Friendship and Victimisation in Children with Autism Spectrum Disorder in Support Mainstream Secondary Schools - a Mixed Method Analysis

Background: School-age children with autism are reported to have poorer well-being outcomes than their mainstream peers (Begeer et al., 2017). Despite research consistently indicating that children with autism are less likely to be socially included than their mainstream peers (Adams et al., 2014; Little, 2001), there is little research exploring the experiences of friendships and victimisation in autistic students in school.

Objectives: To compare well-being and social outcomes of students with and without autism attending mainstream secondary schools with an autism support base.

Methods: A mixed-methods approach was used to explore well-being and experiences of friendships and bullying of 27 young people with autism and 33 typically developing children. Groups were matched in terms of gender ratios and by age (mean age= 12 years, 1 month; SD = 0.69).

Self-report questionnaires included the Subjective Happiness Scale (Lyubomirsky & Lepper, 1999); Self-Report Coping Scale (Causey & Dubow, 1992); and the self-report Strengths and Difficulties Questionnaire (SDQ; Goodman, 2001). Qualitative information was gathered adapting interview items from module three of the Autism Diagnostic Observation Schedule (Lord et al., 2000; as per Rowley et al., 2012).

Results: Regression analyses were conducted to examine how far coping styles and sources of support predicted well-being. Analyses revealed no interaction between Autism/comparison groups and impact of coping styles on SDQ total problem score or happiness, so both groups are considered together (regression statistics, SDQ: Adj r2 =.15, p =.005; Happiness: Adj r2 =.23, p <.001). An Internalising style of coping predicted increased difficulties (β =.35, p =.002). Statistically significant predictors of happiness were using a Problem Solving style of coping, a positive predictor (β = .44, p <.001), and Internalising style of coping, which negatively predicted happiness (β = -.25, p =.02). Sources of social support did not predict either well-being outcome.

We also examined the qualitative and quantitative information collected about friendships, bullying and social support. There are no differences between the groups on the incidences of bullying (p =.15), but students with autism were statistically less likely to use peers and close friends for support (p=.01 and p=.001 respectively). Examination of the interview data indicated greater variance in the understanding and experience of friendships amongst students with autism compared to mainstream peers, and incidences of bullying were more likely to be focused on characteristics related to autism than the range of focus for mainstream children (e.g. appearance, accent, "that person bullies everyone").

Conclusions: These findings suggests that all students, not only those with autism, would benefit from support in developing a more problem-focused style of coping with difficulties. Autistic students do have sources of social support, but are less likely to rely on similar-aged peers. This may be related to the variability we observed in understanding and experience of friendship. We suggest that better understanding the factors underpinning this variability will be an important avenue for future research, allowing schools to better target support for students who may be vulnerable to more significant social difficulties.