Autistic adolescents can experience significant challenges in school, including higher rates of absence, exclusion and difficulties with peers compared with non-autistic pupils (John et al., 2022).
Adolescence is an important developmental and educational period; well-being tends to decline, while academic demands increase as young people progress through secondary education. In the UK in particular, the age of fourteen is a critical time when academic pressure intensifies, and happiness related to school may decline. While existing research has highlighted a well-being gap between autistic and non-autistic students, these studies have often focused on the observations of parents rather than what was directly reported by students.
A new study by Greer and colleagues (2026) addresses this evidence gap by analysing self-reported data from over 8,000 adolescents in the Millennium Cohort Study. The study explores whether differences in school-related well-being between autistic and non-autistic young people remain after taking account of academic, social and family factors associated with well-being.
Methods
The study used data from wave six of the UK Millennium Cohort Study, with study participants being fourteen years of age. To compare school well-being, 271 autistic and 8,077 non-autistic young people were included. Well-being was evaluated using student self-reports of positive affect (feeling happy at school) and negative affect (feeling unhappy at school). Regression models compared the two groups first without adjustments, and then after controlling for several key social and academic correlates. These factors included student academic self-concept, i.e., how good they felt they were at learning in a school environment, experiences of bullying, problems in relationships with peers, social support, closeness and conflict with their mother.
Results
Both groups were similar in age (around 13.7 to 13.8 years old). The autistic group was mostly male (about three in four), while the non-autistic group was close to an even split. Both groups were mostly White (85 to 87 per cent), so this wasn’t a notable difference between them, though it still limits how well the findings reflect autistic pupils from minority ethnic backgrounds.
Before taking anything else into account, autistic pupils said they were less happy at school than pupils who were not autistic. On a scale from 0 (not happy at all) to 6 (completely happy), autistic pupils gave school an average score of 4.03, while non-autistic pupils gave it an average score of 4.40.
The results also showed that approximately 27% of autistic pupils said they felt unhappy at school most or all of the time. This was true for approximately 14% of non-autistic pupils. In summary, autistic pupils had more than twice the odds of reporting high negative affect at school.
The researchers then asked a further question: would this gap still be there if pupils in both groups had similarly good experiences of school in other ways? To check this, they took into account things like how confident pupils felt in their schoolwork, whether they had been bullied, how well they got on with other children, how close they felt to their mother, and how well off their family was.
Once these factors were accounted for, the difference between autistic and non-autistic pupils disappeared. This does not establish that these factors caused the original difference, but suggests that they may help explain why school-related well-being differed between the groups. Autistic pupils no longer reported feeling less happy, or more often unhappy, than their peers. The researchers also repeated the original, unadjusted comparison using the whole dataset, including pupils who had not answered every question, and found a very similar gap. This suggests the original difference was not simply down to which pupils happened to be left out of the main analysis.
Finally, the researchers looked only at the autistic pupils to see which of these things mattered most for how they felt about school. One thing stood out: how confident pupils felt about their schoolwork. Autistic pupils who felt capable at maths, English and science were more likely to say that they felt happy about school, and less likely to say they often felt unhappy there.
Feeling unhappy at school was also linked to two further things: being bullied and having more difficulties getting on with other pupils. Other things the researchers looked at, including how much support pupils felt they had from family and friends, how close they felt to their mother, and whether their friends went to the same school as them, were not clearly linked to how happy or unhappy pupils felt.
Conclusions
Overall, this study found that autistic 14-year-olds reported lower positive affect towards school and higher negative affect than non-autistic peers. However, once a set of factors linked to school well-being was accounted for, including academic self-concept, bullying, peer problems, family relationships and socioeconomic status, autistic and non-autistic pupils reported similar levels of school well-being. Among autistic pupils specifically, academic self-concept was the factor most consistently linked to how they felt about school, alongside bullying and peer difficulties.
This could be seen as a hopeful finding, suggesting that, perhaps, unhappiness at school may not be an unavoidable consequence of being autistic but, instead, it could be linked to other factors that could be changed with the right support.
Strengths and limitations
This study’s greatest strength is the fact that this was a large, nationally representative sample of autistic pupils, rather than the smaller, self-selected groups that most autism research relies on. This is important because who gets included in a study often defines what is found in the study itself. Families who take part in specialist autism research are not always representative of autistic people more broadly. Also, the researchers asked autistic pupils to speak for themselves about how school felt, rather than relying solely on what parents observed. The researchers checked whether missing data had affected their results, and they found that it had not, which adds weight to what they found.
However, we must consider that the data were collected in 2015 and 2016. This was before a pandemic that changed young people’s experience of school. The participants in the study were also predominantly White and male. This is not necessarily an accurate representation of the autistic population, as girls and children from minority ethnic backgrounds are underrepresented in this study. Autism was identified through parent-reported professional diagnosis, which means the findings may not generalise equally to autistic young people who are underrepresented in diagnosis or in population cohorts. The wellbeing measures were brief and not built with autistic pupils in mind. The study therefore provides useful evidence about positive and negative feelings towards school, but cannot capture the full complexity of autistic young people’s psychological well-being or school experience. It is worth asking how accurately two short questions can explore something as complex as how a young person feels about school. Importantly, the study did not measure some aspects of the school environment that may be particularly relevant for autistic pupils, including how well schools met their needs, sensory experiences or whether pupils attended mainstream or special schools. These factors could not therefore be examined as potential explanations for differences in well-being.
The finding that the wellbeing gap faded once the wider set of school, peer and family factors was factored in is hopeful. However, this is not definite proof that changing those things would close the gap. This was one snapshot in time rather than a story followed over a period of years. Therefore, it can’t tell which came first: poor wellbeing or the struggles that surrounded it. Such struggles might, in part, be caused by schools that have not adapted well enough for autistic pupils’ individual needs.
Implications for practice
For teachers, school staff and clinicians, a few practical points stand out. Supporting autistic pupils to feel capable in their schoolwork, not just achieving good grades but feeling confident in their own ability, may be one of the most promising and specific goals for improving their experience of school. This should involve more than encouraging confidence; schools can provide appropriate scaffolding, clear feedback, achievable opportunities for success and teaching approaches that genuinely help pupils’ self-concept step by step.
Bullying and peer difficulties also need to be addressed. This should be understood as a whole-school responsibility rather than solely an intervention for the individual autistic pupil. Existing anti-bullying programmes have shown positive results in general school populations (Ng et al., 2022; Olweus et al., 2019), and this study adds weight to the case for testing whether they work as well for autistic pupils specifically, rather than assuming they do.
Asking autistic pupils directly about their own experience of school, rather than relying on what parents or teachers report, also seems important, given how central self-report was to this study’s findings.
However, it’s important to remember that this study is based on data that is now 10 years old, selected from a sample that does not fully reflect the diversity of autistic pupils, particularly autistic girls and pupils from minority ethnic backgrounds. Anyone using this evidence to inform current practice should take this into consideration and also explore more recent and more representative research as and when this becomes available.
Statement of interests
Steve Calvosa independently drafted this blog, with assistance from AI tools for language refinement and formatting only. The final content was reviewed, edited and approved by the blogger.
Edited by
Dr Dafni Katsampa.
Links
Primary paper
Hazel Greer, Caitlin Williams, Afia Ali, Vaso Totsika (2026) Autistic young people’s psychological well-being in school. Autism. 2026 Apr;30(4):1062-1072. doi: 10.1177/13623613261425010.
Other references
John A, Friedmann Y, DelPozo-Banos M, et al. Association of school absence and exclusion with recorded neurodevelopmental disorders, mental disorders, or self-harm: a nationwide, retrospective, electronic cohort study of children and young people in Wales, UK. Lancet Psychiatry. 2022;9(1):23–34. doi:10.1016/S2215-0366(21)00367-9.
Ng, E. D., Chua, J. Y. X., & Shorey, S. (2022). The effectiveness of educational interventions on traditional bullying and cyberbullying among adolescents: A systematic review and meta-analysis. Trauma, Violence, & Abuse, 23(1), 132 to 151.
Olweus, D., Limber, S., & Breivik, K. (2019). Addressing specific forms of bullying: A large-scale evaluation of the Olweus Bullying Prevention Program. International Journal of Bullying Prevention, 1(1), 70–84.