Navigation – Plan du site

Social space and educational outcomes in Athens

Espace social et résultats scolaires à Athènes
Stavros Spyrellis


Cet article se propose d’étudier si les lycées athéniens révèlent et/ou reproduisent les structures sociales qui les entourent. La méthodologie employée consiste à classer des écoles publiques secondaires selon les résultats scolaires et à les juxtaposer aux inégalités observées dans leur environnement spatial, socio-professionnel et éducatif. La réussite scolaire est mesurée par la performance à l'examen national d'admission à l'enseignement supérieur. Les données individuelles des élèves au niveau de chaque école sont ensuite agrégées afin de démontrer comment la variation peut être liée à des caractéristiques sociales. En utilisant une application SIG et des statistiques d'analyse de correspondance, nous avons produit un indicateur de profil socio-économique et éducatif autour de chaque école. Cette analyse confirme le manque de « démocratisation » du système éducatif athénien. La performance y apparaît comme fortement liée à la profession et la réussite scolaire, tandis que la composition ethnique et le niveau de vie semblent jouer un rôle important.

Haut de page

Texte intégral

1. Introduction

1In countries where social distinctions are particularly apparent, schools in highly segregated neighborhoods are also conceded as de facto segregated, displaying the existence of a strong correlation between residential and educational segregation (Préteceille, 2003; Rhein, Le Pape et Grosbras, 1999). The latter becomes even stronger when policies oblige students to enroll in the local school. However, the formation of expanded middle social classes renders the urban geography of education increasingly complex (J.C. François, 2002; Oberti, 2002, 2005; van Zanten, 2001). This is the case of Athens, where large parts of residential space exhibit a mixed social character (Maloutas, 2004; Spyrellis, 2011). Considering, on the one hand, that education is an important indicator of the quality of social services provided to residents of the same neighborhood and, on the other, that geographical differences in educational attainment historically can be attributed to differences in social composition (Hamnett, Ramsden, & Butler, 2007), it is intriguing to question the extent of educational inequalities throughout Athenian residential zones.

2The Greek educational policy is characterized by lack of choice contrary to other countries as the UK and the US; in London 20% of students attend a secondary school outside their home local authority (Gordon & Monastiriotis, 2007) and in some states in the US the ‘busing’ system forms non-contiguous catchment areas and therefore strong spatial discontinuities throughout the schooling system (Billings, Deming, & Rockoff, 2012). The above mentioned lead to the conclusion that the schooling process is inevitably linked to the sectorization imposed by the Greek catchment area pattern of ‘school districts’, a system similar to the ‘carte scolaire’ in France (Ayed, 2009; J. François & Poupeau, 2008; J.C. François & Rhein, 2007; Oberti, Préteceille, & Rivière, 2012).

3To mitigate educational segregation in a system based on strict catchment areas, equally performing schools and equal social composition should be assured. As S. Bowles and H. Gintis emphasized, exercising education policies in a context of unequal social conditions is impracticable to bring equality (Bowles & Gintis, 1976). In reality, even if one could assume that educational input is absolutely balanced, it is impossible to ensure social homogeneity. As a result, schools, through their performance, register the level of segregation that surrounds them (Oberti et al., 2012).

2. Educational inequalities in Greek society

4Certain social researches in Greece acknowledge the fact that education does not constitute an independent social structure, mostly trailing the general social, economic and cultural interests (Frangoudaki, 1985; Lambiri-Dimaki, 1974; Tsoukalas, 1976). Greek society is a class society (GSEE, 2008; Tsoukalas, 1987) throughout which the family socioeconomic background is probably the main parameter explaining unequal access to higher education. Differences in students’ results according to our analysis are highly associated to their social status and the localization of the school unit that they attend. According to Tsoukalas, the Greek educational system is diachronically, on the one hand contributing to the concealment of the social contradictions it encloses while on the other hand its most important operation is the reproduction of cultural gaps among social classes; 'If there is a country where it could be believable that the myth that social exclusion of the lower classes from the goods of culture can be lifted through educational and cultural mechanisms alone, this is Greece’. (Tsoukalas, 1976).

5Since the Second World War and subsequent to essential educational reforms, going to school and attending higher levels of learning was considered a strong social mobility mechanism (Frangoudaki, 1985). The latter was more prevalent throughout the capital, main accumulator of the agrarian populations that migrated from rural areas. Once in Athens these populations gradually shaped a massive intermediate social class displaying high levels of Upper Secondary Education attainment and a Tertiary Education almost proportional to the rest of Europe. According to T. Maloutas, education - higher levels in particular - served as the main passport to the Athenian middle class (Maloutas, 2007a).

6This perception produced, during the second half of the 20th century, a disproportional over-education especially concerning this vague intermediate class. Main motive of this class was social accession and employment security. Ideal prospect was the access to employment in the public sector, which provided permanence and social recognition. Often, the latter would occur via a long established political clientele system. This ‘special’ relation between voters and politicians resulted in an unrestrained public sector. Indicative of this situation is the fact that in the late 1980's the state employed 70% of the yearly graduates (Tsoukalas, 1987).

7Maloutas states that the significant reduction in recruitment by the public sector since the early 1990's generated further differentiation trends among higher education degrees. A certain devaluation of the degrees linked to the public sector, or degrees which lead to the education education and training of future educators, such as studies in Literature, Political and Social Sciences, Educational studies or studies in Public Administration, was observed together with an upgrade of others such as Economics and Business Studies (Maloutas, 2007a). The latter today is imperative due to the increased demand in the business arena and the introduction of post-industrial economic sectors throughout the Athenian space (Gospodini, 2009; Spyrellis, 2013a; Spyrellis 2013b). The social diffusion of access to tertiary education led to a further augmentation of the enrolments, reinforcing the downgrading of degrees resulting to what is known as the inflation of degrees (Passeron, 1982).

8Before presenting the results of this analysis, some additional information concerning the peculiarities of the Greek educational system should be given.

2.1. Secondary education

9This level of education consists of two degrees: the lower which is the equivalent to Junior High School and it is called Gymnasium, and the higher level, equivalent to Senior High School that is called Lyceum. Compulsory education nowadays, necessitates attendance in Elementary and Gymnasium.

10In 2001, a little over half of the population residing in the Athenian Metropolitan Zone (henceforth “AMZ” – figure1) had completed at least an Upper Secondary Education (51.2%). Moreover, almost 22% abandoned schooling after the Elementary Level, before concluding the current compulsory basic schooling of Lower Secondary Education. The latter can be explained by two factors: firstly due to the educationally homogenous agrarian populations that migrated to thecapital during the second half of the 20th and, secondly, to the fact that before 1980 the compulsory level of education was limited to the Elementary School. Finally an 11.4% of the population did not acquire an Elementary Education Diploma and a 7.5% declared that they were illiterate.

11According to the OECD 2011‘Education at a Glance’ (OECD, 2011), in 2009 on average across OECD countries, the 27% of adults had attained Primary or Lower Secondary levels of education, 44% had completed Upper-Secondary Education and 30% had obtained a tertiary qualification. This report indicates dramatic changes in Greek society concerning schooling and literacy since a 30% increase occurred in 30 years. Moreover, it shows that Upper Secondary Education has become a norm among Greek youngsters.

Figure 1: The Athenian Metropolitan Zone (AMZ)

Figure 1: The Athenian Metropolitan Zone (AMZ)

2.1.1. Private Secondary Education

12It is of great importance to examine the private schools network, 8.5 % of students in 2004, seeking to understand how their dispersion could be taken into account or influence our results. Concerning their operation, private schools in Greece receive no public funding. However, this absence of funding does not prevent the state from exercising control over private education institutions (Dronkers, Felouzis, & Van Zanten, 2010). Furthermore, in Athens these institutions constitute extra-local cores and therefore cannot be linked to their surrounding areas. They consist of a target of the higher social strata, regardless of the place of residence, and their operation is based upon a large network of school transports (Sivignon, Auriac, Deslondres, & Maloutas, 2003). Elite schools are renowned for their performance, 90% of accomplishment in 2010 (To Vima, 31 October 2010), as well as for their social selectivity having a tendency to recycle their clientele throughout the generations (Hadjiyanni et al., 2012). In most cases the expense of such an education and the need for a certain social network in order to be admitted render them inaccessible to most of the intermediate social class. Most of these institutions offer a complete education, from Elementary to Upper Secondary Education; therefore, they rarely absorb students that had started their education in the public network. More and above, the bigger ones - in name, reputation and size - offer the choice of a differentiated syllabus, during the last two High School years, by applying the International Baccalaureate Program which leads to undergraduate studies in other countries and, therefore, is addressed to families of a higher economic potential. However, there is a lower rank of private schools, corresponding to 1.6% of the total amount of students, of a very low educational performance. Such schools function as a shelter in cases of public education dropouts.

13Concerning their dispersion around the AMZ, those of a higher educational performance are to be found around the northeastern higher profile suburbs and those of a poorer performance around neighborhoods of a lower socio-economic status. During the last 20 years, several of the former were relocated in order to modernize their physical infrastructures and therefore to host more students. This was materialized in remote, non-residential zones confirming the assumption that private schools are not dependent on nor influenced by their surroundings

14Last, independently of their performance, some private schools were founded after bilateral agreements between Greece and other foreign countries offering a curriculum influenced by foreign educational systems (French-Greek school, Italian-Greek school, American college etc.).

2.2. Tertiary education in Greece

15In the Greek educational system, candidates for university and technological education institutes are selected after a series of national exams. After finishing their Upper Secondary Education (ISCED 3) (EACEA, 2010; OECD, 2002) at the Senior High School (Lyceum), all candidates wishing to pursue a Tertiary Education (ISED 5 A or B) participate in the National Admissions Examination (henceforth ‘NAE’) organized by the Ministry of Education.

16Students enroll hierarchically in departments, following a decision of the Central Services Department of the Ministry. The latter takes into account both the marks obtained by the candidate, a combination of High-school grades (30%) and NAE results (70%), and the choices the student makes. Our data set shows that from the 29,200 students that graduated in 2004-2005, an 11% did not submit their university entrance application form and therefore did not take the exams.

2.3. Residential relocation and educational strategy

17Residential relocation as a parental educational strategy is expressed by moving towards areas with ‘better’ schools or ‘better’ social composition or finally making attempts to enroll one’s offspring into a ‘good’ public school (Butler & Hamnett, 2007). In Greece, the educational outcome of a school is not a major factor of influence for the choice of residential location. Maloutas indicates that in 1986 only marginally identified schools were considered an important reason of influence and that only the 7% of the participants in a relative survey answered that they had considered it (Maloutas, 2007a). This is linked to the high levels of home ownership and the importance of inter-family relations within Greek society, which often operate as a predicament for moving. An extensive survey conducted by the Greek National Centre for Social Research (N.C.S.R) in 2002 showed that on a 94% only these two factors are considered fundamental (Maloutas, Emmanuel, & Pantelidou, 2006). As education is known to be a major financial concern for Greek society, we can assume that expenses, which in other countries are, destined for relocation, in the Greek case they are used to finance complementary classes and private courses/tuition? helping, therefore, compensate probable inadequacies of the school network.

3. A typology of educational performance

18The first issue of this analysis was to create the typology of the schools relative to the educational outcome. In order to do, we grouped the candidates according to their results in the NAE using 6 performance groups (table 1). Then, based on the repartition in each rank, the schools were classified according to the outcome of their educational service. This procedure resulted in 4 distinctive types of schools.

19In detail, for the academic year 2004-2005, a total of 29, 200 students coming from 310 schools, 281 morning, 17 afternoon and 10 experimental schools, were enroled in the last year of the Secondary Education. The decision to construct our groups according to the NAE results is related to the objective of determining common criteria for the whole of our dataset, an element assured by the fact that the Ministry of Education organizes these examinations centrally. Therefore, we took into account the 88.6% of the students, which in June 2005 filled in and submitted their university entrance application form.

20In order to group the students, we constructed 6 groups of the educational performance (table 1). Based on this categorization, we calculated the number of students per school in each rank and upon these percentages we proceeded with the construction of the typology of the 310 public high school units.

Table 1: The grades according to which the 6 groups of educational performance where formed.



Number of pupils







Very good








Well enough














Source: Greek Ministry of education 2006-2007

  • 1 The fact that the hierarchical classification method is considered as one the most solid methods of (...)

21A Principal Components Analysis (PCA), gave us the opportunity to examine the correlations and oppositions of our data internal structure and, therefore, to locate the main axes of spatial variation throughout the school network. We then determined four distinctive types of educational performance by proceeding to a hierarchical classification1 of the dataset seeing to the first 3 axes of variation, which resulted from the Principal Components Analysis. Their geo-localization (figure 2) resulted in the formation of a synthetic image of the AMZ’s public high school network giving the opportunity to perceive whether schools with relevant outcome tend to be located close to similar schools and to question the existence of clusters of common educational performance.

22The high performance type of school units, which account for 8% of the total, shows an over-representation of pupils that graduated with very good grades or even with distinction. In fact, one out of four students who graduated with distinction came from high performance high schools. Overall, 34% of students graduating from high performance schools had a final grade higher than 16/20 while students who failed to pass the 10/20 show a strong under-representation. Contrary to this, low performance schools, almost double in number of school units (18%), show a strong over-representation of students of very poor educational performance combined with a gradually augmenting under-representation for the rest of the ranks of our performance scale. Indicative is the fact that only 7% of the students scored over 16/20, a percentage almost three times poorer than the general performance in AMZ.

Figure 2: Types of public high schools in metropolitan Athens resulted from a Hierarchical Clustering based upon the 6 rank scale of educational performances in the admission examinations in 2005.

Figure 2: Types of public high schools in metropolitan Athens resulted from a Hierarchical Clustering based upon the 6 rank scale of educational performances in the admission examinations in 2005.

23This analysis revealed two more dominant patterns of a more or less mixed character. The mid-low performance type of schools (40%) constitutes the medium profile of Athenian public high schools and the mid-high schools (34%) show a tendency towards relatively higher education performance. Never the less, even though the differentiation between those two is not that marked, we have decided to keep them both since they do resume different trends.

24Concerning the operational character of schools, 70% of the experimental schools are included in the high performance schools, with those located in upper middle neighborhoods scoring very high, and 100% of afternoon schools located in the lower ranks of the low performance schools.

25A major disadvantage in classifying school performance in Athens is the very common trend of supplementary tutoring; complementary classes by private institutions or private courses. According to M. Kassotakis in the late 1980s, 65% of Higher Education students had taken preparatory courses in private institutions, private courses or both (Kassotakis, 1996). Furthermore, the fact that their role is to reinforce the students’ performance begs the question that we cannot answer since we cannot distinguish the actual contribution of high schools to this performance.

26Unfortunately, the aforementioned cannot be measured from the time when private lessons are usually unofficial, undeclared and untaxed and the private institutes do not provide reliable data of student potential since their clientele is based upon these numbers. To improve on this, one could use the economic survey on the expenses of Greek households; still, it is difficult to define which part of the amount of money declared as spent for educational purposes goes to private extra-curriculum education. Concerning its social character, private tuition is considered a common choice for students of upper social background, when for the intermediate and lower strata preparatory classes in private institutions seem more affordable.

27Although of great importance, this mechanism does not affect our analysis since registering the contribution of the supplementary tutoring as part of high schools performance corroborates to the socio-economic status of students. Undoubtedly, schools do ‘exploit’ the work done otherwise, and therefore results should not be considered to be wholly an outcome of the high-school’s operation. Over and above, each student’s ability to participate and the intensity of the participation in this educational surrogate depends on the economic potential of the family, thus, helping, indirectly, to reveal the strong relation between educational outcome and economic background.

3.1. Educational performance and participation in the National Examinations

28The fact that this analysis sees only to the NEA participants, excluding students that did not participate (11.4%) or did not manage to obtain the average score (40%), suggests the need for further investigation.

29The decision not to participate can be linked to reasons, such as the disbelief of their entering university, choosing to continue abroad or in a private technical or professional institution in Greece or even having no interest to do so. Participation rates fall from 95% in the high performance, to 72% in the lower performance schools. It can be assumed that this 23% percentage point decrease is due to the social and educational background, to the funding capacities or even to the educational performance of the candidate.

30Figure 2 shows that the low performance and also low participation high schools are mainly located around the western parts of the AMZ, those considered of a lower socioeconomic and educational status. Additionally, these neighborhoods show greater concentrations of individuals employed in vocational and technological professions.

31In the past decades, social pressure towards youngsters to participate in the examinations by their family and their social environment was much stronger. Since the 1960s, the greatest prospect, even of very poor families in secluded areas of the country, was to enter higher education. Indicative is that in 1976 only 41% of those who entered a tertiary education institution were 18-19 years old, showing their repetitive attempts, and that in the beginning of the 1980s, the popular strata were represented in the higher education with much higher percentages than in the rest of Europe (Frangoudaki, 1985). Today this pattern is not as evident due to changes in the social and economic standards as well as the educational structure. Of paramount importance was the introduction of a vocational and professional education curriculum and the establishment of non-tertiary post-secondary education solutions for a vocational rehabilitation.

32The limited 5% of graduates from high performance schools that did not participate should be linked to education advantages such as the possibility to study abroad or in private tertiary education institutions.

33In the metropolitan zone, participation and success rate seem to be strongly correlated (r=0.77). In order to comprehend this interrelation, one should concede that before 2006 the Ministry of Education could allocate to candidates the whole number of places attributed to each department independently of their educational performance. Therefore, candidates could gain entrance to a tertiary education institution, ISCED 5A or 5B, even by scoring a final grade below the average of 10/20. In detail, for the academic year 2004-2005, the last student to enter a University scored 9.97/20 and a Technological Education Institute (ISCED 5B) scored 7.38/20. Law changed this situation, in 2006 excluding those students from the admission procedure (official gazette 39, 30 may 2006). As a result, in 2006 the number of students enrolled in private preparatory institutions in order to re-participate in the NEA increased by 30% and the number of students enrolled in public or private post-secondary non-tertiary education institutions increased by 20% (Kathimerini, 2 december 2006).

34A second element that seems to play a role in the high rates of lower performance is that participating in the NAE procedure gives the right to the young male candidates, to ask for a military deferment. Otherwise, every Greek who turns 18 is obliged to enlist for his army service. The deferment from the army gives the opportunity to youngsters to directly enter the professional market, to study in order to participate again in the NAE or to prepare their candidacy for a university course abroad.

4. Relating educational performance and social background

35Relating the educational performance and the residential space of the Greek capital was the last issue to be questioned. It is acknowledged that, groups of a certain social resemblance tend to concentrate spatially and consequences do occur due to the common constraints to which these people are subjected. These points reflect the fact that urban space is not extrinsic to social elements (Marpsat, 1999).

36One should also bear in mind the ‘class effect’, described by Gordon and Monastiriotis as the ‘vicious circle’ of outcome (Gordon & Monastiriotis, 2007). This effect occurs in both the ‘good’ and the ‘bad’ high schools. The ‘good’ high schools come to display even better examinations results, either because students entering get to enjoy other educational advantages, e.g. supplementary tutoring, or because their presence tends to raise the general performance of a school. An analogous mechanism exists in schools identified as ‘bad’. Concerning the latter, different explanations and theories (Gould-Ellen & Austin-Turner, 1997; Jencks & Mayer, 1990) have been cited by researchers in order to identify distinct mechanisms and to explain how a neighborhood affects the lives and behavior of individuals. This social influence led to what Hamnett et al name the ‘school composition effect’ (Hamnett et al., 2007) acknowledging the existence of a ‘tipping point’, as to the proportion of children coming from a (dis)advantaged background, which could turn schools in to ‘to be avoided’ cases.

37In summary, this part of our analysis questions whether or not specific levels of educational outcome can be linked to distinctive types of socio-economic and educational character. To do so we used the 2001 National Census data concerning the socio-economic status (table 2) and the educational status (table 3) of the population.

38To categorize our dataset we used the nine major categories of the European Socio-economic Classes (ESeC) (Rose & Harrison, 2007) and the eight level educational categorization as proposed by the National Statistical Service of Greece (NSSG) (NCSR & NSSG, 2005). In both cases the database was organized on a small scale of 3 500 census tracts, seen on figure 2, a scale which represents the detailed version of the metropolitan zone and is proposed by the NCSR.

39Our interest lies in the relation between school outcome and social background. A major restriction of our educational outcome dataset is the lack of a direct indicator of students’ social background. In other words, the absence of information as to the home area of the students, such as the postcodes used by Hamnett et al (Hamnett et al., 2007) in order to analyze educational attainment in East London, or as the particular individual’s household background (parental occupational and educational status), used by Oberti et al analysis for Paris (Oberti et al., 2012).

Table 2: Distribution of ZMA’s population based upon European Socio-economic Classification (ESeC). Census 2001.

ESeC Classes


Percentage of the active population

SeC 1

Higher grade professional and administrators


SeC 2

Lower grade professional and administrators


SeC 3

Intermediate professions


SeC 4

Small employer and self-employed


SeC 5

Small employer and self-employed in agriculture


SeC 6

Lower supervisory and lower technician occupations


SeC 7

Lower services, sales & clerical occupations


SeC 8

Lower technical occupations


SeC 9

Routine occupations




*Source: NSSG (2001), (Rose & Harrison, 2007)

Table 3: Distribution of ZMA’ population based upon the 7 level educational categorization proposed by the NSSG. Census 2001.



ISCED level

Percentage of the active population


Postgraduate degree


3 %


Undergraduate degree


19 %


Private and Technical degree

ISCED 4 & 5B

14 %


Upper secondary education


37 %


Lower secondary education


10 %


Primary education


15 %


No education

2 %



*Source: NSSG (2001)

40We have been able to improve on this by relating every school to its surrounding social character, taking into account that public high schools draw pupils from their surroundings. We used their geolocalization and assigned the average socio-economic profile of residents in census tracts contained within a radius of 600m around each school. Through a GIS application, these ‘zones of influence’ were used as indicator of the socio-economic status and educational status of the candidates’ neighborhood. This application in order to calculate the character of each of these zones used, in most cases, weighted mean statistics. An exception was made for cases where only one high school per municipality existed and, therefore, students had only one choice. In these cases the average profile of the whole municipality was taken into account.

41This procedure resulted in a concluding database of 310 inscriptions, equal to the number of schools, linking educational performance and average social character. Based on this dataset we proceeded to a correspondence analysis, revealing the associations of catchment areas’ average statuses to school performances, and to the formation of a synthetic version through the use of a typology. This typology, constructed after a hierarchical classification using the first four axes of the data variance, resulted in 8 distinct types of neighborhoods; the dendrogram of this classification explains thoroughly their character (figure 3).

Figure 3. Synthetic mapping of the Athenian Metropolitan Zone after a multivariate analysis which took into account the socioeconomic and educational status of the active population and the performance of public high schools.

Figure 3. Synthetic mapping of the Athenian Metropolitan Zone after a multivariate analysis which took into account the socioeconomic and educational status of the active population and the performance of public high schools.

4.1. Identifying neighbourhoods

42Neighbourhoods, in some primitive, inchoate fashion exist wherever human beings congregate, in permanent family dwellings; and many of the functions of the city tend to be distributed naturally – that is without any theoretical preoccupation or political direction – into neighbourhoods” (Mumford, 1954). What could be a “scholar neighbourhood” in our case is the task undertaken and described below, based on the meanings of statistical analysis and the knowledge of the Athenian human geography. Four types of “scholar neighbourhoods” are depicted, illuminating, therefore, the social mosaic of Greek capital complexity.

4.1.1. High performance Schools

43High performance schools are mostly surrounded by intermediate social strata. In detail, upper middle class neighborhoods, of professionals and administrators, (type 2) dominate followed by mixed middle-class populations (type 3). The latter is the widest social type of the Athenian society, including populations occupied in intermediary professions, small employees and self-employed. Taken as a whole, these zones cover the 37% of our data set presenting an increased social homogeneity (Leontidou, 1990; Maloutas, 2004; Spyrellis, 2013b) and the highest levels of educational performance.

44These neighborhoods are located in the north-eastern and the south-eastern littoral suburban space. On the one hand, they share the same educational standards as the bourgeois neighborhoods, therefore, a mixture of strong and net over representation of high and higher education, but, on the other hand, are of a substantially differentiated social background. These neighborhoods were formed after 1970 in order to fulfill the needs of an emerging middle class, following a massive suburbanization and coastalization trend (Sayas, 2004) and changing severely the social structure of the Greek capital.

45The development of these suburbs was based upon the apartment building pattern giving access to more spacious apartments than those in the centre of Athens, but causing a progressive densification of the Athenian periphery. Areas, which had a late start, mainly type 2 neighbourhoods, kept a less dense development pattern following lower construction coefficients per plot and restrictions as to the amount of storeys per building.

4.1.2. Mid-high performance Schools

46Mid-high performance schools almost monopolize the neighborhoods of very wealthy and highly educated groups (type 1). These bourgeois neighbourhoods of higher occupational groups are located in the inner city traditional upper class quarters and in the furthermost south littoral and northern suburbs. Furthermore, with the exception of the inner city parts, these municipalities are almost exclusively residential (Spyrellis, 2013b). Shaped during the first half of the 20th century, on the standards of the garden-cities, they managed to keep their urban plan almost intact without succumbing to the massive apartment building wave keeping, still today, low population and construction density (Leontidou, 1989; Marmaras, 1991).

47These areas concentrate most of the private schools, which perform better than average, and especially the few elite schools. As a result, public schools, unable to cope with competition end up socially mixed and of variable performance.

48Secondly, the middle classes of our classification surround these schools.

4.1.3. Mid-low performance Schools

49According to our classification the mid-low performance high schools form the average profile schools of the study area. As a result, they are surrounded by most of the residential types we have distinguished. Their strongest representation is observed in the lower-middle class neighborhoods (type 6) to which special attention should be given for they represent the 31% of the data set being also the biggest cluster in the western parts of the capital. These suburbs were formed by the arbitrary housing production which was used in the western parts of Attica during most of the 20th century and especially between 1950 and 1970. During that period, the Athens population exhibited an increase of 37%, concentrating most of the internal immigration wave that occurred. These populations of a labor status were mostly absorbed by these western neighborhoods. The high concentration of lower social echelons is still evident today maintaining its’ social coherence even after the de-industrialization period (1980s-1990s). These parts of the city took their final urban shape during the 1970s and 1980s when poor quality, ex-illegal, constructions were being progressively replaced by apartment buildings, following poor urban planning standards which resulted in low living standards. Today, the populations residing in these neighborhoods are of a poor educational level, Primary or Lower Secondary education, and are mainly occupied in precarious employment and less qualified professions.

50These mid-low performance schools are also systematically found around the central urban areas which were shaped during the 1950-1960 period due to a particular construction development model of apartment buildings, called the anti-parochi system (Marmaras, 1991; Prevelakis, 2000). This production gave, at the time, access to higher living standards and until the 1970s was designated for higher-intermediate and intermediate social groups. The progressive retraction of the latter led to a social change after 1970. The high density building pattern lowered profoundly the living standards and lower-income populations became predominant. The degradation of living standards and the change of social composition around the Athenian municipality (types 4&7) further established a gap between centre and suburbs concerning the type of households. In the 750 census tracts of the central municipality, an average concentration of 28% for families with children by tract is observed when in the suburbs this percentage rises to 42%. On the contrary, the number of one-parent families is almost double in the centre rising from an average concentration of 19% in the suburbs to 32% in the centre. This strong relationship between one-parent families’ proportion and low educational outcome tallies with I. Gordon’s remark that, in the case of the English GCSE examination, the higher the proportion of these categories, the lower the level of educational outcome (Gordon, 1996).

51Today these areas apart from extreme saturation lack of free spaces and low living standards show limited social coherence and are considered to be occupationally and educationally mixed areas. Additionally, the fact that they show an equally distributed profile throughout the occupational scale, with sole exception a slight over-representation of intermediate occupations, corroborates the assumption of vertical segregation trends (Maloutas, 2004). Concerning their educational status, most of the inhabitants have concluded an upper secondary education degree.

4.1.4. Low performance Schools

52Our analysis showed that low performance schools are located around zones of a lower social status. Mainly interpreted by types 7 and 8 of our classification, these parts of the capital show high concentrations of skilled, semi-skilled and unskilled workers of a very low educational status, lower than compulsory education and high levels of illiteracy. These neighborhoods are located in the northwest industrial zone, the poorest parts of the Piraeus port zone and finally in some parts of the central municipality (type 7).

53The latter, only 3% data set, shows the highest connection with low performance schools. Of great resemblance to the central parts we have already presented, this part of Athens first developed during the 1920 in order to satisfy the emerging needs of the newly shaped Athenian bourgeoisie, reached its peak during the 1930s and 1940s when it was progressively transformed into the housing solution for wider social groups, mostly for the upper intermediate and intermediate strata. Finally, during the 1950s and 1960s, this zone had already sprawled and not been in direct vicinity to the traditionally wealthy parts of Athens and became designated to lower social groups. Today it shows the highest levels of concentration for individuals of a non-Greek nationality. It is shown that in Athens the influx of immigrants has contributed to decreasing segregation in the 1990s, (Maloutas, 2007b). The non-Greek residents in AMZ, in 2001 represented the 10.3 % of the total population and more precisely in the center of Athens the 15% of the inhabitants. The anonymity that such urban conditions provide, combined with the real estate over supply, created suitable conditions for these populations. Their concentration is strikingly increased around the type 7 neighborhoods, where these groups represent in some cases a hefty 40% of the population (Spyrellis, 2010). Therefore, low performance High Schools also seem to coincide with high immigrant concentrations. This also corroborates the link between low living standards and educational performance since the immigrants have mainly settled in the worst types of housing supply, oldest, cheapest and least comfortable parts of the privately rented housing stock (Maloutas, 2007b).

5. Conclusions

54This analysis affirmed that the Athenian educational system reproduces the socioeconomic and educational character that surrounds it. High School performance appears to be strongly related to the occupational and educational attainment structure of the capital. This inequality is further fueled by social and urban characteristics such as the composition of the population and the living standards.

55Sole exceptions, due to the differentiated local social structure, are the wealthiest neighborhoods where the level of educational performance is not relayed to social status. The latter is expected since the offspring of the upper social classes is almost exclusively absorbed by the elite private schools, which perform better than public-sector ones.

56Upper middle-class neighborhoods, which should be considered as the ‘top’ of the social rank of the public schools ‘clientele’, surround the best performing high schools. On the other end of the social spectrum, labor class neighborhoods surround the worst. Based on these results, we can come to the conclusion that in Athens highly segregated neighborhoods do demonstrate the highest levels of educational segregation. Given the fact that those neighborhoods are socially homogeneous, we observed that those High Schools do reproduce the social structures that surround them.

57This analysis, moreover, revealed the educational outcome in conjunction with the living standards and the ethnic composition. As has been portrayed the socially mixed, environmentally downgraded and extremely saturated central parts of the capital (type 4&7), of an average socio-professional and educational character show the strongest over-representation of low performance high schools. Additionally, we showed that low performance High Schools seem to coincide with high immigrant concentrations which undoubtedly such concentrations produce/add further complications to the educational service. These two factors are strongly linked since immigrants have mainly settled in the worst types of housing supply, oldest, cheapest and least comfortable parts of the privately rented housing stock (Maloutas, 2007b).

58Another point concerns the respect to “school districts”. The density of the population, mostly in the socially mixed part of the city, results in the location of High Schools within small distances. In numerous cases, schools of different performance were observed in direct contiguity. Therefore, local educational inequalities and relative concentrations of students of different aptitude do occur. The latter can only be explained by assuming the operation of mechanisms to overpass the catchment areas regulation. This means that without relocating, which is expensive and difficult due to the high levels of home-ownership, the families concentrate on the possibility of accessing a ‘better’ public school. This negative pattern of spatial connection sets the question about whether some schools manage to “cream off” the more promising students, according to their background, implicating further the possibility of establishing an ‘meritocratic’ educational system of equal opportunities. A last element, revealing further inequality within the school network, is limited participation in the NAE, of students residing in the neighborhoods of a lower social status.

59Athens, unlike other European metropolises, but similar to great Mediterranean cities, was formed, urbanisticly speaking, after additional expansions to its original plan, dressed by Kleanthis and Schaubert in early 19th century, which is based now on a complex of new neighborhoods formed, first by the Asia Minor refugees in 1922 and second by the post WW II internal migrants. (Myridis, 1982). This pattern seems to be followed by any other, social, economic or political, segregation, as the “scholar neighborhoods” mentioned above.

60The recent aspects of a so called “metropolization” of the Athenian Metropolitan Zone led to various changes in the socio-cultural structure of the city during the last 40 years, due to the political choices undertaken by local and central authorities. Athens seems to be some “prototype” of integrated contradictions designated to be the impetus for this spatial differentiation which, given the particularities mentioned above, seems to be “searching” of two things: its lost “homogeneity” and the “other way to be”.

Haut de page


Ayed C.B., 2009. "La mixité sociale dans l’espace scolaire : une non-politique publique", Actes de la recherche en sciences sociales, No.180, 11-23.

Billings S.B., Deming D.J., Rockoff J.E., 2012, "School Segregation, Educational Attainment and Crime: Evidence from the end of busing in Charlotte-Mecklenburg", NBER working paper series, No.18487

Bowles S., Gintis H., 1976, Schooling in Capitalist America, New York, Basic Books.

Butler T., Hamnett C., 2007, "The Geography of Education: Introduction", Urban Studies, vol.44, No.7, 1161-1174.

Dronkers J., Felouzis G., van Zanten A., 2010, "Education markets and school choice", Educational Research and Evaluation, vol.16, No.2, 99-105.

EACEA. 2010. Organisation of the education system in Greece. Evaluation, European Commission.

EKKE, ΕΣΥΕ., 2005, Panorama of Greek census data 1991–2001. Application under experimental and restricted use at the EKKE.

Frangoudaki A., 1985, Sociology of education. Theories about social inequality in school(in Greek), Athens, Papazisis.

François J.C., Poupeau F., 2008, "Les déterminants socio-spatiaux du placement scolaire", Revue Française de Sociologie, vol.49, No.1, 93-126.

François, J.C., 2002, "Évitement à l’entrée en sixième et division sociale de l ’ espace scolaire à Paris", L’Espace géographique, vol.31, No.4, 307-327.

François J.C., Rhein C., 2007, "Appareil scolaire et division sociale de l’espace francilien", in Mattei M.F., Pumain D. (dir.), Données Urbaines No5, Paris, Anthropos.

GSEE, 2008, Poverty, education and social inequality, Athens, KANEP - GSEE (Center for Development of Education Policy - Greek General Confederation of Labour).

Gordon I., Monastiriotis V., 2007, "Education, Location, Education: A Spatial Analysis of English Secondary School Public Examination Results", Urban Studies, vol.44, No.7, 1203-1228.

Gordon I., 1996, "Family Structure , Educational Achievement and the Inner City", Urban Studies, vol.33, No.3, 407-423.

Gospodini A., 2009, "Post-industrial Trajectories of Mediterranean European Cities: The Case of Post-Olympics Athens", Urban Studies, vol.46, No.5-6, 1157–1186.

Gould-Ellen I., Austin-Turner M., 1997, "Does neighbourhood matter? Recent evidence", Housing Policy Debate, vol.8, No.4, 833-866.

Hadjiyanni A., Kapella A., Korre A., Sirigos S., Spyrellis S., Valassi D., 2012, "Urban segregation and educational inequality in Athens", Proceedings of the Second ISA forum of sociology, Buenos Aires.

Hamnett C., Ramsden M., Butler T., 2007, "Social Background, Ethnicity, School Composition and Educational Attainment in East London", Urban Studies, vol.44, No.7, 1255-1280.

Jencks C., Mayer S., 1990, "The social consequences of growing up in a poor neighbourhood: a review", in Mc Geary M., Lynn L. (eds.), Concentrated Urban Poverty in America, Washington D.C, National Academy.

Kassotakis M., 1996, Access to higher education in Greece, Athens, Grigoris.

Kathimerini - Greek newspaper. 12/12/2006. Article on the results of the NAE of 2006 (in Greek).

Lambiri-Dimaki I., 1974, For a Greek sociology of Education, Athens, E.K.K.E. (National centre for social research).

Leontidou L., 1989, Cities of silence: Working-class space in Athens and Piraeus, 1909-1940(in Greek), Athens, ETVA (Cultural Technological Foundation of the Hellenic Bank of Industrial Development) and Themelio.

Leontidou L., 1990, The Mediterranean city in transition: Social change and urban development, Cambridge, Cambridge University Press.

Maloutas T., 2004, "Segregation and Residential Mobility: Spatially Entrapped Social Mobility and Its Impact on Segregation in Athens", European Urban and Regional Studies, vol.11, No.3, 195–211.

Maloutas T., 2007a, "Middle class education strategies and residential segregation in Athens", Journal of Education Policy, vol.22, No.1, 49–68.

Maloutas T., 2007b, "Segregation, Social Polarization and Immigration in Athens during the 1990s: Theoretical Expectations and Contextual Difference", International Journal of Urban and Regional Research, vol.31, No.4, 733–758.

Maloutas T., Emmanuel D., Pantelidou M., 2006, Athens: Social structures, practices and perceptions(in Greek), Athens, E.K.K.E. (National centre for social research).

Marmaras E., 1991, The Urban Building in Interwar Athens. (in Greek), Athens, ETVA (Cultural Technological Foundation of the Hellenic Bank of Industrial Development).

Marpsat M., 1999, La modélisation des effets de quartier aux États-Unis : une revue des travaux récents, Populations, vol.54, No.2, 303–330.

Mumford L., 1954, The neighbourhood and the neighbourhood unit, Town Planning, vol.24, 256–270.

Myridis M., 2010, The structure of the Greek Higher Educational System: a mapping approach (in Greek), Proceedings of the eleventh National Cartographic Conference, Greece.

Myridis M., 1982, Transformations spatiales et mutations sociales en milieu urbain: Athènes de l'entre deux guerres, Ph.D. thesis, Denis Diderot University, Paris.

OECD, 2002, Education at a Glance 2002: Oecd Indicators, OECD Publishing.

OECD, 2011, Education at a Glance 2011: OECD indicators, OECD Publishing.

Oberti M., 2002, Choisir son quartier, choisir son école, in Penven, A., Bonny Y., Roncin C. (eds.), Au coeur de la cité, Rennes, Presses universitaires.

Oberti M., 2005, "Différenciation sociale et scolaire du territoire : inégalités et configurations locales", Sociétés contemporaines, vol.59-60, No.3, 13-42.

Oberti M., Préteceille E., Rivière C., 2012, Les effets de l’assouplissement de la carte scolaire dans la banlieue parisienne, Paris, Sciences Po-CNRS.

Passeron J.C., 1982, "L’inflation des diplômes. Remarques sur l’usage de quelques concepts analogiques en sociologie", Revue Française de Sociologie, vol.23, No.4, 551-584.

Prevelakis G., 2000, Athènes. Urbanisme, culture et politique, Paris, Harmattan.

Préteceille E., 2003, La division sociale de l’espace francilien. Typologie socioprofessionnelle 1999 et transformations de l'espace résidentiel 1990-99, Paris, Observatoire Sociologique du Changement.

Rhein C., Le Pape A., Grosbras P., 1996, Division sociale de l’espace et inégalités de scolarisation, in Rapport de recherche au Plan Construction Architecture, Paris, Ministère de l’Equipement.

Rose D., Harrison E., 2007, "The European socio-economic classification: a new social class schema for european research", European Societies, vol.9, No.3, 459-490.

Sayas J. 2004, "An exploration of the social and spatial division of labour in the Athenian urban space", The Greek Review of Social Research, vol.113, 167-206.

Sivignon M., Auriac F., Deslondres O., Maloutas T., 2003, Atlas de la Grèce, Paris, CNRS et La Documentation française.

Spyrellis S., 2010, "The shaping of residential segregation by historical, political and social events. The case of Greater Athens area", Proceedings of the 12th New Housing Researchers’ Colloquium, Istanbul.

Spyrellis S., 2011. "Social categorization and economic sectors : the case of metropolitan Athens", in Kalogirou S., (Ed.), Proceedings of the 17th European Colloquium on Quantitative and Theoretical Geography, Athens.

Spyrellis S., 2013a, "The role of economic sectors and occupational hierarchies in shaping housing segregation in metropolitan Athens (in Greek)", vol.21, Geographies.

Spyrellis S., 2013b, Division sociale de l’espace métropolitain d'Athènes, facteurs économiques et enjeux scolaires, Ph.D. thesis, Denis Diderot University, Paris.

To Vima - Greek newspaper. (8/11/2010). The best public high schools Greece (in Greek).

Tsoukalas K., 1976, Dépendance et reproduction: Le rôle social des appareils scolaires en Grèce (1830-1922). Ph.D. thesis, Sorbonne, Paris.

Tsoukalas K., 1986, State, society and employment in post-war Greece (in Greek), Athens, Themelio.

van Zanten A., 2001, L’école de la périphérie: scolarité et ségrégation en banlieue, Paris, Presses Universitaires.

Haut de page


1 The fact that the hierarchical classification method is considered as one the most solid methods of spatial analysis was the main reason for making this choice.

Haut de page

Table des illustrations

Titre Figure 1: The Athenian Metropolitan Zone (AMZ)
Fichier image/png, 251k
Titre Figure 2: Types of public high schools in metropolitan Athens resulted from a Hierarchical Clustering based upon the 6 rank scale of educational performances in the admission examinations in 2005.
Fichier image/png, 668k
Titre Figure 3. Synthetic mapping of the Athenian Metropolitan Zone after a multivariate analysis which took into account the socioeconomic and educational status of the active population and the performance of public high schools.
Fichier image/png, 635k
Haut de page

Pour citer cet article

Référence électronique

Stavros Spyrellis, « Social space and educational outcomes in Athens », Cybergeo : European Journal of Geography [En ligne], Espace, Société, Territoire, document 745, mis en ligne le 21 octobre 2015, consulté le 25 mai 2017. URL : ; DOI : 10.4000/cybergeo.27265

Haut de page


Stavros Spyrellis

Docteur en géographie associé à l’UMR 8504 Géographie-cités, Équipe PARIS
13 rue du Four
75 006 Paris

Haut de page

Droits d’auteur

© CNRS-UMR Géographie-cités 8504

Haut de page