Introduction

Improving schools and developing inclusion remain inescapable goals of political agendas and educational reforms internationally (Ainscow, Booth, & Dyson, 2006; OECD, 2015). In this respect, various authors have drawn on the "journey toward inclusion" metaphor. According to Nguyen (2015), inclusion is a journey that questions societies' values and policies. In Confronting Marginalisation in Education, Messiou (2012) offers an interesting framework in which to promote inclusion and sets out that professionals and students are constantly and continuously embarking on voyages of collaboration. The voyages seek to improve schools' capacities through the twofold aim of responding to diversity and of putting into effect inclusive values (Echeita, 2006).

Nowadays, inclusion is focused on the pragmatic idea of education as an emancipatory element that is able to transform children's and young people's lives (UNESCO, 2016b). Inclusion is a process that aims to find effective ways of responding to diversity and identify and remove barriers of presence, participation and achievement, with particular emphasis on those groups of learners who may be at risk of marginalisation, exclusion, or underachievement (Ainscow, Booth, & Dyson, 2006). At the same time, inclusion seeks to eliminate exclusionary processes from education and neg ative attitudes or responses to diversity in relation to race, economic status, social class, sexual orientation, ethnicity, religion, gender, language, and attainment, as well as with regard to disabilities (Ainscow, 2015; Messiou et al., 2016; UNESCO, 2009). In fact, one key aspect of inclusive education is that it responds to student diversity by creating learning environments and opportunities for all (Spratt & Florian, 2015). In this sense, the European Agency for Special Needs and Inclusive Education (2015) emphasizes the importance of empowering teachers to promote a truly inclusive education. Even so, making schools effective for all is probably one of the biggest international challenges of education today (Ainscow et al., 2012b).

The question of how schools can be more effective and inclusive has been widely discussed in recent years (Ainscow, 2012; Echeita et al., 2014; Hehir & Katzman, 2012; Intxausti, Etxeberria, & Bartau, 2017; Miles & Ain scow, 2011; Save the children, 2008). There is a growing number of tools, guides and resources being published that support inclusive education (UNESCO, 2015b; UNESCO, 2016a; UNESCO, 2017). The scholarly community has also been clearly influenced by the ethics of school effectiveness and the tendency to make measurements (Sammons & Bakkum, 2011). In fact, there is growing interest in assessing how the response to diversity is actu ally taking place in school settings (Arnaiz & Azorin, 2014). In most cases, the assessment of schools is conducted using tools (questionnaires and scales) that have been designed specifically for this purpose. In a reviewof the literature to search for instruments aimed for teachers, Azorin (2017a) reports that research focuses on: cultural diversity beliefs (Chiner, Cardona, & Gómez, 2015; López & Hinojosa, 2016; Vázquez-Montilla, Just, & Triscari, 2014); effective factors for inclusive education (Brandes et al., 2012; Kitsantas & Mason, 2012; Sharma, Loreman, & Forlin, 2012); measures to responding to diversity in schools (Alvarez et al. 2002; Dominguez & Pino, 2009; Ferrandis, Grau, & Fortes, 2010; González, 2012; Martinez, 2013); attitudes toward diversity and inclusion (Beacham & Rouse, 2012; Colmenero, 2006; Forlin et al., 2011; Montánchez, 2014; Vélez, 2013); inclusive education opinions (Bravo, 2013; López, Echeita, & Martin, 2009); and teacher training for the response to diversity (González-Gil et al., 2013; Lledó, 2009; Pegalajar, 2014).

In brief, teachers have always faced the challenge of how best to respond to the differences among students (Azorin, 2016; Cornejo, 2017; Messiou, 2017; Messiou & Ainscow, 2015). Therefore, it is important to have tools that enable us to examine this in greater depth. It was with this idea in mind that the Themis Inclusion Tool (Azorin & Ainscow, 2018) came into being. In this sense, Themis can be presented as an example of a range of similar instruments like the Index for Inclusion (Booth & Ainscow, 2011) or the Manchester Inclusion Standard (Moore, Ainscow, & Fox, 2007). Themis seeks to bring together the main issues of responding to diversity within the paradigm of inclusive education. The tool help teachers to reflect on how they are responding to diversity and, from the point of view of inclusion, identify what lines of change could be implemented to develop contextualized improvement plans. The scale covers three dimensions: contexts, resources and processes. A brief summary of the dimensions and the topics of the different items of each are presented below.

Contexts: This refers to the circumstances surrounding the schools. It is inspired by the ecology of equity, which advocates undertaking changes for greater inclusion within schools, between schools, and beyond schools (Ainscow et al., 2012a). In terms of school improvement, one key issue is context specificity and the institutional, social, geographical and political panorama (Chapman et al., 2012). Thus, the items of this first dimension are related to the socioeconomic situation, cultural diversity, education policy, leadership, pro inclusion values, preventing discriminations, teach er-student relationship, collaboration among teachers, family-school links, community engagement, and networks between schools.

Resources: This dimension is associated to the previous one and refers to personal, institutional and local resources. In this sense, it is important to evaluate the resources schools have available for inclusion (Valenzuela, Guillén, & Campa, 2014). The items of this dimension go into training resources, human, material, technological and physical resources, as wellas the consideration of the school as a resource, and community resources.

Processes: This dimension focuses on the presence, participation and achievements of children (Ainscow, Booth, & Dyson, 2006). The items of this last dimension stimulate thinking about celebration of diversity, teaching planning, the education process, variety of methodology, flexible heterogeneous groups, organization of times and spaces, support, and evaluation and transit between stages.

Aims of the Research

Method

Data analysis

Following the influential work of Muijs (2011) about doing quantitative research in education, data were analyzed using the SPSS 23.0 statistical software. Table 1 below shows the cut-off points for delimiting the profiles (barely inclusive, towards inclusion and inclusive) of the dimensions: contexts, resources and processes:

Table 1: Cut-off points

A barely inclusive profile corresponds to a school that needs to under take deep reforms to overcome and remove the barriers to inclusion. When a school has this type of profile this means that during the processes of reflection, self-evaluation or the diagnosis more weaknesses than strengths were detected in the response to student diversity. This implies a need for a change in direction and putting into effect actions or plans aimed at developing more inclusive practices in these schools.

A towards inclusion profile places a school as being on the way towards inclusion. While there may still be plausible inclusive practices to bring in, the focus in these cases is more on attending to those aspects which can still be improved in the light of the opinions expressed by the teachers. It is likely that a school with this profile will achieve inclusion if it undertakes ad hoc collective projects to improve the weaknesses detected.

An inclusive profile is associated with the paradigm of an effective and inclusive school. The strengths outweigh the weaknesses. Thus, a school with this profile reflects on cultures, policies and practices that converge towards an education that advocates the analysis of its contexts, resources and processes, that combats discriminating practices, that creates welcom ing communities and that seeks to compensate inequalities and fosters the construction of an inclusive society while working to achieve a true education for all scenario.

Results

Below we offer the results of the research according to the research aims.

Aim 1. To identify the main strengths and weaknesses deriving from the valuation the teachers make of the attention to diversity in their schools according to the three dimensions used in the study: contexts, resources and processes.

In terms of strengths, Table 2 shows the ten items that were most highly scored by teachers, with scores of between 4 and 5 points, where 4 = agree and 5 = totally agree.

Table 2: Strengths identified according to item, and percentages

The strengths revealed by the analysis correspond to the dimensions of context (items 8, 9, 12 and 17) and processes (items 44, 46, 47, 60, 61 and 64). The first strength identified corresponded to the development of inclusive values. Elsewhere, teachers take into account not only the result but also the process and identify themselves with values associated with inclusion. Similarly, when they plan their teaching, they take all their students into consideration and use various evaluation tools; they also think that there is a good atmosphere of coexistence within their school and periodically review and update their teaching program. Finally, teachers admitted that they design back-up and support activities and encourage families to get involved in the teaching-learning processes of their children and inform students and their families about their process from one educational stage to the next.

Table 3 below shows the ten items worst rated by teachers (with scores of 1 or 2 points: 1 = totally disagree; 2 = disagree), and these constitute the weaknesses found in attention to diversity.

Table 3: Weaknesses according to item, and percentages

The weaknesses found lie in the dimension of contexts (items 4, 15, 19, 20, 22 and 23) and resources (items 25, 26, 31 and 33). The weaknesses reflect no networking between schools at any level; a poor teacher participation in projects linked to inclusion; and a strong disagreement with the means available under current law to address attention to diversity. Other weak nesses that appear are the need for more specialists to attend to diversity, for tighter bonds between associations and for the involvement in the school of stakeholders related to volunteer groups. Teachers also indicated that there were insufficient computer stations for all students and that co-teaching activities were scarce. The lowest percentage was for the inventories of materials that are not made periodically and the scarce collaboration with other institutions in the surrounding area.

Aim 2. To diagnose the schools' profile (barely inclusive, towards inclusion and inclusive) according to the responses of the teachers for each dimension.

Figure 1: Percentages by profiles and dimensions

The scores given by the teachers for contexts place schools mainly (76.1%) in the profile called "towards inclusion," followed by 16% in "inclusive" and 7.9% in "barely inclusive". In terms of the resources dimension, 63.1% fit into the "towards inclusion" profile, 36.7% appear as "inclusive", and only 0.5% are "barely inclusive". In the case of the processes dimension, the percentages are closer, with 55.2% showing an "inclusive" profile and 42.6% a "towards inclusion" profile. The other 2.2% are in the "barely inclusive" profile.

Aim 3. To detect whether there are statistically significant differences for the following variables: type of school (state and private), location (rural, urban and peri-urban) and scholar stage (nursery, primary and secondary).

Table 4 below shows the frequency and the percentages of teachers whose responses are associated to each of the profiles according to the type of school.

Table 4: Significant differences according to type of school

The information in Table 4 above shows that there are statistically significant differences in both the processes and resources according to the type of school. Specifically, the percentage in processes for the inclusive profile for private schools was 72.6%, versus 51.6% for state schools and for resources it was 63.2% for the former versus 31.1% for the latter.

Table 5 below shows, for the three dimensions, the frequencies and percentages of teachers whose responses matched each profile according to location.

Table 5: Significant differences when taking into account the location of schools

From the data in Table 5, the deduction is that there are no statisti cally significant differences in terms of location. Finally, Table 6 gives the frequency and percentages of the teachers' responses for each dimension according to the scholar stage.

Table 6: Significant differences according to the school stage

The table above shows that within the inclusive profile there are sta tistically significant differences for the three dimensions during secondary schooling, where the lowest percentage is for contexts (7%), followed by resources (20.7%), and processes (37.5%). In this case, it was nursery and primary school teachers who perceived their schools to be more inclusive than secondary school teachers did, with the latter rating the practices in their schools lower in all the dimensions.

Discussion

The aim of this study was to explore the response to the challenge of student diversity in nursery, primary and secondary schools in the Region of Murcia. It delved into the strengths and weaknesses that teachers face in their response to diversity and, therefore, in their journey towards inclusion. Other studies exist in this context that address teachers' self-assessment education processes (Arnaiz & Azorin, 2014) and the implementation of improvement plans (Arnaiz, Azorin, & Garcia, 2015). However, it is not only important to evaluate the processes in schools in order to be able to adopt measures to change and improve, but also to study the contexts the schools find themselves in and the resources available to them torespond to diversity. As laid out in the theoretical grounding in this paper, research has been conducted in Spain to ascertain teachers' concepts regarding the inclusion process and to identify segregation, integration and inclusion stances (López et al., 2009). Our research, however, offers schools another type of diagnosis about their current situation and the improvements that can be made in terms of inclusion, which can be considered to be an original contribution to this area of knowledge. Far from seeking to distinguish good and bad schools, the aim is to draw up a map of the current state of the response to diversity on the basis of the profiles established. From there, work can be undertaken to (1) alleviate the weaknesses detected in the schools categorized as barely inclusive; (2) seek new formulas to pave the way forward for those schools fitting into the towards inclusion profile; and (3) maintain the good practices in effect in the inclusive schools. In this, we agree with Darretxe et al. (2013) that, in order to advance effectively in inclusion, one has to start from the current situation of each school. Hence, it is necessary to create interruptions in which to reconsider the day-to-day work in the schools and explore ways of developing more inclusive practices (Ainscow, 2005).

Noteworthy among the main strengths found was how much teach ers identified with values associated with inclusion and that their praxis fosters inclusive values among students. Indeed, the furtherance of these values is considered key in developing more inclusive schools (Booth & Ainscow, 2011). Another positive factor to emerge from the findings is that teachers take into account all their students when planning their teaching and periodically review their teaching programme. Among the items that scored most highly was the readiness of families to involve themselves in their children's education. In this sense, building collaboration bridges with families contributes to the school-without-walls paradigm proposed by Santos-Guerra and De La Rosa (2013), where families have an active and leading role to play in the school institution. Another strength to appear from the findings was the transit between scholar stages. There is a ped agogical current today known as "scholar transit," which supports the need to accompany students and their families during this process (Perry, Dockett, & Petriwskyj, 2014).

Regarding the weaknesses detected, it was observed that a large number of schools did not engage in support and collaboration networks with other schools on their area. Although there is work going on to set up these type of networks in Spain, there is still much to be done before they become part of the reality of our education system. However, in other countries, like the United Kingdom, this is a well established practice in schools (Ainscow, 2016; Muijs, Ainscow, Chapman, & West, 2011). According to Azorin and Muijs (2017), networks can be seen as conduits for educational change and inclusion development, and also as a "mirror" wherein the present and future of education takes place. Other weaknesses detected had to do with the links between schools and communities, local associations and volunteers. Schools have to overcome barriers and look beyond the school gates, which means increasing collaboration with local communities (Azorin, 2017b; Dyson & Todd, 2011). One feature that high lighted the scarce collaboration culture in the teachers surveyed was the absence of co-teaching actions. In this way, authors like Rodriguez (2014) have emphasised the benefits of co-teaching as a strategy for improving education and improvement, and others, like Ainscow (2011), insist on the importance of promoting a more collaborative culture within which teachers support one another.

In terms of the profile of the schools in our survey, it was noted that, within the dimensions of contexts and resources, the scores were mainly within "towards inclusion". This reflects that teachers value their schools highly and believe that, in general, their institution is on the right path towards developing more inclusive practices. As regards processes, the highest percentage was for the "inclusive" profile. Nevertheless, these data should be viewed with caution, since many of the items in this dimension are direct questions about the actions of the teachers, so the social desir ability of providing suitable answers needs to be considered.

Finally, some statistically significant differences were observed for pro cesses and resources according to the type of school, with a more optimistic perception of inclusion being shown by teachers in private schools than in state schools. This is different to the findings of research by León and Arjona (2011) in private schools in Andalusia, which reported insufficient resources for inclusion that did not adapt to the students' needs. There were no significant differences for location. However, the results obtained in the study of Callado, Molina, Pérez, and Rodriguez (2015) demonstrate that rural areas do not take full advantage of the context they are in to favour inclusion processes and continue to develop proposals that are merely integrative. To conclude, differences were observed within the inclusive profile for the three dimensions in secondary schools compared to nurs ery and primary, indicating that these are more inclusive than secondary schools. This is in agreement with Bravo (2010), who reported that there were more inclusive practices in primary than secondary schools.

Conclusion, Limitations and Strengths of the Study

As is apparent throughout this paper, improving schools and developing inclusion are two of the aims on the world agenda of education (Ainscow, Booth, & Dyson, 2006). So how schools can overcome the barriers to better inclusion is a recurring issue in the literature. However, in order to be aware of the strengths and weaknesses in schools and, in particular, those of theprofessionals who work there (leadership teams and teachers) efficient tools are needed to detect the positive and negative aspects of attention to student diversity. This is important to ascertain what is being done well and what requires improvement and also to make teachers more aware of their actual situation. In this sense, Themis proved of great help in fostering reflection by teachers about the contexts, resources and processes that underpin their education and affect inclusion (Azorin & Ainscow, 2018).

Thinking on the overall research aim associated to the exploration of the teacher response to the student diversity challenge in schools, in terms of the strengths and weaknesses in the schools of Murcia, the conclusion drawn from our analysis is that the main strengths are to be found in the processes dimension, which indicates that teachers value their own teaching highly. In contrast, most weaknesses were detected in the contexts dimension, and especially the area of community engagement with the school, and vice versa. This points to a need for collaborative practices that bring together the social and professional capital in the schools' sur rounding area and to foster more exchanges with other schools, institutions, associations and volunteer networks (Azorin, 2017b).

The profile "barely inclusive" was scored low in the three dimensions considered (contexts, resources and processes), while "towards inclusion" was scored highest for contexts and resources, and "inclusive" for pro cesses. The conclusion in this respect is that the schools in this study and the teachers responding are in the main on the way towards inclusion. This is encouraging but it makes clear too that work is still necessary to get the "inclusive" profile into first position for all the dimensions.

Similarly, it is unquestionable that variables such as school type, loca tion and scholar stage demand different ways of addressing inclusion, so the response to diversity must be treated according to the educational context and the conditioning factors (Echeita et al., 2014).

Finally, the sample was chosen on the basis of accessibility, which may be considered as a limitation as there were only 3 private schools compared to 35 state schools. An evaluation of statistically significant differences between the two types of schools would probably require a more balanced sample to compensate some of the assertions in both categories. Likewise, the total population (545 teachers) came from a specific area, although it should be stated that this was a pioneer study for the Region of Murcia, where it is hoped that future studies will comprise larger samples and hence greater representativeness. At the same time, there has to be a recognition of the limitations of only having teacher perceptions while there are other instruments, such as the Index for Inclusion (Booth & Ainscow, 2011), that encourages schools to compare the views of different stakeholders, including students and parents.