Variables affecting the attitudes of teachers’ towards inclusive education in Kosovo

Abstract Inclusive education is a priority and a challenge for education systems. The success of the inclusive practices’ implementation depends on the attitudes of teachers’ towards inclusive education. This research reflects the current state of pre-university teachers’ attitudes towards the inclusion in Kosovo, as well as it examines the relationship between teachers’ attitudes and selected background variables and self-efficacy. The research conducted a quantitative design and included 499 teachers from 40 schools in Kosovo. The instruments that were used to collect the data are Teachers’ Attitudes towards Inclusion Scale (TAIS) and Teacher Efficacy for Inclusive Practices (TEIP). Data were analyzed using descriptive statistics and inferential statistical tests. Results show positive attitudes of teachers’ towards inclusive education. Secondary level teachers have more positive attitudes towards inclusion than those from primary ones. Correlation results found a negative relationship between teachers’ and educational qualification and experience of teachers with students with special needs. The factors that predict teachers’ attitudes towards inclusion are self-efficacy for inclusive practices, professional qualification and work experience with children with special needs. The findings highlight the need to develop teaching skills and provide systematic support for teachers, while placing them as a priority in education agendas to ensure the success of inclusion.

Abstract: Inclusive education is a priority and a challenge for education systems. The success of the inclusive practices' implementation depends on the attitudes of teachers' towards inclusive education. This research reflects the current state of pre-university teachers' attitudes towards the inclusion in Kosovo, as well as it examines the relationship between teachers' attitudes and selected background variables and self-efficacy. The research conducted a quantitative design and included 499 teachers from 40 schools in Kosovo. The instruments that were used to collect the data are Teachers' Attitudes towards Inclusion Scale (TAIS) and Teacher Efficacy for Inclusive Practices (TEIP). Data were analyzed using descriptive statistics and inferential statistical tests. Results show positive attitudes of teachers' towards inclusive education. Secondary level teachers have more positive attitudes towards inclusion than those from primary ones. Correlation results found a negative relationship between teachers' and educational qualification and experience of teachers with students with special needs. The factors that predict teachers' attitudes towards inclusion are self-efficacy for inclusive practices, professional qualification and work experience with children with special needs. The ABOUT THE AUTHORS Donika Koliqi is a PhD candidate in Education Sciences, at the Faculty of Education, University of Prishtina "Hasan Prishtina", Kosovo. She is an assistant professor in the subject of inclusive education in all programs of Faculty of Education in Prishtina. Her research interest focuses on Inclusive Education and Teaching and Learning Theory. She has a doctoral thesis focused on the field of inclusive education and has also published scientific papers in the same field. Naser Zabeli is Professor in Inclusive Education and Effective Teaching and Learning in the Department of Pedagogy at the Department of Pedagogy at the Faculty of Education, University of Prishtina in Kosovo. His research interests are in inclusive education and contemporary teaching.

PUBLIC INTEREST STATEMENT
One of the trends of contemporary education in the last three decades is inclusive education, respectively, the inclusion of children with special needs in schools. Different studies have concluded that the successful implementation of inclusive practices depends on the attitudes of teachers, as well as the factors that influence their attitudes. However, no research has been carried out in Kosovo that identifies the factors which can have an effect on teachers' attitudes towards inclusiveness. Consequently, this study aimed to identify teachers' attitudes and the factors that predicted them. Teachers have positive attitudes towards inclusive education, and three variables have been identified that can predict their attitudes, including educational qualification, self-efficacy, and experience with children with special needs. Based on these empirical results, representatives at the central and local level should focus educational policies on restructuring the teacher study programs, on the organization of inclusive training and appropriate support, so that teachers possess the knowledge and skills to implement good inclusive practices.

Introduction
An issue which has followed the educational system historically is the fact that children with special needs have been denied the basic right to education in regular public institutions, as their families take care of them or they are sent in special schools. Although education is considered an important fundamental right for all people without exception (Mezquita-Hoyos et al., 2018), however when it comes to children with special educational needs, society has consistently focused on their disabilities rather than on their potential (UNESCO, 1994). Inclusive education is an educational philosophy that has managed to change the education organization of children with special needs while integrating them in regular schools (Van Mieghem et al., 2018). According to Opertti et al. (2014), "entails the openness, willingness and competencies to understand, embrace, and support the diversity of learners' profiles, circumstances, needs, styles, and expectations as a powerful source for democratizing and enhancing learning opportunities, processes, and outcomes" (149). This shift as a global movement for three decades (Amor et al., 2018), has been developed internationally based on four core ideas: (i) Under the influence of the Universal Declaration of Human Rights (United Nations Headquarters, 2006) education is considered as a fundamental right of every human being; (ii) The Salamanca Declaration and Framework for Action UNESCO (1994) promoted and supported the education of children with special needs in regular schools; (iii) The World Education Forum in Dakar (UNESCO, 2000), education focused on all marginalized groups; (iv) Currently, inclusion has managed to be set as a priority agenda in the reform of education systems, claiming to provide conditions for ensuring quality education for all children (Opertti et al., 2014). Moreover, the European Union recommends and promotes as a common value the education of all children from an early age, providing them with equal opportunities for active participation in society, which increases "understanding of the European identity" (Official Journal of the European Union, 2018, p. 5). All these documents have affected the development of the inclusion process. Therefore, it is expected that all education systems should be reformed ensuring access, participation and achievement of all students (Ainscow & Sandill, 2010).
Inclusion has been claimed as a program which helps regular schools decide, accept and adapt to children's diversity (Rapp & Corral-Granados, 2021). According to UNESCO (2015) it still remains a challenge of education systems and it is undeniable that they are still far from achieving inclusion. Although in many countries education policies have been reformed regarding inclusion, there are still schools that cannot guarantee the inclusion of all children with special needs in regular classes (Peters et al., 2005), pra ende ballafaqohen me vështirësi të ndryshme (Mónico et al., 2018). The inclusive environment can improve or obstruct the learning process of the child with special needs, therefore this environment must offer a sense of security, belonging, pride as well as favorable conditions for learning (Bucholz & Sheffler, 2009). Creating such a learning environment is a complex process that includes the development of policies which ensure inclusiveness, awareness and inclusive attitudes, cooperation between key factors, human and financial resources, appropriate infrastructure, as well as professional development of staff (Shevlin et al., 2009). Molina Roldán and others (2021) have added that from these environments the children benefit, such as the formation of attitudes, cognitive and social development.
The most important dimension for the development of inclusive schools are the attitudes of factors in educational settings and their approach to diversity (Ainscow & Sandill, 2010). Attitudes include the beliefs, views, feelings of the individual towards a person, object, phenomenon, etc. (Alkhateeb et al., 2015), and consist of three components: cognitive, affective and conative (Majoko, 2017). The affective component represents the emotional aspect, respectively positive, negative or neutral feelings towards an object. The conative component includes the action towards an object, and the cognitive component includes the evaluation of an object based on beliefs and convictions about it (Judd et al., 1991). The teachers' role in the implementation of inclusive education has been assessed as the most important aspect, with the probability that if they are negative they can be an obstacle to successfully achieve being part of such a school. Cultivating positive teacher attitudes (Urton et al., 2014), while overcoming "negative" attitudes, and raising teacher awareness remain one of the essential requirements for achieving full inclusion (UNESCO, 2009).
Teachers' attitudes towards inclusive education have been researched over the years (Avramidis & Norwich, 2002;Barnes & Gaines, 2017;Czyż, 2018;Dorji et al., 2019;Gyimah et al., 2009;Saloviita, 2018). According to the findings, teachers' attitudes can be positive (Avramidis & Norwich, 2002;Barnes & Gaines, 2017), neutral (Czyż, 2018;Ross-Hill, 2009) and negative (Mukhopadhyay, 2014). Being so, they depend on numerous indicators (Avramidis & Norwich, 2002). Based on research, Avramidis and Norwich (2002) have identified three main groups of factors that can influence teachers' attitudes: factors related to the teacher himself; child-related factors and also environmental factors. Current research focuses on factors that are related to teachers: gender, age, educational qualification, teaching experience, teaching experience with a student with special needs, and previous training on inclusive education.
Factors that affect teachers' attitudes towards inclusive education are demographic ones (Majoko, 2017;Parasuram, 2006). The research results of Butakor et al. (2018) have shown that older teachers have more negative attitudes towards inclusive education, compared to those who are younger. According to research female teachers have more positive attitudes towards inclusive education than male teachers (Boyle et al., 2013;Butakor et al., 2018;Gyimah et al., 2009;Tsakiridou & Polyzopoulou, 2014). Teacher training programs have been assessed as an important indicator of teachers' attitudes towards inclusive education. Research has shown that teachers with a master's degree have more positive attitudes than those with a bachelor's degree (Tsakiridou & Polyzopoulou, 2014), but there is research which concluded that teachers with a bachelor's degree were more positive towards this specific group of children than those with a master's degree (Dorji et al., 2019). The results of research conducted with teachers from Chhabra et al. (2010), and Majoko (2017) show that teachers who have attended training on inclusive education have more positive attitudes as they are also more emotionally attached to children with special needs. The research of Tsakiridou and Polyzopoulou (2014) found that secondary school teachers were more negative in terms of attitudes towards inclusive education, compared to preschool and primary school teachers. Previous studies also show that it should be considered longer contact of teachers with children known to be with special needs and the opportunity of working with children with special needs because these aspects make them more experienced inferring them to be more positive towards inclusive education (Boyle et al., 2013;Dorji et al., 2019;Parasuram, 2006). Referring to teachers' confidence in their abilities according to Sharma et al. (2012) is of great importance for the success of inclusion. In the research of Weisel and Dror (2006) self-efficacy was the key factor influencing teachers' attitudes. Similarly, in the research of Savolainen et al. (2012), it has been found a positive correlation between the highest level of self-efficacy and their attitudes towards inclusive education, while there has been detected a weak correlation between them in the research of Saloviita (2018).
Teachers' positive attitude towards inclusion influences the acceptance, adaptation and the change of teaching approach according to the needs and interests of each child in the classroom (Weisel & Dror, 2006). Therefore, attitudes studies can provide policymakers with information on the factors that implicate teachers' attitudes, concerns, and needs, taking inclusive practices to the next level.

Inclusive education in Kosovo
Special education in Kosovo started in 1950, organized in four special schools at the pre-university level, with a small number of students with special needs (Ministry of Education, Science and Technology,,2010). Later in 1991, there were four special schools and four attached classes part of the main schools, with 400 students with special needs and 60 employees (Aliu et al., 2003). By 1999, the education system in Kosovo was facing difficulties including special education (Ministry of Education, Science and Technology, 2010). After 1999, the education system was reformed, including the education of children with special needs, from segregated to inclusive. Extensive support was provided by international agencies including UNICEF, Save the Children and the Finnish Support for the Development of Special Education in Kosovo (FSDEK), which assisted in drafting documents, provided training, introduced comprehensive good practices and provided didactic materials (Behluli & Zabeli, 2014). The main documents that ensure the right to inclusive education are based on the Universal Declaration of Human Rights (United Nations Headquarters, 2006)  As an educational philosophy, inclusion is relatively new in Kosovo that aims "to include all children regardless of race, gender, ethnicity, disability or any other difference in the nearest regular school" (Ministry of Education, Science and Technology, 2014, p. 4). There are several ways to realize the education of children with special educational needs in pre-university education: in resource centers, special schools, in attached classes, in resource rooms and in regular classrooms within public schools (Ministry of Education, Science and Technology, 2015). There have been findings that the number of children with special needs attending education in resource centers and special schools is 349, while in regular schools 3903 (Ministry of Education, Science and Technology, 2020).
The Ministry of Education, Science and Technology (MEST, 2017) has worked for the realization of inclusive education by establishing laws and educational policies, however the education system in Kosovo has still got many difficulties in completing the concept of inclusion in educational institutions (Save the Children, 2017). One of the challenges is the mindset of understanding the concept of "child disability" from a medical perspective, and not from a social or field perspective (UNICEF, 2017). Generally, in Kosovo, the level of awareness and implementation of inclusive education is low (Zabeli et al., 2021). Inclusive culture in educational institutions continues to be prejudicial, which in fact affects expectations, process, quality and achievements. In this context, low awareness and attitudes towards inclusion, especially of teachers, have slowed down the process of achieving inclusion in practice. One of the main barriers faced by the inclusive process in Kosovo is the attitudes of teachers.
Inclusion as a current and challenging topic for the pre-university education system in Kosovo has been little studied. One of the researches at the national level is the qualitative one of Zabeli et al. (2021), regarding the perceptions based on the experience of teachers. The research found that teachers have general knowledge about inclusion but not specific information according to special needs; have the belief that inclusion has a positive effect on children with special needs from a social point of view, but not on a cognitive one. The other qualitative research is the one of Zabeli and Gjelaj (2020) with 10 preschool educators, eight of whom showed positive attitudes towards inclusion and a positive impact on the overall development of the child with special needs, while two of them were skeptical about academic achievement. No quantitative studies have been undertaken to provide a more comprehensive picture of attitudes towards inclusive education. Meanwhile, no national study has identified the factors that may shape their attitudes.
This scientific paper aims to explore the Kosovar context in relation to inclusive education. The two main goals of the current study are identifying the attitudes of teachers and exploring the factors that influence their formation, providing data on overcoming rejection attitudes and achieving a successful inclusion in the Kosovo education system.

The aim of the study
The aims of this study were to examine (a) General teachers' attitudes towards inclusive education; (b) Association between teachers' selected background variables and attitudes toward inclusive education; (c) Relationship between self-efficacy and teachers' attitudes about inclusive education. The study addresses a current topic and the focus of numerous and important research regarding the development of the pre-university education system, according to the contemporary concept of inclusive education. Research on inclusive education in Kosovo is minimal, especially on teachers' attitudes and the factors that influence their formation. The findings of the current research provide opportunities to generate more detailed information on teachers' attitudes, identify difficulties and address their needs, developing the process of inclusive practices. The key research questions were: (1) What are teachers' attitudes towards inclusive education in Kosovo?
(2) Is there a relationship between teachers' selected background variables and attitudes towards inclusive education?
(3) Is there a relationship between teachers' self-efficacy and attitudes towards inclusive education?

Research design
The research conducted a quantitative design, through a survey questionnaire. This design is used in case the research problem is the identification of factors that affect a certain result and measured by survey wich "provides a quantitative or numeric description of trends, attitudes, or opinions of a population by studying a sample of that population" (Creswell, 1994, p. 17).

Participants
Participants in the study were 738 in-service teachers. Teachers were from grade levels 1 to 12, from public schools in Kosovo. Total 40 schools from six regional districts were selected randomly. The number of teachers was 738, and 499 of them responded to the questionnaires (see Table 1).

Data collection instruments
The survey questionnaire was used to collect data, and it was divided into three parts. The first part gathered background information about the teachers. The second part of survey include the "The Teacher Attitude towards Inclusive Education Scale" (TAIS) scale (Saloviita, 2015) to measured teachers' attitudes towards inclusive education. The reliability of TAIS scale in other studies has changed from Cronbach's α = .81 to α = .90 regardless of the samples they have had (Saloviita, 2015). This scale consists of 10 statements, which are measured by 5 points of the Likert scale (1-strongly disagree, 2-disagree, 3-I am neutral, 4-agree and 5-completely agree). The scale includes four factors: "inclusion as value, expected outcomes, child rights and teacher workload" (Saloviita, 2018, p. 6). TAIS scale results were interpreted using values: 1 in 1.79 represents "strongly disagree"; 1.80-2.59 represents "disagree"; 2.60-3.39 represents "neutral"; 3.40-4.19 represents "agree"; 4.20-5.00 represents "completely agree".
The third part of the survey has used the Teacher Efficacy Inclusive Practices (TEIP) scale , to measure a relationship between teachers' self-efficacy and attitudes towards inclusive education. The reliability Cronbach's alpha of the TEIP scale was 0.93 . This scale has 18 items, with a 6-point Likert scale from "strongly disagree" to "completely agree" and include three main domains: efficacy in managing behavior, efficacy in inclusive instruction, and efficacy in collaboration. The questionnaire was translated into Albanian language. The TAIS scale reached the alpha coefficient a = .534. To achieve acceptable reliability, three items were removed, wich reliability coefficient reached a = .70 (see Table 2). Level of reliability of the TEIP scale reached a = .941 which indicates an excellent reliability of the scale (see Table 2).

Data collection procedure
The researchers obtained permission from principals of the schools to conduct the research. The principals provided the e-mail addresses of the teachers. The cover letter has included information for the research and the ethical part. The questionnaire was worked out in Google Form format, attached to the e-mail and sent to the teachers, which was completed on a voluntary basis by them.

Analyzing of data
Statistical analysis was performed through the SPSS Statistics program (Version 26). Data were analysed using Cronbach's alpha, descriptive statistics, t-tests, Cohen' d, analysis of variance (ANOVA) and Correlation procedures. The Cronbach's alpha is used to measure the reliability of the instrument and its scales. Descriptive statistics (frequencies, percentage, mean, standard deviation) were used to describe the data (attitudes, gender, age, educational qualification, teaching experience, teaching experience with children with special needs, previous training on inclusive education). Inferential statistical t-test was used to analyze the differences between males and females in relation to attitudes towards inclusion (gender). Another inferential statistical test, analysis of variance (ANOVA), was used to compare the results of three and more than two gropus (age, city, educational qualification, educational level of work, teaching experience, teaching experience with children with special needs, previous training on inclusive education). Pearson's correlation was used to determine the correlation between self-efficacy and teachers' attitudes, and to determine correlation between teachers' attitudes towards inclusive education with educational qualification, teaching experience, teaching experience with children with special needs, previous training on inclusive education. The multiple linear regression was used to predict the value of the dependent variable, through independent variables.

Level of teacher' attitudes towards inclusive education
Teachers' attitudes towards inclusive education were measured through the TAIS scale. The results of the research showed that teachers (N = 499) "agree" with inclusive education (see Table 4), so they have positive attitudes M = 3.55, SD = 0.58 (see , Table 3).
Descriptive analyses were used to describe and to compare the average between the items and to summarize the data. Descriptive statistics of teachers' attitudes towards inclusive education based on the TAIS scale are presented in Table 4. According to the responses, the items "A child with special educational needs should be transferred to a special education classroom in order not to violate his/her rights" were on average (M = 3.12, SD = 1.08), and "The best result is achieved if each child with special educational needs is placed in a special education classroom that best suits him/her" (M = 3.33, SD = 1.10) which achieved the lower average. The statements "Integrated children with special educational needs create extra work for teachers in mainstream classrooms" (M = 4.23 SD = .66), and "The learning of children with special educational needs can be effectively supported in mainstream classrooms as well" achieved the higher average (M = 3.63, SD = .87).

Level of teachers' self-efficacy towards inclusive practices
Teachers' self-efficacy towards inclusive practices was measured through the TEIP scale. The results showed high self-efficacy of teachers (N = 499) (M = 5.11, SD = 0.51) (see Table 5).    There was a statistically significant difference between educational level of work groups as determined by one-way ANOVA (F = 5, 706 Sig. < .000). A Tukey post hoc test revealed that secondary school teachers had slightly more positive attitudes (N = 194) (M = 25.31, SD = 4.03), than primary school teachers (N = 131) (M = 23.25, SD = 4.11) with p = 0.00. However, there was no statistically significant difference between the other groups.  Valid N (listwise) 497

Analysis of demographic variables data
Source. Field Data, 2020.
Koliqi & Zabeli, Cogent Education (2022) Table 6 presents the analysis of Chi-square test between dependent and independent variables. The results indicate that teachers' age, educational level of work, main regions, trainings for inclusive education, work with children with special needs, and self-efficacity are statistically significant variables to change the teachers' attitudes towards inclusive education. Whereas gender and work experience are not statistically significant variables (see Table 6).

The multiplate Pearson's correlation between teachers' attitudes towards inclusive education and background variables
The multiplate Pearson's correlation was run to determine the relationship between teachers' attitudes towards inclusive education, age, educational qualification, teaching experience, teaching experience with children with special needs and previous training on inclusive education. Coefficient between educational qualification and attitudes towards inclusive education is small negative relationship r = -0.130. There is a statistically significant small negative relationship between educational qualification of teachers and their attitudes towards inclusive education (r = -0.130, N = 495, p = 0.004). As well, there is a statistically significant small negative correlation between teaching experience with children with special needs and they attitudes towards inclusive education (r = −0.142, N = 497, p = 0.001).According on results there were no statistically significant association between teachers' attitude towards inclusive education with age (r = 0.052, N = 498, p = 0.248), between training on inclusive education and teachers' attitude towards inclusive education (r = −0.045, N = 499, p = 0.313), between teachers' attitude towards inclusive education with teaching experience (r = 0.039, N = 499, p = 0.387).
Between educational qualification and teaching experience with children with special needs (r = 0.043, N = 493, p = 0.336). The coefficient between age and educational qualification is a small negative relationship (r = −0.133), because the correlation coefficient is less than 0. Based on these results, there is statistically significant small negative relationship between teachers' age and educational qualification (r = −0.133, N = 494, p = 0.003). Pearson's correlation coefficient between age and training on inclusive education is negligible linear association (r = 0.201). Based on these results, there is a statistically significant negligible linear association between (r = 0.201, N = 498, p = 0.000). Between age and teaching experience, there is a high linear association (r = 0.741). Based on these results, there is a statistically significant strong linear association between (r = 0.741, N = 498, p = 0.000). Coefficient between age and teaching experience with children with special needs is negligible linear association (r = 0.131). Based on these results, there is a statistically significant negligible linear association between (r = 0.131, N = 496, p = 0.004).
Pearson's coefficient between educational qualification and training on inclusive education is small negative relationship (r = −0.227). There is a statistically significant small negative relationship between teachers' educational qualification and training on inclusive education (r = −0.227, N = 495, p = 0.000). Coefficient between educational qualification and teaching experience is small negative relationship (r = −0.096). There is statistically significant small negative relationship between teachers' educational qualification and teaching experience (r = −0.096, N = 495, p = 0.033). Between training on inclusive education and teaching experience, there is negligible linear association (r = 0.226). Based on these results, there is a statistically significant negligible linear association (r = 0.226, N = 499, p = 0.000). Coefficient between training on inclusive Source. Computed from the Field Data, 2020.
Note. p < 0.05 means there is a statistically significant relationship. education and teaching experience with children with special needs is linear association (r = 0.122). Based on these results, there is a statistically significant negligible linear association (r = 0.122, N = 497, p = 0.006). Pearson coefficient between teaching experience and teaching experience with children with special needs is linear association (r = 0.125). Based on these results, there is a statistically significant negligible linear association (r = 0.125, N = 497, p = 0.005).

The correlation between self-efficacy of teachers' and their attitudes towards inclusive education
The Pearson's correlation was run to determine the relationship between attitude and self-efficacy. Correlation coefficient of attitudes and self-efficacy by itself is good correlation (r = 1.0). Coefficient between self-efficacy and teachers' attitudes is a medium-strong linear association (r = 0.71). Based on these results, there is no statistically significant relationship between teachers' selfefficacy and their attitudes towards inclusive education (r = .71, N = 499, p = 0 .007; see Table 7).

The multiple linear regression
The multiple regression is used to predict the value of the dependent variable, through independent variables. The results show that the independent variables predict significantly the value of the dependent variable (F (3.189), p < 0.01) (see Table 8).
According to the results teacher's education qualification (p = 0.024), work with children with special needs (p = 0.002), and self-efficacy (p = 0.033) adequately predict the value of their attitudes' regarding inclusive education. The other independent variables, gender, age, trainings on inclusive education do not predict the value of teachers' attitudes (see Table 9).

Discussion and conclusion
This study has analyzed teachers' attitudes towards inclusive education and the factors that influence the formation of their attitudes. In general, Kosovar teachers have reflected positive attitudes towards inclusive education. Whereas, among the eight factors, the only variable that has shown significant differences in teachers' attitudes is the educational level of work. Moreover, teachers' education qualification, experience work with children with special needs, and selfefficacy predict the value of their attitudes. Therefore, the current study has identified some new insights regarding inclusive education.
Teachers in the current research have positive attitudes towards inclusive education, namely they "agree" with this practice. This finding corresponds with the findings of other studies (Barnes & Gaines, 2017;Zabeli & Gjelaj, 2020). However, according to the Official Gazette of the Republic of Kosova (2013), teachers in Kosovo have low level awareness for inclusive practices. Moreover, the teachers in the research conducted by Zabeli and Gjelaj (2020) have expressed that they do not have the professional competences and the necessary support to address the demands of the inclusive class. Although the teachers were not ready to work in the inclusive classroom, they were asked to implement it (Adewumi & Mosito, 2019). It is likely that teachers in Kosovo, since the beginning of the implementation of this practice, have not been prepared and continue to lack adequate competencies. In addition, teachers are challenged by the high number of students in the classroom, the lack of training on inclusion and the lack of human support Zabeli & Gjelaj, 2020). According to Boyle et al. (2015), if teachers believe that they do not have professional skills and do not have enough resources, this can hinder the achievement of the goal of inclusion. It is recommended that policymakers prioritize the development of inclusive policies, while school management should identify and continuously support the teachers' needs to achieve the full inclusion of children with special educational needs.
Concerning background variables in the research, there were no differences identified in the attitudes of teachers according to their gender, similar to the research findings of Chhabra et al. (2010). In general, female teachers have more positive attitudes than male teachers (Tsakiridou & Polyzopoulou, 2014;Butakor et al., 2018). Moreover, in the study of (Hadjikakou & Mnasonos, 2012), female teachers felt more prepared to work in inclusive classes and according to them, with a little commitment, they can include all types of children's disabilities. The findings of the current research did not show a positive relationship between the attitudes of teachers depending on age. However, research has shown that teachers between the ages of 20 to 30 have more positive attitudes towards inclusion, compared to those who are older than 40-50 years (Boyle et al., 2013). It is likely that this result derives from the completion of studies recently and from their enthusiasm in the first steps of practicing the profession as a teacher.
The findings of the current research did not show significant differences in the attitudes of teachers depending on the attended training on inclusive education, and negative correlation between them. These findings are similar to the results of other research (Ahmmed et al., 2012); however, according to Van Mieghem et al. (2018) the continuous professional development of teachers based on successful practices and experiences influences their attitudes. Educational qualification is a crucial factor in teachers' attitudes (Tsakiridou & Polyzopoulou, 2014); however, the results show no differences and even no correlation between them. But, they can predict the teachers' attitude, and through education programs they may have more information and are more prepared for inclusive classroom. Mónico et al. (2018) point out that if teachers have the knowledge and skills to work in the classroom with inclusive pedagogy, in all likelihood they will hold more positive attitudes towards it. Therefore, it is important that teacher education institutions design programs that provide knowledge and develop skills in applying methodology that responds to the diversity of students with special needs.
The research results did not identify differences between teachers with different levels of professional qualification. However, programs that prepare teachers are important for the formation of attitudes and preparation as inclusive teachers (Kraska & Boyle, 2014). It can be concluded that the higher the educational level of the teachers, the more positive are their attitudes towards inclusive education.
Regarding teachers' education level work differences were identified between the attitudes of teachers, respectively, teachers of secondary school level had slightly more positive attitudes, compared to their primary-level counterparts. Barnes and Gaines (2017) also found that teachers from secondary school have a more positive attitude than their counterparts who work with children from other educational levels.
According to Marshall et al. (2002), and Parasuram (2006), international studies show that less experienced teachers have more positive attitudes towards inclusion than those who are part of the educational system much longer. On the other hand, the current research found no differences between the attitudes of teachers with reference to work experience, and also no association between them. This may be due to the lack of experience of teachers' in inclusive classrooms and the lack of support during this process, as this relatively new practice is still in the early stages of implementation in regular schools in Kosovo. The attitude of the teacher' towards children with special needs can be formed under the influence of the experience the teacher has with them. Therefore, an influential factor in teachers' attitudes towards inclusion is their work and contact with children with special needs (Dorji et al., 2019;Saloviita, 2018). However, the research findings are not in line with other studies, and also a small negative correlation between teaching experience with children with special needs and the attitude towards inclusive education. In the research of Yeo et al. (2016), negative attitudes of teachers were associated with stress caused by challenges and difficulties in adapting to individual needs, while positive attitudes were related to skills of working with students and new information that the teacher has reached working with them. As well, in this study teachers' experience work with children with special needs can predict their attitudes, because they can get to know children, understand that they can achieve and with a little dedication can help their development.
In the current research, the teachers had high self-efficacy, but in terms of the correlation between self-efficacy and teachers' attitudes towards inclusion, research has found a weak correlation. Similarly, the research of Saloviita (2018) did not find a positive correlation between them, while the research of Savolainen et al. (2012) found a positive correlation between the highest level of self-efficacy and their attitudes towards inclusive education. Regarding the results teachers have high self-efficacy for inclusive practices but have indifferent attitudes to inclusive education. One of the reasons may be the extra work of teachers' in inclusive classrooms and classroom management, as well their skepticism in the academic achievements of children with special needs. On the other side, results show that teachers' self-efficacy can predict their attitudes. If teachers can successfully face difficulties in inclusive classrooms, their self-efficacy increases and their attitudes can be more positive.
Other findings of the research were also the correlations between demographic variables. Small negative correlations were found between teachers'age and educational qualification, educational qualification and training on inclusive education, teachers' educational qualification and teaching experience. Negligible linear relationship between was between age and training on inclusive education, age and teaching experience with children with special needs, and training on inclusive education, training on inclusive education and teaching experience, training on inclusive education and teaching experience with children with special needs, teaching experience and teaching experience with children with special. Only between age and teaching experience was strong relationship. The results did not show a strong correlation between the variables; however, they are an important part of shaping teachers' attitudes.
All these data illuminate the real challenge of achieving the inclusive agenda in Kosovo. It must first be understood that the equal and quality inclusion of children with special needs "is key to the formation of human capital and their participation in social and economic life" (WHO & The World Bank, 2011, p. 226). Inclusive education as a social construct and right of each child, should not be left to spontaneity or individual initiatives, but should be mandatory, not only legally but also in practice. In this context, Engelbrecht and Savolainen (2017) concluded that developing inclusive education requires a dynamic and collaborative approach between policymakers, school management and teachers. In order to develop inclusive policies in practice, the policymakers should put as a priority in their agenda the development of educational policies that ensure the advancement of inclusive practices. The research findings will help managers of the institution in teachers' preparation programs how to evaluate the current programs and make the improvements that are necessary. Also, programs should provide practical information on how to approach different individual needs, ensuring the success of all children. Moreover, school institutions, in addition to access to children, must provide training for inclusiveness and the provision of human and physical resources for teachers. The trainings should offer programs that expand the theoretical knowledge of teachers on the philosophy and main components of inclusiveness, but also to develop inclusive pedagogical skills, to be competent for the optimal development of children with special needs, and to meet the needs of their special education. Furthermore, teachers must possess the skills to use Universal Design for Learning, which responds to the diversity of all children in the classroom.
These circumstances are those which will raise the awareness of teachers to a higher level about their vital role towards inclusive quality education. At this matter lies the importance of research exploring attitudes and how teachers value their position in the process and what they need to achieve successful inclusion.

Future study
Recommendations for future researchers are (a) exploring the attitudes of school principals, students, student parents, supportive and special teachers; (b) identification of child and environmental factors; (c) researching the problem from a qualitative point of view.

Limitations
This study has some limitations as listed further: (a) the relatively small sample size in the survey; (b) research focus specifically on teaching attitudes in service; (c) quantitative research has limited the exploration of perceptions and experience through qualitative research, interview.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author's Contribution
All authors have been actively involved in the whole process of conducting the research and each of them has contributed to the preparation of the final product. Therefore, all authors are responsible for the content of this manuscript.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Availability of Data and Material
The data sets used and analyzed during the current study are available from the corresponding author on reasonable request.