Резюме. In the paper attention is drawn to the phenomenon of early school leaving, which is regarded through the prism of the stratification mechanisms in the field of secondary school education in Bulgaria. A special emphasis is placed on educational segregation, related to a widening of the gap between different types of Bulgarian schools. The authors present the data of an empirical study on the opinions of secondary school students in Bulgaria about the early school leaving phenomenon, attempting to highlight a number of risk determinants that trigger school failure as a consequence of certain life and educational trajectories. It is emphasized that both a profound understanding of the risk factors, associated with early school leaving, and potential remedies should be considered in the context of a much broader socio-structural framework and a horizon of sociocultural meanings, including those, related to ethnicity. The paper concludes with a description of some main tools for prevention within the socio-existential perspective.

Ключови думи: education; early school leaving; inclusion

Introduction

In the context of today’s post-industrial, globalizing society, the problem of the access of children to school education and early school leaving (ESL) is of particular importance and relevance to the EU countries. This is the result of the impact not only of the macro-social processes, which reflect the specific trends at the societal level, related to the construction of the knowledge-based economy and society in Europe, but also a consequence of the demand for effective ways of raising such socio-economic markers as competitiveness, sustainability and growth potential. The increased interest in the problem is also influenced by the indispensable intention for a general humanization of the social relations and the emphasis on the existential well-being of each individual.

Assessing the potential benefits of individual learning and the potential loss of school dropouts, a number of researchers adhere to traditional, economically determined perceptions. Within this so-called investment model a common vision has been accepted, which reveals higher benefits from schooling among individuals from vulnerable, disadvantaged groups and general increase of returns for all social groups and strata since 1980s. Besides interpreting human capital and its growth only in terms of “one-dimensional skill” and “productivity” the alternative conceptions view schooling as such a mechanism, which generates different types of experience and affects many dimensions of skills.

Schooling not only affects income but also provides individuals with the opportunity and capacity of making better decisions about health, marriage, and parenting style. It improves multi-tasking or time management skills, offers broader opportunities for the development of critical thinking and social competencies. Schooling promotes trust and civic participation. There are very strong arguments that compulsory schooling affects overall life satisfaction (Oreopoulos, 2007). That is why it is essentially important to extend our knowledge about the early school leaving phenomenon and its predictors.

ESL in Bulgaria: state of affairs

At EU level the term “early school leaving” describes all forms of leaving education and training before completing the upper secondary level. It includes those who have never enrolled and those who have dropped out of education and training1). According to the EU Commission Staff Working Paper (2011) ESL is typically caused by a cumulative process of disengagement as a result of personal, social, economic, geographical, education or family-related reasons (p. 11). It reveals a very complex and nonhomogeneous nature, which differs from one country or region to another. The EU Council Recommendation (2011) stresses upon that ESL are a process rather than a one-off event; it can be prevented best if the first signs of this process are recognised (p. 10).

In the context of the great social significance of the ESL the Europe 2020 strategy has outlined the priority objective of reducing early school leaving from 14.4% in 2009 to less than 10% by 2020. Despite the active aspirations of Bulgaria to adhere to this goal by committing to 11%, the overall situation in the country seems to be rather alarming. The survey results show that the number of dropouts has increased significantly in 2015 (21.146 students out of a total of 749.094) and represents 2.82% as opposed to 2.36% of the previous 2014. No improvements were reported in the following year. In 2016 a total of 21.171 pupils out of 741.235 left school, which represents 2.86%2). Obviously, taking into account the number of dropouts, there is a slow but steady negative tendency, although there is a numerical increase of a separate category of children who leave the school institution due to their departure abroad. In addition, there is an increase compared to the previous year of the number of students at primary school age, who leave school because of family reasons, and a similar increase in the number of students from secondary school, who leave because of their reluctance to study. As a result, between one fifth and one quarter of the Bulgarian children aged 11 – 14 do not attend secondary school.

It is necessary first to take into account the fact that in the case of ESL students the situation in Bulgaria is contradictory. There are attractive schools with a large influx of enrolling students where dropouts are non-existent. There are also school institutions where, despite the existence of a number of difficulties, the problem of early school leaving is solved successfully enough. At the same time, there are schools that report very high levels of ESL students. Recognition of the significant differences and imbalances, related to the dropout phenomenon should be reflected in precisely defining the different groups of schools with a particular focus on the most vulnerable educational institutions in order to construct the design of interventions according to their specifics and the profile of each of them.

A number of studies in Bulgaria show that the over-concentration of ESL students is characteristic of two types of schools as a whole.

1. Schools which are the only schools in a settlement. The percentage of nonenrolled secondary school pupils who have completed primary schools has almost doubled there. Such a situation is often explained by the fact that in some of the more conservative and patriarchal communities in Bulgarian society the physical closeness of the school is one of the key prerequisites for attendance and the absence of a school in the settlement leads to dropping/ not enrolling in the school.

2. Segregated Roma or Gypsy schools (New (2011 – 12) qualifies them as socalled “ghetto schools”) in which the number of dropouts is extremely high: in Roma neighbourhoods in big cities, and sometimes in smaller ones, there are schools in which nearly all pupils are from the Roma minority3). School segregation can reveal itself both as inter-school segregation and intra-school segregation with separate Roma classes, but in all its dimensions it is a special form of discrimination which, in Roma case, overlaps with lower quality education (Curcic et al., 2014; Rostas & Kostka, 2014; Downes, 2015). Even in this group of schools the dynamics varies considerably – a positive trend can be found in some of the “Roma schools”, where students continue to be enrolled in secondary education. Regional disproportions are also very serious. Moreover, these disproportions are not directly determined by an economic point of view.

The complexity of the problem calls for a careful consideration of a wide range of causes and risks regarded through the prism of stratification processes, which lead to the expulsion not only of individuals, but in a sense of certain subgroups out of the system of school education. In this respect the following main factors need to be understood:

– Differences with regards to ESL according to gender, academic performance or achieved education levels;

– The socio-economic background or a proxy, such as neighbourhood information;

– The migration or minority background and/or mother tongue of the learner.

For implementation of a profound vision on ESL it is advisable to take as a starting point an analysis of the national, regional and local specificities on the basis of a multi-dimensional and multi-disciplinary approach, which should take particular account of the student voice4).

Methodology and methods of the research

In order to define the key risk factors for secondary school dropout in Bulgaria a special survey was carried out focused on the students’ opinions. The methodology included a questionnaire survey. When processing the obtained data, a percentage analysis and a ranking-scaling method were used on the basis of SPSS Statistics. The main objective of the survey was to explore the views of Bulgarian students on the phenomenon of ESL in order to highlight the profile of the individual, correlating with the motivation for dropout.

Aims: 1. To reveal the risks that increase the propensity of Bulgarian students to ESL, with a particular focus on those SES influences and ethnic background which have the most important effect. 2. To identify the interventions and tools for prevention that appear to contribute to the inclusion of at-risk young people in the school community.

Research methods: survey, statistical processing of empirical data with Statistica SPSS.V. 16.0, pertinent analysis, correlation analysis.

Participants

The contingent includes 152 students who are studying in secondary schools. Participants were in 5th, 6th and 7th grade (11 – 17 years old), recruited from 3 regular schools in the Stara Zagora region (two of which were Roma schools). This choice was predetermined by the fact that the level of ESL in Bulgaria is usually the highest among students in these grades and among students from segregated schools. Besides Stara Zagora region is one of the fourth areas with over-concentration of ESL. All students completed the questionnaire voluntarily within school hours. The ethical rules for undertaking research with children were applied. All parents had given their informed consent for their child to participate in the study.

Procedure

The questionnaire is composed of a series of close-ended questions (29 items), the items grouped into the following subscales, aim to identify the different risks for ESL: I. Family profile and living conditions; II. Students’ health, work experience and access to equipment; III. School attendance and school difficulties; IV. Participation in extracurricular and out-of-school activities; V. Conflicts at school; VI. Help with personal/school issues and teaching methods; VII. Relationship with parents; VIII. Factors contributing to success and wellbeing at school; IX. Life goal orientation and reasons for dropout. Each item of the questionnaire contains unequivocal statements and does not allow for differences. The sample is exhaustive and does not claim to be representative. Piloting procedures have not been done.

The data are not normally distributed, necessitating the use of nonparametric techniques in the analysis. The statistical significance of differences in the responses is verified by a Kruskal-Wallis nonparametric k-test (Man-Whitney Post-Hock Test)

– a nonparametric alternative to a single-factor dispersion analysis for comparison of three or more groups. The corrected value of (p) of Bonferoni is used.

Results

The presented analysis selectively covers part of the data obtained. The “ethnicity” factor is defined as the presumed leader in the analysis of the answers to questions 1 to 6 (inclusive). The distribution of respondents by ethnic group is shown in Tabl 1.

It should be noted that those respondents who choose the “other” option in terms of their ethnicity are representatives of a specific sub-group “Turkish Roma” (Roma who identify themselves as Turks), so it is no coincidence that their answers are so close to those of Roma children.

The distribution of respondents by place of residence is as follows: 119 students (78%) are from cities and 33 (22%) are from villages. The distribution by gender includes 81 male (53.3%) and 71 (46.7%) female learners.

One of the risk factors that causes ESL could be related to a certain type of family. In order to identify it, respondents were asked question 1 “What family do you live in?”. Data indicate that students of biological families with two parents (78.8%) predominate, followed by those living with their grandparents – 9.9%. Third, there are the respondents who live in a single-parent family (7.9%) and fourth - those in a family with a step parent – 2%. The share of those living in a foster family is insignificant – 0.7%. One respondent indicates that he has no family (0.7%). In terms of ethnicity, the distribution of responses shows that the majority of respondents (75% persons of Roma origin and 87% persons of the dominant ethnos) live in families with two parents and a relatively equal number of individuals from both ethnic groups live in single-parent families. At the same time a sufficient number of respondents from the Roma minority – nearly 16% say they live only with grandparents against 3% of students with Bulgarian origin (Table 2).

The next question (“What education does your mother have?” – Question 2) is focused on the mother’s educational status, which is regarded to be one of the factors, influencing school dropout. Ordinarily, parents’ low educational status seriously increases the level of risk for ESL. The data shows that mothers with secondary education predominate – about 35%. Of these, the mothers of Roma minority children comprise nearly 33% and mothers of the dominant ethnicity – 42%. These are followed by mothers with higher education – 23.5%. Here, the differences become particularly significant. Mothers with higher education of Bulgarian ethnicity are almost 49% and Roma mothers – less than 6%. There are 31.6% of mothers with primary and lower secondary education (13.4% and 18.8% respectively), the majority of them being Roma. It’s clear that a great share of Roma mothers fall in the category of low educated people – of whom 18.8 with primary and almost 29% with lower secondary education. 9.4% of the mothers are without education and they are still mainly of non-Bulgarian origin (Table 3)

The distribution of the responses to a question related to the father’s education (“What education does your father have?” – Question 3) demonstrates that fathers with secondary education predominate, too - about 39,2%. Of these, the fathers of Roma origin are 15% and those of the dominant ethnicity – nearly 52%. They are followed by fathers with low secondary education – 23,6 % (Roma people – 33,8 % and people with Bulgarian origin – 10,3%). There are 20,9 % of fathers with higher education among which fathers of Bulgarian ethnicity are nearly three times more. The total number of fathers with primary education is 8,6% and they are only from minorities. Without education respectively are 7.4% of fathers and the absolute majority of them are Roma people.

The fourth question is “Do Your Parents Work?” with three variants of answers a) yes, both of them; b) yes, one of them; c) no, both are unemployed. The question checks out parents’employment status and indirectly the family’s purchasing power and quality of life. The distribution of data clearly shows that employed parents (72.4%) predominate, followed by families with one working parent (23.4%) – (Table 4).

As it can be seen from the data, almost a quarter of the respondents’ families have only one working parent, and the total share of such families of Roma origin is almost twice as big as that of Bulgarian ones, and makes about 42%. In addition, there are families (minorities only), where both parents are unemployed (3.9%). The risk of ESL in this case is growing steadily and is confirmed by research, which has found that lower employment rates associated with an increased propensity to drop out of school, especially among ethnic minorities (Rumberger, 1983).

Although the results of the survey, as revealed in question 5, show that the majority of respondents (88.2%) do not have to work to support their families, the number of Roma children is twice as big as that of the children with Bulgarian ethnicity, which often or occasionally need to work – 4 and 6 persons respectively, which makes almost 14% of their total contingent.

An interesting result indicates that most of the students surveyed declare that they need to work but at the same time they come from families with two working parents. This also reflects the specificity of the socio-economic situation in Bulgaria, where the number of the “working poor” is high.

As far as Question 6 is concerned (“Having a computer or tablet at home”) the distribution of data shows that the majority of respondents (87%) have this type of device at their disposal. At the same time, 22% of the Roma children surveyed did not have a computer compared to 3% of the students with Bulgarian origin (Table 5).

Here is another challenge faced by children from marginalized minority communities. It reflects the increased use of such learning aids as a result of the development of new cultural transmission media, which seriously reduces the total influence of school culture. So the decadence of the school-culture monopoly may thus work to maintain and even increase the disadvantage of lower-class children, as their parents frequently lack the cultural and financial resources to use these media as educational resources.

It can be concluded that the three questions, in the answers of which there are statistically significant differences concerning ethnicity (Nos. 2, 3, 6), the difference is between the Bulgarians and the two other ethnic groups, with the predominant force effect of factor from typical to greater than typical.

The impact of a number of the risks, related to school, ethno-cultural specifics, etc. is also reflected in other results obtained during the study. For example, question 11 (“What is difficult for you at school?”) is indicative enough. The data shows: almost half of the respondents (49%) associate the emerging difficulties with the lack of good discipline, and such opinions almost doubled the number of the students with Bulgarian ethnicity compared to those of Roma origin.

Secondly, there is a response regarding the complexity of the curriculum – and the situation is radically different: out of the 33% who share this opinion, the majority are Roma students (58%). Often, this category of children, having a number of educational deficits – due to low speech culture, insufficient cognitive and learning skills, etc. find themselves in a more difficult position, which negatively reflects on their academic success and increases their chances of dropout. Poor Bulgarian language skills (a factor mentioned only by the minority of students – a total of 9 respondents) also increase the level of risk. Of those who associate their learning difficulties with irregular attendance of lessons and gaps in their knowledge (about 4%), only minority students are involved.

Regarding participation in extracurricular forms of work (question 16), the distribution of responses based on the ethnic background of respondents appears to be more balanced. The data show that in most extracurricular activities children from minorities are actively involved, and in the dance groups and music-vocal groups, for example, the number of children from the Roma ethnicity is twice as big. To a great extent this is influenced by the specificity of the sample, which includes the children of the segregated Roma School in Chirpan, where the organization of extra-curricular forms of work is carried out at high level. But here too, some worrying trends can be highlighted. Firstly, the extremely low number of respondents who participated in the social-psychological trainings or counselling (3 persons in total, 2 of which are of Roma origin) and secondly, a significant number of people – almost 42% who declare that they do not participate in such activities. This is the group which involves children at a very high risk of dropout because they fail to develop enough trust and interest in the school community.

Even more dramatic is the situation with the participation in additional forms of education (question 17), where the number of systemically non-participating students reaches 44%, which is twice the share of minority representatives. This greatly reduces the educational chances of these children and limits the opportunities for their development.

The answers to question 18 regarding the reasons for students’learning difficulties and low success rate are interesting enough. Of all the respondents who explain their difficulties with non-attendance (12.5%), almost 90% are representatives of the minorities. The irregular completion of one’s homework made it difficult for 16% of the respondents to study, among which is the overwhelming majority of the minority groups. The same applies to the students’ short attention span. Chronic and other illnesses, as a factor that makes learning difficult for them, have only been indicated by minority children (4% in total).

On the question of the reasons for ESL (question 24), the opinions of the interviewed students largely coincide. In their view, school dropout and ESL is the result of a low interest in learning (almost 37% of respondents have chosen this option). Another common reason stems from the difficulties in learning (23% of all students). A more significant difference is evident when students relate the reason for ESL to early marriage – such an opinion is shared mainly by the respondents of the minority origin (12 persons against 4 of the Bulgarian ethnicity), which is in line with the cultural-specific prerequisites for dropout.

For Question 28 (“What do you like most about school?”) the respondents from all ethnic groups most often answer that their friends are the main reason for them to attend school – almost 50%. A large proportion of Roma children give their teachers a preference (of the 17% having opted for this answer, 68% are Roma). This is probably due to a number of circumstances – such as respect for the teachers’ professional qualities, their personal charm and authority, high appreciation of their support in training, etc. The same can be seen in the analysis of the answers to question 29 (“What helps you be successful in school?”). Most of the respondents receive the biggest support first from their families - 41%, secondly – from teachers (almost 28% of the total number of respondents) and thirdly – from their friends – 15%, with almost the same percentage of respondents relying mostly on their own strength.

The distribution of the responses based on ethnicity again highlights some specific points. The Roma children rely heavily on the support of their teachers while the students with Bulgarian ethnicity rely on themselves. And this circumstance puts into focus a number of reasons, objective and subjective, which can refer to both the ethno-cultural and social-status characteristics, as well as to individual-personal ones (control locus, etc.). What is important here, however, are the enormous expectations addressed by the Roma children towards their teachers, and the huge potential that lies here.

Discussion

In the light of the family reasons for ESL in Bulgaria, it is important to comment on some issues, related to the type of family. Data from respondents’ answers indicate that a small but significant number of students live in single parent families or live only with grandparents. This generally reveals two negative trends.

The first one is the increasing number of single parent families. Children, who are raised by a single parent, turn out to be much more vulnerable to a number of risks, due to the fact that such type of family is often characterized by a higher level of financial insecurity, poor sustainability, increased disorganization, and serious threats to poverty and social exclusion. Children of single parent households may develop social and academic problems; they are more likely to become teen parents. So it’s not surprising that in terms of family structure, the dropping-out children often come from broken homes or single-parent families (Pong & Ju, 2000).

The second tendency is closely related to external, international migration, mainly influenced by economic reasons. Parental migration, usually followed by leaving children in the care of elderly grandparents, leads to a sharp decline in parental control and attention. A great number of studies show that custodial grandchildren may encounter greater risk of behavioral and emotional difficulties than children in general. They may suffer from insufficient interaction with parents, family conflict, uncertainty about the future, and societal stigma. Besides, custodial grandchildren may experience greater mental health difficulties than children in general (Ghuman et al., 1999). Children under the age of 18 living with relatives fared worse than children living with biological parents on most measures of behavioral, emotional, and physical well-being. They were more likely to have caregivers with symptoms of poor mental health themselves (Billing et al., 2002). In addition, Roma children often find themselves in a much more difficult situation faced with challenges due to the low level of their grandparents’education and their poor work and professional habits.

Undoubtedly, the strongest factors that place children at risk and contribute to their ESL, unconditionally remain those directly related to the family (Kumpfer & Alvorado, 2003). These factors usually include dysfunctional home life, the lack of parental engagement, low interest and low parenting expectations, use of non-national language in the home environment, mobility, ineffective or abusive parenting. The level of risk for the Roma children is substantially higher here as well. A similar situation is observed in terms of peer pressure.

The analysis of data, concerning students’ family profile, living conditions and other characteristics, shows that a significant share of the students can be qualified as children of families with low SES. As for respondents of Roma origin, this is true for the majority of them. Roma students often live in poverty, their parents have a lower level of formal education and are less economically active, they come more often from large families or from single-parent families and their families owe less often their household dwelling (Garaz, 2014). Thus, poverty itself becomes one of the most risky socio-economic factors for dropout (Surdu et al, 2011). A number of financial and other difficulties faced by poor households usually exert a negative influence on both the conditions in which children are raised and the way in which their socialization proceeds.

Compared to other EU countries the situation with poverty in Bulgaria still remains hard: the accumulated statistical data on incomes and living conditions5) reveal a very disturbing picture. NSI said that there were 1 665.3 thousand people below the poverty line in 2017, or 23.4% of the country’s population. Compared to the previous 2016, the relative share of the poor population increased by 0.5 percentage points (Table 6).

The highest share of the poor was among the Roma ethnic group – 77.2%. Among Roma people, the highest percentage was among the unemployed – 39.9 %. According to statistics 73.2% of Roma people with primary or no education were poor, whereas there were no poverty stricken people with higher education. In 2017, 29.2% of children aged 0 – 17 years were at risk of poverty. In 2017, eight out of ten children (80.1%), whose parents were either primary or non-educated, live in poverty. With respect of the above, the research literature emphasizes the fact that ESL rates are substantially higher for students from lower socioeconomic backgrounds. The process of “withdrawing from school” for them can be explained both in terms of the frustration-esteem model and by applying the participationidentification model, which focuses on student’s involvement in school, unifying behavioral and emotional elements” (Finn & Rock, 1997).

While there is considerable evidence indicating that children from low SES backgrounds are more likely to leave school without graduating, there is little to suggest what exactly it is about being poor that renders children prone to dropping out. Several possibilities exist: inadequate parenting; poor schools or teachers; schools with high levels of truancy; pressures to augment family income, accompanied with a view that schooling has limited economic returns; peers with low aspirations; poor nutrition and health; and too few role models in the community (Audas & Willms, 2001).

One of the specifics of dropout, which stands out permanently in Bulgaria, proves to be linked to the transitive nature of the Bulgarian society and triggered by a number of systemic crises of the transition, leading to extremely difficult life conditions of the vulnerable groups. Besides, the transition from a socialist to a market economy has increased many of the factors contributing to the Roma social exclusion, which leads to formation of a Romani underclass (Sobotka & Vermeersch, 2012; Brown et al, 2013; Rostas & Kostka, 2014). As a result of the complex socio-economic situation, in-deep social differentiation and inequality, concentrated enclaves of lasting poverty are formed, encompassing significant social subgroups and communities, with markedly marginalized and ethnized profiles, as well as persistent tendencies towards encapsulation and self-closing.

Growing up in the context of this unfavourable social context, children from such minority communities often turn out to be the product of the so-called ‘peripheral’ socialization, characterized by externally imposed social closure, limited economic and cultural resources and therefore scarce social capital in terms of both aspirations and values, and social networks (Broccolichi & Van Zanten, 2000). This is the perspective of the predominant part of the Roma minority children in Bulgaria: every 6th Roma child is left out of the education system even before the end of the elementary stage… 12 – 19 year old Roma who never attend school or drop out of school without completing primary education are 21.9% (Zahariev et al., 2013: 37 – 39).

Undoubtedly, the overall increased risk of dropout for the children from this community is due to many other socio-economic factors, largely related to their low social status, high level of internal and external migration, geographical distancing from the educational institutions, lack of good transport-logistical opportunities for going to school, family multiplicity and deterioration of the housing environment. The same applies to yet another risk predictor, such as the use of child labor.

Besides these factors, the sociocultural and community-environmental factors should be highlighted with a particular emphasis on the peculiarities of the Roma culture - more archaic and traditional in its values-normative foundations, with a dominant role of the family-clan or micro-community group. In this separated and closed cultural world, seeking to preserve its autonomy in response to the expansion of modernization, the identity continues to be formed mainly through the ethno-cultural traditions and customs, religious beliefs and values, as well as through the absorption of the professional crafts, firmly rooted in the structure of the everyday life.

All these specifics determine a rather conservative, reserved and distant attitude of Roma people towards institutional education, as well as their fears that the school not only “plucks” the Roma child from his habitual way of life and the familyclan circle but also leads him out the framework of parental control. In this sense, certain factors exert a markedly negative impact. Among these are: the low status of education among a significant part of the Roma population, the more archaic forms of consciousness, specific gender-role behavior leading to the gender predisposition and limitation of girls’ access to education, the inconsistency and ambivalence of the Roma pupils’ subjective perceptions about the value of education and its significance for their personal life, the vector of their emotional attitude towards school and the school environment, which is often regarded through the prism of the alienated and stigmatized attitude by others.

Last but not least, there are school factors influencing the dropout: the learning environment and the overall climate in the school, the structure of the school and its organizational and functional features, the relationships with teachers and pupils, the teaching programs and teaching methods, the available resources and the material and technical support of the learning process, the accepted system for evaluation, etc. Factors such as poor relationship with teachers, unjust and oppressive practices, unrelated learning experiences, lack of autonomy, education/ exams issues and social and academic exclusion, are among the ones mentioned most often (Cefai, Downes & Cavioni, 2016).

Limitations

One limitation that occurs in the study is investigating solely the students’ opinions, related to ESL without comparing them to the information provided by teachers.

The sample was taken from three schools in two municipalities only and bearing in mind the rather small size of the sample, any generalization of the results must be made with caution. The insufficient sample did not allow for a more differentiated approach to the research and the analysis of the data according to the types of schools existing in Bulgaria (urban v/s rural, metropolitan v/s provincial, elite v/s “ordinary”, schools with a predominance of Bulgarian students v/s Roma schools, etc.)

There is an additional limitation concerning the experience of the participants in the survey aged 12 – 13 years. Since the phenomenon of dropping out of school is very complex and difficult to analyse students cannot be expected to reflect upon the various aspects of its essence. The use of mixed method designs would help overcome this limitation.

Finally, longer-term investigations with repeated data collection are needed to examine how these risk factors and their influence on student’s personality change over time, or with students’ age.

Implications

In the future, the mixed method designs – research interviews, observations, focus groups with students, parents, teachers and experts, as well as any other developmentally appropriate tools need to be recommended. Further studies should be conducted under more controlled implementation conditions, and with a particular focus on how different facets of SES can affect different risks of ESL. Repeated investigations in different types of school and among different age categories of students may in itself constitute a more focused and specialized forms of intervention.

Main tools for the prevention

Researchers’ reviews of “effective” schools found that organizational factors, including “clear school goals,” “rigorous academic standards,” “order and discipline”, student-teacher ratio, high expectations of teachers for all students, high levels of collaboration were positively related to school performance (McDill et al., 1985).

There are many similarities among the interventions, including their emphasis on changing the student, beginning with a personal-affective focus and then shifting to an academic focus, and their efforts to address alterable variables (Lehr et al., 2003).

In accordance with the Bulgarian specifics of the ESL phenomenon the most important elements of a sustainable and comprehensive strategy for prevention should include:

– creation of unifying coordinating bodies –inter-sectoral dropout prevention councils;

– local and regional adaptation;

– sustainable funding to support long-term interventions, especially for schools in segregated areas;

– development of specific outreach pre-school/school programmes and family social services;

– youth coaching and cross-age tutoring as a nationwide measure to support students at risk;

– explicit attention to social and emotional learning;

– further development of the centres for inclusive education;

– emphasis on vocational learning and practically-oriented forms of education;

– educational desegregation.

Conclusion

Educational segregation is the last of the proposed strategic steps that are directly related to the education policy towards the Roma education in Bulgaria. The ambivalent nature of this policy can be traced back to the socialistic period, with the emphasis on equality of opportunity and at the same time the building of the segregated schools in the Roma neighbourhoods. During the postcommunist period, the educational integration course was officially announced, but there is still a high level of segregation: according to MES data (old data, as this type of data was last collected 10 years ago) there are 106 Roma schools in Bulgaria. In this type of school, both material conditions and a number of other indicators are often worse than those of other schools. Moreover, such Roma schools are not only ethnical, they are dominated by students with low socio-economic background. It is a systemic cause for consistent high ratios of educational failure and ESL.

To sum up, the desegregation of the Roma education should become the mainstay of national education strategies in order to ensure equal educational opportunities for all children along with comprehensive socialization. The key components in terms of preventing ESL within the framework of the socio-existential perspective should be participation and experiencing a sense of belonging and engagement in the school community that will eventually increase the social and psycho-emotional well-being of the individual and the social capital.

NOTES

1. Reducing early school leaving: Key messages and policy support Final Report of the Thematic Working Group on Early School Leaving November 2013, https:// www.bmb.gv.at/schulen/unterricht/ba/twg_report_final_25671.pdf?61ec4c [visited on 07.04.2018].

2. Data from the National Statistical Institute’s surveys - Dropouts by degrees and reasons for leaving – http://www.nsi.bg.

3. Kolev, D., Krumova T. (2017). Students non-continuing secondary education: how many, where, why and how? http://amalipe.com/files/publications/SO.pdf [visited on 03.04.2018].

4. EU Council recommendation on policies to reduce early school leaving com (2011) 19 final SEC (2011) 98 final SEC (2011) 97 final SEC (2011) 96.

5. EU-SILC, NSI, 2017.

6. Audas R., Willms J. D. (2001). Engagement and Dropping out of School: A life course perspective. Working Paper for the Applied Research Branch, Strategic

Policy, Human Resources Development, Université du Québec – Chicoutimi. http://sbisrvntweb.uqac.ca/archivage/15292281.pdf [visited on 12.04.2018].

7. Billing A., Ehrle J., Kortenkamp K. (2002). Assessing the New Federalism Project. Washington, DC: Urban Institute; Children Cared for by Relatives: What Do We Know About Their Well-Being?

8. Brown, P., Dwyer, P. & Scullion, L. (2013). The Limits of Inclusion? Exploring the Views of Roma and Non-Roma in Six European Union Member States. http:// www.romasource.eu/userfiles/attachments/pages/167/rs-finalresearchreportfull-2013-en.pdf.

9. Downes, P. (2015). Early School Leaving Prevention and Engaging Parents from Ethnic Minority and Migrant Backgrounds: Key Issues and Guiding Principles Across 9 European City Municipalities. European Union, European Regional Development Fund, Urbact Secretariat, Paris.

10. European Commission Staff Working Paper (2011). Reducing Early School Leaving. Accompanying document to the Proposal for a Council Recommendation on policies to reduce Early School Leaving. 26th January 2011.

REFERENCES

Allen-Meares, P., Washington, R. O. & Welsh, B. L. (2000). Social work services in schools. Boston: Allyn & Bacon.

Broccolichi, S. Van Zanten, A. (2000). School Competition and Pupil Flight in the Urban Periphery, Journal of Education Policy, 15(1), pp. 51 – 60.

Cefai, C., Downes, P. & Cavioni, V. (2016). Breaking the cycle: a phenomenological approach to broadening access to postsecondary education. European Journal of Psychology of Education, 31(2), 255 – 274.

Curcic, S., Miskovic, M., Plaut, S. & Ceobanu, C. (2014). Inclusion, integration or perpetual exclusion ? A critical examination of the decade of Roma Inclusion, 2005-2015. European Educational Research Journal, 13, 257 – 267.

Finn, J. D. & Rock, D. A. (1997). Academic success among students at risk for school failure. Journal of Applied Psychology, 82(2), 221 – 234.

Garaz, S. (2014). Helping the marginalised or supporting the elite? Affirmative action as a tool for increasing access to higher education for ethnic Roma. European Educational Research Journal, 13, 295 – 311.

Ghuman HS, Weist MD & Shafer ME. (1999). Demographic and clinical characteristics of emotionally disturbed children being raised by grandparents. Psychiatric Services. 50:1496 – 1498.

Kumpfer, K.L. & Alvorado, R. (2003) Family-strengthing approaches for the prevention of youth problem behaviours. American Psychologist, 58, 457 – 465.

Lehr, C. A., Hanson, A., Sinclair, M. F. & Christenson, S. L. (2003). Moving beyond dropout prevention towards school completion: An integrative review of data-based interventions. School Psychology Review, 32, 342 – 364.

McDill, E. L., Natriello, G. & Pallas, A. M. (1985). Raising standards and retaining students: the impact of the reform recommendation on potential dropouts. Review of Educational Research, 55, 415 – 33.

New, W. (2011 – 2012). Stigma and Roma education policy reform in Slovakia. European Education, 43, 45 – 61.

Oreopoulos, Ph. (2007). Do dropouts drop out too soon?). Wealth, health, and happiness from compulsory schooling. Journal of Public Economics, 91, pp. 2213 – 2229.

Pong, S. L. & Ju, D. B. (2000). The effects of change in family structure and income on dropping out of middle and high school. Journal of Family Issues. 21(2), 147 – 169.

Rostas, I. & Kostka, J. (2014). Structural dimensions of Roma school desegregation policies in Central and Eastern Europe. European Educational Research Journal, 13, 268 – 281.

Rumberger, R. W. (1983). Dropping out of high school: the influence of race, sex and family background, American Educational Research Journal, 20, 199 – 210.

Sobotka, E. & Vermeersch, P. (2012) Governing Human Rights and Roma Inclusion: can the EU be a catalystfor local social change? Human Rights Quarterly, 34(3), 800 – 822.

Surdu, L., Vincze, E. & Wamsiedel, M. (2011). Roma School Participation, Non-Attendance, and Discrimination in Romania. Bucharest:Vanemonde.

Zahariev, B, Yordanov I. & Decheva, Y. (2013). Lost Future? Research on phenomena of school dropout. Sofia. UNICEF.