How do Children Value Educational Inclusion? Opinions from Primary Schools

Dancing with the concept of inclusion: The North-South divide

The word inclusion is part of the vocabulary of people linked to the world of education, from those who do it from their offices in high-level international institutions, to those who practice their profession as educators in a rural school somewhere in a small corner of the world. But what is the meaning of inclusion? Is it the same for the high officials at institutions as for the professional in their daily work? The same discourse can have different interpretations because the contexts are also different, as can be seen in the Unesco (2020) report on education in the world, which shows that while there are universal mechanisms contributing to educational inequality, "others are specific to social and economic contexts, as the covid-19 pandemic has laid bare" (p. 6). The scope of this problem even reaches experts who adhere to inclusion and then go on to ignore it in their research work (Messiou, 2017).

Inclusive education is often defined as overcoming barriers to learning and development for all children (Alcáraz-Garcia & Arnáiz-Sánchez, 2020). However, the levels of application and implementation of inclusive education vary from country to country (Magumise & Sefotho, 2020; Sánchez et al., 2019) and involve various problems in countries of both the Global North and the Global South (Armstrong et al., 2011; Unesco, 2020). The deep economic gap between one and the other is evident, as covid-19 has recently revealed. In Latin America and in many other parts of the world, many people survive on a subsistence economy. They go out every morning to sell their modest agricultural output in the streets of their town or city and, obviously, it is impossible for them to meet the minimum standards of protection for those that the pandemic forces. The divide that marks the distance between the Global North and South has its repercussions in all areas and, of course, in education.

In the context of southern countries, inclusion refers to providing support at an ordinary school to students with disabilities (Slee, 2014). In the most disadvantaged communities, inclusive discourse is present in the policies, but must confront conceptions and practices that, subtly, keep exclusion alive (Hansen et al., 2020; Slee, 2019). Placing some students in specific categories (the so-called special educational needs, hereafter SEN) marginalizes other students who do not fit into any predetermined category (Magumise & Sefotho, 2020). Furthermore, the presence of students with SEN does not guarantee their inclusion at all because relations with the rest of the members of the educational community may not reach the expected and desirable degree of coexistence, empathy, and solidarity (Marchesi, 2014). Although in the Ibero-American educational systems great progress has been made in inclusive policies, much remains to be done until all its participants perceive the usefulness of this way of thinking about education (Giménez, 2018).

In northern countries, the neoliberalism, and the interest in promoting standardization in education (Azorin 2016; Liasidou & Symeou, 2018) make inclusion a great challenge also for rich countries. International agencies such as the United Nations (UN) and Unesco have made great efforts to defend inclusion as a fundamental principle of schooling. But the reality continues to show that, in many northern countries, what really counts are school performance and measurable results (Armstrong et al., 2011). Walking towards a school truly grounded in the principles of inclusive education will require new spaces for reflection, collaboration and co-construction shared between social and educational institutions (Fiuza-Asorey et al., 2021; Parrilla et al., 2018).

Inclusion in Spain

The Spanish Educational System has evolved from segregation to integration and the current laws in force recognize that the educational system is inclusive (Agramunt et al., 2018). However, given the lack of social and pedagogical consensus (Giménez, 2018; Gómez, 2021), it can be affirmed that in fact real change is still to come.

Spain has the highest rate of early school leavers in the European Union, above other less developed countries that have recently joined (López et al., 2016), which is easily linked to the impact on children at school of the fact that since 2009 the population living in poverty has continued to grow (Save the Children, 2015).

Education in Spain is governed by the Organic Education Law (in Spanish, LOE), by the principle of attention to diversity, that gives priority to common education for all students -the so-called "ordinary education"-, but the possibility of receiving a special education if necessary (art. 4.3). The Educational Administrations have the responsibility of providing resources to the students who require "an educational attention different from the ordinary" (art. 71.2) and, more specifically, to those students who "[...] require, for a period of their schooling or throughout it, certain specific educational support and services on account of disability or serious conduct disorders" (art. 73).

Campoy (2011) highlights two serious problems arising from this law: (a) diversity is synonymous with disability and (b) it is proposed, on days deemed appropriate, to segregate those who cannot be adequately cared for in an ordinary centre. Recent research conducted by Alcáraz-Garcia and Arnáiz-Sánchez (2020) has highlighted the reality in Spain: in the last ten years, 13.54 % of students with special educational needs have been enrolled in specific special education centres. In this way the situation of exclusion is maintained since not only is disability understood as a problem, but also the line that marks segregation is blurred according to the "terms determined by the Educational Administrations". This, in practice, means the emergence of inequalities based on the place of residence.

In the last year, a new education law has come into force: Organic Law 3/2020, of 29 December, which amends Organic Law 2/2006, of 3 May, on Education (Lomloe). This law repeals the Lomce, which was in force until the 2020/2021 academic year. The Lomloe (2020) has a threefold objective: modernising the education system, enhancing equity, and guaranteeing educational inclusion. It may be necessary to wait for the current academic year to see whether the law really justice to the purpose does for which it was created in terms of educational inclusion, although some positions are already critical of the Spanish and international commitment that this law makes to the development of more inclusive education (Echeita, 2021; Gómez, 2021).

Inclusive education should be understood as a reform that responds to the diversity of all students (Ainscow & Messiou, 2018; Echeita, 2021). In Spain, the changes in the principles, objectives and values of the educational system are subject to the lack of consensus among the Autonomous Communities when applying these measures, and to the concern about the efficiency of the results (Gómez, 2021; Gutiérrez, 2019). These problems shake the four pillars of twenty-first century education based on constructivism: learning to know, learning to do, learning to live together/learning to live with others, and learning to be (Delors, 1996). Therefore, there is a distancing from inclusive philosophy (Sanz & Serrano, 2019).

Context of the study: The voice of the students

Inclusive education can never be the result of the enactment of a law but will only germinate because of the effort of a community and all its members (Andújar & Rosali, 2014; Corral, 2019). The culture of inclusion should not be improvised but is a fundamental aspect to work with in each school organization (Curieses, 2015). For this reason, it is necessary to open debates and discussions on how to make schools become more inclusive and democratic spaces (Sierra et al., 2019) because each educational community has a unique reality (Echeita, 2021).

Encouraging schools to work to be more inclusive (Ainscow & Mes-siou, 2018) means creating educational policies and daily practices that create school cultures that boost participation, where everyone learns, thinks, and makes decisions together (Väyrynen & Paksuniemi, 2020). The success of the educational system cannot be restricted to the good intentions and investments of the State. The voice of teachers has often been thought of as the main lever for change towards a more inclusive education (de Boer et al., 2011; Echeita, 2021; Giménez, 2018), especially in primary education (Mitra & Serriere, 2012). However, the voice of students is gaining more visibility on the national and international scene (Azorin, 2016; Urbina-Garcia, 2019). Like families, boys and girls have the right to be heard (Ainscow & Messiou, 2018; Grande & González, 2015; Messiou & Ainscow, 2015), because as Messiou (2018) points out, their voices are an important instrument in moving towards more inclusive policies, as inclusion means raising the voices of all members of the education community.

Therefore, listening to the voices of the students is the objective of the present investigation. Rarely does it happen in Spain that attention is paid to the boys and girls, and their insight. Nevertheless, experiments have been conducted, such as those reported by Durán et al. (2005) in Madrid, Catalonia, and the Basque Country, or by Coronado and Belarte (2018) in Valencia. All of them put into practice the Index for Inclusion of Booth and Ainscow (2015) or, more recently, use the Acade Guide (Arnáiz & Azorin, 2014; Escarbajal et al., 2020) or the Themis (Azorin et al., 2019). This experiment in Galicia refers to listening to the voices of 211 primary school students.

Participants' demographic information

A quantitative design was adopted. A descriptive-exploratory study was carried out to analyse the variables related to inclusive politics, culture and practices that correspond to the dimensions of inclusion. The study also had a transversal nature, since it examined a specific population at a certain time, observing the educational reality, the behaviour of its dimensions, axes, and the relationship and frequency between them (Hernández et al., 2015).

An intentional non-probability sample of 211 students aged 6-12 years (M = 8.27; SD = 1.640) was selected. Ninety-two were boys (43.602 %) and 119 girls (56.398 %). They were enrolled in Primary Education in six public centres of the Autonomous Community of Galicia, attending to their diversity in both urban-rural and population distribution by provinces, being distributed between the academic years: first^t (n = 37; 17.535%), second (n = 50; 23.696%), third (n = 52; 24.645%), fourth (n = 25; 11.848%), fifth (n = 25; 11.848%) and sixth (n = 22; 10.427%).

Instrument

The material used to collect information was a translation and adaptation of the Index for Inclusion (third version) (Boot & Ainscow, 2015). It consists of a self-applied questionnaire, with 29 items that evaluate on a 4-point Likert-type scale (1. Strongly disagree; 2. Disagree; 3. Agree; 4. Strongly agree) the degree of agreement with a series of statements regarding three dimensions:

1. Culture: creation of a community in which all members share inclusive values and all of them have a role in the participation and collaboration processes.

2. Policies: actions are proposed aimed at centres introducing inclusion in their policies and in management processes.

3. Practices: actions that show the deployment of culture and inclusive politics or philosophy.

Table 1 presents the values of the mean, the standard deviation and the Cronbach's alpha evaluating the consistency of the items by dimension. The resulting values indicate an excellent internal consistency of the complete instrument, i.e. considering all the items of the three dimensions (α > .913), and good internal consistency of its component dimensions; that is, for culture (α > .720), for policies (α > .876) and for practices (α > .762). In table 1 is also presented the alpha if typified elements are considered.

Table 1: Distribution of the results and internal consistency coefficient of the Index for Inclusion

^a Standard deviation.

Source: own elaboration by means of IBM-SPSS v27.

The Index is a tool or set of materials designed to facilitate the development of inclusive education in our schools, the purpose of which is not to serve the purposes of the research, but to build collaborative school communities that foster high levels of achievement for all learners. Therefore, neither the authors nor their translators into different languages have incorporated data on its psychometric properties in any of its versions, but on complementary studies that have used this tool have obtained similar results in the analysis of its validity, maintaining the structure in three dimensions (culture, policy, and practices) that refer to a more general concept (Menino-Mencia et al., 2019; Sánchez et al., 2019), and also similar in terms of reliability (Fiuza-Asorey et al., 2021).

Procedure

The questionnaires were applied in six Public Centres of the Autonomous Community of Galicia. Ethical considerations including information and consent for the study in families have been observed, respecting the confidentiality and protection of the data of the participants. Given that the instrument used is not only for research purposes, but also for self-improvement of the centres and of teaching, the centres were offered the possibility of their personnel getting involved in the analysis and evaluation of the barriers to inclusion. A deadline of one week was estimated for the application and delivery of the questionnaires.

Data analysis

The data analysis was performed through the IBM SPSS 24 Statistical Package. For a significance level of 95 %, the normality assumption was not fulfilled (K-S test p < .001). However, knowing that the normality assumption can be assumed for large samples (n > 100) (Moore, 2004) and, after determining the equality of variances through the Levene statistic (p > .001), it was decided to resort to parametric tests. Specifically, the following tests were used: T-Student to analyse the global Inclusion Index and its three dimensions, establishing comparisons by sex; Simple Analysis of Variance (ANOVA) to study the differences by age (from 6-12 years old) and by grade (first-sixth year); in the latter case, they were regrouped in three groups. Finally, Pearson's Correlational Analysis and simple linear regression analysis were used to check the relationship between the dimensions of the Index and with respect to age.

The confidence level was established at p < .05. Additionally, the effect size was calculated through Cohen's d (1988), considering the interpretation of small effect when d = .200, medium when d = .500 and high when d = .800.

Description of the mean and deviation values in inclusive culture, policies, and practices

In general, the students were closed to agree with the statements indicated in the Index (M = 2.503, SD = 0.443), with the dimensions inclusive culture (M = 2.666; SD = 0.569) and policies (M = 2.463, SD = 0.536) standing out. The inclusive practices obtained ratings that approximated the "Disagree" category (M = 2.380, SD = 0.466).

The central tendency values (mean, median), dispersion (standard deviation), position (percentiles) of each dimension of the index, and the total are presented in table 2, differentiating by sex, age, and grade.

Table 2: Descriptive statistics: central tendency and dispersion values

^a Mean; ^b Standard deviation.

Source: own elaboration by means of IBM SPSS v27.

Table 2 shows, in a descriptive level, a favourable trend for girls in relation to the perception of inclusive policies and practices and, in general, in terms of inclusion in their centres. There are hardly any variations by sex in the dimensions of inclusive cultures. These differences were, in all cases, less than one point. Regarding age, the younger (6-7 years old) valued more positively the inclusive culture and policies, compared to a more negative assessment (with values around point 2 'Disagree') among the 11 -12-year-olds. The valuation of inclusive practices increases from 6 to 8 years old and decrease again until 12. In relation to the academic year, there was a decline in the valuations, coinciding with the age.

Differences by sex, age, and grade in relation to the values of inclusive culture, policies and practices

The statistical T-Student showed that there were no statistically significant differences in relation to sex, neither in the total value of the Index (t₂₀₉ = 1.077; p =.283), nor in its dimensions: inclusive culture (t₂₀₉ = -.5 45; p = .587), policies (t₂₀₉ = -1.624; p =.106) and practices (t₂₀₉ = .541; p = .589). Everything points to the fact that the small differences observed were due to the variability of the sampling (standard error), thus assuming the absence of differences in the means by sex of the population.

The ANOVA test revealed the presence of statistically significant differences regarding age, among students of six age groups, in their perception of inclusion (F =54.444; p < .001; d = .432) and in inclusive culture (F = 38.683; p < .001; d = .381), policies (F = 30.077; p < .001; d = .450) and practices (F =29.300; p < .001; d = .317). The effect size was moderate or medium.

The data in Table 3, which presents the results of the post-hoc test with Bonferroni correction, led to the conclusion of the presence of differences by age in the three dimensions and in the total.

Table 3: Multiple comparisons: post-hoc tests for the variable age, with Bonferroni correction

^a Only statistically significant results are Included. ^b p < .05; ^c p <.001

Source: own elaboration by means of IBM-SPSS v27.

Table 3 shows that boys and girls aged 6-7 obtained higher means than the rest in inclusive culture and policies, with no differences between them. The same happened with children aged 7-8, whose answers were above those aged 9-10 and 11-12. In inclusive practices, there was a significant difference between young people (aged 6-9) compared to older students (aged 11-12). It seems that, as age advances, the perception of the inclusion in the centre decreases. These changes in perception were progressive, while there were no differences between close ages (between 6-7, 7-8, 8-9, 9-10 or 11-12 years old). The effect size of those differences was moderate, although it was low in several cases (d < 0.200).

An ANOVA analysis performed for the academic year reported statistically significant differences both for the total (F = 151.284; p < .001; d = .340) and for its dimensions: inclusive culture (F = 116.666; p < .001; d = .413), policies (F = 82.312; p < .001; d = .312) and practices (F = 85.352; p < .001; d =.381). The association between the course and the perception of inclusion in terms of culture, politics and inclusive practices was confirmed.

Table 4 shows a more positive perception of inclusion among the first/second-year students compared to those in fifth-sixth year, especially regarding cultures, where the greatest mean differences were observed with respect to the group of fifth-sixth year (M differences = 1.087) and in the total (M differences = 0.895). Differences were also found between third-fourth year, although they were less notable and even not significant in terms of practices (p = .999) The magnitude of the effect of the differences was moderate in most cases.

Table 4: Multiple comparisons: post-hoc tests for the variable academic year, with Bonferroni correction

^a p < .05; ^b p < .001

Source: own elaboration by means of IBM-SPSS v27.

Relationship between the dimensions of the Index and the age

The data in Table 5 demonstrates the existence of significant and positive relationships between the dimensions of the index; these were moderate (r > .40). The age was included in the analysis, which allowed to confirm a significant, negative, and strong relationship of this variable with the dimensions of the index.

Table 5: Correlations between the Index for Inclusion and age

^a p < .001 (2-tailed).

Source: own elaboration by means of IBM-SPSS v27.

With the intention of checking how age affects, differentially, each of the three dimensions of the Index, as well as its overall effect, table 6 shows the value of the coefficient of determination (R² adjusted), which indicates the percentage of variance of the criterion variable that is due to the predictor variable (age). In addition, information is included on the weight of the predictor variable in each of these variables. All regression values are presented as positive, with the constant (age) being a negative value, i.e. age acts as a predictor of a lower level of inclusion in the centre in all its dimensions.

Table 6: Simple linear regression model with the predictor variable: age

Note. ^a p < .001

Source: own elaboration by means of IBM-SPSS v27.

The results indicated that age presented a significant linear relationship with the scores obtained in the Index. Specifically, age manages to explain 60.6 % (r² = .608; r² adjusted = .606; F = 323.76, p < .001) of the total score obtained in the Index (Y = 4.03 - .182X), the 52 % of the dimension inclusive cultures (Y = 4.48 - .0.217), the 39.1 % of the policies (Y = 3.95 - .0.177X) and the 37.7 % of the practices (Y = 3.65 - .152X). The significance coefficients of the predictor variable in each dimension and in the total were significant (p < .001). The t-test showed the possibility of generalizing the results to the population.