Research Evidence Relating to Inclusive Education

13.6.4 Educational achievement and psychosocial development

In general, methodologically sound studies have come up with mixed results, the majority reporting either positive effects or no differences for inclusion. (Some would argue that if there are no differences, this is also an argument for inclusion: why have segregated education programmes when they are no better than placement in regular classes?) The following is a representative sample of research carried out in this area.

A Canadian study of 3rd grade students with ‘at risk’ characteristics (e.g., learning disabilities, behaviour disorders) compared the impact on achievement of a multi-faceted inclusive education programme. The intervention group (N=34) received all instruction and support in general education classrooms, while the comparison group (N=38) received ‘pull-out’ resource room support. The intervention group also received a programme that included collaborative consultation, cooperative teaching, parent involvement and adapted instruction in reading, writing and mathematics. The comparison group continued using general education teaching methods characterised by whole-class instruction and minimal cooperation between the general and special teachers. Significant effects were found in the writing scores for the inclusive education group. The general education students were not held back by the presence of the at-risk students in the classroom; on the contrary, their reading and mathematics scores benefited from the additional interventions offered by the programme (Saint-Laurent et al., 1998).

A USA study addressed the effects of an inclusive school programme on the academic achievement of students with mild or severe learning disabilities in grades two - six. The experimental group comprised 71 learning disabled students from three inclusive education classrooms. In these classrooms special education teachers worked collaboratively with general education teachers, each student’s programme was built upon the general education curriculum, and instructional assistants were used to support the SWSEN. The control group of 73 learning disabled students were in classrooms which were to become part of the inclusive programme, but in which the students received traditional resource class programmes. Results showed that the students with mild learning disabilities in the inclusive classrooms made significantly more progress in reading and comparable progress in mathematics, compared with those in the resource classes. Students with severe learning disabilities made comparable progress in reading and mathematics in both settings (Waldron & McLeskey, 1998).

In a study carried out in Hawaii, the effects of placement in general education classrooms or in self-contained special education classrooms on the social relationships of students with severe disabilities were reported. Nine matched students were studied in each of the two placements. The results showed that those who were placed in the general education classrooms had higher levels of contact with non-disabled peers, received and provided higher levels of social support, and had much larger friendship networks (Fryxell & Kennedy, 1995). These results were echoed in growing research evidence suggesting that children who attend special schools are more likely to experience bullying than children who attend mainstream settings, and that inclusive education is a key factor in reducing or eliminating bullying (Rose et al., 2011). As noted by Cologon (2013), too, research evidence also suggests that genuinely inclusive education allows SWSEN to experience greater social interaction and build and develop friendships they may not have encountered otherwise (Finke et al., 2009; Antia et al., 2011), and to engage in less disruptive behavior (Finke et al., 2009; Mogharreban & Bruns, 2009; Stahmer et al., 2011).

One of the most comprehensive studies of the effects of inclusive programmes on the development of social competence in students with severe disabilities is that reported by Fisher & Meyer (2002). In a matched-pairs design, 40 students were assessed across two years of inclusive versus self-contained special education classrooms. Those in the inclusive programme made significant, albeit small, gains on measures of social competence, compared with students in self-contained classrooms

Several Dutch studies have found better academic outcomes for SWSEN in inclusive classrooms, compared with those in segregated settings (de Graaf et al., 2013; Peetsma et al., 2001). For example, another Dutch study reported on the differences in academic and psychosocial development of at risk students in special and mainstream education. It was found that those in special education classes did less well in academic performances and that these differences increased as the students got older. In psychosocial development, variables such as social behaviour and attitudes to work also favoured students in regular classes (Karsten et al., 2001).

A UK study compared the outcomes for adolescents with Down syndrome of similar abilities but educated in mainstream or in special schools. The results showed no evidence of educational benefits for those in segregated settings, despite the higher teacher-student ratios. Those who attended their neighbourhood mainstream schools made significant gains (two-three years) over their special school peers in expressive language and in academic achievement (Buckley, 2006). Note, however, that this study has not been published in peer-reviewed journals.

There is substantial evidence that inclusive education enhances the communication and language development in both SWSEN and their non-disabled peers (Finke et al., 2009; Fisher & Shogren, 2012; Hart & 
Whalon, 2011; 
Stahmer et al., 2011).

A 2004 study in England showed that the presence of relatively large numbers of SWSEN (not analysed by category) in ordinary schools did not have a negative impact on the achievement of general education learners at the local education authority level. Rather, attainment seemed to be largely independent of levels of inclusive education. Other factors, such as socio-economic status, gender, ethnicity and language, seemed to be much more significant. Furthermore, the researchers found evidence that SWSEN were making good progress academically, personally and socially. They also found some evidence (chiefly in the views of teachers and pupils) that inclusion can have positive effects on the wider achievements of all learners, such as on their social skills and understanding. On the other hand, they also found some indications that having special educational needs might be a risk factor for isolation and for low self-esteem (Dyson et al., 2004).

Another English study produced similar results, finding no evidence that the presence of higher proportions of learners with special educational needs (also not analysed by category) in secondary schools lowered the performance of general education students. Indeed, as with the previous study, many educators in those schools believed that the inclusive education strategies used actually contributed to improved overall educational achievement (Rouse & Florian, 2006).

The impact of inclusion on the achievement of general education elementary school students was also investigated in a US study reported by Sharpe et al. (1994). Two groups were studied: 35 students whose classes included five students with learning disabilities, and 108 who had no classmates with special educational needs. Measures of academic achievement were taken over a three-year period at three points: pre-inclusion, inclusion and post-inclusion. The researchers found no significant differences between the two groups of learners on basic skills of language arts, reading and mathematics. Certainly, there was no evidence of any decline in the academic or behavioural performances of learners in the inclusive setting.

Similar findings were reported in a recent Canadian study. Friesen et al. (2009) analysed data from British Columbia to compare the performance of successive cohorts within every public elementary school in B.C. (as measured by the change in individual test scores between grades 4 and 7), to see if the proportion of disabled peers makes any difference to the achievement of non-disabled students. They concluded that ‘Attending school with a higher percentage of students with disabilities is found to have only extremely small and statistically insignificant effects on the reading and numeracy achievement of non-disabled students’ (p.1).

A range of other studies confirm the previous findings that students who are not disabled benefit academically from inclusive education, with equal or better academic outcomes compared to those in non-inclusive settings (Dessemontet & Bless, 2013; Farrell, et al., 2007; Kalambouka et al., 2007; 
Kliewer, 2008; Odom et al., 2011).

Mixed and negative findings. In one of the earliest meta-analyses, 50 studies compared general (i.e., inclusive) and special class placements. It was found that placement in general classes resulted in better outcomes for learners with mild mental retardation, but poorer outcomes for students with learning disabilities or behavioural/ emotional problems (Carlberg & Kavale, 1980).

A comprehensive review of inclusion research involving students with autism also reported mixed results. In one set of studies, those who were fully included (a) displayed higher levels of engagement and social interaction, (b) gave and received higher levels of social support, and (c) had larger friendship networks. This was counterbalanced, however, by another study that found that these students were more frequently on the receiving, rather than the giving, end of social interactions. The review also described a study in which the effect of inclusive education, compared with segregated education, on the language ability of autistic students was evaluated. The fact that there were no differences between the two placements was interpreted as supporting inclusion, since segregated placements were shown to be of no benefit (Harrower & Dunlap, 2001).

Peetsma et al. (2001) reported on a longitudinal study on the effects of inclusion on the academic and psychosocial development of Dutch students with mild learning and behavioural difficulties. The results were that, after two years, only a few differences in development were found: students made more progress in mathematics in inclusive settings, but school motivation developed more favourably in special schools. After four years, students in regular schools had made more progress in academic performance, whereas there were no differences in psychosocial functioning. However, a small–scale qualitative study, which was incorporated as part of the major study, showed that students with psychosocial problems made somewhat better progress in special education than in regular education, pointing to the need to pay attention to the psychosocial development of students with mild disabilities when they are placed in inclusive settings.

Several studies have found that quality of instruction, rather than placement, is the most important predictor of student achievement. For example, in one study of mathematics achievement of students with hearing impairments, placement in regular or special classes did not seem to impact on achievement. Rather specific features of quality placement included a supportive teacher, regular and extensive reviews of material, direct instruction and a positive classroom environment (Kluwin & Moores, 1989).

These findings were echoed in a report by Ofsted (2006) on English provisions for SWSEN. It considered that the most important factor in determining the best outcomes for pupils with learning difficulties and disabilities was not the type but the quality of the provision. Effective provision was distributed equally in the mainstream and special schools visited, but there was more good and outstanding provision in resourced mainstream schools than elsewhere.

One final point of mixed evidence can be found in a report from the European Agency for Development in Special Needs Education (2003). This suggested that inclusion generally works positively at the primary school level, but serious problems emerge at the secondary level. This was attributed to increased topic specialisation, the different organisation of secondary schools, and the increasing gap between the achievement of SWSEN and other students with age.