analysis the team include all types of SEND need, and combine all levels of support (children with Education Health and Care Plans or statements of SEND and those with SEND support or support at School Action or School Action Plus). The analysis is conducted on the cohorts of children who reached the end of secondary school (Key Stage 4) in 2014, 2015 and 2016. Reforms to the SEND Code of Practice (which governs the decisions that schools can make about SEND) were underway during this period, therefore both ‘old’ and ‘new’ SEND codes are included.

What other questions will the research ask?

The full research project, which is due to complete in December 2018, will consider a longer time-frame so that trends can be assessed, will examine different levels of SEND and different types of need individually, and will link the identification of SEND to a range of risk factors that predict the likelihood of SEND.

The research team will use this framework to ask questions about when and where SEND is most likely to be under-identified (when children who have SEND are not recognised by the system), or over-identified (when children who do not have SEND are mistakenly assessed as having SEND). They will study whether some groups of children (such as disadvantaged children or black children) are more likely to be identified with some types of SEND than other children, or less likely to be identified with other types of SEND.

Another strand of the research will compare children’s SEND status as recognised by schools with whether or not they have received care from NHS Child and Adolescent Mental Health Services; this strand will be based on a sample of data from South London.

Explore our projects

Search projects

We improve people’s lives by funding research that informs social policy, primarily in EducationWelfare and Justice. We also fund student programmes that give young people skills and confidence in science and research.

We offer our grant-holders the freedom to frame questions and enable new thinking. Our research must stand up to rigorous academic scrutiny, but we understand that to be successful in effecting change, it also needs to be relevant to people’s experience.