The project reported here has been carried out as a part of the evaluation of the comprehensive reform in Norwegian primary and secondary education. The research question for the project has been «What is the relationship between measures of special education and student outcomes in upper secondary education?«

The analyses and interpretations in this report rest on comprehensive data material. We have sought to examine our research question using both qualitative and quantitative data. The quantitative material consists of four different datasets. The first contains data on 1804 students, grouped as whole classes, making it possible to compare special students with other students. Then there are three other sets of data collected from surveys of: (1) 336 teachers, (2) 85 principals, and (3) seven leaders of the county administrations of upper secondary education. Each group of respondents provided their views on different aspects of special education. The data were collected when the students were in their first year of upper secondary education.

The qualitative material comprises data collected from interviews with 15 special education students. We also interviewed their parents, teachers and principals. The data were collected at two different points, when the students were in their first year of upper secondary education and then again in their second year. Combined, the data material provides a rich and valuable source of information on special education.

In the quantitative data the students are divided into four groups. a) non-special education students (n=1327), b) students in mainstream classes receiving extra support (n=201), c) students in mainstream classes receiving special education (n=130), and d) students in special classes (n=146).

In the international and Norwegian research and literature on special education the debate on segregation, integration and inclusion is important. Most Norwegian research on special education after Reform 94 has focused on this. The main conclusions from this research are that students receiving special education while belonging to mainstream classes, have achieved better grades and qualifications than students receiving their special education while belonging to special classes (Kvalsund og Myklebust 1998, Kvalsund m.fl. 1998, Kvalsund 1999, Myklebust 1999, Markussen 2000, 2001, 2004). These findings were somewhat qualified by Markussen et.al. (2003) who concluded that as important as the inside-outside perspective, was the presence of a high level of pedagogical and didactical reflections in the schools, and a very close follow up of the students. Some of the research also showed that the mainstream students did not lose academically when having special students in their classes (Grøgaard 2000, 2002).

Also in the international debate and research literature there has been much attention given to the question of segregation, integration and/or inclusion. We mention some important contributions: Special education for the mildly retarded - is much of it justifiable? (Dunn 1968), Meeting Special Needs in Ordinary Schools (Hegarty1989), The Illusion of Full Inclusion (Hallahan & Kaufmann 1995), Mainstream or special? (Jenkinson 1997), Inclusive education at Work (OECD 1999), and Included or Excluded? (Cigman 2007).

The terms are difficult to define, and as far as we can see, a common definition upon which everyone has agreed, does not exist. Even if it is not correct, and we are violating the basic thinking within the inclusion education movement, and focusing on the student rather on the relationship between the school and the student, we have chosen to use the term included, when a student belongs to a mainstream class, whether he/she receives special education inside or outside the class, and segregated when he/she belongs to a special class.

The comprehensive Reform 2006 was launched beginning August 2006. Preceding the reform there was a committee suggesting the removal of paragraphs on special education from the Education act, and instead put extra emphasis on Adjusted Education, meaning that every student should get his/her education adjusted to his/her abilities. This suggestion did not survive the political debate and when the Parliament decided on the new reform, the paragraphs on special education in the Education Act remained unchanged. But simultaneously the politicians, the Government and The Parliament, signalled to the education system that they actually wanted less special education and more adjusted education. By doing this, they sent a double message: Keep on doing as you have always done, but we want less special education! It is within this context our research project has been carried out.

The proportion of students receiving special education is considerably lower in the Student preparatory program (1 percent) than within the vocational tracks we are examining (10-12 percent) (health and social care, industry and techniques and restaurant and food-processing). The situation is the same for those receiving extra help and support, 5 and 12-18 percent, respectively. Most of the students, around 40-50 percent, receiving special education or extra help and support received this in one subject only. Most of special education given to those belonging to mainstream classes is given in small groups outside the classroom, and only a small portion is given as one-to-one lessons.

From our data we find that there are more students receiving special education or extra help and support:

These relationships between different background variables and receiving special education or extra help and support are informative. We do not suggest or imply causality, but we would argue that these relationships indicate that some of the difficulties creating need for special education or extra help and support have been created through the lives special education students have been living in their families, in the education system and in society; the need for special education is partly socially constructed (Skrtic 1991, Lundeberg 2008).

We asked the teachers, the principals and the leaders of upper secondary education administrations some questions on their opinions and attitudes towards different aspects of special education. They all agreed that it shall be possible to deliver special education within mainstream classrooms, but they do not consider that it adequately serves the needs of all students, and consequently they believe it should not be applied to all. Many of the teachers agree that special education carried out within mainstream classes gives a greater outcome socially than academically, both for special and other students.

We asked the principals and the county leaders of the upper secondary education administrations to tell us about the pedagogical ideology or philosophy upon which the implementation of their special education is resting. None of them could tell us about such a pedagogical basis or platform.

Through our interviews with 15 students and their parents, teachers and principals, we have observed clear examples of failure in lower secondary education to address need adequately . We met students who did not achieve to the levels of their capabilities during lower secondary, because their learning needs were not adequately met. The result, when they progress to upper secondary education, is a need for special education, a need that possibly might have been avoided if lower secondary had acted otherwise. We have seen upper secondary schools trying their best to repair the damage done previously, with varying success. This qualitative material does not point to segregated or included settings as the best solution when it comes to producing the best results; we have in our meeting with these 15 students, observed both solutions producing success and failure. What it does suggest is that a higher level of pedagogical and didactical reflection in the schools coupled with a very close monitoring of the students, as Markussen et.al. (2003) suggested, may produce better results. But as we have not identified such a school or practice in our data, we have not been able to track such relationships.

A systematic comparison through a regression model shows that students receiving special education while belonging to a mainstream class achieve considerably better grades at the end of first year of upper secondary education, than students in special classes. Those with grades in the lowest decile in lower secondary and in mainstream classes, gain 0.89 points on a scale from 0 to 6 over students in special classes. And those who achieve just below the middle in lower secondary gain 1.95 points, all else equal.

The conclusion is that students included in ordinary classes achieve better grades at the end of the first year of upper secondary than students in special classes. A possible explanation is that the students included in ordinary classes, contrary to those who are segregated, attend classes with a higher level of overall achievement. Mainstream students set the level which the special students have to reach.

There are some other factors influencing the results (grades) at the end of the first year of upper secondary education for students with special education. The most important of these, apart from whether the students are included or segregated, are the grades from lower secondary and which study program they are doing.

We have also studied how the achievement of mainstream students responded to the inclusion of special education students in their classes. Students in vocational tracks tend to gain having special education students in their classes. A vocational class of 12-15 students is able to carry 15-20 percent of special students, without any effects on the achievement levels of mainstream students. We observe a win-win-situation. Within the student preparatory program we were not able to conclude due to weakness in the data material.

One way to understand special education is to look at it as a means for society to help students who will not be able to succeed in mainstream education without support. To help these students certain measures are implemented, and as already mentioned, the research question for this project was «What is the relationship between measures of special education and student outcomes in upper secondary education?» We have studied these aspects of special education provision:

In our analyses we have found that most of these factors have a minimal effect on the outcome of special education. We have found that the variables with the strongest predictive power are grades from lower secondary, study program attended and whether special education is segregated or included. It is very interesting to notice the strong effect on the outcome of first year of upper secondary education, of belonging to special classes, other things being equal. And it is also interesting to notice that there is no effect on the outcome depending on whether special students belonging to mainstream classes receive some of their special education outside the class or not. The qualitative study supports these findings. Some students with special education belonging to mainstream classes experience success, others experience failure, and this is independent of whether they leave class for some lessons or not.

We have also studied special educational policy and the pedagogical platform upon which special education is supposed to be based. We find mixed signals from central authorities. Not changing the Education Act when creating a comprehensive reform, they say that special education shall remain unchanged, but at the same time they clearly communicate that they expect less special education. These mixed signals confuse the practitioners, and no clear policy or practice can be observed. The teachers, the principals and the county leaders of the upper secondary education administrations are not able to tell us about a common understanding of the educational political platform or the pedagogical platform for the implementation of special education. The mixed signals from the politicians result in mixed understandings and mixed practice.

Our findings are in accordance with previous research on special education in Norwegian upper secondary education following Reform 94. Then, as in this project, the results showed that inclusion proved best solution for the majority, but qualitative studies showed that segregated solutions could function for some students, especially when accompanied by a high level of pedagogical and didactical reflection in the schools, together with close monitoring of the students. So far we cannot track any changes in special education in upper secondary education after the comprehensive Reform 2006.

Which implications should these findings have? In our view the results of this project support an argument for reducing the number of special classes in Norwegian upper secondary education. Those attending these classes are in a weak learning environment. Ambitions are low as are self confidence and expectations, and no one is pushed forward by their classmates or by the environment. The implication is clear: reduction in the use of special classes.

If this happens, more students have to be included, and we have seen that special education students in both academic preparatory programs and vocational tracks will gain from this, as well as mainstream students in vocational tracks. When it comes to mainstream students in the academic preparatory program, we are, due to weakness in the data material, not able to conclude on this matter. Based on this we recommend more inclusive practice in the vocational tracks. In the academic tracks, we have to recommend further research in order to analyze the possible effects on the achievements for mainstream students of having special education students in their classes.

When it comes to the question of whether special education students belonging to mainstream classes should receive all their special education inside the class, or whether they should receive parts of it in small groups or in one-to-one-settings, the results of this project give no clear answer. In the qualitative data we see that both strategies may produce good results, and the quantitative data is not able to separate these groups when it comes to results from the first year of upper secondary education. The data do not give us a direction for any suggestions, but we may say that when using segregated strategies, these should be implemented carefully. It is very important to listen to the students. If they want their special education inside mainstream classes, they should. Our conversations with the students, who have, against their will, been placed outside the class, have convinced us that this does not create motivation and eagerness to learn.