National education policy in England (and more generally across the UK) pursues avowedly inclusive aims, but within the context of a highly demanding 'standards' agenda which focuses on meeting targets for raising the attainments of students to specified levels. Not surprisingly, this apparently twin-track approach has led to some concerns as to how schools can reconcile the imperatives to which they are subject and whether increasing their inclusiveness might not at the same time reduce their capacity to produce good outcomes for their students. Although there are substantial reviews of research on the impact of inclusion for pupils with special educational needs (SEN), to date little is known about the impact of inclusion on the academic and social outcomes for pupils without SEN. This is an important aspect of the inclusion debate at a time when questions are being raised about the viability of inclusion (Ofsted, 2000) and when some teachers are expressing concerns about the increased inclusion of particular groups of pupils, especially those with emotional and behavioural problems.
The aim of this review is to explore empirical evidence about the relationship between the inclusiveness of a school and the outcomes it produces for its student population, especially the population of students without special educational needs. This is a significant gap in the evidence base which is currently being used by educators to inform the inclusion debate. The danger of leaving this gap unfilled is that policy and practice will be developed on the basis of an enthusiasm for inclusion or an antipathy towards it, neither of which is informed by robust evidence.
The aims of the review are closely linked to those of a major study that we recently completed for the DfES on the relationship between inclusion and pupil achievement in English schools (Dyson et al., 2004) and findings from this study complement those of the systematic review. However, the finding of this major UK study have not been included in the present systematic review since the study was still in progress at the time of the review.
This review seeks to answer the following question:
What evidence is there that the inclusiveness of schools has impacts on outcomes for the students without special needs in those schools?
There is a large volume of literature on the impact of inclusion for the pupils included rather than those without SEN and this falls outside the scope of this review. 'Population inclusivity' is defined in this review as what the Audit Commission has recently called 'presence' (Audit Commission, 2002). This refers to the inclusion in a regular school population of students who in otherwise comparable schools might be placed outside the mainstream. It excludes evidence relating to inclusion of pupils with English as an additional language (EAL) or the inclusion of pupils from ethnic minorities. In addition, although very important, the review ignores evidence of the impact of inclusivity on the following groups: teachers, head teachers, managers and other school-related staff, parents and carers of pupils with SEN and/or of pupils without SEN. The review also only focuses on pupils aged 5-16.
Identifying and describing studies
We reviewed studies which met the following criteria:
- They are in English (given limitations on available resources).
- They report on the results of empirical research rather than being purely theoretical or exhortatory.
- They are concerned with the phases of compulsory schooling.
- They report the outcomes of the 'intervention' (i.e. an increase in population inclusivity).
- They report these outcomes in relation to students without SEN or whole-school populations (but not simply in relation to students with SEN).
- They report robust evidence of the impact of the intervention:
- through a longitudinal study of one school or
- by comparison with a similar but less inclusive school (with a lower level of population inclusivity) or
- by comparison between different conditions within the same school (such as more and less inclusive classes) or
- by some other equally robust means.
- They are concerned with the impact of inclusion on pupils' personal, social and/or academic outcomes.
Searches through electronic databases constituted the main strand of our search strategy. These covered journal articles, books and book chapters, conference papers and proceedings, theses, dissertations and reports. The strategy was refined in consultation with members of the Advisory Group and members of staff at the EPPI Centre. A set of agreed search terms to guide the review was subsequently tested out in a particular database and new terms were added or existing terms were altered or removed. When the Review Group was satisfied that, by using the devised set of terms, all the available items stored in the databases could be identified, the search was run in all relevant databases.
The inclusion/exclusion criteria were applied to all studies and, when there was uncertainty about the inclusion of a study, the criteria were refined and clarified further. This resulted in a set of very specific criteria and, consequently, the mapping exercise led to the identification of a small number of studies. All studies in the descriptive map were also included in the in-depth review.
The framework adopted for synthesising studies viewed the impact of population inclusivity according to the following variables:
- area of key impact (academic or social outcomes)
- type of school (primary versus secondary schools)
- nature of SEN (adopting four main categories):
- cognition and learning
- behavioural, emotional and social development
- sensory and/or physical needs
- communication and interaction
After completing the searches and selecting publications that met the criteria, the number of studies included in the review was reduced to 26, all of which were subjected to the data-extraction process. All these were evaluation studies, 15 of which were 'naturally occurring' and 11 'researcher manipulated', involving some form of experimental design. The majority of these studies were carried out in the United States of America (USA) (N = 22); there were also two studies from Australia, one from Canada and one from Ireland. There was a slight preponderance of studies in which the included pupils experienced difficulties in the area of cognition and learning, although pupils with other types of difficulties were also mentioned. The majority of the studies (N = 21) focused on academic outcomes and these were measured in a wide variety of ways, including class tests, national examinations and teacher ratings.
Virtually all the studies (N = 21) focused on the outcomes of inclusion for primary-aged pupils. The nature of the inclusion experienced by the pupils with SEN was described in different ways. In some studies (N = 16), this was described as the proportion of pupils with SEN in a mainstream class, whereas in others (N = 14) it was described as the number of hours per week (or day) that a child with SEN spent in a mainstream class. Some studies described inclusion in both ways.
Many of the 26 studies focused on more than one type of outcome. Although 12 focused solely on academic outcomes and five addressed social outcomes, there are nine that considered both academic and social outcomes. This means that the 26 studies yielded more than 26 findings. Indeed there are 40 findings from the studies that relate to one or other or both of these outcomes.
The majority of studies (N = 24) reported the results of including pupils with difficulties in the area of cognition and learning, although studies, more often than not, refer to groups of pupils with a variety of SEN. Therefore, it is difficult to provide direct conclusions regarding the impact of including pupils with a specific type of SEN on the academic and/or social or other outcomes of all school pupils. However, there seem to be more negative outcomes reported when pupils with emotional/behavioural difficulties (EBD), as a main or additional difficulty, were included compared with the other types of SEN. There were no studies reporting negative outcomes for the majority of pupils in school/class when pupils with physical and/or sensory and communication difficulties were included.
Of these 40 different findings, nine (23%) indicate that there was a positive academic and/or social impact on non-SEN pupils of including pupils with SEN. Six (15%) suggest a negative impact, 21 (53%) a neutral impact and four (10%) suggest mixed impacts. Taken as a whole, the findings indicate that placing children with SEN in mainstream schools is unlikely to have a negative impact on academic and social outcomes for pupils without SEN.
Further analysis indicated the following:
- Findings are slightly more positive for academic rather than social outcomes.
- At secondary level, where there are very few studies, outcomes are slightly more mixed.
- Some of the findings suggest that the inclusion of pupils with SEN in primary schools can have a positive impact on the achievement of their mainstream peers, particularly if the support offered to the pupil with SEN is well managed.
There is no evidence about the impact of placing pupils with SEN in mainstream schools on achievement in different curriculum areas. Hence none of the findings indicate that the 'inclusion effect' is more or less serious for any one particular curriculum area.
There is nothing in this review or in the DfES study in which we have been involved (Dyson et al., 2004) to suggest that the commitment to inclusion in relation to pupils with SEN is likely to have a significant impact on overall levels of attainment in mainstream schools. This suggests that the government, LEAs and schools should have no concerns about pursuing the inclusion agenda. However, policy-makers should pursue inclusion policies in an informed way, consulting with all relevant stakeholders at all times. In addition, the lack of studies in the secondary sector suggests that schools and LEAs should pursue the inclusion agenda with some caution and, where possible, commission research that can explore this complex area in more depth.
In relation to practice, this review suggests that schools, parents and LEA professionals should have no concerns about the impact of inclusion on achievement, especially in primary schools. This applies across all four categories of SEN. However, these studies and other research reviews (Harrower, 1999; Farrell, 2000) indicate that successful inclusion does not occur in a vacuum. Parents, teachers and pupils need to be fully committed to the idea; programmes of work have to be carefully planned and reviewed regularly; and support staff need to work flexibly as a team and receive appropriate support and training.
There are a number of implications for further research that arise from this review. In particular, the lack of studies that focused on secondary schools indicates that this is a key area that future studies should address. In addition, further studies could focus on the impact of including larger groups of students with SEN. There is also a need for more longitudinal research that could trace the relationship between inclusion and the achievements of non-SEN pupils over time. Finally, more studies should focus on the views of pupils without SEN about inclusion. Given current interest in involving users in planning, carrying out and evaluating research, it is surprising that so few studies actually focus on the pupils' views.
Dyson A, Farrell P, Polat F, Hutcheson G, Gallanaugh F (2004) Inclusion and Pupil Achievement. London: DfES.
Farrell P (2000) The impact of research on developments in inclusive education. International Journal of Inclusive Education 4: 153–162.
Harrower JK (1999) Educational inclusion of children with severe disabilities. Journal of Positive Behaviour Interventions 1: 215–230.
Ofsted (2000) Educational Inclusion: Guidance for Inspectors and Schools. London: Ofsted.
This report should be cited as: Kalambouka A, Farrell P, Dyson A, Kaplan I (2005) The impact of population inclusivity in schools on student outcomes. In: Research Evidence in Education Library. London: EPPI Centre, Social Science Research Unit, Institute of Education, University of London.