Segregation in Basic School in Haiti, Reflecting the Social Relations of Inequality

This article aims to establish, on the basis of a mixed explanatory sequential estimate, how actors of the educational system (teachers and school managers) use academic and social mechanisms to distance strong students from the weak by means of categorizing or distributing them unevenly and differently in the school space. On the theoretical level, while relying on Bourdieu's structuralist constructivism, cultural discontinuity theory and systemic discrimination as a framework for analysis, this study combines three approaches to the sociology of education (functionalist, conflictualist and the approach of rational choices) as epistemological base, to try to understand the dynamics of school segregation and the mechanisms, by which it reflects the social relations of inequality. From a methodological point of view, it begins an empirical research conducted in the Western Department, starting; on the one hand, from a survey conducted by questionnaire with 303 actors (100 teachers, 95 parents, 13 managers of institutions and 95 students), and four group interviews with five teacher groups, four individual interviews with principals. Using quantitative and qualitative methods in an explanatory mixed sequential quote process, quantitative data analysis is followed by qualitative analysis to consolidate and explain the quantitative results.


Introduction
This article addresses the issue of the relationship between school segregation and social relations of inequality. It aims to establish that school segregation observed in the basic school of the Western Department of Haiti is the reflection of the social relations of inequality. The Haitian school system, in its manner of functioning and its method to separate social and school (P. E François, 2016;Tardieu, 2015;Joint, 2005;Bourdieu, 1966), does not treat all the students the same way (Tardieu, 2015;P.E François, 2010;Joint, 2005). Therefore, quality education for all is far from being a reality in Haiti despite the commitments made by the Haitian government. Indeed, the Haitian school system is made up of a set of unevenly effective educational institutions, some of which are more efficient than others, offering different education to the students depending on their socio-economic background (Tardieu, 1990(Tardieu, , 2015Joint, 2005; P.E. François, 2010;D. Pierre, 2012). The system is unfair, as those researchers have already pointed out. It fails to allow all to have access to a good quality education, whereas since the reform of JC Bernard in 1982 to national plan of education and formation (PNEF, 1997(PNEF, -2007 passing by the conference of Jomtien in 1990 and the World Forum of Education which took place at Dakar in 2000. The Haitian Government advocates access and quality education for all. Unfortunately, schools with good quality education are not numerous and are distributed in the country according to the social class of students, which Coleman et al. (1966) call systemic discrimination of supply. There are areas that have establishments of better quality and therefore yearn by the majority. Some (favored and moderately favored) families take ownership of these facilities for their children, while other disadvantaged families are content with schools of poor reputations offering substandard education. Thus, students who are socio-economically disadvantaged do not benefit from the same quality of education as the others and become disadvantaged in terms of both their learning and their learning outcomes (Joint, 2005; P.E. François, 2010;D. Pierre, 2012;Tardieu, 2015;Felouzis et al., 2016).
This paradoxical situation of unequal differentiated distribution of school skills and knowledge (Felouzis et al., 2016;P.E. François, 2010;D. Pierre, 2012;Joint, 2005;Tardieu, 1990Tardieu, , 2015, in terms of quality education for all advocated by the state, suggests a lot of questions. The first are those relating to school segregation practices: in a system that values quality education for all, can there be practices of school segregation? Then, those related to quality education for all and equity: Is it possible to promote equity and quality education for all when children from disadvantaged socio-economic backgrounds are concentrated in reputable non-performing schools with doubtful schooling methods and children from favored and middle-class families are welcomed into high-performance and prestigious institutions? Subsequently, the problems linked to the question of the influence of social relations of inequality in regard to school segregation: can school segregation be related to the social relations of inequality?
The article is structured around five parts. The first one questions the state of the art in this area. The second briefly addresses the methodology. The third presents the findings of the study on "Linking School Segregation and Social Relationships of Inequality". The fourth begins the discussion of these results and the last part focuses on the conclusion.

School Segregation: What We Talking about
The contributions in this section question the relationship between school segregation and social relations of inequality. Indeed, at first glance, the notion of school segregation seems to refer to the theoretical finding that it is not the product of chance, and it can also be explained by the socio-occupational category to which the parents of students belong, because the socio-economic position of each plays a big role in school segregation. In the sense that each member of a society belonging to a social category (or social bloc to borrow the concept of Alain Bihr) has an income, a social status and a social position places it in a relative position, either in a dominant or dominated position, relative to others in society (Bourdieu, 1979). Coleman (1990), in his explanation of the functioning of social systems, postulates that the social system has two components: actors and events that are connected by two types of links: control and interest (degree to which well-being of an individual is affected by the realization of an event).

Social Relations
Unlike Coleman (1990), Pfefferkorn (2007) in addressing the issue of social structure emphasizes that it can be understood as a complex dynamic interweaving of all social relationships. The author, according to a Marxist point of view, states "social reality is the unity resulting from the organization of all social relations, a unity that in no way excludes contradictions between them and therefore implies no closure of this reality about itself". In this vein, Bihr (2012) points out that the social hierarchy presents itself as a kind of continuous scale of positions (determined by wealth, prestige, education, power, etc.) giving the best birth to layers, social blocks or social strata (Bihr, 2012, p.18) and thereby establishing social relations of inequality or asymmetric.
For his part, Bourdieu (1984) constructed a theory of social space. This theory explains the logic of the formation of social groups from the modes of hierarchization of societies, the lifestyles and struggles of these social groups and the way in which social hierarchies and groups are reproduced. In its logic, Bourdieu (1984) defines society as an interlocking of autonomous fields, defining each of the specific modes of domination: economic, political, cultural, artistic, sporting, religious, etc. Each field, the author continues, is at the same time a field of forces. It is marked by an unequal distribution of resources and therefore there is a relationship of forces between dominant and dominated. There are competitive relationships between social agents and each field is characterized by mechanism for capitalizing legitimate resources that are unique to it (Bourdieu, 1984).
The diversity of ideas emerging from this analysis of social structure has allowed us to understand that Haitian society, as a human collectivity structured by a hierarchy, seems to be composed of individuals and social groups, with opposite interests, having asymmetrical and conflicting social relations that establish the bonds that constitute the social life of this society. They are hierarchical (social groups) and maintain relations marked by the domination of some and the differences, or even inequalities, existing between members of different groups (Hurbon, 1987;Labelle, 1987;Tardieu, 1990;Joint, 2005;François, 2010;Pfefferkorn, 2007;Pierre, 2012;Bihr, 2012). As a result, Haitian social reality is similar to a conflictual social space structured by dominant positions (Bourdieu, 1984), and includes a set of asymmetrical social relations between social groups whose main challenge is unequal access to social services. Resources and these relationships have a practical meaning. Social relations therefore have two components: an economic component (income) and a cultural component (the level of graduation).
This analytical distinction and that of sociologists in the area of social stratification, allowed us to propose the following composition: in Haiti there are three major social blocks with conflicting interests. An easy-to-use social block that speaks French correctly (Labelle, 1987;Tardieu, 1990Tardieu, , 2015Hurbon, 1998;François, 2010;Pierre, 2012) holds the means of production (economic, social and cultural) and it dominates the politico-socioeconomic and cultural field: it is the social block of the privileged; an intermediate social block (it is the social block of medium or medium level favored) whose position in the various social fields contributes to perpetuating the domination of the social block of the privileged; and finally, the great mass with its culture of a predominantly creole-speaking social category, with its labor force and constitutes the social block of the disadvantaged or of modest origin.
Haitian society is therefore hierarchical with different social groups and having unequal access to the country's valued resources (material, social and political, cultural) (Labelle, 1987;Bihr, 2012;Dorvilier, 2012;Tardieu, 2015). In short, the constitutive hierarchy of this social training thus generates a system of inequalities preventing social cohesion and thus leading to the accumulation of advantages or handicaps (Pierre, 2012;Bihr, 2012;François, 2010;Pfefferkorn, 2007;Tardieu, 1990Tardieu, , 2015 and contributing to the reproduction of the asymmetrical social relationship between those who possess and those who do not (Bihr, 2012, p.22).

School Segregation Factors
The theoretical factors and currents relating to school segregation are many and varied. Plenty of studies have attempted to answer the question of factors of school segregation from several angles. Some authors attempt to explain the links between school segregation and social segregation (Riegert, 2016;Merle, 2012;Duru-Bellat, 2004, etc.). Others focus on school segregation and residential segregation (Delvaux and Serhadlioglu, 2014;Ly, 2011…), school segregation and ethnic segregation (Felouzis et al., 2009;Merle, 2012…), academic segregation and school segregation (Riegert, 2016;Merle, 2012…), school segregation and social composition of students (Piketty, 2015) etc.
Our initial analysis of this literature review suggests that there are a number of factors (non-school and academic) that act as causes and effects (Conger, 2005;Dronkers and Levels, 2006;Walsemann and Bell, 2010;Park and Kyei, 2010) school segregation and that it acts on the trajectory of students and academic success (Riegert, 2016;Merle, 2012;Ly, 2011;Duru-Bellat, 2003).
In fact, Payet (1998), in school segregation: a sociological perspective on violence in schools, stresses four factors of school segregation: the existence of a private educational sector, the development of the consumerist practices of families (search for quality institutions), autonomy of institutions and their evaluation, ethnicity of the school market (inflation of the value of the school property).
For their part, Felouzis et Perroton (2009), is growing up among peers at school: ethnic segregation and social reproduction in the French educational system, on the basis of the nationality of the pupils and their given names as an indicator of their ethnic origin, highlight the idea of the complexity of social reproductive patterns in schools by showing the weight of peer groups in the socialization process of adolescents and state that the quantitative criterion of segregation. The strongest was the ethnic origin of the students, which was much more than their social origin. Merle (2012), points out that some factors other than geographical or urban segregation explain school segregation. To substantiate his thesis, the author, unlike Payet (1998), goes beyond the analysis of . In his research on school segregation, based on a threefold approach (historical, international comparisons and segregation measures), Merle (2012) highlights the existence of four types of school segregation: 1) Sexual segregation which emphasizes social sex relationships and explains the long-term segregation of girls and their different paths from boys. The school reflects not only the social roles of sex and occupational segregation, but also the centre of their reproduction (Merle, 2012, p.10); 2) ethnic segregation takes into account foreign pupils and their distribution in schools: the French school organization establishes a segregation based on the color of the skin and the ethnic origin and results from the concentration of immigrant populations or children of immigrants in the suburbs and working-class neighborhoods (Merle, 2012, p.11); 3) social segregation, which consists of the separation of pupils according to their social origin. According to Merle (2012), this segregation dimension is weak when there is little difference in the way students are grouped according to their social recruitment, instead it is strong in the opposite situation (Merle, 2012, p.15) and; 4) academic segregation, which takes into account four forms: difference in educational level between pupils, distinction between schools according to the average level of competence of their pupils, the way in which classes are set up, options and sections reserved for the best students, and the existence of the private/public intersector. The author reports that the strength of the links between ethnic, academic and social segregation shows that segregation is cumulative, and justifies the generic use of the term "school segregation" (Merle, 2012, p.18). Riegert (2016) corroborates Merle's analysis (2012) by using the exposure index to measure social and school segregation. Indeed, in his thesis on "School Inequality, Segregation and Peer Effects", on gender issues in the school, Riegert (2016) has described quantitatively the extent of social and school segregation in secondary education. The author highlights a strong difference between the social environments attended by the middle school and high school students according to their social background or level of education.
All in all, it remains true that school segregation is not only determined by academic segregation, residential, geographic, social or ethnic segregation…, and it builds (school segregation) power relationships that place social groups of students in asymmetrical or unequal social relationships and as a result, one wonders if it is not also a reflection of the social relations of inequality.

Effects of School Segregation
To explain the effect of school segregation on learning, Felouzis et al., (2016), in school inequalities and education policies, focus on three inequality vectors: composition effects (being schooled in a school or class on a low average has negative effects on their learning), the effects of school climate (school values, learning environment, school culture versus student culture, motivation to learn) and the quality effects of teaching (level of education and experience of teachers, nature of mobilization of school teams) (Felouzis et al., 2016, p.8). According to the authors, "these three vectors (composition effects, school climate and quality of schooling) are strongly interconnected and together form the quality of education offered to students." The Organization for Economic Co-operation and Development (OECD), in 2012, in Equity and Quality in education: supporting disadvantaged students and schools also drew the attention of European countries on the harmful effects of school segregation on student success. Wherefore, she states, in addressing both layers of equity 'fairness and inclusion', which correlate with education, […] students with low parental education, low socio-economic status has higher risk of low performance. These two dimensions of equity, fairness and inclusion, overlap. Often, low performance converges in specific population groups; disadvantaged students are at higher risk of low performance than their more advantaged peers. (OCDE, 2012, p.16) The Council of Europe, a policy paper entitled 'Fighting against school segregation in Europe through inclusive education', published in September 2017 attests that school segregation reduces students' chances of acquiring skills and violates their right to education.
School segregation is one of the worst forms of discrimination. It seriously undermines the students' rights whose learning opportunities are undermined by isolation and lack of inclusion in schools (Council of Europe, 2017, p.5) It not only discriminates children and violates their right to receive education on the basis of equality, but it also reduces their chances of acquiring, alongside other people, the essential skills of life. (Council of Europe, 2017, p.13) Other authors such as Reardon, Yun and Eitle (2000), Conger (2005), Dronkers and Levels (2006), Walsemann and Bell (2010), Park and Kyei (2010), Merle (2012) are all unanimous in recognizing that school segregation has other determinants and can produce perverse effects. In this sense, Dronkers and Levels ask the following question: Do school segregation and school resources explain region-of-origin differences in the mathematics achievement of immigrant students? To answer this question, they conducted a study in which Levels and Dronkers (2007) show that: (…) Ethnic school segregation affects the scholastic outcomes of native and immigrant students from some regions of origin more than those of immigrant students from other regions. The analysis shows that neither ethnic, nor socioeconomic, school segregation explains the lower mathematics achievement of immigrant students from Latin America, Northern Africa, and Western Asia (Dronkers & Levels, 2007, p.435-462) In addition, from a study linking an ethnographic survey of two academic inspections in France, a survey by interview and the exploration of statistical data on a cohort of students entering second class in 2001, Laforgue (2004) shows that school administration contributes to the production of school segregation. He points out that: (…) deconcentrated school authorities' practices can have an impact on the degree of schools' diversity. In particular, the members of the Rectorate and the Academic Inspectorate make decisions concerning the offer of training and the sectorization of the institutions, which clearly have segregative effects. In addition, he has also shown that the issues of school administration are not reduced to logic external to the school institution. The production of segregation by the administration is also due to the fact that it focuses on issues of management of structures and personnel (Laforgue, 2004) Other studies emphasize that school segregation can influence the behavior of segregated students and affect their aspirations and academic success. With this in mind, Walsemann and Bell (2010) attempt to establish the relationship between "integrated schools, segregated curriculum: Effects of within school segregation on adolescent health behaviors and educational aspirations": (…) they examined the extent to which within-school segregation, as measured by unevenness in the distribution of Black and White adolescents across levels of the English curriculum (advanced placement-international baccalaureate-honors, general, remedial, or no English), was associated with smoking, drinking, and educational aspirations, which previous studies found are related to school racial/ethnic composition. "Our results provide evidence that within-school segregation may influence both students' aspirations and their behaviors" (Walsemann & Bell, 2010, p.1687-1695 In contrast to these studies that focus on the effects of school segregation either on the differences of origin or the factors that produce segregation, Reardon, Yun and Eitle (2000) analyze its effects on ethnicity and social groups. In this sense, they are interested in: "The changing structure of school segregation: Measurement and evidence of multiracial metropolitan-area school segregation, 1989-1995". This study focuses on: (…) aggregate patterns and trends in segregation among white (non-Hispanic), black, Hispanic, and Asian public-school students in 217 metropolitan areas during the period 1989-1995. We first describe a set of methodological tools that enable us both to measure the mutual segregation among multiple racial groups and to partition total metropolitan-area school segregation into geographic and racial components. Then we use these tools to examine patterns and trends in metropolitan-area school segregation. We find that the average levels of multiracial school segregation have been unchanged from 1989 to 1995, but that this stability masks important shifts in the geographic and racial components making up average levels of total metropolitan school segregation. In particular, segregation between non-Hispanic white students and all other students has increased, on average, while segregation among black, Hispanic, and Asian student groups has declined. In addition, the contribution to average levels of total metropolitan segregation due to between-district segregation has grown, whereas the relative contribution of within-district segregation has declined (Reardon, Yun, & Eitle, 2000 p.351-364) To base the effects of school segregation on the chances of success in school, some studies compare two types of school segregation: in schools between students and between immigrants themselves. Thus, Conger (2005) examines racial segregation among racial groups between non-immigrant and immigrant students. Indeed, his article examines: (…) ethnic segregation, defined as segregation among racial groups as well as between native-born and immigrant students, across elementary school classrooms in New York City. Specifically, the study compares patterns in within-school segregation across ethnic groups, grades, boroughs, and years. Current levels of within-school segregation are also compared to levels of across-school segregation and to levels of segregation that result from three simulations where students are assigned to their classrooms: (a) randomly, (b) to achieve complete ethnic segregation, and (c) according to their prior year test scores. Results indicate that racial segregation across schools is far greater than racial segregation within schools; however, the segregation of immigrants' within-schools is equal to the segregation of immigrants across schools. Within-school segregation cannot be entirely attributed to random processes or to the use of ability grouping practices, particularly in the case of black and Hispanic segregation. Finally, segregation within-schools varies considerably across the five boroughs and declined during the second half of the 1990s (Conger, 2005, p.225-244) School segregation also causes performance gaps between different groups of students. Indeed, Park and Kyei (2010) undertake a comparative study in eighteen (countries) and we quote: (…) our comparison of performance gaps in mathematics between native and immigrant students across 18 countries shows considerable cross-national variation in the size of performance differences between native and immigrant students. We hypothesize that countries with higher levels of school segregation by family socioeconomic status (SES) and immigrant status should display stronger effects of immigrant status. In countries with high levels of school segregation, immigrant students should be more likely to be sorted into schools that are poorer and/or have higher concentration of immigrant students, which may in turn negatively affect immigrant students' achievement. Two-level hierarchical linear models show that countries' native-immigrant gaps are systematically related with levels of school segregation by family SES but not with levels of school segregation by immigrant status. The finding implies that redistributing key educational resources to disadvantaged schools to moderate the negative consequences of attending low SES schools may facilitate educational integration of immigrant students (Park & Kyei, 2010, p.207-228) This point around the authors' work on school segregation and social relations of inequality makes it possible to draw four main conclusions. The first shows that social groups in Haiti do not have the same social positions, assets and dispositions and are linked by asymmetrical social relations, conflict or domination or even inequality and the question of the link between school segregation and social relations of inequality is not addressed in Haiti. Secondly, all of this work illustrates the breadth and variety of research conducted on school segregation and has produced a vast body of theoretical and empirical literature, whose purpose is to explain the logic of socially differentiated schools, and to put the emphasis on emphasis that is placed on the factors, perverse effects and impact of school segregation on failures, performance, success, the relationship between ethnicity and school segregation, and students' academic backgrounds. These studies suggest that the factors of school segregation are much more environmental and external to the school system and that these effects of externalities have a substantial effect on the overall dynamics (Piketty, 2015, p.81). The school participates in school segregation for the purpose of building academic excellence as a tool for social categorization and imposes it on all its members and users as an organization that has the power to construct the real, as the legitimate definition of objective reality (Perrenoud, 1984, p.16-17). This analysis gives us the opportunity to grasp the complexity of the social phenomenon "school segregation" and to suppose that there is a possible logical link with the "social relations of inequality". Thirdly, this review confirms that school segregation is implemented differently and according to the context in which the actors find themselves, and although the interest in the phenomenon is rather recent in Haiti, it is a subject of study, already ancient in France and in American and Anglo-Saxon sociological literature, etc. It represents an obstacle to the collective learning of citizenship and living together in Haiti. As a result, school segregation, as a social construct, is a collective action. Fourthly, even if this review of the literature reveals a certain convergence between the socio-economic and cultural conditions of parents, ethnicity and the level of academic competence of children, as vectors of school segregation, one observation is essential: the essential work does not focus on the relationship between school segregation and social relations of inequality. Since then it is no less true that school segregation is not only determined by academic segregation, residential segregation, geographical, social or ethnic ... it builds (school segregation) power relations that put groups students in asymmetrical social relations or inequality and as a result, it is questionable whether it is not also the reflection of the social relations of inequality.

Materials and Methods
In this article, we used the mixed explanatory sequential estimate method (quantitative and qualitative approaches). The choice of these approaches is justified, because on the one hand, school segregation is a multi-dimensional and therefore complex subject that covers aspects which go well beyond the school setting. On the other hand, our research question requires several data sources (primary, secondary, big data) and mixed research methods are considered as intermediate solutions between two epistemological positions: the search for an explanation for the link between school segregation and the social relations of inequality and the search for truths multiple to explain the nature of this link.
The two approaches (quantitative and qualitative) are combined sequentially and the results are used to meet the objective of establishing that school segregation observed in the basic schools of the Western Department of Haiti is the reflection social relations of inequality. However, let us underline that the study is not general in scope since not all respondents were randomly selected, and schools were intentionally selected from sampling by reasoned choice, so it is a non-probability sample.
This research, as we stated before, was carried out in two stages: a) quantitative and b) qualitative. "We start with quantitative data collection and analysis, followed by a qualitative phase to explain quantitative results (triangulation and complementarity objectives).
An empirical research conducted in the Western Department, starting, on the one hand, from a survey conducted by questionnaire with 303 actors (100 teachers, 95 parents, 13 managers of institutions and 95 students), and four group interviews with five teacher groups, four individual interviews with principals For the quantitative phase, collection instrument consists of four questionnaires, which have been pretested. A student questionnaire was developed along with a teacher questionnaire, a school manager questionnaire and a parent questionnaire. Thirteen schools of different types (congregation, communal, community, lay, Protestant, public) from basic education in grades 6 to 9 in the Western Department agreed to participate in the survey in order to select respondents.
Four hundred and fifty (450) questionnaires were sent to these participating schools and administered to parents of primary school students, pupils from the sixth to the ninth year of basic education and teachers of basic education who have been invited by the school manager to participate in the survey. Three hundred and three (303) participants returned completed questionnaires (approximately 67 % of respondents).
Eight individual and semi-structured group interviews were held. The aim is to invite certain persons involved in the quantitative survey to comment on the same themes of the quantitative phase, namely school segregation, social relations of inequality, social composition of school and class, social discrimination, socioeconomic status of parents. In fact, the qualitative stage carried out more than 500 minutes of discussion with school managers and teachers. Indeed, we then conducted four individual semi-structured interviews with school managers and four group interviews with groups of five teachers on the topic "school segregation and social relations of inequality".

Quantitative Analysis of Results
As part of this survey, we selected and administered a questionnaire to a sample of one hundred (100) teachers to verify their perceptions of the practice of school segregation. Initial results show that they all practice school segregation, mainly using the grades obtained by the students to categorize them. However, only 15 % of teachers say they use school segregation in their class (often or quite often), while 83 % of respondents say they do not use this practice often and only 2 % did not answer the question. Table 1 shows the distribution of schools by type and practice of teacher segregation.
In order to practice school segregation, teachers in eleven (11) out of twelve (12) schools confirm that they have used two school segregation mechanisms: academic mechanisms (student grades) and social mechanisms (social origin, language of instruction or gender). Indeed, teachers in 11 out of 12 schools use the academic level of pupils (score obtained) as an academic mechanism to separate them and only one (1) a school where teachers use social background to categorize students. As can be seen from Table 2, schools are distributed by type and according to the criteria for the classification of pupils by teachers. Source: survey on linking school segregation with social relations of inequality The quantitative phase reveals that all schools responding to survey questions practice school segregation (often, quite often or occasionally). Indeed, school segregation is implemented from a set of mechanisms that are used to discriminate students in high, medium and low, according to respondents. These mechanisms are the grade obtained (academic segregation), social origin (social segregation), language or sex. With regard to the academic mechanism, 77% of the respondents (teachers) practice segregation in their classes. The second mechanism used to segregate school is social origin (social segregation). Nearly 10% of teachers use the social background of students as a criterion for categorizing them. The last two mechanisms that respondents argued were the language of instruction (French) and sex that are used as tools of school segregation.
The survey reveals that students do not attend the same schools. The majority of parents (80%) choose Congregational and prestigious schools based on the quality of education offered.
Respondents, to justify their segregationist behavior, put forward several basic reasons: On the teacher side, the reasons are diverse. Some want to:  Provide mentoring for some students who they describe as weak (38 %);  Guarantee excellence at school (26 %);  Categorize students (14.5 %); For their part, some school administrators seem to want to:  Stimulate students (61 %);  Facilitate learning (23 %);  Better deal with strengths and means (8 %).
Regarding the practice of school segregation, it was found from the data collected from survey participants that respondents (teachers and school administrators) segregate school and distribute students in different school spaces with different types of school segregation and different treatments through school segregation mechanisms. Bourque (2008), Keslassy (2010) and Dhume (2014) believe that discrimination is the first mechanism to exclude, humiliate or classify others and treat them differently; in our case, it is the weak pupil. However, Bourque considers as mechanisms to exploit the difference the fact of "unequally treating a person or a group of people on the basis of a criterion of unlawful discrimination ..." (Bourque, 2008, p.83, quoted by Bissonnette, 2017). In the survey we conducted, the results reveal that teachers and school administrators separate or discriminate students according to certain criteria or mechanisms (academic and social). This approach corresponds to the different forms of school segregation described, first, by Merle (2012), such as academic segregation or social segregation: […] According to Merle (2012), social segregation is the separation of students according to their social origin. This dimension of segregation is weak when the methods of grouping students differ little according to their social recruitment, strong in the opposite situation and the academic segregation takes into account different forms: difference in academic level between students, distinction between schools according to the level of average competence of their pupils, etc. ... (Merle, 2012, p.15) Then, by Riegert (2016) who corroborates Merle's analysis (2012) using the exposure index to measure social and school segregation. Indeed, in his thesis on "School inequalities, segregation and peer effects", dealing with the issues of diversity in the school, Riegert (2016) quantitatively describes the extent of social and educational segregation in secondary education. The author highlights a strong difference between social environments attended by high school students according to their social background or their educational level. He observes that: student from an affluent background has an average of twice as many students in his class, also from a well-off background, as a student from a popular or average background. Similarly, the best students on average have twice as many students in their class who are also very well-educated than low-and middle-level students (Riegert, 2016, p.9) Finally, by Felouzis et al. (2016), in school inequalities and education policies that define school segregation as: the separation of students into structures, streams, classes, even groups of different levels according to their personal characteristics (social origin, gender, ethnic origin, etc.) (Felouzis et al., 2016, p.38) By practicing school segregation, respondents to the survey (teachers and school administrators) thus satisfy two conditions mentioned by Van Zanten (2012), cited by Felouzis et al. (2016) to talk about segregation at school: the first is that we must highlight that the school concentration of certain populations results not only from external factors, but at least partly from the specific action of institutions and school actors (Van Zanten, 2012 cited by Felouzis and al., 2016, p.38, cited); the second is that, to talk about school segregation, we must be able to show that spatial segregation at school leads to the production of specific forms of school and social inequality and exclusion (Felouzis et al., 2016, p.38).
Besides, we found that the language of instruction is used to separate students in the classroom, which can have perverse effects on their learning and their right to education (Reardon, Yun and Eitle, 2000;Joint, 2005, Dronkers and Levels, 2006, Walsemann and Bell, 2010, Park and Kyei, 2010, Merle, 2012, Council of Europe, 2107. Indeed, French, a colonial heritage and the official language of the Republic of Haiti, is a language of social distinction and domination (Hurbon, 1998, p.217-238) reflecting the social relationship of inequality between the three large social blocks (favored, intermediate, disadvantaged). As a result, this language occupies a prominent place not only in official administration and communication, but also in teaching both as a language and as a school discipline. Nevertheless, for Creole-speakers attending school, the French language separates them from other students and isolates them from the basic learning that conditions their academic success and creates academic inequalities between those who master it and those who cannot master it.

Qualitative Analysis of Results
School segregation is a topic that is the subject of almost all the speeches of those interviewed. This topic is addressed in different ways by respondents. According to the participants, the school makes a differentiated distribution of pupils in the school space according to certain school or non-academic mechanisms that may be academic (classification according to grades or school performance) or social (gender, language, social origin, socioeconomic status, etc.). School segregation evokes the image of the translation of social relations (of domination or inequality) in the school space by considering that it manifests itself when being pupils, classified by the school system in distinct social groups, endowed of a differentiated social valuation, separated in the classroom or in the school space and brought together little by little.
Normally our society already evolves in a segregationist system; schools are not spared this phenomenon. In our schools or in the schools, there are several groups of individuals and the schools are taking a path, perhaps, that is not educational. Sometimes we can say that such and such a student does not react, we classify him as such (Maintenance E-01) ».
Others believe that school segregation exists and takes into account the socio-economic status of students, because some parents of students have access to resources that others do not. As such, not only are students being categorized, but also groups are not receiving the same treatment: "Often, our country, divides the class of people. Even at school, when you have parents who have good means to pay on time, to buy materials for their students, these kinds of students can have different treatment than others who have no means (Maintenance E-03)".
Does not enter who wants high performing schools with prestigious reputation offering a good quality education (Tardieu, 2015;D. Pierre, 2012;P, François, 2010), because the selection is intense and highlights criteria that can only be met by the favored and moderately favored groups with a strong cultural heritage. These criteria, according to Joint (2005), have a single objective: to select students whose parents can afford to pay tuition fees and supervise their children (Joint, 2005, p.200). Indeed, when asked how they recruit their quality students, the respondents said: "Through evaluations that have nothing to do with traditional evaluations, for example, asking children "what do you know about?" "Define" and so on, we're not interested in that because children can study, but how to use what they've studied". And so, through assessments that we do by putting children in the production situation. And when they are in a production situation, we evaluate their ability to integrate into development, their ability to go and obtain the knowledge needed to respond to this development, their ability to work as a team and that is extremely important and from there we evaluate their productions. We also take them to self-assessment so that they can correct themselves, so that they can achieve their goals and the goals of the community (Interview E-04) ".
The survey indicates that school segregation through the social origin mechanism and discriminates against students according to the educational level and socioeconomic status of their parents. The supervision of the family, their social position, their economic and cultural capital contribute to accentuate school segregation since it is those supervised by their fathers and mothers who obtain good results, according to the respondents.
"…There are students whose parents are mathematicians, engineers, the child works during the holidays, we give them the responsibility of supervising the others who are much weaker and during this time, I give them a lot of exercises too so that they can move forward (Interview  ".
According to the interviewees, students are treated according to their means. The child who pays well for schooling has a better treatment than the one who has trouble paying.
"…There are children who have all their documents, who pay the school well and these students can have many more values compared to other students. We do not do it explicitly, but implicitly, there is always that because we cannot classify all the children in the same category (Interview E-03) ".
Students do not have the same opportunities to succeed. The Haitian school, those we analyzed in the Western Department, uses either the social origin or the academic mechanism to categorize or arrange them in the classroom, according to respondents. In addition, teachers discourage weak students while encouraging the best students to become their preferred cooperators (Joint, 2005, p.332): "Listen, we are human, it happens sometimes, the teacher acts in terms of friendship, economic and intellectual. For example, a child who is quite talented in the classroom, he enjoys a certain privilege in the classroom even though there may be some difficulties, but the child who is gifted, one puts him in front of to the child who takes time to understand (Interview E-01)".
Some categorize students by putting the strong in the same class: "… There are schools that do the opposite, all able students are in the same class and that is why there is class A, B, C, etc., Interview E-03".
The study reveals that the language of instruction is also a mechanism of school segregation. Interviewees think that the fact that the language of instruction is French puts Creole-speaking students in a linguistic position that separates them and does not promote their learning, and therefore they do not benefit from the academic skills and knowledge taught: "… We find ourselves with people who are torn apart and linguistically, we will see that we will say to the individual: express yourself. This means in a subtle way that Creole is forbidden. In some schools, tokens were given to those who speak Creole on the school yard. It is a form of linguistic repression that excludes the learner from the learning process (Interview E-02)".
According to survey participants, the French language requirement at school can cause learning difficulties for Haitian students. The fact that students do not master the language of instruction (French) puts them in a situation of discomfort that conditions their participation or not in the classroom. French separates students, isolates them and stifles any desire to participate in their learning: "There are several factors, for example, there is language factor, that is to say, difficulty of speaking. Because if you speak French and then the child does not speak French, he will not to participate, he is not going to ask a question, he is not going to say that he has not understood, he is not going to do anything, he will just sit down and wait for it to end (Interview E -05)".
In the sense of Merle (2012) and Riegert (2016), our interviews make it possible to analyze the question of school segregation as the effect of certain academic and social mechanisms. In fact, Merle's research (2012) identified four forms of school segregation: differences in the level of schooling among pupils, distinction between schools according to the average level of competence of their pupils, mode of class formation and the existence of options and sections reserved by law for the best students, and the existence of the private/ public intersector. The testimonies of teachers and school administrators lead us to believe that school segregation is practiced in schools through these mechanisms and there is a link between it and the social relations of inequalities. In our interviews, interviewees let us know that: The analysis of the issue of school segregation in the literature available shows a recurrence of the phenomenon that would be related to various factors and the results of this research are very different in terms of the determinants of school segregation. Some mention non-academic factors and others point out school factors. Delvaux and Serhadlioglu (2014) focus on residential segregation as a factor in school segregation and Payet (1998), Laforgue (2005, Merle (2012) focus on school strategy in class composition and the social selectivity of school choices as determinants of school segregation. For her part, Duru-Bellat (2004) claims that segregation is caused by social segregation, which is not different from the results of our interviews. It seems that there are a number of factors that influence school segregation and whose respective weight is not easy to isolate: first, it (school segregation) reproduces the social inequalities of success, knowledge acquisition and progression, the school organization and how it works.
The analysis of school segregation has led many other researchers to focus on the "social relations" dimension. These researchers , Oberti 2007, Joint 2005, Van Zanten 2004 emphasize how this dimension structures school strategies, especially those of upper-and middle-class families. This research shows that the school strategies of parents from the upper classes are highly contextualized, because they are part of socially differentiated school spaces (Pourpeau, 2008). Depending on their place in the social space, families do not have access to the same school offer (Oberti, 2007, Duru-Bellat, 2003 or the same school markets , and some schools have a social composition of upper and middle class students compared to others who have children from modest or disadvantaged backgrounds (Joint, 2005).
Second, the theory of cultural discontinuity (Bernstein, 1975, Willis, 1978, Bourdieu and Passeron, 1966, Lahire, 2008cited by Felouzis et al., 2016 and the theory of systemic discrimination (Coleman et al., 1966) based on a similar finding in the very logic of hierarchical societies (Felouzis et al., 2016) states that school reproduces and accentuates school inequalities by means of all kinds mechanisms, including school segregation. The first shows that, since school is indifferent to differences, it reproduces social relations by means of school segregation by reflecting the social hierarchy. The second focuses on the differentiated distribution of education provision by type of student; so, the school reflects not only the social relations of inequality but also accentuates them through school segregation using academic and social mechanisms. In this vein, according to these theories, the school has a responsibility in the perpetuation of the social relations of inequality (Bourdieu, 1966, p.336).
According to Bourdieu and Passeron (1985), there is an interaction between school culture and family culture. According to Felouzis et al., (2016), this interaction can be explained by the theory of cultural discontinuity.
According to the results of our research, the respondents state that "the supervision of the family contributes to accentuate the segregation of school since it is the students who are supervised by their fathers and mothers who obtain good grades and are considered as gifted scholars. Thus, the social position of parents, their cultural and economic capital influence the separation of students and schools from one another. In addition, according to the respondents, the majority of parents of prestigious lay and congregational students are educated (at least a bachelor's degree), they accompany students in their studies and provide material supervision compared to the rest of the parents who have barely any education, third cycle of fundamental ".
In the interviews, teachers and school administrators point out that students who have a lack of mathematics have filled them with the help of their parents who are engineers. The same observation is also pointed out by French teachers who stipulate that students who do better in French are those who are supervised by their parents and practice at home.
Since students' parents are unequally socioeconomic and culturally unequal, students themselves will be unequal (Felouzis et al., 2016, p. 12). If the school is indifferent to differences, it only reproduces the social relations of inequality (Bourdieu, 1966). Boudon (1973), for his part, thinks that school inequalities are rooted in the very structures of social stratification.
According to the respondents, in addition to the above-mentioned theories, the way in which the survey participants (teachers and school administrators) select and distribute students in the school space only reproduces school inequalities. In short, social inequalities through school segregation mechanisms are as follows: It is clear that such a teaching system can only function perfectly as long as it manages to recruit and select those who are able to meet the demands that it imposes objectively. That is to say, as long as it is addressed to individuals endowed with cultural capital (and the ability to make this capital grow) that it presupposes and consecrates without explicitly demanding it and without transmitting it methodically. The only test he can really feel is not, as we see, that of numbers, but that of the quality of the taught (Bourdieu, 1966, p.340) Based on the mechanisms used (academic and social) to prioritize or categorize students into strong, medium and weak (or gifted, less gifted and mediocre) and according to the socio-economic status of their parents (favored, moderately favored and disadvantaged), it seems that there is a link between school segregation and the social relations of inequality. Indeed, behind school segregation, as a source of production of social and educational inequalities (Bourdieu, 1966, Felouzis et al., 2016, are read the social relations of inequality whose school is at school, both a product and a reproducer (Bourdieu and Passeron, 1970).
All in all, according to our observations, school segregation supposes the existence of a division, that is to say, the separation between the dominant social group and the dominated social group. It is this division that is at the heart of school segregation and founds it as an anti-pedagogical principle: in a well-organized and well-organized segregated school, the high-achieving pupils are divided into the weak or the average, so that there are two parts , namely those who hold the head of the class on the one hand, and weak students on the other hand. We could try to say that there are those who govern didactic and pedagogical activities (governing party) and those who suffer the misdeeds of school segregation (governed part).
The school system contributes directly to reproducing the social order of society by maintaining and legitimizing the social position of origin. All in all, from the conflictualist perspective, school segregation, which consists of discriminating between students and performers, the average and the weak, or schools between performers, quasi-performers and weak, only reflects the image of the social hierarchy (favored social bloc, intermediate and disadvantaged), by analogy. Indeed, according to Baudelot and Establet, "the school is said to be unique, neutral, egalitarian and meritocratic, whereas in fact it divides each generation between those who are destined, by their social origin, to join the bourgeoisie or the proletariat from which they come " (Baudelot and Establet, 1975, p.92).
Instead of meritocratic consent, school segregation forces us to think of pedagogical practice as social relations of inequality. Therefore, the pedagogic situation is no longer conceivable as a means facilitating the learning of all in the classroom, but as a tension or pedagogical power, in other words as opposition of two social groups: those who are successful and those who are not. If one expresses this opposition in rigorous terms, one will say that school segregation is a pedagogical principle of domination by which the school institution, as a pedagogical power, opposes all the segregated (scholars).

Discussion
Quantitative and qualitative analyses of the results revealed that school segregation is practiced in the participating elementary schools through academic mechanisms (student achievement level as measured by the mark obtained by these or the performance of schools) and social mechanisms (socio-economic status of students, language of instruction, sex).
These analyses and interpretations also indicate that, all else being equal, school segregation, especially through the basic school, is the superficial image that faithfully reproduces the social relations of inequality. This confirms the hypothesis of this article.
The implications of these results are the subject of the following discussion, which draws on the results of the research and work carried out by the functionalists, the conflictualists and the approach of rational choices of Boudon in particular in the field of the sociology of education, as the basis for the arguments that justify our assertion.
The school system assigns to each social group of students positions that place it in a hierarchical structure according to the social structures of society. Indeed, for functionalists, particularly Parsons (2004), we must not only inculcate in students the dominant values of society (in Demba, 2010, p.41), but also the student selection and guidance is one of the school functions that responds to the needs of society by distributing them in different social roles. So school, by practicing school segregation, only distributes the pupils in the different classroom roles (strong, average, weak) between those who manage the didactic activities and those who undergo them. In this functionalist perspective, school segregation becomes a powerful mechanism used by the school to prioritize and classify students in order to respond to the demands of a structured and hierarchical society where social relations are asymmetrical or unequal. Davis and Moore (1945) go in the same direction. For them, society must divide students into different positions in order to get them to perform the tasks related to these positions (Davis andMoore, 1945 in Cazeneuve, 1976, p.158). As part of our study, the school, through school segregation, separates them into strong, medium and weak or talented, average, mediocre).
Thus, the results of the research of the functionalists give us the possibility of positioning the results found in the study of the establishment of a link between school segregation and social relations of inequality.
Secondly, school segregation, as an act of discrimination and classification of social groups of pupils, corresponds to the logic of social control and reproduction of social relations of inequality. Indeed, for the conflictualists Passeron, 1970, Baudelot andEstablet, 1975), "the school is an agent of reproduction of relations of domination or relations of inequality. This reproduction takes place through a pedagogical action that is symbolic violence " (Bourdieu and Passeron, 1970, p.57). The school system must reproduce the social order and legitimize the social position of origin. All in all, the school segregation generated by the school must reflect, through the mechanism of school segregation, the social relations of inequality. According to Bourdieu and Passeron (1970), school only reproduces inequalities or social relations of inequality. It is from this perspective that it divides students into gifted, less talented and mediocre to reflect the image of the social hierarchy (social block of the favored, moderately favored and disadvantaged).
To be convinced of this, let us take Bourdieu's theory of constructivist structuralism, which emphasizes the weight of existing social structures. The author brings into play two key notions in this theory: habitus and fields. According to Bourdieu, habitus, as a generating principle of practices, is defined as "structured structures predisposed to function as structuring structures", and tends to reproduce the structures of which it is the product (Bourdieu, 1980). It is in this sense that Bourdieu (1980) states that habitus appears as a kind of social structures of our subjectivity, which are first formed through our first experiences (primary habitus), then our adult life (secondary habitus). This is the way in which social structures imprint themselves in our heads and bodies by internalizing exteriority (Bourdieu, 2000). It reproduces when it is confronted with usual situations and it can be led to innovate when it is faced with new situations (Dortier, 2004, p.58).
The action of teachers, school administrators and parents of pupils is therefore, at the end of Bourdieu's theorization, fundamentally the product of the objective structures of the school world in which they evolve, and which shape in them a set of dispositions, which will structure their ways of thinking, perceiving and acting (Bourdieu, 1987).
It is rather about principles of reproduction, that is to say, the habitus is a reproducer of social structures of which it is the product (Bourdieu, 1987). School segregation is to some extent the habitus of school system respondents (teachers, parents, school managers, etc.). In the same way, the field contains structural mechanisms of domination reproduced from generation to generation (Bourdieu, 1979).
School segregation, from this perspective, finds its founding in the legitimization of the classification schemes of the social hierarchy).
The act of classifying students (in strong, medium and weak), according to the results, from the perspective of Bourdieu (2000), is an act of reproduction of the social hierarchy. Thus, school generals school segregation as a powerful mechanism for social relations of inequality. The results of the analysis seem to confirm this trend.
Furthermore, the school system has an ideological reproduction function, that is to say, it contributes to social control Passeron, 1970, Baudelot andEstablet, 1975) and a certain social order (Durkheim, 1999;Parsons, 1951) by transmitting to new generations the conceptions of this hierarchical society. From this perspective, the school system reproduces the social relations of inequality whose main mechanism is school segregation. Then, as Pitts (1990) quoted by Demba (2010, p.45), the social order would result from the imposition of norms and values by those who hold power. In the same vein, Bowles and Gintis (1976) in Demba (2010) consider the school as an instrument of reproduction of relations of domination or social relations of inequality, selection of individuals and social segregation (Demba, 2010, pp. 45-46). The school would therefore reproduce the hierarchy of social positions by using school segregation as a mechanism. Moreover, the social order is based on the power of control of dominant social groups who use the school to reproduce their position of domination, in accordance with their particular interests (Bourdieu and Passeron, 1970). The school is no longer a vector of progress and individual mobility, but an instance of social control and reproduction of social relations of inequality (Duru-Bellat, 2004).
Finally, from another point of view, Boudon's rational choice approach allows us to confirm our results concerning the link between school segregation and the social relations of inequality. Rather than analyzing the school and particularly the question of school segregation as a functional fact or a question of reproduction of the social relations of inequality, we can consider it from the perspective of the rational choices of Boudon (1973), such as the social construct that engages survey participants, such as teachers, students, school administrators, and school policies. Because, according to the results of our study, there is a school policy of certain institutions aiming to gather in the same school space of the students according to their school level or their social origin, under the pressure of the parents of pupils. Boudon (1973) in "unequal opportunities, social mobility in industrial societies" has shown that school inequalities are rooted in the very structures of social stratification. In this sense, in a hierarchical society where social relations are asymmetrical, school segregation, as a mechanism of discrimination across the school, can only be a reflection of these social relations of inequality.

Conclusions
At the end of the analysis of the results in our study, we have seen that schools are categorized (Catholic, Protestant, secular, community, municipal, presbyteral, public, etc.). Moreover, not only do these schools offer different qualities of education. One is preferable to the other, but they are also socially differentiated, because the poor communal, community, public, presbyteral and secular schools concentrate social groups of pupils, majority of the poor, whereas the independent, Protestant, independent, congregational, secular schools welcome mainly favored and intermediate students (Joint, 2005, D. Pierre, 2012, Tardieu, 2015, according to the results of the study. This makes it clear that school segregation is the superficial image that faithfully reproduces the social relations of inequality.
On the basis of our results and the theories mentioned above, school segregation is the superficial image that faithfully reproduces the social relations of inequality, especially through the basic school, all things being equal.
Nevertheless, the results of our research should be interpreted with caution, since some authors believe that school segregation is not necessarily the result of an established student separation policy. A first objection may be raised to the representativeness of the sample interviewed, because for practical reasons we only interviewed key respondents from the education system (Pupils, parents of pupils, teachers and school managers). In this sense, social desirability could be a possible explanation. The study is also not general in scope since not all respondents were randomly selected, and schools were intentionally selected from sampling by reasoned choice, so it is a non-probability sample. Respondents were inclined to respond in the socially valued sense.
It should also be noted that the other departments are not the subject of this research. This link is observed in the department of the west; does it exist in the other nine departments of the country? Unfortunately, we cannot answer this question in our research. As a result of our study, a first to our knowledge at the level of the Haitian fundamental school, we would like similar studies to be carried out in several departments in order to confirm or refute our results.
Our research offers also the opportunity to explore a few possible avenues. It would be appropriate and even useful to conduct such research in all departments of the country; then calculate the inter-institutional and intra-institutional school segregation index and look for the correlation between the two variables to verify whether or not there is a cause-and-effect relationship.