Norway is regarded as a country with a successful inclusive education because the vast majority of pupils receive their special needs education (SNE) within ordinary educational settings. However, the Norwegian primary and lower secondary schools are facing substantial criticism because of an extended use of teacher assistants within SNE. A steadily rising number of assistants may seem rather incomprehensible given the fact that research confirms that teacher competence significantly impacts the quality as well as the outcomes of SNE. In Norway, each SNE decision depends upon explicit statements from an expert assessment. This indicates that the expert assessments may have a substantial impact on the quality and the quantity of SNE. These expert assessments are withheld from public access. Therefore, the purpose behind this study is to access a selection of expert assessments to explore their statements regarding teacher competence and recommended educational settings for SNE. The results indicate that the experts have developed a certain professional practice which may both explain and reinforce the extended use of assistants within SNE in Norway.

This article represents a follow-up analysis of a study published in this journal last year, and parts of the data material and results from the previous study (Hannås 2018), will also be presented and discussed in this article.

Introduction
Inclusion is an overarching principle for all education in Norwegian schools. Even though there are discussions regarding how to implement inclusive education in practice, the organisation of SNE has changed considerably since the Salamanca Statement was adopted in 1994. While SNE previously often was organised through various activities outside of the ordinary classroom, most pupils now receive SNE as an integral part of a general education. The number of pupils placed in special groups is considered low, and hence Norway is recognized as a country with success regarding the organisation of SNE (Wendelborg & Tøssebro 2010).

Though the situation respecting educational settings is mainly characterized as positive, critical questions have been raised concerning the quality of SNE (Haug 2015). Research reveals that teacher competence has a significant impact on educational quality (Baumert et al. 2010; Shulman 1986). Despite a higher learning potential within SNE when compared to general education, several studies indicate that this is not always reflected through the pupils' benefit from the SNE (Haug 2015). Recently, greater attention has been paid to what is described as “an extensive use of teacher assistants” in SNE in Norway (Bele 2010; Hausstätter & Takala 2008; Hannås 2018). Critical voices claim that the pupils with the most comprehensive special needs often are being taught by the staff members with the lowest competence at the school, i.e. by employees without formal competence in pedagogy, in SNE, nor in the relevant teaching subjects (Children's Ombudsman 2017; Nordahl et al. 2018).

Both scope and content, i.e. quantity and quality, of SNE in Norway are regulated through national legislation and public documents. Every application for SNE is dealt with individually and according to common legal procedures. Both the pupil, parents, teachers, expert authorities—and when necessary others—are involved throughout the process preceding a SNE provision. A key document within this process is the expert assessment (EA). According to the law, the EA must document the pupil’s individual needs regarding both scope and content of the SNE provision (Ministry of Education and Research, MER 1998). The EA possesses a special position and status, because the school principals are obliged by the law to respond – in one way or another – to the statements and recommendations from the experts. Yet, the EAs are withheld from public access and unavailable for scrutiny by anyone other than the ones involved in the process. Therefore, it might be of interest to explore and analyse these documents. Since educational settings often are referred to in positive terms while teacher competence of SNE is a subject of criticism, it may be intriguing to scrutinise how these two aspects are dealt with and assessed within the EAs.

In an article analysing processes similar to an SNE application in Norway, R.M. Emerson and S.L. Messinger elaborated a theory identifying and describing common key aspects that seem characteristic for these kinds of processes. The theory highlights the impact of the contribution from the experts that are involved in the process. The experts are highly skilled professionals and often act as the authorities’ official representatives when dealing with cases and processes. Their assessments have further proven to be deciding factors regarding the results and outcomes of these processes. The theory thus indicates that the EAs may have a substantial impact on both the quality and outcome of a pupil’s SNE (Emerson & Messinger 1977).

Given their extraordinary status and position, the aim of this study is to explore a selection of EAs in order to identify and analyse the content of relevant statements and recommendations regarding the need for teacher competence and preferable educational settings for SNE. The research questions to be investigated are:

What kind of competence do the EAs recommend for the SNE?
What kind of educational settings do the EAs recommend for SNE?

Earlier research
Since the Salamanca Statement (1994), the effort to shut down special schools and reduce the number of pupils receiving SNE in segregated educational settings has been an international trend. Although there is no common agreement on how to implement the principle in practice, inclusion is seen as an ongoing process in order to respond to the diversity of pupils in mainstream classrooms (Ferguson 2008; Reindal 2016; Ainscow 2005, Göranssson & Nilholm 2014). In addition to the physical location, researchers also see inclusion as referring to the pupil’s “sense of belonging”, i.e. to the pupils’ subjective feelings of being accepted and included among others (Ainscow et al. 2006; Armstrong, Armstrong, and Barton 2016; Haug 2017a, 2017b; Nilholm & Göransson 2013; Nordahl et al. 2018). The principle further relates to a common access to an equal education, which includes the right to an adequate and adapted education. This means that the diversity of individual needs among the pupils must be taken into consideration (Haug 2017b). According to Haug, placement seems to be the most frequent criterion to avoid segregation, while the quality of education has a lower priority in inclusive education. He emphasises that inclusive education is not simply a question of where the SNE takes place, but also a question of various aspects regarding the quality of the SNE itself (Haug 2017b).

Teacher competence has a strong impact on pupils learning, development, and achievements; and some researchers point at the teacher as the most important single factor regarding a pupil’s educational outcomes (Baumert et al. 2010; Hanushek 2014; Hattie 2009). Both the level and the content of the teacher’s education may have a severe impact on the pupil’s achievements (Darling-Hammond 2000). Formal qualification in SNE is particularly valuable when it is linked to general education, and a close collaboration between the SNE teacher and other teachers (class or subject teachers) is a prerequisite for succeeding in creating inclusive educational environments (Nordahl et al. 2018).

Researchers discuss what kind of teacher competence that ought to be preferred or requested for SNE (Brownell, Sindelar, Kiley & Danielsson 2010; Haug 2017b; Florian 2008, 2014). Previously, learning about SNE mainly focused on different kinds of learning disabilities. However, recent research indicates that pupils with special needs benefit most from general teaching strategies (Brownell et al. 2010, Haug 2017b; Florian 2014). Thus, a co-teaching model, where a SNE teacher and the class teacher both teach all the pupils in the classroom together, appears to be a model that benefits all pupils (Cook & Friend 1995; Friend et al. 2010; Scruggs, Mastropieri & Mc-Duffie 2007; Thousand & Villa 2007).

The use of teacher assistants (TAs) to manage the growing diversity of pupils in mainstream classrooms has become extensive in several countries (Moran & Abbott 2002; Takala 2008). Even though the TAs cannot be responsible for teaching alone, several researchers are concerned because they often are highly involved in the teaching and learning processes (Bennett, Ng-Knight & Hayes 2015; Webster et al. 2012; Hannås 2018; Sadler 2014; Takala 2008). Norway is one of the countries with an extensive use of TAs in SNE. Though TAs are not supposed to take full responsibility for SNE, research indicates that they often seem to have a considerable responsibility for the implementation of SNE in Norway (Nordahl & Hausstätter 2009; Bele 2010; Haustätter & Takala 2008; Hannås 2018). Some point to the extensive use of TAs as a significant explanation why pupils who receive SNE often seem to fail in realizing their learning potential (Nordahl et al. 2018). Haug (2017c) confirms that the quality of SNE is closely related to the competence invested in SNE and claims that formal competence in SNE seems to be a decisive factor in making the SNE ‘work’. A kind of co-teaching including an assistant instead of a SNE teacher, beside the class teacher, seems common in Norway (Hannås and Strømsvik 2017). Thus, researchers question the lack of competence requirements for those who plan, carry out, and assess SNE in Norwegian schools. They claim that this may contribute to a poor learning outcome for the pupils who receive SNE (Nordahl & Hausstätter 2009).

Background
7.8 percent of the pupils in primary and lower secondary school in Norway receive SNE (Norwegian Directorate of Education and Training, NDET, 2017). Most of these pupils receive part-time SNE, and the vast majority (75 percent) receive 7 hours or less of SNE per week (Festøy & Haug 2018; Haug 2015). In 2017, approximately 0.6 percent of the pupils (3,993 out of 629,300) were in special schools, or in special classes within ordinary schools (Nordahl et al. 2018, 99; NDET 2017).

In recent years, the authorities have registered within which educational settings the SNE has been provided. The information is systematised into the four categories: mainly alone; mainly in groups of 2-5 pupils; mainly in groups of 6 or more pupils; and mainly in ordinary class. Table 1 (below) shows the distribution of SNE between the different educational settings in 2017-2018 (Nordahl et al. 2018).

Table 1 shows that 40.2 percent of the pupils who received SNE, did so mainly in ordinary classes. This represents a significant increase since 2014 when 28 percent of the pupils received SNE within their ordinary class. The table further shows that a total of 46.9 percent of the pupils received SNE mainly in smaller groups, and that 12.9 percent of the pupils received it mainly alone with an assistant or a teacher (Nordahl et al. 2018, 98).

There is no similar registration indicating how much of the SNE that is carried out by TAs. Available statistics, however, confirm that the use of TAs in SNE has been increasing for a long time. From 2010 to 2017, the increase exceeded 80 percent, and some researchers estimate that approximately half of SNE is actually being carried out by TAs (Nordahl et al. 2018).

Available data further shows that 68.3 percent of the TAs in schools are working within SNE, and that 6.9 percent are working within general education. Although there are some uncertainties connected to the figures reported, the difference between general and special needs education may seem remarkable, especially considering that SNE constitutes only 7.8 percent of the education as a whole (NEDT 2017; Hannås 2018).

According to the Norwegian Education Act, all education in lower and secondary school must be adapted to the pupils’ abilities and aptitudes, and pupils who do not benefit sufficiently from the ordinary education are entitled to SNE. Pupils and parents may individually, or together with the school request an assessment of a pupil’s need for SNE. The decision on a subsequent application for SNE relies on the EA regarding the pupils’ individual learning abilities and need for SNE. The assessment is carried out by an independent expert authority, The Educational and Counselling Service (PPT). The EA describes the pupil’s individual learning potential and learning difficulties. It assesses the pupils need for SNE, recommends whether the application for SNE should be denied or granted, and eventually how much and what kind of SNE the pupil is entitled to (MER 1998).

The Norwegian Directorate for Education and Training (NDET) has published official guidelines for SNE, specifying the details of what the EA is supposed to provide. The guidelines confirm that PPT has to recommend an exact number of hours for SNE per school year for each case, and also state that:

An assessment considering the amount [of the special needs education] includes how the special needs education should be organised by using different kinds of educational settings. It can, for example, point out the need for specific competence or extra personnel (SNE teacher competence or assistant), if it is necessary to ensure that the pupil receives a sufficient education (NDET 2014:52, my translation).

The guidelines further state that the EA should primarily advise educational settings that may contribute to the pupil’s group or class affiliation (ibid 2014). They emphasise that there are certain restrictions limiting the schools’ possibilities to organise the education by dividing pupils into groups based on their school performances or other criteria such as gender or ethnicity (MER 1998). National regulations, however, allow a school to let a group of poor readers of the same grade receive intensive training in reading outside the class for a certain period of time, assuming they return to the class soon after the period (Official Norwegian Report 2014:7, 25).

Based on the EA, the school owner (i.e. the school’s principal) decides whether to grant or reject the application for SNE. If the application is granted, the decision must specify how much and what kind of SNE the pupil is entitled to. In case of any disparity or deviation from the EA, the decision also has to explain how the school intends to answer to and fulfil the recommendations given by the EA within the general education. The principal’s decision may be appealed and is legally binding in case of a lawsuit (MER 1998).

No specific competence is required for teachers who are teaching SNE (MER 1998). It is possible to gain bachelor’s and master’s degrees in SNE, but the majority of the teachers with formal SNE competence have either a teacher education including 60 ECTs (European Credit Transfer and Accumulation System) of SNE, or a master’s degree including 90-120 ECTs of SNE. The curriculums at the different institutions offering SNE education do not share a common core of content (Hausstätter & Takala 2008).

The official requirements for a teacher’s formal qualifications and competences are basically the same within SNE and general education. Per these requirements, all teachers have to possess an adequate and formal education within the subjects they are teaching. Individuals who are not qualified for teaching, in certain cases may assist in SNE as well as in general education, assuming they receive relevant supervision from a qualified teacher (NDET 2014:28). Thus, the principal may exempt those who teach SNE from the general competence requirements. However, educational authorities emphasise that a principal’s individual decision must clearly state whenever the SNE includes the use of TAs (NDET 2014:28). It is also the principal’s responsibility to ensure that the school staff always possesses the competence considered necessary to secure the quality of the education provided by the school (MER 1998; NDET 2016).

The close link between the EA and the principal’s decision, implicates that the content and statements within the EAs may have a significant impact on the principals’ decisions, as well as on the quality and quantity of the subsequent SNE provision. The unique position makes it particularly interesting to explore and scrutinize the EAs’ statements and recommendations regarding organisation and competence in SNE. Hence, this article is based on a qualitative study including analyses of relevant content within a limited selection of EAs.

Method, empirical data and analysis
This study is part of a single embedded case study of SNE competence and educational settings in Norway. A case study is a suitable method for exploring a specific phenomenon or practice, such as teacher competence and educational settings of SNE (Yin 2014). Following an approval from the Norwegian Data Protection Officer for Research, and in collaboration with the chief municipal education officer and the principals at all primary and lower secondary schools in a large Norwegian city, the parents of all children who received SNE in 2014-2015 were invited to attend in the investigation. A request to the pupils and their parents resulted in 69 written consents. The empirical material subsequently collected from the schools includes four different kinds of documents; EAs, principals’ decisions, individual education plans (IEPs), and annual reports from each of the 69 cases. In two cases, PPT’s EAs were lacking. Since the purpose of this study was to gain a better insight into what statements or recommendations the experts from PPT actually provide regarding educational settings and the need for competence in SNE, the analysis focuses on the 67 EAs written by the professional experts at PPT. Most of the EAs in the material consist of 3-5 pages, but several contain 5-7 pages. The empirical data is analysed through an inductive approach using qualitative content analysis (Elo & Kyngös 2008). The EAs were numbered from E1 to E67. Then all relevant information regarding teacher competence and educational settings was identified, coded, and categorized. Below follows a presentation of the results from the analysis, accompanied by excerpts showing some typical examples of statements and assessments from the most frequent categories.

Results
The results below are presented according to the two research questions; starting with the findings from the analyses of the assessments regarding teacher competence, followed by the findings from the analyses of the assessments regarding educational settings of the SNE.

PPT’s assessments regarding competence needs in SNE
The EA analysis revealed that it is often hard to determine which competence PPT actually recommends. Some EAs do not explicitly address or assess the issue, while others mention it implicitly, through statements concerning other matters, such as content or organisation of the SNE. This caused some difficulty throughout the interpretation of codes and categories, which may have led to some inaccuracies in the final results.

The codes identified regarding different competences are: SNE teacher, teacher, assistant, and none (no information). Table 2 reveals the number of EAs that recommend each of the competences.

Table 2 shows that the majority of the EAs (43) mentioned or recommended teacher competence for the SNE. Secondly is assistant, which appears in 40 of the EAs. Finally, it shows that the code SNE teacher is mentioned in fewest of the EAs (17), and that 18 EAs do not contain or mention any specific competence at all.

The analysis further revealed that the different codes regularly occur in various combinations, i.e. categories. Table 3 presents the different combinations of competences that occur in the material and the number of EAs per each competence category.

Table 2 showed that teacher is the most frequently mentioned competence in the EAs. Furthermore, Table 3 reveals that although teacher is mentioned in 43 EAs, in most cases (22 EAs), it appears in a category which also includes the code assistant, and in 13 EAs it occurs in a category including both assistant and SNE teacher in addition to teacher. In 6 EAs, teacher appears as the only competence mentioned. Finally, only 2 EAs mention a combination of the two codes teacher and SNE teacher.

Assistant, which is mentioned in 40 EAs, occurs together with teacher in 22 EAs, and together with both teacher and SNE teacher in 13 EAs. Additionally, assistant is mentioned as the only competence in four EAs, and in one EA it is mentioned together with SNE teacher.

While 18 EAs do not mention any specific competence at all, the code SNE teacher is mentioned in a total of 17 EAs. In 13 of these, it occurs in a category which also includes teacher and assistant. 2 EAs mention both SNE teacher and teacher, and 1 EA mentions a combination of SNE teacher and assistant. Finally, SNE teacher occurs as the only competence code mentioned in 1 EA.

Table 3 shows that most of the EAs (22 of 67) recommend or mention a combination of assistant and teacher competence related to the SNE. The excerpts below (E6 and E2) are two typical examples of assessments from this category. The excerpt from E6 concerns a boy in the 5th Grade, who, according to the EA, struggles with concentration and learning difficulties in mathematics, reading, and writing. The EA recommends a total of 228 hours of SNE per year. The second excerption (E2), concerns a boy in 2nd Grade who has specific speech challenges.

E6: In Norwegian, English, and mathematics, NN needs extra training by an additional teacher/assistant in order to learn what is being taught. In these situations, the assistant needs supervision from a teacher, because the teachers have the main responsibility when it comes to teaching.

E2: The recommended scope is 44 hours of SNE alone, either together with the contact teacher or with an assistant.

The EAs within the second largest category (18 of 67), have no specific assessments or recommendations related to the question of competence at all. The excerpt below (E18) is a typical example from this category. The assessment concerns a girl in 10th Grade with ADHD (Attention Deficit Hyperactivity Disorder) and persistent difficulties in mathematics.

E18: A scope of 114 hours of SNE seems appropriate to ensure that the pupil receive an equal and satisfactory education in mathematics. PPT is also aware of the fact that the class, additionally, is divided into smaller groups two hours per week in mathematics. This should be sufficient considering NN’s special needs.

The third largest category (13 of 67) contains EAs which mention all three competence codes. Below is a typical example of an assessment within this category. The excerpt (E60) applies to a boy in 8th Grade who struggles with complex social and medical problems that cause extensive learning difficulties. The EA recommends a total of 1064 hours of SNE per year.

E60: To achieve and maintain a positive learning curve, NN needs a lot of extra training, preferably alone—with an assistant/teacher/SNE teacher—or in a small group.

PPT´s assessments regarding educational settings for SNE
The analysis of the EAs’ statements concerning educational settings resulted in the following codes being identified: ordinary class, smaller group, alone, other solutions, and none (no recommendation). Table 4 reveals the amount of EAs recommending each of the educational settings (in numbers and percentages).

Table 4 shows that smaller group, which is recommended in 53 of the EAs, is the most frequent code regarding educational settings in the material. Next follow the two codes ordinary class and alone, which both are recommended in 38 EAs, while 9 EAs do not contain any specific comments related to educational settings (none).

The analysis further showed that the codes for different educational settings also occur in various combinations, i.e. categories. Table 5 reveals the appearance and distribution of different codes and categories related to educational settings in the empirical material.

Table 5 reveals that smaller group, which is the most frequently mentioned code in the material (53 EAs), occurs in the largest category (21 EAs) together with the two codes ordinary class and alone. It is also mentioned in the second largest category (13 EAs) together with ordinary class, and in the third largest category (11 EAs) together with the alone. 9 EAs, which do not contain any statements regarding educational settings, constitute the fourth largest category; and the fifth contains 6 EAs where smaller group is the only code mentioned.

In brief, Table 5 shows that smaller group, which is mentioned in 53 EAs, occurs in five different categories, each containing various combinations of different educational settings. Ordinary class which is mentioned in 38 EAs, is included as one of the codes in five different categories. In one EA, it is the only code present. Alone which is also mentioned in 38 EAs, is likewise included in six different categories, and in two EAs it is the only code present. As mentioned above, nine EAs do not mention or refer to any educational setting. Finally, other solution is mentioned in three EAs: two belong to a category including three other codes, and in the third it is the only code mentioned in the EA.

The most frequently mentioned category of educational settings (21 EAs) includes three codes: ordinary class, smaller group, and alone. The excerpt below (E22) is an example of a typical assessment from this category. It applies to a boy in 7th Grade who has ADD (Attention Deficit Disorder) and concentration difficulties, but otherwise strong learning ability. The EA recommends a total of 142 hours SNE per year, either by a teacher or an assistant:

E22: The school has to assess whether the [special educational] support should be organised in—or in relation to—the ordinary class, or if some parts of it should be organised alone with a teacher/an assistant, or in a smaller group of pupils.

The second largest category (Table 5) includes 13 EAs, which all mention both ordinary class and smaller group as recommendable educational settings for the SNE. The excerpt below (E15) is a representative example from this category. It applies to a girl in 9th Grade with problems related to dyslexia and dyscalculia. The EA recommends a total of 882 hours of SNE, distributed over three years, from 8th to 10th Grade.

E15: The school has to assess whether the support should be organised within or in connection to the ordinary class, or if parts of the education are most appropriate when given in a smaller group of pupils.

The third largest category includes 11 EAs, which all contain recommendations including both small group and alone with a teacher or an assistant. An example of a typical statement from this category is that the SNE “may be organised individually [alone] or in a smaller group, together with 3–4 other pupils” (E38, boy 10th Grade).

Discussion and conclusion
According to Norwegian legislation, the EAs are not obliged to, but they may, recommend specific competence or extra personnel, if the professional experts consider it necessary in order to ensure that the pupil may achieve a sufficient educational benefit. This may explain the lack of expert assessments and recommendations concerning the need for competence in approximately a quarter of the EAs in the material. It may also help to explain some of the typical statements concerning the need for competence in other EAs. The results show that teacher is the most frequently mentioned competence in the material, closely followed by assistant, and that they are both mentioned in more than half of the EAs (Table 2). The content analysis, however, revealed that teacher usually appears in a context where the EAs emphasise that the assistants have to receive guidance from a professional teacher. Similarly, the analysis further revealed that, although SNE teacher is mentioned in approximately a quarter of the EAs, all except one of the EAs—besides SNE teacher—also mentions teacher and/or assistant in their statements regarding competence for SNE. Representing typical examples from the largest categories, the excerpts (E6, E2, E18 and E60) reveal that the professional experts in general tend to include several different competences in each assessment, without preferring or recommending any one of them in particular. By leaving several options open, the experts create large a room for manoeuvre, letting the principal decide which one of the mentioned competences to invest in the SNE. By doing so, the expert assessment is designed to support the final decision, no matter which of the aviliable options the principle chooses to implement.

With respect to the question of which educational setting to choose for the SNE, the content analysis reveals a similar trend within the experts’ assessments. The excerpts, representing typical assessments (E22, E15 and E38), indicate that the experts’ statements tend to mention several different educational settings without arguing for or recommending any of them specifically.

The main result from the study of EAs thus indicates that the professional experts in general tend to provide several options and leave the final decision, both regarding which competence to invest in and which educational setting to implement, to the principal. However, the principal’s decision may be influenced by several different conditions, such as his/her own professional background and knowledge of SNE and inclusion, or by the current practical and economic situation at the school (Nordahl 2018:107).

The professional experts at PPT are often considered as advocates for pupils who need special needs education. For that reason, most people would presumably expect them to voice the pupils’ needs for special support in their assessments, both regarding the need for SNE teacher competence and appropriate educational settings. However, the analyses in the study reveal that the professional experts’ assessments in fact, instead may serve to support, facilitate, encourage, or legitimise a decision regarding SNE that appears most convenient given the current circumstances at the school. Because of the EA’s unique position and decisional influence on the SNE provision (Emerson & Messinger 1977; MER 1998), an expert practice like the one revealed above, may ultimately also serve to maintain and reinforce the extended use of assistants, which is the reason why SNE in Norwegian schools are being exposed to a lot of substantial criticism (Nordahl & Hausstätter 2009; Bele 2010; Haustätter & Takala 2008; Hannås 2018; Nordahl et al. 2018; Haug 2017c).

Earlier research indicates that SNE in Norway seems to have adapted well to the overarching principle of inclusive education regarding questions related to educational settings, but it often questions the quality—as well as the pupils’ outcome—of the SNE. Maybe the awareness and focus on educational settings, in terms of physical location, has overshadowed other implications of the principle related to the need for professional expertise regarding both inclusion and SNE. This study suggests that this may partly be explained by a certain practice among the professional experts in PPT who assess the pupils’ individual needs related to SNE. Given the severe impact of teacher’s competence on the quality of SNE, one way of enhancing SNE quality might be to instruct PPT by requesting clearer and more specific assessments regarding recommendable or preferable teacher competences in the EAs. The lack of assessments regarding the need for teacher competence applies to the need for SNE competence as well as to the need for competence regarding what inclusive education means—and how to implement it—in practice.

This study explores a selection of expert assessments with the aim to enlighten and scrutinize the professional experts’ assessments regarding the need for competence and recommendable educational settings for SNE. The results indicate that a certain practice developed among professional experts may contribute to an explanation of the extended, and still growing use of assistants in SNE in Norway. This is however a small study, and to my knowledge, it is the only one so far analysing the content of EAs and scrutinizing the statements from the professional experts at PPT. Hence, there is a great need for more research. This includes the need for more extensive studies in order to investigate whether the professional practice described above is widespread within other regions and applies to the country as a whole, or not. Furthermore, I chose to explore the expert assessments because of their unique status and potentially crucial and decisive impact on the SNE, and because of the lack of similar analysis of these documents. The result indicates that it might be useful to reconsider which kind of assessments to request through the expert assessments. It will be interesting to explore and compare relevant content in the subsequent documents, i.e. the principals’ decisions, the individual education plans, and the annual reports, that follow the 67 EAs in the cases involved in this study. However, there is still a great need for further research, including more classroom studies focusing on educational practices, in order to develop a better knowledge of how to implement an inclusive SNE in practice.

References

Ainscow, M. 2005. “Developing inclusive education systems: What are the levers for change?” Journal of Educational Change, 6(2), 109-124.
Ainscow, M., T. Booth & A. Dyson. 2006. Improving schools, developing inclusion. Routledge.
Armstrong, F., D. Armstrong & L. Barton. 2016. Inclusive education: Policy, contexts and comparative perspectives. (2nd ed). Routledge.
Baumert, J., M. Kunter, W. Blum, M. Brunner, T. Voss, A. Jordan, U. Klausman, S. Krauss, M. Neubrand & Y.-M. Tsai. 2010. «Teachers’ Mathematical Knowledge, Cognitive Activation in the Classroom, and Student Progress. American Educational Research Journal, 47(1), 133-180. doi:103102/0002831209345157
Bele, I.V. 2010. «Lærernes egenvurdering av spesialpedagogisk kompetanse – og viktige kilder for kompetanseutvikling». [Teachers self-evaluation of their own competence in SNE]. Norsk pedagogisk tidsskrift, 94(06), 476-491.
Bennett, M., T. Ng-Knight, & B. Hayes, B. 2017. «Autonomy-supportive teaching and its antecedents: differences between teachers and teaching assistants and the predictive role of perceived competence”. European Journal of Psychology of Education, 32(4), 643-667. doi: 10.1007/s10212-016-0321-x
Brownell, M. T., P. T. Sindelar, M. T. Kiely & Danielson, L. C. 2010. “Special Education Teacher Quality and Preparation: Exposing Foundations, Constructing a New Model”. Exceptional Children, 76(3), 357-377.
Children’s Ombudsman. 2017. Uten mål og mening? Elever med spesialundervisning i grunnskolen. [Without goals and meaning? Children who receive special needs education in primary and lower secondary school]. Barneombudets fagrapport. Oslo.
Cook, L. & M. Friend, M. 1995. “Co-Teaching: Guidelines for Creating Effective Practices”. Focus on Exceptional Children, 28(3), 1-16.
Darling-Hammond, L. 2000. “Teacher quality and student achievement, A review of state policy evidence”. Education Policy Analysis Archives, 8 (1), 1-44.
Elo, S. & Kyngös, H. 2008. The qualitative content analysis process. Journal of Advanced Nursing 62(1), 107-115. DOI: 10.1111/j.13652648.2007.04569.x
Emerson, R. M. & S. L. Messinger. 1977. «The Micro-Politics of Trouble”. Social Problems, 25(2), 121-134.
Ferguson, D.L. 2008. “International trends in inclusive education: the continuing challenge to teach each one and everyone”. European Journal of Special Needs Education, 23 (2), 109-120.
Fagerlid, A. R. F., & P. Haug. 2018. «Coherence in conditions for learning in special education”. European Journal of Special Needs Education, 1-16.
https://doi.org/10. 1080/08856257.2018.1542227
Florian, L. 2008. “Inclusion: special or inclusive education: future trends”. British Journal of Special Education, 35(4), 202-208.
Florian, L. 2014. “What counts as Evidence of Inclusive Education?” European Journal of Special Needs Education, 29(3), 286-294.
Friend, M. L. Cook, D. Hurley-Chamberlain, D. & C. Shamberger. 2010. “Co-teaching: An Illustration of the Complexity of Collaboration in Special Education”. Journal of Educational and Psychological Consultation, 20(1), 9-27.
Göransson, K. & C. Nilholm, C. 2014. «Conceptual diversities and empirical shortcomings–a critical analysis of research on inclusive education”. European Journal of Special Needs Education, 29(3), 265-280. Doi:10.1080/08856257.2014.933545.
Hannås, B. M. 2018. “Hvordan sikre kvalitet i spesialundervisningen? En empirisk undersøkelse av Pedagogisk-psykologisk tjenestes sakkyndige vurdering?». [How to ensure the quality in special needs education?]. Psykologi i kommunen, 53 (1), 69-85. Drammen.
Hannås, B. M, og C. L. Strømsvik. 2017. «Tidlig innsats, økt lærertetthet og tolærersystem» In: Hannås (ed.) Pedagogikk, trender og politikk i klasserommet. Empiriske og teoretiske studier av pedagogisk praksis [Pedagogy, trends and politics in the classroom]. Gyldendal Akademisk. Oslo, Norway.
Hanushek, E. A. 2014. “Boosting Teacher Effectiveness”. In What Lies Ahead for America’s Children and Their Schools, edited by C. E. J. Finn and R. Sousa. 23–35. Stanford, CA. Hoover Institution Press.
Hattie, J. 2009. Visible Learning. A Synthesis of Over 800 Meta-analyses Relating to Achievement. Routledge.
Haug P. 2015. “Spesialundervisning og ordinær opplæring”. [Special needs education and regular education]. Nordisk tidsskrift for pedagogikk og kritikk, 1. 1-14. http://dx.doi.org/10.1758/ntpk.v1.121

Haug, P. 2017a. «Spesialundervisning, læringsmiljø og inkludering». [Special needs education, educational environments and inclusion]. FoU i praksis [elektkronisk ressurs], 1, 41-62. Retrieved from https://utdanningsforskning.no/artikler/spesialundervisning-laringsmiljo-og-inkludering/
Haug, P. 2017b. “Understanding inclusive education: ideals and reality”. Scandinavian Journal of Disability Research, 19(3), 206-217.
Haug, P. 2017c. «Kva spesialundervisning handlar om, og kva funksjon den har». [What is special needs education about, and how does it work?]. I Haug, P (red.). Spesialundervisning. Innhald og funksjon. 385-411. Oslo. Det norske samlaget.
Hausstätter, R. S. & M. Takala. 2008. «The core of special teacher education: a comparison of Finland and Norway”. European Journal of Special Needs Education, 23(2), 121-134.
Ministry of Education and Research. 1998. Lov om grunnskolen og den vidaregåande opplæringa (opplæringslova). The Education Act. Oslo. Kunnskapsdepartementet. Retrieved December 2018, from https://lovdata.no/dokument/NL/lov/1998-07-17-61
Moran, A. & L. Abbott. 2002. “Developing inclusive schools: the pivotal role of teaching assistants in promoting inclusion in special and mainstream schools in Northern Ireland”. European Journal of Special Needs Education, 17(2), 161-173.
Norwegian Directorate for Education and Training. 2014. Veilederen spesialundervisning [Guidelines for special needs education]. https://www.udir.no/laring-og-trivsel/sarskilte-behov/spesialundervisning/Spesialundervisning/
Norwegian Directorate for Education and Training. 2016. Education Mirror Norway. Retrieved from http://utdanningsspeilet.udir.no/2016/wp-content/uploads/2016/10/Utdanningsspeilet_2016_en.pdf
Norwegian Directorate for Education and Training. 2017. Education Mirror Norway. Retrieved from http://utdanningsspeilet.udir.no/2017/
Nilholm, C. & K. Göransson 2013. Inkluderande undervisning. Vad kan man lära av forskning? [Inclusive education. What can we learn from research?] Fou skriftserie nr. 3. Stockholm. Specialpedagogiska skolmyndigheten.
Nordahl, T., et al. 2018. Inkluderende fellesskap for barn og unge. [Inclusive community for children and young people]. Ekspertgruppen for barn og unge med behov for særskilt tilrettelegging. Fagbokforlaget.
Nordahl, T. & R. S. Hausstätter. 2009. Spesialundervisningens forutsetninger, innsatser og resultater. [The prerequisites, efforts and results of special education]. Rapport nr 2 fra prosjektet: Gjennomgang av spesialundervisning, evaluering av Kunnskapsløftet. Elverum: Høgskolen i Hedmark.
Official Norwegian Report. 2014. Elevenes læring i fremtidens skole. Et kunnskapsgrunnlag, [Pupils' learning in the future school. A knowledge base]. 7. Oslo: Kunnskapsdepartementet. The Norwegian Ministry of Education.
Reindal, S. M. 2016. “Discussing inclusive education: An inquiry into different interpretations and a search for ethical aspects of inclusion using the capabilities approach”. European Journal of Special Needs Education, 31(1), 1-12.
Saddler, H. 2014. “Researching the influence of teaching assistants on the learning of pupils identified with special educational needs in mainstream primary schools: Exploring social inclusion”. Journal of Research in Special Educational Needs, 14(3), 145-152.
Scruggs, T. E., M. A. Mastropieri & K. A. McDuffie. 2007. “Co-teaching in inclusive classrooms: A Metasynthesis of Qualitative Research”. Exceptional children, 73(4), 392-416.
Shulman, L. S. 1986. “Those Who Understand: Knowledge Growth in Teaching”. Educational Researcher, 15(2), 4-14.
Takala, M. 2007. “The work of classroom assistants in special and mainstream education in Finland”. British Journal of Special Education, 34(1), 50-57.
Thousand, J. S., A. I. Nevin & R. A. Villa. 2007. “Collaborative teaching: a review of the scientific evidence”. The SAGE Handbook of Special Education. London: Sage.
Webster, R., A. Russell & P. Blatchford, P. 2012. Reassessing the impact of teaching assistants: How research challenges practice and policy. Routledge. doi:10.4324/9780203151969.
Wendelborg, C. & J. Tøssebro. 2010. «Marginalisation processes in inclusive education in Norway: a longitudinal study of classroom participation”. Disability & Society, 25(6), 701-714.
Yin, R., K. 2003. Case study research: Design and methods. 3rd. ed., Newbury Park: Sage.