Autistic girls and school exclusion: Perspectives of students and their parents

Abstract

Background and aims

If a child’s behaviour does not conform to school policy or causes harm to either peers or staff, they may be temporarily or permanently excluded from school. Whilst it is unlawful to exclude children due to their needs, school exclusion is common amongst children with special educational needs, including autism. Currently, little is known about experiences of school exclusion from the perspectives of autistic students and/or their parents. This is particularly the case for girls on the autism spectrum.

Methods

Semi-structured interviews were conducted with eight autistic girls and their parents (seven mothers, one father). Interviews explored experiences of mainstream schooling; alternative educational provisions that were offered (if any); the school exclusion process; and the girls’ current educational provision. As well as asking the girls and their parents about positive and negative aspects of their past and current experiences, participants were asked to reflect on areas for potential improvements.

Results

Interviews were analysed using thematic analysis and three key themes emerged from the data: inappropriate school environments (including problems with the sensory environment, difficulties when placed with inappropriate peers and general pressures of mainstream classrooms), tensions in school relationships (including problems with staff and peers, alongside a general lack of communication), and problems with staff responses (including a perceived lack of understanding of the girls’ needs and a lack of appropriate support being provided, resulting in ‘battles’ between parents and schools).

Conclusions

The themes and subthemes that emerged from the interviews were not unique to autistic girls. Indeed, issues such as inappropriate school environments, a lack of staff understanding and breakdowns in relationships have been repeatedly raised by parents and young autistic people (mostly boys) in other studies, albeit in different environments. Nevertheless, the results highlight that more needs to be done to positively influence the direction of the girls’ educational journeys.

Implications

To improve the inclusion of autistic girls, it is recommended that educational establishments be proactive in developing inclusive environments, build positive relationships (both in and outside of the classroom) and, if exclusion is unavoidable, better support students both before and after the process.

It has long been thought that autism spectrum disorder (henceforth, autism) predominantly affects males, and a fairly consistent male prevalence bias of 4:1 has been reported in the literature (Kreiser & White, 2014; Lai, Lombardo, Auyeung, Chakrabarti, & Baron-Cohen, 2015). However, these ratios fluctuate across the autism spectrum, ranging from 2:1 in those with intellectual disabilities (Holtmann, Bölte, & Poustka, 2007) to 5.1:1 in those without (Kim et al., 2011). Several reasons for this discrepancy have been proposed, including variations in behaviour and interests between autistic ¹ boys and girls (Hiller, Young, & Weber, 2014; Hsiao, Tseng, Huang, & Gau, 2013; Koenig & Tsatsanis, 2006; Mandy et al., 2012), under- or misrecognition of autism in females (Lai et al., 2016), genetics (Lai et al., 2011), developmental differences across the genders (Lai et al., 2015) and a male assessment bias (Russell, Steer, & Golding, 2011).

It is thought that an early feminine advantage (Koenig & Tsatsanis, 2006) enables autistic females to assimiliate with typically developing peers (Hiller et al., 2014; Sedgewick, Hill, Yates, Pickering, & Pellicano, 2015). However, this appears to diminish with the onset of adolescence (McLennan, Lord, & Schopler, 1993; Ranson & Byrne, 2014), when physical, psychological and social changes cause difficulties in sustaining friendships (Hsiao et al., 2013) and understanding social conflict (Sedgewick et al., 2015). Despite such factors rendering this group vulnerable to social isolation (Dean et al., 2014), little is known of the experiences of adolescent females, particularly in relation to education.

Although the social requirements of the adolescent classroom provide ample opportunities to identify atypical behaviours, autistic girls are frequently overlooked (Hendrickx, 2015). Whilst autistic females can give the illusion of coping (Hendrickx, 2015; Riley-Hall, 2012; Wagner, 2006), a lack of support in the increasingly pressured environment of secondary school (attended between the ages of 11 and 18 years) poses a risk to their well-being. Unstructured time, where social fluency is paramount (Riley-Hall, 2012), is key in exposing the behaviours of autistic females (Hiller et al., 2014). However, teachers may be unaware of the inconspicuous behaviour of females (Osler, 2006), due to their more amenable classroom demeanour (Wagner, 2006), their tendency to ‘camouflage’ their social communication difficulties (Attwood, 2007; Lai et al., 2016) and a lack of behavioural problems (Dworzynski et al., 2012; Hiller et al., 2014; Mandy et al., 2012). These behaviours may also place them at risk of bullying, isolation and rejection (Dean et al., 2014; Sofronoff, Dark, & Stone, 2011). This not only limits their opportunities to succeed, but may contribute to school exclusion.

In England, the exclusion of pupils from school can take the form of a fixed period exclusion (where the child is temporarily removed from the school) or a permanent exclusion (expulsion). Such steps may be taken if the child’s behaviour does not conform to the school’s behaviour policy or causes harm to peers or staff (Department for Education, 2012). Despite it being unlawful to exclude children due to their needs (Department for Education, 2012), children with special educational needs (SEN) account for up to seven of every ten permanent exclusions in England (Department for Education, 2015). This is a significant problem for autistic students, with 45% experiencing school exclusion (Ambitious About Autism, 2016). These figures may relate to teachers finding autistic students hardest to include (Humphrey & Symes, 2013), citing anxiety over training, knowledge and provision to successfully facilitate inclusion (Glashan, MacKay, & Grieve, 2004; Witmer & Ferreri, 2014).

With regard to girls and school exclusion, most of the literature comes from typically developing students. Osler, Street, Lall and Vincent (2002) questioned 81 girls regarding their experiences of school exclusion and found that poor peer interactions impacted upon the direction of their education. For autistic females, who may be predisposed to anxiety (Albano & Krain, 2006), the role peer-conflict has upon exclusion is potentially significant. In another study, Lloyd and O’Regan (1999) investigated the views of 20 adolescent girls with experience of exclusion in a variety of Alternative Provisions (AP: school provisions for children who cannot access mainstream education). Whilst many girls in Lloyd and O’Regan’s (1999) study held negative recollections of mainstream education, positivity was expressed about their experiences of AP.

There are some specialist educational provisions for autistic children who have been excluded from school (see Brede, Remington, Kenny, Warren & Pellicano, 2016), but the most common AP after exclusion is a pupil referral unit (PRU) (Lawrence, 2011). PRUs provide education to children from a variety of backgrounds, including excluded students, those at risk of exclusion, those that are ill or those lacking appropriate school placement (Michael & Frederickson, 2013). Children with SEN, including autism, are frequently educated in PRUs (Department for Education, 2016). PRUs have the benefits of having high staff to pupil ratios and a smaller, more personalised environment. However, they also cater for a high number of students with social, emotional and mental health issues and exposure to these students may encourage ‘challenging’ or ‘troublesome’ behaviours in autistic students (Kaplan, 1982). Although PRUs are considered short-stay schools, with the ultimate aim of transitioning children back into mainstream education (Hart, 2013), this is not always achievable (Lawrence, 2011). In addition, there has been no research assessing the impact of PRUs on autistic females.

To date, much of the research on the educational experiences of autistic students relies on the views of parents and professionals, with children unable to input on decisions affecting their own lives. This is most evident regarding secondary education, where there is little effective consultation with students (Ben-Arieh, 2005; Clarke, Boorman, & Nind, 2011). Yet students provide unique insights into the barriers to, and facilitators of, their educational experiences. Humphrey and Lewis (2008), for example, found that anxiety, sensory-sensitivities, bullying and low self-image contributed to negative secondary school experiences in their sample of 20 autistic boys. More recently, Dillon, Underwood and Freemantle (2016) reported how 14 autistic secondary school students (11 boys, 3 girls) were positive about their environment and appreciated good relationships with staff; reiterating how calmer atmospheres and positive staff–pupil relationships are potentially crucial to inclusion (Brede et al., 2016; Lloyd & O’Regan, 1999; Makin, Hill & Pellicano, 2017; Michael & Frederickson, 2013; Robertson, Chamberlain, & Kasari, 2003).

The present study aimed to explore experiences of school exclusion amongst autistic girls. Few studies have consulted autistic girls to determine what contributes to their educational experiences. This is despite research into women and girls, as well as education and inclusion, being highlighted as priority areas by the autism community (Pellicano, Charman, & Dinsmore, 2014). Thus, by capturing the voices of excluded autistic girls and their parents, it was hoped that insight into their experiences of secondary school education would be provided, adding to knowledge concerning this under-studied population.

Method

Participants

The study was advertised by contacting PRUs and special schools across South East England, and via charitable organisations (e.g. the National Autistic Society, the Autism Education Trust). Autistic girls were invited to participate if they were of secondary-school age (11–18 years) and had experienced exclusion from mainstream secondary education. A total of eight parent–child dyads took part in the study (including seven mothers and one father). Six of the girls lived in single parent (maternal) families and all but one had siblings. Of the two families whose siblings also held additional diagnosis, only one was diagnosed with an autism spectrum condition. The girls were aged 12–17 years (mean = 14.75; SD = 1.51). At the time of the study, six had received a formal diagnosis of an autism spectrum condition, one was undergoing assessment (and subsequently received a formal diagnosis) and one was assessed (by a clinician) as being on the autism spectrum, but costs prohibited her parents from securing an official diagnosis. For the seven girls who had received a formal diagnosis of an autism spectrum condition, the first author reviewed their educational files (comprising psychologist reports and medical records) to ensure that the girls met DSM-5 (American Psychiatric Association, 2013) or ICD-10 (World Health Organisation, 1993) criteria for an autism spectrum condition. For the one girl who had not received a formal diagnosis, her personal records were indicative of an autism spectrum condition. None of the girls were in receipt of a Statement of SEN or an Education, Health and Care Plan (both official documents that record the educational needs of young people in England), despite parents attempting to seek formal acknowledgment of their children’s needs. Four of the girls were diagnosed with additional conditions: all four had attention deficit hyperactivity disorder (ADHD); two had oppositional defiance disorder (ODD) and one had dyspraxia. These diagnoses were received in early childhood, before receiving an autism diagnosis.

Seven of the eight girls were currently situated in PRUs, with one awaiting placement at an AP due to recent (parental) withdrawal from school. All girls had experienced exclusion in some form. Five had experienced permanent exclusion; of these, four had been placed in other schools (i.e. not the school they had been excluded from) before their permanent exclusion. This was undertaken as part of a strategy called a ‘managed move’, in which pupils are sent to a different school for a fresh start, usually with a set of achievable targets (Bagley & Hallam, 2016). Similarly, one girl had experienced a failed managed move, and was currently situated at the PRU, as it was felt that this provision was most suited to her needs. Two girls had self-excluded, with one citing ongoing mental health issues as a contributing factor.

Materials and procedure

Ethical approval for the study was obtained via the Departmental Ethics Committee at UCL Institute of Education. Semi-structured interviews were conducted in calm environments, at either the resident PRU (n = 5) or at their home (n = 3). Given the sensitivity of the interview topic, and the fact that they had not met the interviewer (KS) before, girls were provided with the option of having their parent present during the interview. This proved popular, with seven of the eight girls opting for this approach. Whilst this may have influenced results (e.g. potentially, the girls could have been less open about their experiences with their parent present), the overarching priority was to ensure that the girls felt safe and comfortable during the interview. Six of the girls excused themselves from their parent’s interview (at their own request), with only two opting to remain in the room. In all cases, parental interviews followed the girls’ interviews.

Interview schedules were developed after consulting similar literature (Humphrey & Lewis, 2008; Tobin et al., 2012). Interviews began with general rapport building, in which the purpose of the study and key ethical information was explained to the participants. In all interviews, participants were presented with open-ended questions, followed-up with prompts for further information, if required. Parents and girls were questioned about: (a) the current classroom environment (positive and negative aspects; whether it differs to previous mainstream environments; and views on the staff in the current educational provision); (b) experiences of exclusion from school (the factors leading to the exclusion, and the exclusion process itself); (c) prior expectations of their current educational provision (knowledge of the reasons why the girls were placed in their current educational provision; views about the options offered to them, if any; and how they felt about these options); and (d) the girls’ previous experiences of mainstream education (positive and negative aspects; their views on the staff in mainstream schools; and how they felt mainstream education could be improved to meet the needs of the girls). Parents were questioned in the same broad areas but were also asked to reflect on the impact of the autism diagnostic process upon their child’s educational journey.

Interviews were audio recorded and transcribed verbatim. Interviews ranged in length from 14 to 57 min for the girls (mean = 31.38, SD = 15.35) and between 30 and 72 min for the parents (mean = 53.13, SD = 13.56). The length varied based on the nature of the girls’ experiences (e.g. number of exclusions, the range of APs or strategies offered prior to exclusion), participants’ levels of sociability, and participants’ willingness to provide additional information in response to questions.

The resulting data were analysed, by two of the authors (FS and LC), using thematic analysis (Braun & Clarke, 2006). Data were interpreted within an essentialist framework (to report the experiences, meanings and reality of the participants), using an inductive (‘bottom-up’) approach (i.e. without integrating the themes within any pre-existing coding schemes, or preconceptions of the researchers). This involved the authors independently familiarising themselves with the data, by reviewing the transcripts to develop an initial set of themes and sub-themes. The authors met several times to review these themes and subthemes, discussing them using a semantic approach (identifying themes at a ‘surface’ level, without theorizing beyond the actual content of the data).

Results

Despite the girls and their parents having some differing perspectives regarding the girls’ education experiences, similar themes were found across the entire sample, with (a) inappropriate school environments, (b) tensions in school relationships and (c) problems with staff responses all contributing to educational experiences (see Figure 1).

Figure 1.

Educational experiences of excluded autistic girls: Themes and subthemes.

Theme 1, Inappropriate School Environments, was featured across both participant groups, with respondents citing the ‘impersonal’ environment of mainstream as being contributory to their experience. Difficult sensory environments were commonly discussed, with one parent remarking that this was akin to ‘psychological enclosure’ in which ‘you are surrounded by all these people, all this different noise…a big room and lots of people in it’. Students similarly expressed concerns regarding sensory demands impacting on their learning experience: ‘it’s very difficult to learn when there is a lot of noise going on’. Students also cited class size as a barrier: ‘it’s very hard [to access support] when there’s so many children in one place’. Parents reiterated these concerns: ‘with 30 in a class they don’t offer one to one support’; and felt it impacted on the girls’ experiences: ‘there’s a lot of people, there’s a lot of fighting for attention, whether it be in the playground or in the classroom’. In contrast, there was a sense of relief in regards to smaller environments with girls explaining that they were ‘really good ‘cause you do get to learn… [you’re] more focused on what you’re doing’. Yet transitioning to this environment might be challenging: ‘its smallness could bring more attention on me and I’m not used to that’.

Although the intimacy of the PRU was a positive aspect, inappropriate classroom peers presented concern for the girls, who were taught alongside students with behavioural issues. Girls found this distracting due to ‘dramas’ that took ‘learning time away from the pupils who need it’, whilst others found these students to be ‘naughty’ and to ‘bully’ them. Parents echoed these comments, explaining that ‘if one of those people [is] disrupting the class it affects everyone and it isn’t fair’. Parents felt that little could be done in this regard, as PRUs were for students ‘who are naughty at school and misbehave’. Girls felt that the situation could be improved through greater levels of teacher-student interactions, as teachers could ‘try to [do] something the students want to do but at a learning level as well’. Whilst not all girls and parents shared this view, it highlights that unsuitable peer groups may threaten – and offset – the benefits of smaller environments such as the PRU. Indeed, this was a key concern amongst parents, who were anxious about how these influences would ‘impact on’ their child ‘because you hear all these horror stories about PRUs’.

Despite these concerns, the girls valued the alternative opportunities that the PRU offered (e.g. ‘work experience’, ‘animal class’, ‘offsite activities’). Girls felt that such activities helped them progress and would ‘open up opportunities’ for them. Parents also sensed value in this type of provision: ‘going off site and doing things that she wouldn’t normally do…she seems a lot happier’. The girls were also pleased that PRUs were flexible to their needs; for example, one student could ‘listen to music to calm down or to concentrate’. Another felt that the staff ‘understand where I’m coming from and when I’m angry they leave me alone’, resulting in areas of conflict being removed: ‘they don’t moan at you as much…when you’re late or when you’ve got a bit of wrong uniform’. Girls also alluded to adaptations not being accommodated in mainstream education. As one (albeit not at a PRU) noted: ‘The value of being somewhere where you feel safe, comfortable cannot be underestimated’. Parents were also pleased that the curriculum, despite yielding similar outcomes to mainstream education, appeared to be tailored to their child’s needs: ‘in mainstream school there isn’t any of that, it’s normal curriculum and it’s not necessarily helping each individual child, whereas at the PRU it’s much more geared to the types of situation that they are in’.

To the girls, the PRU provided ‘really fun ways to learn’ and alleviated some of the pressures in mainstream education. The girls were happy that they did not have to ‘constantly remember things’ or ‘catch-up’. When questioned why this was, students referred to the staff: ‘[they] don’t put pressure on you as much, they’re really relaxed’. Girls also felt able to ask for support in smaller environments: ‘I can say “I don’t get this” and they’ll go through it with me … it’s easier because it’s in a smaller class’. The pressure of mainstream education was most evident in the responses of girls who had experienced self-exclusion, with one remarking: ‘I put so much pressure on myself because they were putting pressure on us’. Some indicated a degree of performance anxiety in class, indicating a need for support: ‘I felt stupid for putting my hand up in front of 30 people and asking for help…people look at you in a funny way’. Although these fears were apparent in mainstream education, one girl remarked that they felt it was better in smaller classes as ‘not everyone judges you’.

Parents also expressed concerns about the effect of pressure on the girls in mainstream environments, stating that educational messages of: ‘if you are not in school every lesson, every day, working really hard, you will fail’ lead to detrimental mental health and propagate a culture whereby targets are more important than the individual child: ‘they had targets they needed to reach, she wasn’t performing to what they needed her to do’. Another parent explained: ‘Because the school is under pressure, they pass it down to the teachers, who are under pressure who pass it down to the students who are under pressure and that is just toxic’. This ‘toxic’ culture was also evident regarding attendance, with both parents and girls commenting on its impact. In one case, a child was set an unachievable attendance target that jeopardised a managed move: ‘I was supposed to get 96% attendance but I’m quite sick and I had 2% attendance…And then I had 60% but I had a really severe injury and I had two weeks off’. The parent complained that even though her daughter’s emotional well-being had improved, this had hindered a potentially successful transition: ‘if that managed move was better, she’d still be at that second school. If the pressure wasn’t so hard and the attendance [target] wasn’t so high, she would still be there’.

Although parents of self-excluded girls were ‘desperate’ for their children to attend an educational provision, they lamented the focus on attendance: ‘[it is] incredibly stressful to get a text every morning to say that your child has not attended school’. Some felt that mainstream provision did not fully understand the reasons behind the students’ absence, with one parent threatened with prosecution until the child received medical intervention. Another parent highlighted the lessened pressure in this regard when their child attended the PRU: ‘I don’t feel so much of the pressure about if she’s having a bad day and ringing up and saying she can’t come in today … I felt was going to break under at the other places’.

This example of parent–school understanding also highlights the importance of Theme 2: Tensions in School Relationships. Staff relationships were found to be one of the most influential aspects regarding the girls’ educational journeys. Girls were keen to share positive experiences and all enjoyed it when teachers took an interest in them as individuals. One appreciated it when ‘they always say “Morning” and see how you are throughout the day’ and another was pleased with how ‘they always ask me what I’ve done and how I’ve got on’. This approachability seemed to inspire teacher confidence and enabled those who felt unable to seek support (in mainstream) to do so: ‘I can kind of ask them anything…we have a bond’. Students felt this friendliness levelled the power dynamic: ‘because they were more like students, than actual teachers’. It also helped them feel valued: ‘they’ll make time for you, which I really like because it makes you feel like they actually care’. Girls viewed mainstream teachers who had similar attitudes favourably: ‘I did like a few of the teachers … there are some teachers that I became quite close with’; as well as those who offered support: ‘there’s one teacher there who’s always try[ing] to help’.

Despite these positive relationships, there were several examples of staff–student conflicts. Some were relatively minor: ‘she just nitpicks…I sit sideways with my back on the wall and she doesn’t like that’. Other girls expressed general dislike, with comments such as: ‘the teachers annoy me’, ‘they didn’t understand me’ and ‘some [teachers] were really grumpy’. Others tried to justify the reasoning behind their response, giving further insight into their relationship with staff: ‘Some of them I really hated…with mainstream schools especially there’s so many more different teachers…there’s just more of an option for having people you don’t like’; or commented on a breakdown of relationships: ‘some of them did try to help but then they gave up and I didn’t like that’. Girls also alluded to teacher expectations causing conflict: ‘they had high expectation of me and they were like “why didn’t you finish that work” and I’m like “I wasn’t feeling very well today” and they’re like “I don’t care”…’ Some teachers exposed students to ridicule in front of their peers ‘and often make an example of you in front of the class’. This meant girls felt threatened, as if the teachers were on: ‘a power trip and feel like they need to impose themselves on everyone and you need to do exactly what they say’. This placed barriers in their relationships, and it caused the girls to internalise their experiences: ‘I felt that they were all judging me when I got into lessons and they’d all give me funny looks’; causing apathy in establishing relationships: ‘if you want to ask them a question they would ignore you…then it made me feel a bit low because I wouldn’t ask for help cause they wouldn’t really listen’. Despite these setbacks, girls still desired a relationship with their teachers with one despairing that: ‘they just really didn’t get to know me’. Others commented that they liked feeling valued by the staff: ‘being listened to…it’s nice when teachers get to know the students because you get to know what they like’, which they felt unachievable with such negative interactions. Parents were acutely aware of the importance and nature of student–teacher relationships and gave insight into areas that the girls neglected. One parent reflected on how her child was exposed to shouting as ‘she hadn’t covered her history book in plastic’, which resulted in her withdrawal from education. One parent also recalled an incident in which their child had disclosed their SEN but was told ‘no, you’re just lazy’.

Both girls and parents were also aware of the impact peer relationships had on the school experience. Bullying was an issue, with examples of social bullying: ‘I was a bitch for talking to other girls and being friends with them and that I was abandoning them and betraying them’; and physical bullying: ‘someone threw my lunch round the room’. This caused the girls to retaliate: ‘I’d throw chairs at them, only once but then that boy had been winding me up’; or withdraw: ‘I ended up not responding at all because I didn’t feel like entering that situation’. Parents were also aware to the impact of bullying: ‘she was bullied daily, which ended up with her reacting, her reactions weren’t great’; and its relation to exclusion: ‘obviously the bullying, her assaulting people or people doing to her ended up with her showering the teacher [with pencils]’.

One parent was also concerned about how their daughter’s desire to make friends put her in risky situations: ‘she was being groomed by an older pupil and told to [do] naughty things like hold my baccy [tobacco]’. Girls valued belonging: ‘nice to know you belong in a group’. Nevertheless, adolescence is a difficult time (for anyone) and the social difficulties associated with autism may result in social isolation. One girl felt ‘angry and annoyed’ at her inability to make friends. Another tried to escape social rejection by making herself: ‘sound better…so that they wouldn’t bully me’, which was unsuccessful. However, girls did enjoy the social aspect of their school lives and were pleased if they found positive, albeit small, social groups in which they identified: ‘I gained two very special friends … I have very close relationships with them individually’.

Whilst the girls’ experiences came from their everyday interactions, parents’ main source of feedback came from the school. Communication was a key sub-theme, with parents calling for improvements across all educational provisions (including mainstream schools and PRUs). Parents disliked negative contact, citing how they ‘dread[ed] the phone calls from mainstream, because it’s only if there is an issue’ and were eager for ‘positive confirmation as well, you want your children to be happy’. Parents were also dismayed when educational provisions did not listen to their feedback, especially regarding their children’s vulnerabilities: ‘I specifically said that she wasn’t allowed to go out at lunchtime … I drove past her when she was out one time’. In one instance, a lack of consideration for a parental request for help went unnoticed: ‘I called them up desperate for help, and they could have told me [about what was happening] in any one of those conversations and they didn’t’, leading to the child’s eventual exclusion from the school. Parents also felt a sense of isolation once their child was excluded: ‘I did feel like we were just left to do whatever… there was no communication at all’; having to rely on their own initiative to secure resources: ‘I’ve had to phone them up and tell them that she has a right to an education and I need some work sent home for her’. If communication is effective, parents feel reassured about their children’s progress: If I had a problem with her at home that would affect her day at school, I know I can ring up the school … And if there is a problem at school they can contact me so the communications always there. One parent stated: ‘I think it’s the line of communication that makes you feel supported and it’s always open which is very important in any aspect of life’.

Girls and parents voiced concerns in regards to Theme 3: Problems with Staff Responses. Girls felt that the staff’s lack of understanding of their individual needs contributed to their experience. For example, staff demonstrated limited understanding of the girls’ coping strategies: ‘I had music playing, it gave me the distance I needed … I think that was one of the first places conflicts with authority took place, because you aren’t allowed music in school’. This was cited by the students as contributory to their exclusion. One girl explained that she was ‘out of lessons walking the corridor sometimes blasting music out’. Another girl related how promises of support were not upheld: the head of year said “I promise if anything else happens there I’ll send you home” … I had a very severe panic attack when I was in the office … They phoned up the teacher and said “I have her here in full meltdown mode”… and they just said “go back to your lesson, you’re over reacting” … '. The parent was similarly frustrated at the limited understanding displayed in this instance, stating: ‘[they] really don’t understand how she’s feeling at all…they would just say…“stop making yourself sick, we’re not going to send you home”…’. The parent of another child, who was also severely affected by anxiety, tried to rationalise the experience: ‘[whilst] they tried very hard to help … the lack of understanding … really hurts them’. This was often attributed to limited teacher knowledge as to why anxiety was expressed; as one girl explained: ‘most innocuous things to everyone else … they were normal but they affected me much worse than they affected everyone else’. Another girl acknowledged that ‘teachers can’t have knowledge of how every classroom dynamic works and how every relationship between students work’ but felt that the lack of this understanding ‘makes you vulnerable’.

Experiences considered to be ‘frankly terrifying’ were common amongst the groups, with students and their parents feeling that these incidents encouraged the externalisation of anxiety and were contributory factors to school exclusion: ‘[we] need somebody in a mainstream school that is knowledgeable of children with special needs and understands them’; ‘know how to deal with people like me, with my disabilities, my problems and other people’s problems and what they have as well’. However, there are concerns as to the number of SEN-experienced school staff: ‘just one person in a massive secondary school who was absolutely drowning’. The girls noted that if they had access to ‘someone they felt comfortable talking to’, it would have made a difference to their ability to succeed: ‘I feel that would have helped me so much’.

Despite calls for help, school support was deemed unhelpful. Girls who were provided with the opportunity to use SEN support were either chastised for using it too often: ‘I got told off for going to the place where I got support too much’; felt isolated by the experience: ‘[they] shove you in the corner of the isolation room’; or felt that it was ineffective: ‘I ended up not using them because I found something more effective’. Some parents felt that schools were frightened by the level of support required: ‘she does need a lot of support’; or placed barriers in accessing appropriate support: ‘I had asked if she could go somewhere on her own to do her [practice exams]…they [said] they’ve changed the rules, you can only have it if you have medical evidence’. However, in-class support assistants were appreciated by some of the girls, who reported that these staff ‘don’t just benefit me [they] benefit the class’. Conversely, some girls felt that they interfered with their learning: ‘she would be doing the work for me but not telling me how to do it’.

Similarly, support during transitions (i.e. prior to, or following, school exclusion or withdrawal) was viewed as either unsuitable or non-existent. Reduced timetables (in which students are given a bespoke timetable, generally reduced in contact time) were applied to several of girls, as a strategy to alleviate the more social parts of the curriculum. Whilst this was successful for some: ‘[it] completely cut out the social side of education, so in free periods or times where I didn’t have lessons I could leave the school, which was really valuable’; others found that it was unhelpful: ‘[they] gave me options to do a part-time timetable but that just wasn’t going to work for me because I’m quite all or nothing’. Managed moves were also used as a tactic to remove the child from their current educational provision, to give them a fresh start, but the outcome of this was compromised by setting unachievable targets on attendance or due to altercations that led to their exclusion, for example: ‘there was this girl there who was supposed to be my friend, she was on the [autism spectrum], and she told a teacher that I [swore at her]’. A student was also given temporary respite at a PRU, but their parent noted that the original school she attended was averse to her return and kept providing conflicting information: ‘they kept messing her about saying yes she can go back, no she can’t’.

Following exclusion from school, parents expressed concern about the long delays (up to 3 months) that it took to access AP, with no set work provided in the meantime: ‘They didn’t contact us to find out how she was doing or anything…they just sort of washed their hands of her, dumped her and said that’s where you are going and that was that’. Parents were also anxious as to the impact this had on their children’s well-being: ‘it’s not acceptable … it affects you socially, mentally, emotionally and physically … she wasn’t socialising, she wasn’t learning’. Only two girls were offered any support through the transitional period, but these arrangements were unsuitable. One girl stated: I got the choice of returning to [school A] which is my first mainstream and being home tutored … 5 hours a week, I’ve been told I’m quite smart and … I need mental stimulation so I decided I give the PRU [a chance]. The other girl remarked: ‘home schooling or home tutoring [were offered] but those things just didn’t really work ‘cause I couldn’t get out of bed to go downstairs and do work’.

Notably, parent battles were prominent in trying to secure appropriate school for their children with one parent stating: ‘We didn’t get any support and anything that was initiated was initiated by myself’. Parents had to contact agencies themselves due to limited school support: ‘I spoke to [the] education inclusion [service] myself on several occasions to try and find a best way forward for her’. However, in some cases, parents were unsure of the effectiveness of the support: ‘if I couldn’t understand what was going on what hope did they have really?’ Some parents also felt that barriers were placed in the way of achieving success for their child, without considering their needs: ‘[the school] set the bar in the same place for a child with no difficulties…it was just setting her up to fail’. Parents of autistic girls frequently face challenges in securing an accurate diagnosis, and understanding what this means for their child, and may result in parents feeling isolated: ‘never been any support for us as parents’; ‘because I didn’t have that support I did feel failed. We were just left on our own to deal with it basically’. A strong desire for support, to have someone to talk to ‘just even a shoulder to cry on’ was indicative across all parent interviews. Additionally, one parent identified that it is also important for parents ‘to understand teachers a bit better, because I think parents don’t understand how much pressure teachers are under’. It was thought that this could improve working relationships and potentially avoid an oppositional stance when communication breaks down.

Discussion

Interviewing excluded adolescent autistic girls and their parents provided insights into which factors contributed to the girls’ educational experiences. We found that three key themes (inappropriate school environments, tensions in school relationships and problems with staff responses) influenced the direction of the girls’ educational journeys.

With regard to school environments, participants valued the intimacy of the PRU. Whilst PRUs are not designed for (and should not be used as a destination for) vulnerable autistic children, it was seen to offer some protection from the sensory demands found in mainstream education (Lloyd & O’Regan, 1999; Michael & Frederickson, 2013). For autistic children, an inability to effectively digest the sensory environment may cause disruption to daily routines (Schaff, Toth-Cohen, Johnson, Outten, & Benevides, 2011) and lead to withdrawal or aggression (Bogdashina, 2016). It has also been linked to academic achievement (Ashburner, Ziviani, & Rodger, 2008). Mainstream schools should therefore be mindful to the sensory needs of their students to enable effective inclusion.

Flexible environments were beneficial in this regard. When provisions were more accepting of the girls’ use of coping strategies and actively sought to alleviate barriers (e.g. uniform, lateness), engagement increased (see also Dillon et al., 2016). The girls and their parents were also appreciative of the experience-driven curriculum of the PRU, which provided opportunities for the girls to actively use their interests and take part in work experience. Such opportunities help socio-cultural development (Stern, 2012) and could protect against poorer outcomes in later life (Howlin, 2013). Whilst it would be harder to implement such initiatives in mainstream schools (which have higher student numbers than provisions such as PRUs), there may be scope to preferentially offer such opportunities to vulnerable students (including autistic girls) in mainstream schools.

Academic pressures also impacted the girls’ experiences. Girls frequently referred to performance anxiety and, whilst this is not solely an autistic trait, a feminine predisposition to anxiety (Albano & Krain, 2006; McLean, Asnaani, Litz, & Hoffman, 2011) may be enhanced by the social difficulties experienced by autistic students. Parents were equally aware of the pressure faced by their children, lamenting that schools were more concerned with meeting attendance and achievement targets than their children’s well-being. Although Sammons et al. (2014) suggest a positive correlation between attendance and achievement, schools should aim to understand reasons behind absence, rather than applying pressure upon individuals who are at higher risk of school refusal (Heyne & Sauter, 2013). Educational establishments should also not ignore other statistically significant indicators of achievement (e.g. consistent behaviour policies, an emphasis on learning and the value of pupils) in their ethos (Sammons et al., 2014). If this is not addressed, these vulnerable individuals will not achieve inclusion.

Although adapting the environment is helpful for autistic girls, the quality of interactions in educational provisions was crucial to the success of inclusion, as highlighted in the second theme, ‘Tensions in School Relationships’. Whilst concerns have been raised about the impact of social difficulties on peer rejection (Hebron & Humphrey, 2014; McLennan et al., 1993; Sofronoff et al., 2011), autistic girls in this study were not socially de-motivated (cf. Chevallier, Molesworth, & Happé, 2012) and enjoyed the company of small peer groups (see also Sedgewick et al., 2015). To aid in identifying appropriate peer groups and preventing girls from experiencing the long-term effects of social-isolation (Dean et al., 2014; Hart, 2013), it may be helpful for schools to develop children’s social skills using socially motivated educational strategies (Moyse & Porter, 2015; Myles & Simpson, 2001). Yet whilst peer conflict was commonly reported in the interviews, it was the attitudes of school staff that were of particular significance to the process of exclusion.

Teachers and school staff who demonstrated flexible, accepting and approachable attitudes were viewed as more inclusive than those who did not. Teachers have considerable influence over the school experience, for they (outside of parents) are one of the most significant presences in a child’s life (Hewitt, 2007; Woolfson & Brady, 2009). When teachers harbour poor pre-conceptions of difference, it can lead to detrimental outcomes for SEN children (Sciutto, Richwine, Mentrikoski, & Niedzwiecki, 2012) and enhance conflict (Wainscot, Naylor, Suitcliffe, Tantan, & Williams, 2008). Further, whilst teachers’ prior knowledge of SEN may encourage attitudes towards inclusion (Avramidis, Bayliss, & Burden, 2000; Park & Chitiyo, 2011; Robertson et al., 2003), this may not be integrated within their practice (Woolfson & Brady, 2009). It is therefore important for efforts to be directed towards ensuring that teachers have positive attitudes towards their autistic pupils and that this translates into practice. If not, the negative interactions evident in the girls responses will continue to feature, and could have detrimental knock-on effects (e.g. on their mental health) (May, Cornish, & Rinehart, 2014).

Communication was central to the parental experience. Parents reported that when efficient communicative strategies were used, this increased their confidence in the school’s ability to include their child (see also Tobias, 2009; Tobin et al., 2012). Despite the benefits of establishing relationships with parents, parents reported a distinct absence of efforts to do so across educational provisions. Parents felt that schools often communicated poorly or were unresponsive to requests for support. Further, some parents began to fear incoming school phone calls due to their negativity (see also Dillon et al., 2016). If schools perpetuate negative interactions with families, parents feel isolated and worry that they are considered ‘bad’ parents (Tobin et al., 2012). Although busy educational environments (especially in mainstream schools) may prohibit consistent contact, schools must include families in their children’s educational experiences. Communication is indicative of satisfaction for SEN families (Porter, Georgeson, Daniels, Martin, & Feiler, 2014), so schools should persevere in establishing good working relationships.

This absence of communication highlights an undercurrent of the third theme, ‘Problems with Staff Responses’. Girls reported that staff were often unable to understand their needs, which led to conflict and restricted access to support. While it may be easier to identify crisis in autistic boys (due to their more overt presentation) (May et al., 2014; Wagner, 2006), schools must not judge children based on preconceptions. While the current study did not assess the girls’ anxiety, many in the sample expressed fears over performance, peer acceptance and an increased perception of difference during classroom activities (see also Helverschou, Bakken, & Martinsen, 2011; White, Bray, & Ollendick, 2012). For autistic girls undergoing the stress of adolescence, this can potentially lead to detrimental outcomes, such as the externalisation of this internal stress (Kleinhansa et al., 2008). This can manifest in withdrawal, panic attacks or aggression, which could be missconstrued as behavioural problems (Helverschou et al., 2011) and possibly lead to exclusion. It is important, therefore, for schools to adopt more inclusive strategies for autistic students (especially girls) to allow them to cope without fear of retribution. Whilst both the girls and their parents expressed a desire for more targeted support, it was felt that an increased presence of knowledgeable staff that could listen and understand their needs would have been beneficial. Although supporting autism is to become a stronger feature in initial teacher training in England (from 2018; Department for Education, 2016), this needs to address areas in which autistic students are most vulnerable to improve inclusion for all.

It is also important that educationalists address the quality of support offered before and after exclusion. Girls and parents both felt that, currently, what the schools provided was either unhelpful or non-existent. Before exclusion, girls experienced either reduced timetables, managed moves or a combination of these. Whilst reduction in contact time was initially successful, it did not prevent their withdrawal from education. Similarly, girls on managed moves were initially positive about this option, but were given unachievable targets that eventually contributed towards their exclusion. Following school exclusion, support from the school largely vanished, with only two children being offered home tuition (which was unsuitable for their academic needs). This lack of consideration regarding transitional stages further isolated families from educational opportunities and heightened the anxiety of the girls and their parents. It is therefore recommended that, if exclusion from school is the only option, post-exclusion transition plans are developed; to help mitigate against the substantial negative outcomes associated with school exclusion.

Whilst the results of the current study highlight several important factors regarding the school exclusion experiences of autistic girls, the themes identified from the interviews are not specific to girls. Although there is a strong need to focus research efforts on girls and women on the autism spectrum (Pellicano et al., 2014), inappropriate school environments, a lack of staff understanding and breakdowns in relationships have been repeatedly raised by parents and young autistic people (mostly boys) in other studies, albeit in different environments (e.g. Brede et al., 2016; Makin et al., 2017). Future studies could investigate if/how variables such as social anxiety and friendship affect exclusion in girls compared to boys, as while the overall themes may be similar, the fine-grained content may differ. For example, while both boys and girls on the autism spectrum can face difficulties with peer relationships and mental health, the form these difficulties take are gender-dependent; with girls facing more relational conflict and anxiety, as in the neurotypical population (Albano & Krain, 2006; Crick & Grotpeter, 1995).

It is also important to address the limitations of this research. Girls were difficult to source; given the problems with accurately diagnosing autism in females (Gould & Ashton-Smith, 2011), and the fact we were sampling a very specific sub-group of this population (autistic girls who had been excluded from school). This resulted in the inclusion of one student who had not received a confirmed autism diagnosis. This is, however, likely to be a common occurrence in autistic girls who are frequently undiagnosed (either through a lack of identification, costs or due to issues with a male biased presentation; Gould & Ashton-Smith, 2011). The lack of a confirmed diagnosis does not necessarily mean that this participant should have been excluded from the study; professionals had recognised her high levels of autistic traits, but her parents had opted not to pursue the referral further.

It is also likely that the girls taking part in this study – who had confirmed diagnoses and were excluded from mainstream provision – represent a specific sub-group of girls on the spectrum. This is because many autistic girls are overlooked altogether, and those who are identified may remain in their original educational provision. Whilst this limits the generalizability of these findings, the fact that the themes identified from the interviews so strongly echo those of earlier studies (e.g. Brede et al., 2016) supports the idea that they are not entirely unique to this group (despite these girls potentially having a more ‘severe’ presentation and symptomology than other autistic girls).

A further limitation is the absence of a comparison group of typically developing girls to distinguish nuances in the autistic experience. A study on exclusion experiences of typically developing girls (Osler, 2006) highlighted similar themes to those identified in this study (e.g. interpersonal conflict, lack of support). However, without a comparison group, it is not possible to determine whether factors such as autistic symptoms, co-occurring mental health issues or even teacher’s attitudes or behaviours towards autistic girls, differ significantly from the neurotypical student population.

Conclusion

Consulting excluded autistic girls and their parents permitted insights into a relatively understudied area and several recommendations can be proposed. First, research needs to explore how mainstream schools can adopt the flexibility seen in AP into their practice. This will not be easy, as mainstream schools have less specialised teachers and larger class sizes (and therefore larger teacher–pupil ratios). However, it appears to be essential for successful inclusion. Second, educational provision should actively promote positive, inclusive attitudes in their staff. Importantly, this needs to come from school leaders, who are best placed to influence the ethos of their staff (e.g. Charman et al., 2011; Horrocks, White, & Roberts, 2008). Schools must also try to establish positive relationships with families; not only to reduce the burden placed on them but also to access their valuable knowledge (McNerney, Hill & Pellicano, 2015). Finally, it is recommended that – if exclusion from school is the only remaining option – educational establishments develop efficient transition plans (for before and after exclusion). This should be implemented for all children, but especially those on the autism spectrum; to protect these individuals and their families from the detrimental impact of exclusion from school.