136 Sodobna pedagogika/Journal of Contemporary Educational Studies Let./Vol. 71 (137) Št./No. 1/2020 Str. /pp. 136–154 ISSN 0038 0474 Anikó Fehérvári Dropping out in Hungary: teachers’ perceptions Abstract: The theoretical background of this paper is provided by studies on dropping out and leaving school early . The article also builds on various international and Hungarian data collections, presenting the Hungarian situation and trends, and the student, organisation, school and system factors that influence dropping out. The paper provides an analysis of data from a questionnaire-based survey conducted in 2018 in 83 schools. Schools were chosen based on the rate of students at risk of dropping out. The study aims to explore the school (teaching) practices in these schools and identify risk factors that may increase or decrease the chances of leaving school early . The paper focuses on one dimension of this complex research problem: the question of the school and teachers taking responsibility. The results indicate that teachers’ perceptions of problems are proportionate to the rate of students at risk of dropping out in their schools. Teachers believe that schools and teachers have an important role in preventing student dropout. However, they consider dropout to be a bigger problem in schools other than their own, and they believe that other actors (parents, society, the media) are more responsible than teachers for student dropout. Keywords: dropping out, teachers’ perception, Hungarian public education UDC: 37.091 Scientific article Anikó Fehérvári, PhD., Eötvös Loránd University, Faculty of Education and Psychology, Address: 1075, Kazinczy Budapest, Hungary, Hungary; e-mail: fehervari.aniko@ppk.elte.hu Fehérvári 137 Introduction The issue of early leaving education and training is not just about school per- formance or failures; it is about educational and social inequalities, poverty and marginalisation. Education deficit has multiple negative impacts on the individual as well as on society. Persons with low levels of education are more likely to have less income, more exposed to the risk of unemployment, and may develop health problems more frequently. They generate less tax income for society and incur greater healthcare expenses, and may also augment crime rates. International data confirm that education is a lucrative investment for both the individual and society (OECD 2017). The data also affirm that persons with higher levels of education are more health conscious and active, and are less easily manipulated by power. According to medium-term labour market forecasts (CEDEFOP 2016; EUROGOUND 2015), by 2025, the demand for low-educated, low-skilled workers for manual or routine jobs will decline in Europe, including in Hungary . The demand for workers with secondary-level education will stagnate, while the demand for degree holders and highly skilled workers will rise. Therefore, the risk of unem- ployment among young people with a low level of education is much higher than among those with secondary and higher education. The same applies in Hungary. Hungarian studies also reveal a significant difference between the employability of vocational school graduates and secondary vocational school graduates in favour of the latter group (Fehérvári 2015a; Makó 2014). The secondary school final ex- amination (matura) is a dividing line in the labour market. Without it, the risk of unemployment increases exponentially (Nagy 2010). Based on these scenarios, countries (including Hungary) in which a relatively higher proportion of inhabitants have low educational attainment face slower economic growth and poorer competitiveness. Employability data indicate that in Hungary large groups (those with low education and skills, persons with disabilities, etc.) are left out of the job market. The European Union aims to increase the rate of employment to 75% by 2020 and improve employability. To this end, one of the goals of the Europe 2020 strategy is to slash the rate of those with low educational attainment. The strategy envisions reducing to below 10% the average proportion 138 Sodobna pedagogika/Journal of Contemporary Educational Studies Fehérvári of 18-24 year olds in the EU leaving education and training without secondary or vocational qualifications. According to the official definition used in the European Union, an early school leaver from education and training refers to a person aged 18-24 who has completed at most lower secondary education (ISCED 3 skilled worker or secondary school final examination) and is not involved in further education or training. The indicator ‘early leavers from education and training’ is expressed as a percentage of the people aged 18-24 with such criteria out of the total population aged 18- 24. 1 This is an outcome indicator which measures the effectiveness of education. The age group examined has already exited public education, as its members are older than the compulsory education age. According to Eurostat data, in 2002 the average rate of early leavers was higher than in Hungary, but the trend was downward, while in Hungary the first decade of the 2000s was characterised by volatility and stagnation. In the EU28 the rate of early leavers was an average of 10.8% in 2018; in Hungary the same rate was 12.5%. Twenty-one member states scored better than Hungary and only five member states scored worse 2 , but in five of them the trend is improving. Eurostat also compares the proportion of early leavers from education and training according to the degree of urbanisation in particular countries, with regions classified as cities, towns and suburbs or rural areas. The 2018 figures show a significant difference in Romania, Bulgaria and Hungary among the EU28: in each of the three countries, early leaving is a problem typical of rural areas. According to Eurostat data (2018), in Hungary , this means that the rate of early leavers in the most rural region is double that of the capital and the Central Region. The gap is growing ever wider (2002-2018): the proportion of early leavers from education and training is shrinking or stagnating in the richer regions, while it is stagnating or growing in the poorer regions 3 . Dropout data have been collected in Hungarian public education statistics since 2014, so it is possible to identify which type of education is the most affected and when (i.e., at which grade) dropping out occurs (Fehérvári 2015b) 4 . The data show that dropping out of full-time education is not very frequent in primary schools and grammar schools, affecting approximately only 1% of students. Conversely , in vocational schools and particularly in bridging programmes, attrition is significant: in 2016 the dropout rate was 15% and 37% respectively. The figures refer to one school year and are accumulated in the course of the three-year training period, 1 The statistical indicator is calculated by dividing the number of early leavers from education and training, as defined above, by the total population of the same age group in the annual Labour Force Survey (LFS) by Eurostat. The indicator is essentially a low-end estimate, as a certain proportion of the age group 18-24 in education will never acquire secondary education in future. Moreover, the LFS does not classify a young person as an early leaver if they participate in any kind of education, including that which does not lead to a qualification. 2 Bulgaria, Italy, Romania, Malta, Spain 3 The early leavers indicator is an outcome measure. For a process measure, the dropout indicator provides information on the progress of students within the school system. 4 A student is considered dropped out if they are included in the register of students on the same day of the previous year but not on 1 st October of the reported year (in other words, their student status ceased), and they have not acquired a secondary qualification. 139 Dropping out in Hungary: teachers’ perceptions which means that in a vocational school the dropout rate can be as high as 35-40% (Varga 2018). Causes of dropping out Systematic analyses addressing the dropout problem (Lyche 2010; Rum- berger 2012) identify individual, family and institutional factors in the background. Witte et al. (2013) add a community factor (Table 1). Analysts point out that drop- ping out comes at the end of a long process which is simultaneously influenced by multiple factors. Due to its complexity and temporal change, it is difficult to examine, as static and two-variable analyses can easily lead to the error of rein- forcing existing stereotypes without actually highlighting the various interaction effects or the dynamic nature of the problem (Smeyers 2006). Personal factors School-related factors Scholastic performance, skills, year repetition Type and structure. Resources: operator, com- pensating for disadvantages Attitude: • Commitment (involvement): learning, social • Deviance • Attendance School (teaching) practices: involvement in learning processes, motivation, school climate, relations, year repetition practice, expectations, parent-school communication Background: • Past experience (pre-primary school, school successes and failures) • Health status, disability • Work while at school • Family: o structure (single-parent or family with many children) o attitude o demography o economic and cultural resources (poverty, illness, educational attainment) Communities (institutional resources, e.g., child protection, parent relations, social networking) Belonging to a community: • traits of peer at the next desk • discrimination, segregation Table 1: Causes of dropping out (Lyche 2010; Rumberger 2012; Witte et al. 2013) Among individual factors, scholastic performance and year repetitions should be highlighted; the latter is the strongest precursor of dropping out. It is also im- portant to note that school performance is often greatly influenced by family back- ground factors (ethnicity, the parents’ educational attainment, income, economic and cultural capital, etc.). Also, underachievement is not necessarily caused by a lack of skills; it can be caused by a lack of motivation and involvement, boredom or a lack of social relationships in the school. While performance is a highly sensitive 140 Sodobna pedagogika/Journal of Contemporary Educational Studies Fehérvári indicator , countless factors can be behind worsening performance or underachieve- ment (Lamb et al. 2010; Lyche 2010). International research findings indicate that school structure seems to be the most important group of school-related factors, and there are differences in the risk of dropping out in terms of the operator responsible for the school (private, public or church), and in terms of the compensatory effect of the school (which school or school type is able to make up for family disadvantages). Regarding school-related factors, teaching practices have a major impact. Other important factors include involvement in learning processes, participation, interest in learning, motivation and a sense of belonging to a community. It is worth noting that student commitment is partly related to the school’s organisational characteristics: a highly impersonal, hierarchic school environment is less conducive to students’ commitment, and dro- pout is more frequent. The quality of the school climate, and the school’s internal and external relations are also crucial. A strained climate and poor teacher-student and school-parent relations also increase the risk of attrition. It is to be emphasised that school factors have a greater impact on students in schools where the student population has a lower socioeconomic background compared to schools with a more favourable breakdown of students. In the category of institutional factors, the practice of failure and year repetition should be highlighted. Research find- ings prove that the rate of dropouts is higher among repeaters, and older students are more likely to leave secondary school without a qualification. Another factor increasing the risk of dropout is if the school is undemanding of students, in other words, it lowers its expectations of certain students. School effectiveness studies mention high expectations as one of the keys to effectiveness; expectations should be met by students as well as teachers (Rumberger 2012; Wilson et. al 2011; Witte et. al 2013). Hungarian studies also found that dropout is the result of a longer process in the course of which risks become conspicuous already in the primary school years (Liskó 2003). Students’ learning path, their socioeconomic background and ethnicity were identified as the main factors. Although the process starts before entry to the school system – during the primary school years – the actual event (dropping out) generally happens in sec- ondary education. It is typically linked to vocational schools, and there is a big difference in dropout rates between secondary school types (Fehérvári 2008). The obvious reason is that progress paths are different by type of education. Another characteristic Hungarian feature is that differences in skills are closely related to students’ socioeconomic background, as proven by international (PISA, TIMSS) student performance studies (Ostorics 2015). Consequently, dropout studies fo- cused on the school progress of Roma youths (Kertesi and Kézdi 2013a; 2013b), and those exploring vocational training (Fehérvári 2008) deserve special attention. These studies find the following to be the main individual and family attributes of young people dropping out of education and training: parents with low edu- cational attainment, large families, Roma ethnicity, school failures (poor grades, failed subjects and years, etc.), absenteeism, vision of the school more negative than that of the average student, attitude to learning more negative than that of the average student, and inappropriate choice of vocation or school. Poor motiva- 141 Dropping out in Hungary: teachers’ perceptions tion, absenteeism and inadequate preparation for classes are major risk factors of dropping out. Besides exploring individual characteristics, Hungarian researchers also analyse the roles and responsibilities of schools. They underline the importance of educational practices, as it is often the case that early leaving is not the student’s but the school’s decision. Berényi (2015) classifies vocational schools into two cat- egories of action: instrumental and expressive. She argues that instrumental schools make an effort to push out problem students, while expressive schools accept the composition of their student population as a fact of life and consider it their duty to offer remedial means to help them catch up. Research studies targeting institu- tional effects also try to determine the factors that help a school retain its students and make it more successful than another school with a student population of the same characteristics. The findings of V arga (2015) reveal that schools that are more successful in teaching their students are characterised by an inclusive approach, a higher level of staff training, a wider range of educational services, a more extensive and closer-knit network, and greater willingness to be innovative compared to other schools. Széll (2015) also reports that teachers’ educational attainment (BA, MA, postgraduate), their professional experience and their willingness to participate in further training are important factors. In addition, he points out that schools that are more effective in teaching their students experience a lower rate of staff turnover: in other words, the school’s retentive force also applies to staff. There are three main types of education policy interventions in early school leaving: prevention, which comes mainly at pre-primary and primary school level; intervention focused on secondary schooling; and compensation, which is aimed at providing early school leavers with qualifications (European Commission 2015). The EU’s development policies stress the role of schools in combating early school leaving. It should also be emphasised that all this is embedded in a close intersectoral cooperation: the EU considers education policy interventions effective if they are coordinated with the employment, health and social policy sectors and rely on the collaboration of the micro, mezzo and macro levels. Besides the inter- sectoral approach, data collection is another significant horizontal aspect. A priority development area in data collection is the introduction of an early warning system which allows for the tracking of student performance and other factors on an indi- vidual level (for instance, failure, year repetition, absences and lack of motivation) which are precursors, and can therefore help identify students at risk of dropping out and promote intervention. Methods Research question The study explores teachers’ perceptions of and attitudes to dropout, and the role and responsibility they attribute to themselves regarding the problem. For the purpose of the study, our definition of dropping out was as follows: dropping out 142 Sodobna pedagogika/Journal of Contemporary Educational Studies Fehérvári is when a student leaves school (their student status with the educational institu- tion ceases) without acquiring the qualification appertaining to the given level of education and training. Participants This paper presents the first results of a questionnaire-based survey of 83 primary schools participating in a teacher training project 5 launched in 2018 aimed preventing dropout 6 . The questionnaires were sent electronically to the target groups of the project, consisting of teachers and principals of primary schools in four counties 7 and Budapest. The sample contains schools that fall into the most vulnerable third of the schools in each county in terms of dropout, based on the early warning system. Hungary has operated an early warning system since 2016. The indicator provides information about students at risk of dropping out based on data gathered from grades 5-8 of primary schools. The early warning system helps to identify students at risk of dropping out, and based on the warning, school-level or student-level intervention can be planned. Internationally, early warning factors of dropping out include boredom, bullying, exclusion, behavioural change, depression, repeat of a grade, absenteeism and deteriorating academic achievement (Tomcsit et al. 2014). The Hungarian early warning system monitors the three latter factors. Schools collect and report these data to the competent educational training and consultancy centres so that the proportion of students at risk of dropping out can be assessed and education policy interventions be planned. The population surveyed consists of teachers in 83 schools with the highest dropout risk in their respective regions. The total number of teachers participating in the study was 2,555 and the number of respondents was 1,255 (response rate of 49%). Instruments and data analysis The questionnaire focused on investigating the push/pull factors of dropping out. We focused on the factors identified in the exploration of the theoretical back- ground; the main sets of closed-ended questions were related to organisation and teaching practices and had already been used in previous Hungarian research studies (Sági 2015; Széll 2015; Tomcsik et al. 2014). The main topics of the questionnaire were as follows: school climate (15 items), school attachment (20 items), teacher 5 The training indirectly contributes to dropout prevention by enhancing teachers’ methodological preparation. As the study did not aim to evaluate the training project, the time of survey is irrelevant from the point of view of the training schedule. 6 Supporter: Human Resources Operational Programme, Hungary , No. 3.1.2-16-2016-00001; Title: Methodological renewal of public education to reduce early school leaving 7 Vas, Győr-Moson-Sopron, Zala, Borsod-Abaúj-Zemplén counties 143 Dropping out in Hungary: teachers’ perceptions competence (10 items), education goals and effectiveness (10 items), learning process (10 items), teachers’ views and attitudes (dropping out, inclusion and expectations – 15 questions). The presentation of the results is based on factor analysis, ANOVA and descriptive statistics. This paper analyses the data revealing teachers’ views on dropping out. It ex- amines the indicators that respondents identify as dropout risks, the responsibility they attribute to themselves and their school, and the ways in which their views are influenced by school climate. We focus on teachers’ views partly due to limita- tions of space, and partly because the discourse on the dropout problem is mostly dominated by reference to stereotypes and individual, student-related reasons for dropping out. This paper aims to point out that the role of teachers and schools is at least as important. Results The set of questions about dropping out tried to find out how big a problem the responding teachers perceive dropout to be in Hungarian education in general, and in their schools in particular. 86% of the responding teachers considered dropping out to be a problem in Hungary, yet they believed that it was less so in their own schools. Only 23% replied that it was equally a problem in their own school. We applied analysis of variance to see whether there was a correlation between the proportion of students indicated by the school’s early warning system as being ‘at risk’ and teachers’ perception of the problem (Table 2). Average rate of students at risk N Std. deviation It is not a problem in our school at all. 11.9269 341 8.90687 It is not really a problem in our school. 15.326 518 9.21356 It is rather a problem in our school. 22.5065 215 8.76858 It is a big problem in our school. 25.9567 43 8.36296 Total 16.0797 1117 9.90559 Table 2: How big a problem do you think dropping out is in your school? p = .000 We found that the higher the risk index in a school’s early warning system, the more serious the dropout problem in that school was perceived to be, according to the teachers. According to the data in schools where dropping out was not perceived to be a problem at all, the rate of students at risk of dropping out was 11% on average. In schools where teachers considered dropping out to be a big problem, the average value of the index was 25%. Based on an earlier study (Tomcsik et al. 2014) that aimed to identify the most important indicators of the early warning system, in our questionnaire, from a list 144 Sodobna pedagogika/Journal of Contemporary Educational Studies Fehérvári of 30 items 8 , the teachers had to choose ten that they considered to be the most important early indicators of dropout 9 . In addition, we also asked which of the ten indicators the school was aware of or was monitoring. The ten indicators considered the most important by the respondents are shown in Figure 1. 01 02 03 04 05 06 07 08 09 0 Frequent absences The parents don't follow their child's performance Deteriora ng grades Aggressive behaviour Poor reading skills Overaged Ethnic minority Integraon, le arning and behavioural disorders Living in child care centre Exclusion Monitored, % Cause, % 01 02 03 04 05 06 07 08 09 0 Frequent absences The parents don't follow their child's performance Deteriora ng grades Aggressive behaviour Poor reading skills Overaged Ethnic minority Integraon, le arning and behavioural disorders Living in child care centre Exclusion Monitored, % Cause, % Figure 1: Ten main causes of dropping out and their tracking by the school according to teachers, % (N = 1255) 8 Frequent absences, the parents don’t follow their child’s performance, deteriorating grades, aggressive behaviour, poor reading skills, overaged, ethnic minority, integration, learning and behavi- oural disorders, living in child care centre, exclusion, deterioration by at least 1.1 in a year, parents’ low educational attainment, poor mathematics skills, must work while going to school, parents don’t attend teacher-parent meetings, SEN, lives in a partnership, parents are unemployed, the family lives in poverty , disabilities, low achievement, boredom, chronically ill, single-parent family , orphan, reserved behaviour, immigrant background, raised by grandparents, rich family, two or more siblings 9 Our definition of dropping out was included in the question, so each respondent had the same understanding of the concept. 145 Dropping out in Hungary: teachers’ perceptions The majority of teachers (80%) considered attendance, or absenteeism, to be a crucial indicator. Parental attitude (the parents don’t follow their child’s perform- ance) was also found to be important by the majority (69%). Approximately half of the respondents were of the opinion that deteriorating achievement, aggressive behaviour, poor reading skills and reaching the limit of compulsory education age are early indicators of dropout. The top ten indicators also include belonging to an ethnic minority (Roma students), behavioural and learning disorders, living in a child care institution and exclusion. It is to be noted that indicators such as a deterioration of grades by an average of at least 1.1 over a year, parents’ low educational attainment, poor maths skills, and the need for a student to work and attend school did not make it into the top ten. We highlight these factors because deterioration by 1.1 in academic achievement is part of the early warning system in operation since 2016. The low priority given to poor maths skills is also surprising, as mathematics has been the leading fail subject for many years (Fehérvári 2008). Of the 30 early warning factors, the respondents perceived reserved behaviour, immigrant or rich family background, being raised by grandparents or living in a large family to be the least indicative of dropout. Figure 1 shows the factors that teachers considered to be early warning indicators as well as whether these indicators were monitored by their schools. Schools typically identify aggressive behaviour and deteriorating academic achievement, and do not keep such a close eye on parental attitude and students’ ethnicity 10 . Using exploratory factor analysis, we tried to uncover the underlying struc- ture of the 29 dummy variables. The variables are suitable for analysis, the KMO test value is appropriate (.642), and the significance level by Bartlett’s test is .000. Rotation was applied for ease of interpretation of the factors. The cumulative variance of the analysis is 54.5% (Table 3). 10 In order to protect privacy, schools have not been recording Roma ethnicity since 1993. 146 Sodobna pedagogika/Journal of Contemporary Educational Studies Fehérvári Poor skills Indepen-dent life SEN Academic failure Deviance Non-aggressive deviance Poor family No family Parental attitude Atypical back-ground Raised by grand-parents Poor maths .871 Poor reading comprehension .891 Partnership .669 Works .672 Overaged .478 SEN .755 Integration, learning and behavioural disorders .735 Disabled .489 Achievement deteriorating by 1.1 grades .693 Underachievement .694 Absenteeism .601 Aggressive .614 Ethnicity .474 Multiple siblings .429 Reserved .565 147 Dropping out in Hungary: teachers’ perceptions Exclusion .648 Bored .584 Parents’ low educational attainment .612 Poor family .576 Unemployed family .577 Orphan .716 Living in a child care centre .605 Parental disinterest in child’s achievement .647 Parent does not contact school .774 Immigrant background .668 Rich family .525 Chronic illness .366 Single-parent family .283 Raised by grandparents .780 Table 3: Rotated matrix of factors behind dropping out 148 Sodobna pedagogika/Journal of Contemporary Educational Studies Fehérvári As a result of the factor analysis, 11 factors emerged out of the 30 variables, and present the following internal structure: 1. Poor skills: mathematics, reading comprehension 2. Generally poor academic achievement: poor performance, deterioration of grades by an average of 1.1 over a year 3. Special education needs: integration, learning and behavioural disorders, disability 4. Deviance: absenteeism, aggressive, ethnic minority , multiple siblings 5. Non-aggressive deviance: reserved, excluded, bored 6. Independent life: lives in partnership, works and attends school, older than the compulsory education age limit 7. Poverty in the family background: poor family, parents have low educational attainment, parents unemployed 8. No family: orphan, institutionalised 9. Parents’ negative attitude: parents fail to follow the child’s performance, par- ents do not attend teacher-parent meetings 10. Atypical family background: immigrant, rich family, chronic illness, single- parent family 11. Raised by grandparents The 11 factors fall into three main groups. The first group contains the factors related to academic achievement, or rather failure (a-b). The second includes stu- dents’ personal traits and characteristics (c-f). Here, distinction should be made between special education needs and various deviances, the latter comprising mild, less conspicuous and easy-to-see factors of deviance. The third large group is composed of family background characteristics (g-k), where characteristics of low family status and other family-related deficits affecting dropout constitute separate factors. Aside from identifying the problem and exploring its causes, it is important to determine how great a responsibility teachers attribute to schools in pre- venting dropout. Only 11% of the respondent teachers believed that prevention of dropping out was not part of their job. Based on the teachers’ responses, there is a significant difference in the proportions of students at risk of drop- ping out (p = .000). In schools where teachers did not regard dropout preven- tion as part of their job, the risk index was, on average, three percentage points lower than in schools where teachers considered it to be an important part of their job. We measured on a scale of 0-10 the impact that teachers attribute to schools and their own work in fighting dropout. While they perceived both actors as having a significant effect, they believed that the school’s role was more im- portant than their own: the average score of the school’s importance was 7.8, and their own importance was 7.2. There was no correlation between the degree 149 Dropping out in Hungary: teachers’ perceptions of assuming responsibility and the rate of students at risk of dropping out in the respondents’ schools. A scale of 0-10 was also used to measure which of the 11 stakeholders or factors are most responsible for dropout. The respondent teachers put the greatest responsibility on parents, followed by school principals. They also attributed greater responsibility to society, the state and the media than to themselves, and they be- lieved that pre-primary school and the lower grades were the least responsible. Of the items listed, the student appears in the lower third. 024681 0 Pre-primary school Grades 1-4 The student Grades 5-8 Teachers The media The state Peers Society School principals Parents Figure 2: To what extent do you think the following stakeholders are responsible for students’ dropping out of school? (Scale 0-10, average) Pearson’s correlation indicates that there is a significant relationship between the assessment of impact and responsibility. Respondents who rated their own impact higher also considered the responsibility of teachers, grades 1-4 and grades 5-8 to be greater (the coefficients are .466, .464 and .424 respectively). Analysis of variance was used to compare the impact and responsibility scales with the school climate indicators (Table 4). School climate was investigated using 15 items, including teacher-teacher, teacher-student, teacher-parent and teach- er-school principal relations, as well as an assessment of the school’s internal and external environment. 150 Sodobna pedagogika/Journal of Contemporary Educational Studies Fehérvári School climate indicators Response (scale 1-4 11 ) Average (degree of teacher’s impact, scale 0-10) Average (degree of teacher’s responsibility, scale of 0-10) In this school teachers and students generally have good relations. Somewhat disagree Fully agree 5.4 6.4 - The teachers of the school share common educational values. Somewhat disagree Fully agree 5.4 6.6 6.8 7.0 The school climate is characterised by mutual support. Somewhat disagree Fully agree 5.5 6.4 – The school has good relations with the local community. Somewhat disagree Fully agree 5.6 6.5 – In this school teachers regard parents as partners. Somewhat disagree Fully agree 5.9 6.6 – Most of the teachers in this school consider it to be important that students feel good at schools. Somewhat disagree Fully agree 5.7 6.6 6.2 6.9 In this school students can be involved in making decisions that concern them. Somewhat disagree Fully agree 5.7 6.6 6.5 7.0 In this school teachers can be involved in making decisions that concern them. Somewhat disagree Fully agree 5.8 6.7 6.4 7.0 In most cases parents seek teachers’ professional and pedagogical opinion on their children. Somewhat disagree Fully agree 5.5 6.3 6.2 7.1 Most teachers in this school are interested in the thoughts and opinions of students. Somewhat disagree Fully agree 5.8 6.7 6.8 7.0 If a student needs special assistance, the school provides it. Somewhat disagree Fully agree 5.5 6.7 6.5 6.9 This school is a safe place for students. Somewhat disagree Fully agree 6.1 6.4 6.6 6.8 The principal always discusses the school’s educational goals with the teaching staff and generally takes the staff’s opinion into consideration. Somewhat disagree Fully agree 5.7 6.6 6.4 7.0 In this school teachers regularly discuss their problems and difficulties regarding teaching and education. Somewhat disagree Fully agree 5.7 6.6 - The school provides appropriate extracurricular activities for students. Somewhat disagree Fully agree 5.9 6.4 - Table 4: Relations between school climate and teachers’ impact and responsibility 11 School climate indicators were measured on a scale of 1-4, where 1 was given if the respondent fully disagreed with the statement and 4 if they fully agreed. As the number of respondents who fully disagreed was low, Table 4 shows the averages of the ‘Somewhat disagree’ and ‘Fully agree’ answers. Due to a lack of space, we omitted the ‘Somewhat agree’ data. This does not affect the interpretation, as in each case the figure falls between the other two averages. 151 Dropping out in Hungary: teachers’ perceptions In terms of teachers’ impact, every variable showed a significant connection (p < .015), but the same was less strong regarding teachers’ responsibility: only 9 out of the 15 variables appear to be connected (p < .031). Among the school climate indicators, the biggest difference in the average degree of teachers’ impact was in the perception of teacher-teacher and teacher-student relations. Teachers who saw themselves as having a greater impact on preventing dropout tended to think that teacher-student relations were positive in their school, and the teaching staff adopted common educational values. Regarding teachers’ responsibility , in the set of school climate items, variance was greatest in respect of school heads-teachers and teacher-parent relations. In schools where teachers felt that they were involved in decisions that concerned them, the heads took their opinions into consideration, and parents also sought their opinion, teachers’ degree of responsibility was higher . On the whole, taking both variables into consideration, it can be concluded that teachers feel that they have a greater impact and a greater responsibility in schools where: – the staff shares common educational values – the school regards parents as partners – parents seek teachers’ opinions on their children – it is important for teachers that students should feel good at school – teachers seek students’ opinions – both teachers and students are involved in decisions on matters that concern them – school administrators ask for and take into account the staff’s opinions Conclusions Education deficit has multiple negative impacts on the individual as well as on society; consequently , studies exploring the effectiveness of education systems and schools constitute an important line of research. This paper presents the concepts of leaving education and training early and dropping out, the Hungarian figures and trends in relation to European data, and the most important push/pull factors of dropping out classified into main categories. Relying on data from a questionnaire-based survey of teachers in primary schools participating in a project aimed at prevention of early leaving, we analysed the characteristics that teachers identify as risk factors of dropping out, and ex- plored the impact and responsibility that they attribute to themselves and their schools, as well as how their views are affected by their school climate. We examined these factors in detail because they are the ones that the international literature (Lyche 2010; Witte et al. 2013) underscores as particularly important aside from students’ individual traits and characteristics. One of the findings to be highlighted is that among the early warning indic- ators familiar from the literature, the respondent teachers pointed to attendance 152 Sodobna pedagogika/Journal of Contemporary Educational Studies Fehérvári as the most important one. They perceived parents – specifically, parents’ lack of interest in and failure to follow their children’s academic performance – to be the second most important indicator. Although the prevention project only involved primary schools that the early warning system indicated as the most vulnerable to dropout in their respective regions, the majority of the respondent teachers did not perceive dropout to be a problem in their school. It is conspicuous that the respondents avoid the problem as well as responsibility; it is also worth noting the extent to which they regard the prevention of dropout to be part of their job. At the same time, in schools where the rate of students at risk of dropping out was higher, a greater proportion of teachers perceived dropout to be a problem and also tried to take steps to prevent it. Aside from parents and family , the teachers considered the external environment (society , the state and the media) to be more responsible for dropout than teachers themselves. However, teachers also believed that they and their schools play an important role in preventing dropout. This perception was significantly influenced by the school climate. In schools where the climate was seen as more positive, where teacher-student relations were good and the teaching staff shared common pedagogical values, the respondents felt that they had a great impact on fighting dropout. It is important to realise that the issue of taking responsibility was related to school climate, in particular to school heads-staff and teacher-parent relation- ships. Teachers in schools where teachers were more involved in decision-making saw themselves as more responsible for dropout. In addition, teachers in schools where teachers felt that school heads took their opinions into consideration took greater responsibility for their students’ performance. The findings of the study confirm what we describe in the theoretical part of this paper: teachers’ perceptions, which are influenced by organisational charac- teristics, play an important role in the exploration of dropout. The future direc- tion of our study is to conduct a more complex analysis of these factors and their combined effects. References Berényi, E. (2015). Handling failure and seeking solutions: problem narratives in vocational training. In: A. Fehérvári (ed.). Snapshot of Hungarian Education 2014. Budapest: Hungarian Institute for Educational Research and Development, pp. 187–202. CEDEFOP . (2016). Future skill needs in Europe: critical labour force trends. Luxembourg: Publications Office of the European Union. Eurofound. (2015). European Working Conditions Surveys (EWCS). Retrieved from https:// www .eurofound.europa.eu/surveys/european-working-conditions-surveys (accessed on 31. 07. 2019). Europian Commision. (2015). A whole school approach to tackling early school leaving. European Union. Fehérvári, A. (ur .). (2008). Szakképzés és lemorzsolódás. [Vocational training and Dropping- out]. Budapest: Hungarian Institute for Educational Research and Development. 153 Dropping out in Hungary: teachers’ perceptions Fehérvári, A. (2015a). Career tracking of young skilled workers, 2010–2014. In: A. Fehérvári. (ed.). Snapshot of Hungarian Education 2014. Budapest: Hungarian Institute for Educational Research and Development, pp. 203–218. Fehérvári, A. (2015b). Lemorzsolódás és a korai iskolaelhagyás trendjei. [Trends in early school leaving and dropping-out.]. Neveléstudomány, issue 3 pp. 31–47. Liskó, I. (2003). Kudarcok a középfokú iskolákban. [Failures in secondary schools]. Budapest: Hungarian Institute for Educational Research. Lyche, C. S. (2010). Taking on the Completion Challenge a Literature Review on Policies to Prevent Dropout and Early School Leaving. OECD Education Working Paper No. 53. Kertesi, G. and Kézdi, G. (2013a). Ethnic Segregation Between Hungarian Schools: Long-run Trends and Geographic Distribution. Hungarian Statistical Review, n. 16. Kertesi, G. and Kézdi, G. (2013b). The achievement gap between Roma and nonRoma students in East Central Europe and its potential causes. GRINCOH Working Paper Series, n. 4. Lamb, S., Markussen, E., Teese, R., Sandberg, N. and Polesel, J. (eds.). (2010). School Dropout and Completion: International comparative studies in theory and policy. Dordrecht: Springer. Nagy, P . T. (2010). Utak felfelé. Oktatás és társadalmi mobilitás a 19-20. századi Mag- yarországon. [Education and Social Mobility 19–20. century in Hungary]. Budapest: Hungarian Institute for Educational Research, Új Mandátum Ostorics, L. (2015). A tanulói teljesítménymérések jellemzői, jövőbeni irányvonalai, kritikái. [Characteristics, future directions and criticisms of measuring academic performance]. In: K. Széll (ed.). A tanulói teljesítmények alkalmazhatóságáról. [Applicability of student performance measurement]. Budapest: Hungarian Institute for Educational Research. OECD. (2017). Education at Glance Paris 2018. Rumberger, R. W . (2012). Dropping out. Cambridge: Harvard University Press. Sági, M. (2015). Teaching career patterns. In: A. Fehérvári (ed.). Snapshot of Hungarian Education 2014. Budapest: Hungarian Institute for Educational Research and Devel- opment, pp. 91–107. Smeyers, P . (2006). The Relevance of Irrelevant Research; the Irrelevance of Relevant Research. In: P . Smeyers and M. Depaepe (eds.). Educational Research: Why ‘What Works’ Doesn’t Work Dordrecht. Germany: Springer, pp. 95–108. Széll, K. (2015). Schools’ effectiveness and teachers’ attitudes. In: A. Fehérvári (ed.). Snap- shot of Hungarian Education 2014. Budapest: Hungarian Institute for Educational Research and Development, pp. 108–126. Széll, K. (2015). Iskolai eredményesség a hátrányos helyzet tükrében. [School performance and disadvantage]. Educatio, 24, issue 1, pp. 140–147. Tomcsik, D., Joksimovic, J., Juhász, J. and Mihályi, K. (2014). Early warning systems in six European countries. Budapest: Tempus Közalapítvány. Retrived from http://okta- taskepzes.tka.hu/content/documents/CroCooS/Early_warning_systems_in_six_Euro- pean_countries.pdf (accessed on 31. 07. 2019). V arga, A. (2015). Lemorzsolódás vagy inklúzió. [Dropping-out or Inclusion.] In: A. Fehérvári and G: Tomasz (eds.). Kudarcok és megoldások. [Failures and Solutions]. Budapest: Hungarian Institute for Educational Research, pp. 73–86. V arga, J. (ed.). (2018). A közoktatás indikátorrendszere 2017. [The indicators of Hungarian Pub- lic Education]. Budapest: MTA Közgazdaság- és Regionális Tudományi Kutatóközpont. 154 Sodobna pedagogika/Journal of Contemporary Educational Studies Fehérvári Wilson, S. J., Tanner-Smith, E.E., Lipsey, M.W., Steinka-Fry, K. and Morrison, J. (2011). Dropout prevention and intervention programs: Effects on school completion and drop- out among school-aged children and youth. Campbell Systematic Reviews, 8, pp. 1–62. De Witte, K., J. Cabus, S., Thyssen, G., Groot H.W . and Witte, M (2013.) A Critical Review of the Literature on School Dropout. Educational Research Review issue 10, pp. 13–28. Anikó FEHÉR VÁRI (Univerza v Budimpešti, Madžarska) PROBLEMATIKA OSIPA NA MADŽARSKEM Povzetek: T eoretsko ozadje tega prispevka so raziskave o osipu in zgodnjem opuščanju šolanja, pri čemer se opiramo na različne mednarodne ter madžarske zbirke podatkov , ki prikazujejo trende in razmere na Madžarskem, pa tudi individualne, organizacijske, institucionalne in sistemske dejavnike, ki vplivajo na osip v tej državi. V članku predstavljamo analizo podatkov iz prvega dela empirične raziskave, ki je sicer sestavljajo štiri faze. V letu 2018 je bilo opravljeno anketiranje na 83 šolah, ki so bile izbrane na podlagi deleža učencev z visokim tveganjem osipa. Želeli smo ugotoviti, kakšna so na teh šolah učna ok ol j a te r i de n ti f i ci r a ti d e j a v n i k e tv e g a n j a i n z a š či tn e d e j a v n i k e , k i b i l a h k o p r i p om og l i k v e čj i m a l i manjšim možnostim za zgodnje opuščanje šolanja. V članku se osredotočamo na eno razsežnost tega kompleksnega raziskovalnega problema, tj. na vprašanje odgovornosti šole in učiteljev . R ezultati naka- zujejo, da je stopnja zaznavanja problema pri učiteljih na šoli sorazmern a z deležem učencev z visokim tveganjem za osip. Učitelji so prepri čani, da imajo tako oni kot šola pomembno vlogo pri preprečevanju osipa, toda hkrati menijo, da je problem osipa bolj problem drugih šol in drugih dejavnikov kot pa njih samih (staršev , družbe, medijev). Ključne besede: osip, prepričanja učiteljev , javna šola, Madžarska E-naslov: fehervari.aniko@ppk.elte.hu