Introduction

Discussions about inclusive education have moved through several phases and involved many different discourses (Dyson 1999). For example, arguments for both human rights and effectiveness have been used to justify it. Still, the concept of inclusive education remains somewhat ambiguous; it can and is being used in different ways to serve different political purposes. Although an unequivocal definition might not be possible, a general international consensus exists that school systems should aim to be more inclusive (Mitchell 2005).

One trend is clear in the discussion on inclusive education: whereas much of the earlier discussion originated from within the field of special education, today inclusive education is more widely understood and is not limited to including children with special needs and/or disabilities (Mitchell 2005). Inclusive education is conceptualized more as an effort to build school systems that welcome all children who are currently excluded from education. A primary objective in this effort is to minimize the structural, attitudinal, and pedagogical barriers to learning that today leave many groups of children outside of mainstream educational services. Who these excluded children are varies greatly across countries and educational systems.

A major group excluded from education, especially in poorer countries, is children with disabilities, but many other groups attract attention in today’s inclusive educational policies, including girls, children in crisis, and the children of nomadic groups and cultural or ethnic minorities (UNESCO 2009). When inclusive education is understood as an overarching principle, its goals become parallel to those of UNESCO’s Education for All (EFA) process. In short, inclusive education aims at a good “school for all”; this goal comes very close to the traditional Nordic interpretation of the direction in which schools and special education should be steered (Dahlgren 1984; Eklindh 1996).

Worldwide, a major challenge in developing inclusive educational systems is for schools to manage the goals of increasing access, responding to increased student diversity, and improving the quality of education, all of which are clearly stated goals of EFA. Thus the challenge is to simultaneously improve the equity or equality of both educational opportunities and learning outcomes. The general purpose of this article is to discuss whether and how this is possible. I approach this issue first by briefly reviewing some findings on the effects that inclusive education and special education interventions have on the learning outcomes of students. Second, I take the Finnish education system as an example, discussing its comprehensive education reform and current learning outcomes and comparing them to some major international trends in improving quality in education. I conclude by arguing that inclusive education is a reasonable goal, as excellence and equality are not necessarily competing objectives in educational reforms.

The effects of inclusive education and special educational interventions

As with research on any comprehensive educational approach, it is very difficult to find conclusive evidence on the effectiveness or outcomes of inclusive education. However, inclusive education is sometimes justified because it is more cost-effective than traditional types of services (UNESCO 2005, 2009). More conclusive evidence is needed, especially about the costs and the learning and social outcomes of fully-inclusive educational services and other more traditional types of service provision on a nationwide scale. As many of the findings on inclusive education are related to evaluations or studies of small-scale projects, those findings cannot be directly generalized and used to justify national-level policy decisions. A specific challenge relates to methodology. Although the process of measuring learning outcomes and comparing them internationally is fairly straightforward, explaining the variance in the outcomes remains largely speculative, as few if any studies use experimental control designs and they are usually based on cross-sectional, not longitudinal, research designs.

At a practical level, the problem is a lack of concrete comparisons. For example, we may observe that special education instruction leads to less than satisfactory learning and social outcomes, but we do not know what the individual students’ outcomes would have been without those special education services. As a second example, some may point to the high costs of building a truly inclusive educational system with all reasonable accommodations in place, but we cannot estimate what the costs would be if only traditional services were provided.

However, at least two research perspectives can give us some answers. We have findings on the effects that the level of inclusivity has on learning outcomes, and of different special education interventions. In England, Dyson et al. (2004) used a very large sample of student records (over 500,000) to examine the level of inclusion, which they defined as the number of pupils identified as having special educational needs who were studying in mainstream classes. Did that level have any effect on the school’s social or academic outcomes? Their major finding was that the effects of inclusivity were very small, although slightly negative. Further, most of the negative effect was explained by the effect of socio-economic status: the schools that were more inclusive tended to be those serving students with a more disadvantaged socio-economic background.

Kalambouka et al. (2005) presented somewhat similar outcomes in a review of other studies on the effects that inclusivity had on the academic and social outcomes of education. Although the 40 different findings in the analysis were somewhat mixed, the effects of inclusivity were mostly neutral (53%) and the proportion of findings with a positive impact (23%) was higher than the proportion with a negative impact (15%). The authors concluded that placing children with special educational needs (SEN) “in mainstream schools is unlikely to have a negative impact on academic and social outcomes for pupils without” those needs (Kalambouka et al. 2005, 4).

The second approach is to consider research evidence on the effects of special education. The problem with many experimental studies is that they use relatively small samples that are not representative of the national educational system from which they are taken. One way around this obvious challenge is to collect findings from several studies in a meta-study, which systematically compares the findings (usually expressed as effect sizes) found in different studies and comes to an average effect across a large sample of studies. As the conclusions in a meta-study are based on large quantities of information, they are more powerful indicators of general trends than are the findings in a single study.

Forness (2001) published a meta-study based on several previous meta-studies. One interesting finding was that the interventions directly addressing learning (e.g. direct teaching, reading comprehension strategies, and behaviour modification) seem to be more effective than those targeted to theoretically defined and assumed prerequisites of learning (e.g. psycholinguistic training and modality instruction). Furthermore, contradicting a common expectation, reductions in class size had only a minimal positive effect, and the effect of placement in a special class was not only small, but also negative. The author concludes that “Special education interventions that emphasize ‘education’ are far more effective than ‘special’ education practices that attempt to treat special education students by overcoming negative effects on learning caused by a variety of hypothetical and unobservable constructs (e.g. modality and perceptual-motor factors)” (Forness 2001, 192).

The Finnish education system: Policies for quality and equality of opportunity

The Finnish educational system has received a lot of attention in recent years, mainly because of its positive results in international learning outcomes comparisons, such as the OECD Programme for International Student Assessment (PISA). From the Finnish perspective, the major finding has been that Finnish ninth-grade students are top achievers in reading literacy (PISA 2000), mathematics (PISA 2003), and science (PISA 2006) as reported by the OECD (2001, 2004, 2006). Two interesting observations regarding these results are that the average performance of Finnish students is good, and that the results do not vary much between schools or between students coming from different socio-economic backgrounds (Kivirauma and Ruoho 2007). Thus, the Finnish educational system seems to have succeeded in producing learning outcomes of high quality, along with excellent geographical and social equity.

The growing number of immigrant students, or students whose mother tongue is not Finnish or Swedish (the two official languages), clearly do less well on tests than other students, especially in reading literacy. However, the difference in mathematics is relatively small, and the immigrant students’ mathematics results are very close to the OECD average for non-immigrant students and higher than the average for non-immigrant students in the United States (Itkonen and Jahnukainen 2007). Results on gender equity are less promising; although the difference between the average scores in mathematics and science for males and females is small in Finland compared to the OECD countries, a large difference remains between the scores of males and females on reading literacy. In 2006 this gender gap was the second highest among the OECD countries: 51 points, which corresponds to 0.6 standard deviation units (Kupari et al. 2004; Arinen and Karjalainen 2007).

Another very interesting finding is that the positive difference between the performance of Finnish students and the OECD average is highest among students in the lowest achievement percentiles. As Fig. 1 shows, the Finnish students’ performance is clearly above the OECD average across the range of different achievement levels, but the difference is greatest among students in the lower achievement levels, which are represented as percentile groups in Fig. 1.

Fig. 1
figure 1

Reading comprehension performance of Finnish 9th graders by percentile, compared to the OECD student average in 2003 (Moberg and Savolainen 2006)

Full size image

Many explanations have been offered for the Finnish success story, and several authors have recently discussed the issue. There are no easy explanations, as the result is probably caused by many interacting factors. For the purposes of this paper, however, we can identify some factors that make Finland stand out from many OECD countries. Two perspectives deserve a closer look: the role of educational policies and the role of existing learning support services, especially part-time special education.

Educational policy for equity and excellence

Finland’s development into an educational society is quite recent and is based on the establishment of the Nordic type of welfare state. Four important features of this Nordic model are citizens’ equal rights, the public authority’s responsibility for the welfare of all citizens, the narrowing differences in income and gender equality, and the goal of full employment. The educational system for the welfare state began to develop when compulsory education was extended to 9 years, which led to the comprehensive school reform implemented in Finland in the 1970s, a few years later than in Sweden (Antikainen 2006).

This reform was quite radical. In earlier years, pupils finishing fourth grade were divided into two streams: the grammar school and the civic school. The first was sometimes described as suitable for the “theoretically gifted,” and the second was for the “practically gifted”. The ideology that guided the comprehensive school reform saw education as an important way to carry out broader social reforms. Comprehensive school reform was aimed at increasing socio-cultural, geographical, and gender equity (Kivirauma et al. 2006).

The reform was preceded by active political debate in Finland, just as in Sweden a few years earlier. Interestingly, some of the arguments that arose in Finland in the 1960s were very similar to those in the more recent debate on inclusive education. For example, before the reform was implemented, some politicians feared that a unified comprehensive school would be a threat to the quality of education, especially from the point of view of the most gifted children. Other fears were also expressed: that a comprehensive school would be too demanding for students from disadvantaged backgrounds, that diversity would lead to discipline problems, and that these troubles would be a threat both to the learning of the majority of students and the working conditions of the teachers. The debate over comprehensive schooling was active and quite politicized (Rossi 2007). The difference to today’s debate on inclusive education is that it is not driven so much by clear political interests, but rather by different interest groups like teacher organizations and disabled people’s organizations.

One strong argument in favour of comprehensive schools in Finland and Sweden was that they would better utilize “talent reserves”; critics saw that when school systems divided younger students into the theoretical and practical streams, none of them—especially those from lower social-class backgrounds—got the opportunities to learn according to their full potential (Husén 1962). The comprehensive school reform was gradually introduced in Finland in the 1970s, starting in the northern parts of the country and gradually moving south.

The principles of policy reform that were implemented as Finland’s comprehensive school system developed have proved to be different from those adopted in many other industrialized countries. According to Sahlberg (2007), three major strategies of educational policy reform have been introduced around the world to improve the quality of education: (1) education has been standardized, (2) the focus on literacy and numeracy has increased, and (3) consequential accountability has been introduced. The educational policies in Finland have taken different routes in all three areas. For example, while many countries have introduced centrally-prescribed performance standards for schools, teachers, and students as a mechanism to increase quality, the Finnish system has continued to build on good practices, school-based curriculum development, and the setting of learning targets. In fact, the measurement of student achievement is largely treated as the responsibility of individual schools; students in the Finnish comprehensive school do not take standardized achievement tests. Indeed, the only nationally standardized test is the matriculation examination at the end of 12th grade in the academic track. Furthermore, it is common practise to give numerical grades to students only from the sixth grade onwards. Until then, teachers provide descriptive verbal evaluations. At the end of each school year, parents are invited to come to the elementary school, with their children, to discuss the children’s learning progress; this is also seen as an opportunity for the students to evaluate his/her own work.

Instead of a strong focus on literacy and numeracy, Finnish educational policy has taken a broader approach, giving at least equal value to various aspects of individual growth, creativity, knowledge, and skills. Early support for children with numeracy and literacy problems, however, has been a specific objective of the part-time special education introduced in Finland as a part of the comprehensive education reform (Kivirauma and Ruoho 2007).

The third strategy, consequential accountability, has not been adopted in Finland at all. This method connects school funding to performance, i.e. government support to the school can be reduced if the school does not meet achievement requirements, which are often measured with the standardized testing of students. Instead, the Finnish educational policy has been based on what Sahlberg (2007, 152) calls “intellectual accountability with trust-based professionalism”. In the intellectual accountability approach, schools are accountable for learning outcomes and education authorities are accountable to schools for making their work possible. One reason this kind of flexible policy seems to work well is the “culture of trust” in the Finnish school system, in which “education authorities and political leaders believe that teachers, together with principals, parents, and their communities, know how to provide the best possible education for their children and youth” (Sahlberg 2007, 157).

Some authors have argued that an accountability policy based on the comparative success of schools is likely to use sanctions, such as reducing government support for failing schools. On the other hand, many countries that have demonstrated good outcomes in international comparisons, like Finland and South Korea, have adopted the opposite approach: giving additional resources to schools struggling with achievement because they have more disadvantaged students (Itkonen and Jahnukainen 2007).

The comprehensive school reform also faced many foreseeable challenges, created mainly by the increased diversity. Two approaches were used to counter these challenges. First, teachers were educated and encouraged to adopt principles of differentiated teaching (Kangasniemi 1997). Second, a new system of learning support was created: part-time special education. After the reform, all teacher education was moved to universities; by the end of the 1970s, all teacher training programmes led to a master’s degree, which is still an exception internationally. Elementary school teachers and special education teachers have their own degree programmes, and subject teachers enrol in a 60-ECTS (European Credit Transfer and Accumulation System) pedagogical programme in conjunction with their major and minor subject studies. Raising the academic status of teacher education has probably contributed to the already quite high value given to the teaching profession in Finland that is evident in many studies (see, for instance, Nummenmaa and Välijärvi 2006). In 2008, Helsingin Sanomat, a Finnish newspaper, published interesting anecdotal evidence on the status of teachers. It commissioned a Gallup study of Finnish people aged 15–74 and asked them about the importance they placed on occupation in selecting a spouse or partner. Pollsters asked the interviewees to name their five favourite occupations in this regard. For female respondents, having a husband in the teaching field was in third position, right after doctor and veterinarian; for males, having a teacher for a wife ranked first, before nurse, doctor, and architect (Helsingin Sanomat 2008).

The importance of good teachers for quality education is also emphasized in a recent comparison of 25 school systems worldwide by McKinsey and Company (2007). This study sought to determine the typical attributes of the world’s top ten educational systems to explain why their students perform so well compared to others. Two of the major explanations were related to the quality of teachers. The top ten systems seem to get the “right” people to become teachers because they have enough applicants and use more elective mechanisms to select participants to begin teacher education programmes; the selection usually occurs before the programmes begin. In these countries, teachers’ starting salaries are at or above the OECD average (relative to per capita income), though most of them spend less on education than the OECD average. The other common feature of the top ten systems is that they seem to be able to develop their teachers into effective instructors. McKinsey and Company list some of the ways they accomplish this: the teachers build practical skills during initial teacher training, coaches are placed in schools to support teachers, educational systems manage to select and develop effective instructional leaders, and teachers are enabled to learn from each other.

Increasing the learning outcomes by supporting the disadvantaged

Finland’s comprehensive school reform brought with it changes in the special education system. Special education had been a separate system, but pressures have since increased to bring it closer to regular education. A major change occurred in 1968, when the Comprehensive School Act introduced a wholly new form of special education: part-time special education. It was introduced as an essential part of the new educational policy, as planners foresaw that increased heterogeneity in the 9 year comprehensive school would lead to pedagogical challenges that needed a new kind of attention (Kivirauma and Ruoho 2007). What differentiates part-time special education from the traditional types is that students need not be diagnosed with a problem or disability to be eligible for support. Instead, a student can begin getting part-time special education support immediately when any difficulties arise in school, and both the intensity and duration of the support can vary according to individual needs. The traditional special education support that involves an official decision on eligibility for special education, given usually by local school boards, has continued to co-exist with part-time special education, and both types of support have continued to increase in scope (see Fig. 2).

Fig. 2
figure 2

Development of special needs education services in Finland (Moberg et al. 2009)

Full size image

According to the latest statistics (Tilastokeskus 2008), 8.1% of students have been identified as having special educational needs; these students are given instruction either in special schools or special classes, or in an arrangement that is partially or fully integrated with mainstream education classrooms. Three distinctive features in the development of these forms of services can be noted. First, the major reason for the recent sharp increase in the number of students identified as having special educational needs is the increased support provided in integrated arrangements. Today’s integrated learning arrangements include more than half of all the students who have been identified as having special educational needs. Second, the proportion of students being taught in special schools has clearly been decreasing. Third, part-time special education provision has increased continuously, except for a short decline in the middle 1990s. This small decline can probably be explained by the deep economic recession and banking crises experienced during those years, which forced school authorities to cut back on public spending; special education services were among the first to be reduced. This shows how sensitive special education services are to changes in the economy (Jahnukainen 2003).

Typical of the Finnish comprehensive education system is its profile of giving extensive support services, especially for learning problems, through these two types of special education arrangements. These services, particularly part-time special education, are targeted especially toward the first few school years in comprehensive education and toward supporting pupils in learning basic skills in literacy and numeracy (Kivirauma and Ruoho 2007). It would not be reasonable to say that special education is a separate system of education in Finland today. The vast majority of special education support is provided within the mainstream school campuses, where special education teachers are part of the local school staff and have a strong teacher identity. This is related to the way they are trained in two types of programmes. The traditional form of training is a post-graduate diploma: teachers, who have already graduated, with master’s degrees, can complete special education teacher education programme in a year while working part time in the school. The other type of training is direct master’s degree programmes that qualify teachers both to teach elementary school and to take on various special education positions. Training is not organized in disability-oriented streams, but rather is aimed at providing teachers with a wide variety of knowledge and skills that they can apply in various settings and situations.

Many factors account for the increase in special education in its two forms and they can be discussed from various perspectives, but it is impossible to provide any simple set of conclusive reasons. For the purposes of this article, it is sufficient to conclude that the increase is a developmental process in which the special education system has reacted both to the individual learning needs of pupils and to the pressures created by the comprehensive education system, with its increased academic demands.

Although we must be critical toward this rapid increase in special education support and the trend toward labelling more and more children as having special educational needs, we cannot avoid thinking that the support targeted to disadvantaged students may at least partly explain the PISA results that clearly showed a lower proportion of poorly-performing students in Finland than in the OECD countries on average. Further, Finnish students who perform poorly are still doing far better than their poorly-performing counterparts in other countries (Moberg and Savolainen 2006). Two further arguments provide evidence for this hypothesis. First, the performance of Finnish students has not always been so excellent. For example, earlier comparative studies showed that Finnish students’ performance in mathematics improved gradually, from the early 1980s position that was slightly below the international average to the current top position in PISA (see Sahlberg 2007, 160). Second, a Finnish study (Moberg and Savolainen 2008) compared equally representative and comparable samples of ninth graders who were tested with exactly the same reading comprehension instrument in 1965 and 2005. They found, first, that the reading literacy skills of today’s youth are far better than those of students 40 years ago, i.e. a few years before the comprehensive school reform was implemented. Second, the differences between student performance across the achievement percentiles showed the same pattern as found in the PISA studies; in particular, the students into the lowest percentile categories were doing far better than their counterparts 40 years ago (see Fig. 3). In 1965, only 2% of students had received part-time special education support, which at the time was mainly speech therapy in the early school years. In contrast, as many as 29% of the students in the 2005 sample had received part-time special education support by ninth grade (Moberg and Savolainen 2008).

Fig. 3
figure 3

Reading comprehension performance of Finnish 9th graders by percentiles in 2005 compared to that in 1965 (Moberg and Savolainen 2008)

Full size image

These findings suggest that comprehensive school reform has gradually succeeded in narrowing the achievement gap between stronger and weaker students. This increase in the equality of achievement has taken place during a time marked by increasing special education support, particularly part-time special education support.

Discussion

On the basis of the findings presented in this article, four theses can be offered:

  • Special education can contribute to increased equality in learning outcomes.

  • Teachers and special education teachers play crucial roles as facilitators of inclusive education.

  • Classroom practices are more important than classroom settings.

  • Increasing equity and equality can also lead to increased quality.

Differentiation of education and tracking are well-acknowledged sources of educational inequality; however, inequality may not be caused by differentiation per se, but rather, by the way it is implemented. The common use of special education in Finland can serve as an example, for it seems to be linked to the reduction of inequalities, not their growth (Antikainen 2006). Part-time special education seems to succeed here, perhaps because such support can be started quickly, resources can be used flexibly, and students are not stigmatized or excluded from the mainstream.

Teachers play an essential role in quality education. As noted in the recent study highlighting the role of teachers in the top ten educational systems in the world, “The quality of an education system cannot exceed the quality of its teachers” (McKinsey and Company 2007, 16). When compared to the number of studies on different teaching methods or other educational interventions, there are surprisingly few studies on teacher effects on learning outcomes. However, those few suggest that teachers may have a much stronger effect than other factors like class size (e.g. Sanders and Horn 1998), whose reduction is often proposed as a remedy for increasing quality. Developing good standards for teacher education and getting the “right” persons to become teachers should undoubtedly be an essential part of an educational policy that aims to build quality educational systems.

Interestingly, studies suggest that pupil heterogeneity in a classroom seems to have a smaller effect than does teacher quality (Sanders and Horn 1998; McKinsey and Company 2007). This finding probably helps explain why special class placement as such has not been shown to be an effective intervention across multiple studies (Forness 2001). From this, we might conclude that what actually happens inside the classroom, how a teacher teaches, and what kind of interaction students and teachers have may be much more important in producing good learning outcomes than the class arrangement or setting as such. The large-scale study in England (Dyson et al. 2004) clearly supports this hypothesis: increased classroom inclusivity did not have any meaningful negative effect on the learning outcomes of students. Thus, inclusive classrooms that welcome all learners with different abilities and interests should not be regarded as a threat to the quality of learning.

Finally, striving for equity does not pose a threat to quality; on the contrary, an educational policy approach that aims at equity or more equal educational opportunities can lead to increased quality. Both the analysis of the Finnish educational system reform and its results, and the comparison of the countries with the best student achievement outcomes support this claim. Increased equity was the leading philosophical and practical principle in the reform of the Finnish educational system that led to the current comprehensive school reform with a unified curriculum for the first nine school years. Despite the increased heterogeneity, the school system has become one of the top systems in the world and has simultaneously increased geographical and social equity and narrowed the achievement gap between weaker and stronger learners.

The analysis of the top ten educational systems confirms this idea: all the top systems systematically aimed to deliver a good education to every child. As part of this process, these systems not only set high expectations for what all students should achieve, but they also target resources to the low-achieving students who need them the most. All this requires close monitoring of learning outcomes both at the school and student levels (McKinsey and Company 2007). An extensive learning support system, including part-time special education, is a typical feature of the Finnish educational system. There are good reasons to argue that this extensive learning support is an important factor behind the good learning outcomes of Finnish students. Building the part-time special education system was one of the policy decisions on which the educational reform was built. This approach, together with the decision to follow a policy approach different from many other OECD countries with regard to student testing and accountability, seem to have been effective in producing both equality and quality. These are promising findings that support the goals set by UNESCO’s Education for All process and give hope for more inclusive education in the future.