Children of Turkish Immigrants in Germany and the Netherlands: The Impact of Differences in Vocational and Academic Tracking Systems
by Maurice Crul & Jens Schneider - 2009
Background/Context: Much research is being done on Turkish immigrants and their children in Germany and the Netherlands, but almost always from a national perspective. To compare the situation, for example, regarding educational outcomes across the two countries has proved to be very difficult because of different sets, selection criteria, and time periods for statistical data on immigrant populations. However, those data, which are actually available and comparable to at least some degree, already show how strongly the differences in educational attainment and labor market integration of Turkish immigrants depend on structural and systemic differences in the ways that education is organized in Germany and the Netherlands.
Purpose/Objective/Research Question/Focus of Study: The article analyzes available data on young Turkish immigrants and native-born second generations and their educational success in the two countries with the major Turkish populations in Western Europe. It aims to direct the focus away from group background characteristics, which are actually quite similar, to the influence of institutional arrangements and the way that the educational system facilitates (or not) the educational integration of Turkish youth.
Research Design: The article is based on publicly collected and available data on the Turkish populations in Germany and the Netherlands. This mainly refers to the Dutch SPVA surveys and the German micro-census and Integration Survey.
Conclusions/Recommendations: The findings show that more than group characteristics, systemic and institutional factors can have a decisive role in promoting or hampering the educational and labor market integration of young immigrants and the native-born second generation. The greater openness of the Dutch school system to provide “long routes” and “second chances” shows its effect in significantly higher shares of Turks in higher education. On the other side, the dual system of vocational training in Germany seems to be better suited for labor market integration, especially because apprenticeships are more practice oriented and do count as work experience for later application procedures. The Dutch system also offers better opportunities for girls than does the German system. Yet, the polarization effect between “high achievement” and “failure” of only partial integration success is greater in the Netherlands, whereas the overall advancement is slower, but also less polarizing, in Germany. In this sense, each country could learn something from its neighbor regarding those aspects of the institutional and systemic setting that apparently fail to do the job well enough.
This article compares integration processes in two countriesGermany and the Netherlandsand the role that national context plays in explaining differences in educational outcomes. The importance of the national context has received more attention in Europe than in the United States (Crul & Vermeulen, 2003; Doomernik, 1998; Eldering & Kloprogge, 1989; Fase, 1994; Heckmann, Lederer, & Worbs, 2001; Mahnig, 1998; Muus, 2003; Werner 1994). The American debate has been more restricted, with an emphasis on comparing different ethnic groups in the same city or nationally (see, as the most important studies, Kasinitz, Mollenkopf, & Waters, 2002; Portes, 1996; Portes & Rumbaut, 1996, 2001).1 There have been only a few studies in which the integration of children of immigrants in the United States is compared with children of immigrants in other countries (see, for example, Alba, 2003; Faist, 1995; Mollenkopf, 1999). The national context has mostly been taken for granted (Alba) and, as Reitz (2002) has argued, North American researchers have only recently started to give more attention to the importance of the national context in which immigrants and their children try to move forward.
Probably because of the proximity of European countries, research in Europe has been more cross-nationally oriented. European countries, despite the many connections and the increasing regulatory influence of the European Union, are structured quite differently, as is clear in the case of the two countries presented in this article: Germany and the Netherlands. Although the two nations are immediate neighbors and similar in many respects, the integration of each countrys Turkish communities has taken a very different form. This article aims to explain some of these differences.
THE TURKISH COMMUNITY IN GERMANY AND THE NETHERLANDS
Turkish labor migration has followed similar patterns in Germany and the Netherlands. Germany signed an official agreement on labor migration with Turkey in 1961, and the Netherlands did so in 1964. Spontaneous migration through relatives and covillagers then ensued, later even surpassing the scale of official immigration. The peak of the Turkish labor migration was between 1971 and 1973, years in which more than half a million Turkish workers came to Western Europe, 90% of them recruited by German industries (Özüekren & Van Kempen, 1997). From 1973 on, the economic recession following the oil crisis slowed the demand for labor and prompted the end of official immigration in 1974. Unemployment led many immigrants to return to their home countries, but many others stayed and began to send for their wives and children. Migration took a new upturn in the 1980s and 1990s, when the in-between generation reached marriage age and began choosing spouses from Turkey. The Turkish population in the Netherlands today totals about 300,000 people; in Germany, it is more than 2 million.
European industries particularly needed low-skilled labor at the time, and most first-generation Turkish guest workers were recruited from the lowest socioeconomic strata and had very little education. In the rural areas, where most of them grew up, educational opportunities were generally limited to primary school. In general, first-generation men had only finished primary school, and most women had just a few years of schooling. Because small-scale subsistence farming was the primary activity in these rural communities, education was not particularly important. Sending children to school brought no real advantage in the struggle for existence, and childrens help on the farm was more important. Another reason that education seemed to hold little promise was the nature of the schooling offered. Education in Turkey was not primarily geared toward conveying knowledge that would aid people in their peasant existence, or in breaking away from it. Its main aim was to transmit Turkish national ideology and to promote the cultural integration of the country.2
First-generation Turks in the European labor market therefore represented a vulnerable group whose members had very few alternatives to the factory jobs for which they were recruited. After 1973, economic crises and industrial restructuring put many Turkish immigrants out of work, although a sizable group of first-generation migrants did manage to start their own businesses or help their children to do so.
Comparing Turks in different receiving countries does not necessarily mean that the same population is literally being compared. It is important to take into account internal differences relating to ethnicity, levels of education, and religiosity within the Turkish immigrant populations. Most Turkish migrants came from small villages in central Turkey or along the Black Sea coast; those from larger cities (such as Istanbul, Izmir, and Ankara) are a minority. Some districts in central Turkey delivered enormous numbers of migrants over the years, often dispersed across various European countries. People from the Afyon district, for instance, now reside both in Germany and the Netherlands (cf. Crul, 1994).
The socioeconomic backgrounds of first-generation Turkish labor migrants are fairly similar in the two receiving countries, with some variation. Although labor migrants form the vast majority of the Turkish migrants in Europe, there are also significant groups of refugees who fled political persecution in Turkey or the armed conflict between Kurds and Turks. Most of them arrived in Europe later than the labor migrants.3
We have chosen to examine Turkish immigrants partly because of the sharp socioeconomic contrasts between them and the native populations in most Western European countries. Their socioeconomic status is extremely low. Unlike Turkish Americans, who are generally better educated (Karpat, 1995), most children of Turkish immigrants in Europewho were born in the receiving country or arrived before primary school agegrew up in rather unfavorable circumstances. Family income was often relatively low, and many families lived in substandard and cramped housing. Moreover, this is a group with a quite traditional Muslim background. Turkish immigrants are widely considered to be one of the toughest groups to integrate, so they constantly test the effectiveness of national policies aimed at the integration of newcomers.
THE SCHOOL CAREERS OF CHILDREN OF TURKISH IMMIGRANTS IN GERMANY AND THE NETHERLANDS
In this article, we compare the educational careers of children of Turkish immigrants in Germany and the Netherlands, drawing on all the data available. Ideally, we would use the same indicators of educational status in both countries, including school attendance rates, students educational performance, the distribution of the educational attainment, and the percentage of dropouts and students retained in grade. But unfortunately, not all the relevant data are equally available. Our analysis primarily draws on the micro-census data in Germany and the SPVA4 survey in the Netherlands, but in both countries, we sometimes made use of other surveys. These national data were, of course, collected independently from each other and, as a result, there are some differences. To begin with, the surveys were conducted at different times: The Dutch SPVA survey is from 2002, and the micro-census in Germany is from 2005. The sampling methods differed and will have had an effect on the outcomes. Moreover, parameters (e.g., age ranges) are different. Therefore, all comparisons need to be made with great caution, and interpretations are restricted to domains in which differences are obvious and substantial.
DIFFERENCES IN OUTCOMES
The school careers of children of Turkish immigrants exhibit remarkable differences across the two countries. This is especially clear in the participation of young people in vocationally oriented tracksthe lowest qualifying secondary school type in both countries and in dropout rates. In the Netherlands, half the children of Turkish immigrants follow a vocational track, whereas in Germany, three quarters do.5 At the top end of the educational ladder, in the Netherlands, 22% of the students of Turkish descent are in streams that give direct access to higher education (Jennissens & Hartgens, 2006). The group in higher education is in fact even larger because students can also enter higher education from middle vocational education. In 2005, 29% of Turks aged 1820 were in higher education in the Netherlands (Jennissens & Hartgens).
For Germany, the latest data from the 2005 micro-census show that only 14% of Turks aged 2535 finished a preparatory track that would provide direct access to a university (e.g., Gymnasium), and only 4% would have earned a university degree or the equivalent (Konsortium Bildungsberichterstattung, 2006). In the 1998 EFFNATIS survey,6 only 9% of the second-generation Turks aged 1925 attended a university or a Fachhochschule (i.e., a college for applied sciences). The percentage of successful students of Turkish descent in Germany is less than half that in the Netherlands.
A different picture emerges from analysis of how children actually perform in vocational or preparatory tracks. In the Netherlands, for instance, we found extremely high dropout rates. Of Turkish young people aged 1535, more than one fifth (22%) left school without any secondary school diploma (Herweijer, 2003). In Germany, the percentage was smaller: 18% of individuals aged 2535 (Konsortium Bildungsberichterstattung, 2006). The figures in the Netherlands would in fact be considerably higher if only 2535-year-old individuals were included; Herweijer stated that the difference between the 1525-year-old group and the 2535-year-old group is about 10%. This means that the 2535-year-old group had a dropout rate of around 27%.
The two most important indicators for school success, school performance and dropout rates, thus show contradictory outcomes in the two countries: a lower dropout rate in Germany, but more youngsters of Turkish descent in higher education in the Netherlands.
These differences between the two countries are so large that they can hardly be explained by differences in sampling methods or through the use of concepts and definitions alone. Possible explanations can be found in several domains, which we discuss next.
A dominant way to explain differences between countries has been to look at national models of integration. Usually, three models are distinguished: the model of differential exclusion, the assimilationist model, and the multicultural model (Castles & Miller, 1993). Because Germany has long emphasized avoiding heterogeneity, the country is often associated with the model of differential exclusion, whereas the Netherlands is generally identified with the multicultural model and the acceptance or even promotion of multiculturalism (Castles & Miller).7 Because national models of integration transmit national ideas, norms, and values shaping the interaction (in both ideological and legislative terms) with newcomers and their children, the assumption of this mode of explanation is that these different approaches will also have a substantial effect on the socioeconomic position of immigrants and their children (cf. Brubaker, 1992; Castles & Miller; Joppke, 1999; Koopmans, 2003).
This is certainly true, but the effect of national integration models varies for the different spheres of integration. The models clearly have an influence on naturalization rates and most probably also on the identity formation of the second generation.8 But based on the outcomes presented previously, the idea that a national model of integration also has a straightforward effect on the socioeconomic integration of children of immigrants has to be dismissed. Educational outcomes also seem to be affected by other factors (see also Alba, 2003; Crul & Vermeulen, 2003) that have yet to be identified.
On further inspection, differences in performance can be traced to differences in national institutional arrangements in education and in the varying ways that the transition to the labor market is formalized in the two countries.9 In the case of Germany and the Netherlands, the two national educational systems differ, especially in the duration of schooling, face-to-face contact hours between teachers and students, student selection mechanisms, and types of schooling (for example, apprenticeship tracks).
The first significant disparity is in the age at which education begins. In the Netherlands, the majority of Turkish children start school at age 4; in Germany, children start at age 6. Thus, immigrant children in the Netherlands have about 2 more years of formal education in a crucial phase of their development and socialization, especially with regard to learning the majority language. Another striking difference lies in the number of face-to-face contact hours with teachers during the years of compulsory schooling. Again, this number is significantly lower in Germany, especially in the important first years. Nine-year-olds in German schools have an average total of 661 contact hours, as compared with 1,019 hours in the Netherlands, because children in Germany attend school only on a half-day basis.10 Turkish children in Germany thus receive about 10 hours less tuition per week than in the Netherlands.
A third important difference lies in selection mechanisms for secondary school. Germany generally makes schooling decisions about students when they reach age 10,11 and the selection mechanism channels children into three rather strictly separate school levels (Hauptschule, Realschule, and Gymnasium).12 By contrast, selection in the Netherlands occurs 24 years later. Coupled with the later start and the lower average contact hours, Turkish students in Germany thus have comparatively little time to pull themselves out of their disadvantaged starting position. Moreover, because of the early selection, more students end up in lower qualifying streams (especially Hauptschule, which is the lowest track of secondary education).13 The older selection age in the Netherlands results in higher percentages of Turkish children moving into more prestigious streams.
Considering these factors and numbers, how can we explain the high dropout rates in the Netherlands compared with Germany? In the Netherlands, a considerable number of second-generation Turkish children move into a vocational track (Vbo) at age 12. Their situation resembles that of students of Turkish descent in Germany who move into Hauptschule at age 10. Dropout rates in the Netherlands start to rise at age 15, but at this age, most Turkish youngsters in Germany already possess their diploma, the Hauptschulabschluss. At age 16, children in the Netherlands are still at school full time, whereas in Germany, many have already entered the labor market as an apprentice.
Even in the lower vocational track in the Netherlands, the occupationally related aspect of the apprenticeship is limited: It is called a stage and is more like an internship. Half the subjects that the children study are theoretical subjects, with only the other half devoted to the occupation for which they are being trained. Many students in the vocational track in the Netherlands develop an oppositional stance toward education because of the long days they have to spend in schoolespecially in combination with the difficulty many of them have with theoretical subjects.
But other factors are involved. First of all, the vocational educational stream in the Netherlands is considered a marginal stream within the educational system. Lower vocational education (Vbo) has often been described as the garbage bin of the education system (Crul, 2000). It has to accommodate children with learning problems and all those children who were unsuccessful in the higher streams, often because of behavioral problems. Pels (2001), who has depicted teacherstudent interaction in a Vbo school, counted about 80 admonitions during one math lesson. Crul (2000) has also reported on the prisonlike climate and the regular fights that break out in Vbo schools, sometimes even between students and teachers. The resulting school climate is hardly conducive to good school performances, and consequently, dropout rates in Vbo are very high.
The German equivalent to the Dutch Vbo might be the special schools (Sonderschulen). Unlike the Dutch system, Hauptschule in Germany is part of the mainstream educational system. Especially in rural areas and smaller cities, but also in the southern federal states in general, many children of native-born parents, mainly from working-class or farming families, also follow this track. The educational climate in Hauptschule is not considered as problematic as in Vbo, although in bigger cities especially, the educational level is quite low. As in the Netherlands, the German Hauptschule is also the main school for newly arrived immigrant children.
In Germany, vocational training begins only after the completion of a secondary school diploma, which is in most cases the minimum requirement for an apprenticeship. The so-called dual track includes working as an apprentice in a firm 34 days a week. The apprentice spends the other time in so-called vocational schools (Berufsschulen) where the subjects are closely related to the vocational specialty of the apprentice.
This combination of factors may explain why dropout rates among second-generation Turkish children are so much lower in Germany than in the Netherlandswith far-reaching consequences for other social domains. In the Netherlands, Turkish youngsters who drop out are seriously at risk of becoming an underclass. Unemployment among this group is extremely high, and twice as high as in Germany (Crul & Doomernik, 2003; Worbs, 2003).
In contrast with the Dutch system, the apprenticeship system in Germany seems to facilitate the transition to the labor market for children of Turkish immigrants. Some youngsters of Turkish descent in Germany continue to work for the company at which they started as an apprentice (Böcker & Thränhardt, 2003). Even if they do not, they have 23 years of working experiencea clear asset from the perspective of potential employers. In the Netherlands, youngsters of Turkish descent have to find a place in the labor market on their own and without a comparable apprenticeship that offers skills or, in the ideal case, an education in professional behavior and workplace ethics.
This lack of training also seems to make them more vulnerable to discrimination. There are two possible explanations for the evidence that discrimination plays a more important role in the Netherlands (cf. Crul & Doomernik, 2003). One is the different starting position: In Germany, youngsters of Turkish descent possess a diploma and their employment record as an apprentice, whereas many of the youngsters in the Netherlands have neither a professional diploma nor any practical work experience. The decision to employ someone in Germany is based on school grades, but of course, individual employment records and work certificates play an important role as well. Employment decisions in the Netherlands can only be based on school qualifications, and research in the Netherlands shows that in the choice between immigrant and native youth with the same grades, immigrant youngsters are not given an equal chance (Crul & Doomernik, 2003). The second important difference between the two countries is that youth unemployment in the Netherlands is generally much higher than in Germany. Research shows that discrimination tends to be more widespread when there is tough competition in the labor market; when there is a labor shortage, employers cannot afford to discriminate against qualified potential employees.
The fact that the apprenticeship system in Germany gives youngsters of Turkish descent better opportunities than in the Netherlands does not necessarily mean that it is a perfect tool for integration. There is mounting evidence that Turkish youngsters in particular profit less from the apprenticeship system than their native peers. It is more difficult for them to gain an apprenticeship with good prospects for future work, and they more often drop out of the dual track (Worbs, 2003; von Below, 2003). But, relative to the Netherlands, where there is no such formal apprenticeship system, youngsters of Turkish descent in Germany are in a better position (Crul & Vermeulen, 2003).
The starting position of the first-generation Turks in the two countries was quite similar in many aspects. They mostly came from the countryside and had very low levels of education. But a comparison reveals that the educational experience of the children of Turkish immigrants in the two countries is very different. This is a clear indicator of the importance of institutional arrangements in shaping educational outcomes and the transition to the labor market. Differences in the way that education is structured in the two countries are having a significant effect on how immigrant communities are developing and how they will evolve in the future. In both countries, we can observe an increasing divergence between a well-educated emerging elite and a growing nonqualified underclass. But the degree of this polarization could be quite different in Germany as compared with the Netherlands.
THE GENDER DIVIDE
In addition to looking at the Turkish group as a whole, it is also worthwhile to consider differences in the positions of men and women. Again, the two countries show some remarkable disparities. In this section, we first draw a comparison between Turkish men and women in Germany and the Netherlands. But, because this description of the state of affairs at one moment in time gives a rather static picture, we supplement it with an examination of changes in the situation of young Turkish women in the two countries over time.
THE CURRENT SITUATION
The publication of the micro-census data of 2005 does not give separate information for males and females. The best available data for Germany come from a survey in 2000 done by the Bundesinstitut für Bevölkerungsforschung (BiB) among individuals aged 1830 (von Below, 2003). The BiB survey shows that males were doing slightly better than females of Turkish descent, and naturalized males in particular were doing better than the naturalized females. Of the naturalized males, 31% had a Gymnasium or Fachhochschule diploma, compared with 24% of the females. The percentages for males and females with Turkish nationality were identical, at around 22% (von Below). The BiB figures also show an overrepresentation of females in the category ohne Abschluss (i.e., without any diploma).
In the Netherlands, on the other hand, the latest figures show that girls are overrepresented at the more prestigious levels of the educational system. In 2002, almost two thirds (63%) of second-generation Turkish females aged 1524 were studying at the middle level (middle vocational education) or higher, whereas only 55% of second-generation Turkish males had reached these levels (Gijsberts, 2004).
In the Netherlands, as mentioned earlier, we can see a stronger polarization within the Turkish group. About a quarter of students leave school without any secondary school diploma, but an equally large number are studying in higher education. In Germany, there is more concentration in the lower and middle end of the educational ladder. In both countries, a considerable number of females remain at this level and often do not participate in the labor market after they complete their studies.
The somewhat better overall picture for girls in education changes when we look at labor market participation in both countries. The percentage of Turkish women in Germany aged 2026 who are neither working nor studying is 37%, compared with only 6% of the men (Konsortium Bildungsberichterstattung, 2006; cf. Worbs, 2003).14 This means that about half of those who are not studying are also not working. In the Netherlands, this picture is quite similar: In the 2539 age group, only 41% of Turkish women are working (Hooghiemstra & Merens, 1999). Those women who are inactive generally have no diploma at all or only a lower vocational education diploma, and the majority are homemakers (Tesser, Merens, & Van Praag, 1999). Inactive women in Germany are mostly those who have not finished Hauptschule or those with a Hauptschul-diploma who did not find an apprenticeship position or did not finish it.
The literature contains overwhelming evidence that Turkish parental attitudes toward girls schooling and careers have frequently been ambivalent (Coenen, 2001; Crul, 2000; Lindo, 2000). Coenen has termed such attitudes cultural carryovers from the parents who grew up in rural areas where education for girls was not given much value. Indeed, many Turkish parents have been following the rather traditional definition of the norms and values of their ethnic community of origin, especially in domains such as religious practice, marriage, and gender roles.15 Tightly knit social networks play an essential role in the Turkish community. Strong social control prevails, and the behavior of girls and young women is closely monitored. Both traditional gender role expectations and the prevalence of early marriage pose formidable barriers to the education of Turkish girls. Many leave school early, and most of them become full-time housewives. Read (2004) used the term patriarchal connectivity to describe this phenomenon, whereby women and men are socialized to see themselves primarily as part of a larger kinship structure that privileges male authority and dominance over female educational and professional achievements.
In the Turkish community, this pattern is often related to strong social cohesion, and gossip circulating within tight networks has been portrayed as a strong weapon in keeping members of the community in line (de Vries, 1990). Parents fear gossip in the community about misbehavior, especially by their unmarried daughters, and an early marriage is seen as a way to ward off shame. Another consideration is that parents can reap better short-term payoffs from an early marriage, especially with a family member from Turkey, than from the extension of their daughters educational career. A marriage can bring benefits in terms of both the extended family income (the income of the young couple adds to the family household) and higher status in the community back home among close relatives who are able to send their daughter or son to the Netherlands through the marriage.16
In 2001 in the Netherlands, 15% of the Turkish second-generation women aged 19, and one quarter of those aged 21, were already married, and their first child was usually born soon after (Alders, Harmsen, & Hooghiemstra, 2001).17 Marriage at a young age usually means the end of any educational career, even when a woman is qualified to enter a higher level of education (Crul, 2000). Turkish men also tend to marry at a young age, although they are, on average, older than the women. After marriage, men also often have to leave school or college to provide income for their new family (Alders et al.). In particular, youths with school difficulties or behavior problems may be pushed by their parents to get married and find a job instead of continuing in school.
Cultural factors (for example, traditional gender roles), together with some characteristics of the educational system, seem to drive many Turkish girls out of school and higher education. In the Netherlands, Turkish girls can be found in high numbers in lower vocational training, especially in care work and garment making. A significant number of Turkish parents seem to consider these streams more as preparation for marriage than as training for a career. It is not uncommon for girls to receive a marriage proposal (often when on holiday in Turkey) at age 15 or 16, so some are already engaged when preparing for their exams. Under these circumstances, the incentive to stay in school and finish their studies diminishes. And, more important, parents often do not seem to care if their daughters leave school.
The strong traditional gender divide in the Turkish community seems to block some Turkish women (in comparison with men) from profiting from the relatively open educational system in the Netherlands. The same cultural factors that block social mobility for women in the Netherlands also play a role in Germany, but in a different way and with different results. They more often play a role after Hauptschule, when most girls of Turkish descent in Germany move into dual vocational training. This is when the careers of females and males really start to diverge. Many young Turkish women seem to be pushed out of the dual track because they cannot find an appropriate apprenticeship position or because they do not finish their apprenticeship (von Below, 2003) because of early marriage.
When working, Turkish women in Germany mostly move into traditionally female professions like shop assistant, hairdresser, nurse, or child care worker, often working with a mainly or exclusively female (Turkish) clientele. Most women work full time because part-time jobs are still not very common in Germany. By contrast, in the Netherlands, the majority of the jobs occupied by Turkish women are part time (Hooghiemstra & Merens, 1999). Although most Turkish women in Germany are employed as low- or unskilled workers, they frequently earn more than their peers in the Netherlands. The situation in Germany more often works to promote the financial independence of Turkish women, even though this generally means only a small step up on the social ladder. Young Turkish woman in Germany are less often found in the higher level professions, where they are still pioneers. Compared with Germany, Turkish females in the Netherlands more often pursue higher education. When they enter the labor market, they occupy high-level positions. The symbolic importance of this for a younger generation of Turkish women should not be underestimated. These professional women will set a new example in their communities.
THE LONGITUDINAL PERSPECTIVE
Comparing the in-between generation of the 1980swith younger cohorts of the second generation in the new millennium gives us some clues about educational trends and tendencies among young Turkish women. In both countries, young women of Turkish descent are performing much better than the in-between generation. More are studying and/or have a professional career, and correspondingly fewer women are managing a household as their main activity.
If we analyze the two most important indicators that we have used to define school success, we see that there is a different pace of change in the two countries. Dropout rates declined dramatically in the Netherlands, and Turkish girls in that country are closing the gap with Germany on this important indicator. In 1988, more than three quarters of Turkish girls in the Netherlands left school without a secondary education diploma. Fourteen years later, this figure dropped to around 27%. The dropout rates of women were originally much higher than for men, but in the 2002 survey, the percentages are equal. In the same period, the percentage of young Turkish girls in preparatory tracks for higher education more than doubled. We first saw that females were closing the gap with the males and now we see that their school results are even better.
In Germany, it is difficult to make valid comparisons over time. The micro-census of 1995 cannot be compared with the BiB survey of 2000 because of differences in sampling and in the age ranges of the respondents. It seems, however, from the BiB survey, that differences remain between males and females in Germany that in the Netherlands have been reversed, with the girls having overtaken the boys in terms of positive school outcomes. At the same time, the BiB survey shows that the percentage of Turkish women with a diploma from more prestigious levels of schooling (Abitur or Fachhochschulreife) is rising. About 1 in 5 Turkish women aged 1830 now have a diploma on this levela clear improvement compared with the micro-census of 1995.
But the pace of change is slower in Germany than in the Netherlands. Although more and more Turkish girls in Germany enter into Gymnasium or similar schools, the number of females in these more prestigious forms of education is still small. And the dropout rate among females in Germany is still higher than among males. In the Netherlands, Turkish girls have caught up with, and overtaken, men. More and more Turkish girls are pursuing higher education in the Netherlands, and the pioneering role of the first female Turkish university students is beginning to affect the younger generation. Younger sisters in the family, it seems, are more likely to be allowed to extend their studies (Coenen, 2001; Crul, 2000). This trend may be interpreted as a combination of cultural and generational changes. A girl in higher education is no longer an exception in the Turkish community in the Netherlands, and this makes it easier for parents to consent to their daughters extending their studies. All in all, it seems that girls of Turkish descent in the Netherlands have made more progress in terms of entrance into higher education and lowering their dropout rate than Turkish girls in Germany.
The changing position of Turkish women in the Netherlands shows that differences between countries are not static. The sometimes surprising dynamics in the development of Turkish educational achievement in Germany and the Netherlands are best explained by looking at both the interaction between the actors (e.g., young Turkish women and their parents and siblings) and the institutional arrangements in the respective countries.
The comparisons we have made in this article are not comprehensive, and the data needed for more adequate and in-depth comparisons are missing. Nevertheless, some findings are clear. The position of the children of Turkish immigrants varies widely between the two countries. The main explanation cannot lie in the ethnic community itself, nor in the educational level or social background of the parents, because they are quite similar in the two countries. Although more analytical work still has to be done, and more rigorously comparable data sets are urgently needed, we believe that the information presented here shows that national institutional arrangements for education and labor market transition have a considerable impact on the paths of integration in the two countriesa topic that seems to have been a persistent blind spot in the debate on integration. An interesting finding was that the influential parts of these institutional arrangements seem to be less those policies specifically targeted at migrant youth than the general arrangements prevailing in each country. Institutional arrangements affect ethnic minority groups differently from their native peers, and they also shape the trajectories of men and women in different ways. It is clear that in Germany and the Netherlands, institutional factors have interacted with cultural changes in Turkish communities in those countries. This interaction is especially at play in the Netherlands, where the more open educational system seems to offer better opportunities for young Turkish women.
1. Portes and De Wind (2004) have argued that most scholars disciplinary training is focused on examining migration within a single national context.
2. For a discussion of the role of education in rural Turkey at the time of mass emigration, see Coenen (2001).
3. Rarely mentioned when studying massive labor migration to Western Europe is the fact that, with the exception of Italy, all Mediterranean migrants came from countries that were ruled by dictatorships at that time. This certainly had an influence on the remigration pattern of these groups, especially when comparing Spain or Portugal (redemocratization in the mid-1970s) with Turkey, which experienced a military coup in 1980.
4. The SPVA, spelled out, translates to survey on the social position and use of public services by immigrants and their children.
5. This number combines Haupt- and Realschule, and those students in integrated comprehensive schools (Integrierte Gesamtschulen) following the nonacademic tracks (Konsortium Bildungsberichterstattung, 2006).
6. The EFFNATIS field survey compared children of immigrants in Germany, France, and Britain (Heckmann et al., 2001).
7. See, by way of comparison, Vermeulen (2004) for a detailed discussion on these models and modes of integration.
8. The EFFNATIS field survey showed that children of immigrants in Great Britain and France identified more with the immigration country than did those in Germany. The same tendency was found in the preservation of the mother tongue by the second generation. Children of immigrants in Germany held on to their mother tongue longer than those in France and Britain. However, it is important to note that different groups were chosen in the three countries (Heckmann et al., 2001)
9. Reitz (1998), in his study The Warmth of Welcome: The Social Causes of Economic Success for Immigrants in Different Nations and Cities, was one of the first to point to the importance of differences in educational institutional arrangements regarding the integration of immigrants. Waldinger (2001), in Strangers at the Gates: New Immigrants in Urban America, showed how differences in labor market structures among cities can explain different path of integration of immigrant groups.
10. This has been identified as one of the main reasons for the low German scores in the PISA study. Over the last 2 years, a number of initiatives have been taken, including a nationwide federal €4 billion program for the installation of some 5,000 all-day schools throughout the country (see http://www.bmbf.de/en/1125.php). Of course, it will take some years before this process will show an effect on the educational statistics.
11. The federal states of Berlin and Brandenburg are the only exceptions; there the transition from primary to secondary education occurs at age 12.
12. Exceptions from this are the so-called Gesamtschulen and the integrierte Haupt- und Realschulen in some federal states, which combine different types in one single school. But even here, the transition from one level to the other is not made very easily.
13. The international IGLU (Progress in International Reading Literacy Study) study on elementary schools showed that there is also a significant bias in teachers recommendations for secondary school types. Students with a Turkish background are more likely to receive recommendations for lower qualifying schools, even at the same level of performance as their peers without a migrant background (Bos et al., 2003). A similar pattern applies to Lernbehindertenschulen, schools for students with learning disabilities: Originally designed for handicapped children, they seem to be increasingly used as a reservoir for children whose only disability is a migrant background. As Kornmann and Kornmann (2003) have shown, non-German students are twice as likely to be sent to a Lernbehindertenschule than their German peers.
14. This a particular feature of the Turkish group. Among the entire population with a migrant background, it applies to only 23% of women and 7% of men. Among nonmigrants in the same age group, the numbers go down to 9% of women and 4% of men. These numbers do not distinguish between first- and second-generation individuals. They include, for example, those young women who migrated to Germany for the purpose of marriage (see Note 16). The overall situation of the second generation is significantly better as compared with the first generation. The total percentage of first-generation individuals between 20 and 26 years of age not working and not studying is 18.6%, whereas this is the case only for 8.5% of the second generation (Konsortium Bildungsberichterstattung, 2006).
15. One of the few studies looking at the same mechanisms in the American context is Reads (2004) work on Arab American women. She described how religiosity, gender traditionalism, and homogamy have important consequences for the labor force decisions of the Arab American women. An interesting finding was that the negative effect of religiosity on female labor participation is equally strong for Muslim Arab and Christian Arab women. She concluded, therefore, that being a Muslim is not necessarily synonymous with adherence to traditional gender roles.
16. See Kelek (2005) and the reactions to her book over the controversy about imported brides and grooms from Turkey, and the close links to the strongly gender-related idea of the preservation of Turkishness in the context of emigration to Germany. At the same time, Boos-Nünning and Karakaşoğlu (2005) have underlined that studies about partner choice patterns in Turkish families in Germany frequently reach contradictory results. They concluded that most social scientific research instruments apparently do not provide adequate access to the real lives of girls and young women with migrant backgrounds. Results of this type of study thus have to be viewed with caution.
17. Second-generation young adults make very different decisions about having children than do the in-between generation. Half of the Turkish women from the in-between generation already had a child by age 21 (Alders, 2001; cf. Boos-Nünning & Karakaşoğlu, 2005).
Alba, R. (2003, June) Blurred vs. bright boundaries: Second generation assimilation and exclusion in comparative perspective. Paper presented at the Bellagio Conference on the Second Generation, Bellagio, Italy.
Alders, M. (2001). Allochtone moeders [Migrant mothers]. In J. Garssen, J. de Beer, P. Cuyvers, & A. de Jong (Eds.), Samenleven. Nieuwe feiten over relaties en gezinnen (pp. 107-121). Voorburg: Statistics Netherlands.
Alders, M., Harmsen, C., & Hooghiemstra, E. (2001). Relatievorming van allochtonen [Entering of relations by migrants]. In J. Garssen, J. de Beer, P. Cuyvers, & A. de Jong (Eds.), Samenleven. Nieuwe feiten over relaties en gezinnen (pp. 4353). Voorburg: Statistics Netherlands.
von Below, S. (2003). Schulische Bildung, berufliche Ausbildung und Erwerbstätigkeit junger Migranten. Ergebnissen des Integrationssurveys des BiB [School education, vocational training, and employment of young migrants: Results from the Integration Survey of the Federal Institute for Education Research]. Wiesbaden, Germany: BiB. (Heft 105b).
Böcker, A., & Thränhardt, D. (2003). Is het Duitse integratiebeleid succesvoller, en zo ja, waarom? [Is the German integration policy more successful, and if so, why?]. Migrantenstudies, 19(1), 33–44.
Boos-Nünning, U., & Karakaşoğlu, Y. (2005). Viele Welten leben. Zur Lebenssituation von Mädchen und jungen Frauen mit Migrationshintergrund [Living in many worlds: On the life situation of girls and young women with migration backgrounds]. Münster, Germany: Waxmann.
Bos, W., Lankes, E. M., Prenzel, M., Schwippert, K., Walther, G., & Valtin, R. (Eds.). (2003). Erste Ergebnisse aus IGLU. Schülerleistungen am Ende der vierten Jahrgangsstufe im internationalen Vergleich [First results from IGLU: Pupils' attainment at the end of fourth grade in international comparison]. Münster, Germany: Waxmann.
Brubaker, W. R.(1992). Citizenship and nationhood in France and Germany. Cambridge, MA: Harvard University Press.
Castles, S., & Miller, M. J. (1993). The Age of migration: International population movements in the modern world. London: Macmillan.
Coenen, L. (2001). Word niet zoals wij: de veranderende betekenis van onderwijs bij Turkse gezinnen in Nederland [Do not become like us: The changing meaning of education in Turkish families in the Netherlands]. Amsterdam: Het Spinhuis.
Crul, M. (1994) Springen over je eigen schaduw. De onderwijsprestaties van Marokkanen en Turken van de tweede generatie [Jumping over your own shadow: The educational performances of second-generation Moroccans and Turks]. Migrantenstudies 10(3), 168186.
Crul, M. (2000). De sleutel tot succes. Over hulp, keuzes en kansen in de schoolloopbanen van Turkse en Marokkaanse jongeren van de tweede generatie [The key to success: On support, choices, and chances in the school careers of Turkish and Moroccan youth of the second generation]. Amsterdam: Het Spinhuis.
Crul, M., & Doomernik, J. (2003). The Turkish and the Moroccan second generation in the Netherlands: Divergent trends between and polarization within the two groups. International Migration Review, 37, 10391065.
Crul, M., & Vermeulen, H. (2003). The second generation in Europe: Introduction. International Migration Review, 37, 965986.
Doomernik, J. (1998) The effectiveness of integration policies towards immigrants and their descendants in France, Germany and the Netherlands. Geneva, Switzerland: International Labour Organisation.
Eldering, L., & Kloprogge, J. (1989). Different cultures same school: Ethnic minority children in Europe. Amsterdam: Swets & Zeitlinger.
Faist, T. (1995). Social citizenship for whom? Young Turks in Germany and Mexican-Americans in the United States. Aldershot, England: Avebury.
Fase, W. (1994) Ethnic divisions in Western European education. Münster, Germany: Waxmann.
Gijsberts, M. (2004). De onderwijspositie van vrouwen uit etnische minderheden [The educational position of women from ethnic minorities]. In M. Gijsberts & A. Merens (Eds.), Emancipatie in estafette. De positie van vrouwen uit etnische minderheden (pp. 3158). Rijswijk, Netherlands: Social and Cultural Planning Office.
Heckmann, F., Lederer, H. W., & Worbs, S. (in cooperation with the EFFNATIS research team). (2001). Effectiveness of national integration strategies towards second generation migrant youth in a comparative European perspective: Final report to the European Commission, Bamberg, Germany.
Herweijer, L. (2003) Voortgezet onderwijs, beroepsonderwijs en hoger onderwijs [Secondary education, vocational training, and tertiary education]. In J. Dagevos, M.Gijsberts, & C.v. Praag (Eds.), Rapportage Minderheden. Onderwijs, arbeid en sociaal-culturele integratie (pp. 111142). Rijswijk, Netherlands: Social and Cultural Planning Office.
Hooghiemstra, B., & Merens, J. (1999). Variatie in particpatie. Achtergronden van arbeidsdeelname van allochtone en autochtone vrouwen. Rijswijk, Netherlands: Social and Cultural Planning Office.
Jennissens, R., & Hartgens, M. (2006). Allochtonen in het onderwijs [Migrants in education]. In Integratiekaart 2006 (pp. 1124). The Hague: Netherlands: WODC.
Joppke, C. (1999) Immigration and the nation-state: The United States, Germany and Great Britain. Oxford, England: Oxford University Press.
Karpat, K. (1995) The Turks in America. In S. de Tapia (Ed.), Turcs dEurope...et dailleurs (pp. 612638). Paris: ERISM.
Kasinitz , P., Mollenkopf, J., & Waters, M. (2002). Becoming American: Becoming New Yorkers: Immigrant incorporation in a majority minority city. International Migration Review, 36, 10201036.
Kelek, N. (2005). Die fremde Braut. Bericht aus dem Inneren des türkischen Lebens in Deutschland [The strange bride: Report from the interior of Turkish life in Germany]. Köln, Germany: Kiepenheuer & Witsch.
Konsortium Bildungsberichterstattung. (2006). Bildung in Deutschland. Ein indikatorengestützter Bericht mit einer Analyse zu Bildung und Migration [Education in Germany: An indicator-based report with an analysis of education and migration]. Bielefeld, Germany: W. Bertelsmann Verlag.
Koopmans, R. (2003). Zachte heelmeesters& Een vergelijking van de resultaten van het Nederlandse en Duitse integratiebeleid en wat de WRR daaruit niet concludeert [Desperate diseases need. . . : A comparison of the results of Dutch and German integration policies and what the WRR does not conclude from this]. Migrantenstudies, 37(1), 8792.
Kornmann, A., & Kornmann, R. (2003). Erneuter Anstieg der Überrepräsentation ausländischer Kinder in Schulen für Lernbehinderte [Again an increase in the overrepresentation of foreign children in schools for learning disabilities]. Zeitschrift für Heilpädagogik, 7, 286289.
Lindo, F. (2000). The silent success: The social advancement of Southern European labour migrants in the Netherlands. In H. Vermeulen & R. Penninx (Eds.), Immigrant integration: The Dutch case (pp. 132152). Amsterdam: Het Spinhuis.
Mahnig, H. (1998). Integrationspolitik in Großebritannien, Frankreich, Deutschland und den Niederlanden. Eine vergleichende Analyse [Integration policies in Great Britain, France, Germany, and the Netherlands: A comparative analysis]. Neuchâtel, Switzerland: Swiss Forum for Migration and Population Studies.
Mollenkopf, J. (1999). Assimilating immigrants in Amsterdam: A perspective from New York. Netherlands Journal of Social Science, 36, 126145.
Muus, P. (2003) An international comparison of migration and Immigrant policy with respect to immigrants from Turkey and their participation in the labour market. In L. Haagendoorn, J. Veenman, & W. Vollenbergh (Eds.), Integrating immigrants in the Netherlands: Cultural versus Socio-economic Integration (pp. 1740). Aldershot, England: Ashgate.
Özüekren, A. S., & van Kempen, R. (1997). Turks in European cities: Housing and urban segregation. Utrecht, Netherlands: ERCOMER.
Pels, T. (2001, November). Student disengagement and pedagogical climate. Paper presented at the sixth International Metropolis Conference, Rotterdam, Netherlands.
Portes, A. (1996). The new second generation. New York: Russell Sage Foundation.
Portes, A., & Rumbaut, R. (1996). Immigrant America: A portrait. Berkeley: University of California Press.
Portes, A., & Rumbaut, R. (2001). Legacies. The story of the immigrant second generation. Berkeley: University of California Press.
Portes, A., & De Wind, J. (2004). A cross-Atlantic dialogue: The progress of research and theory in the study of international migration. International Migration Review, 38, 828851.
Read, J. G. (2004). Cultural influences on immigrant womens labor force participation: The Arab-American case. International Migration Review, 38, 5277.
Reitz, J. (1998). The warmth of welcome: The social causes of economic success for immigrants in different nations and cities. Boulder, CO: Westview Press.
Reitz, J. (2002). Host societies and the reception of immigrants: Institutions, markets and policies. New York: Center of Migration Studies of New York.
Tesser, P., Merens, J., & van Praag, C. (1999). Rapportage minderheden 1999 [Report on minorities 1999]. Rijswijk, Netherlands: Social and Cultural Planning Office.
Vermeulen, H. (2004). Models and modes of immigrant integration . . . and where does Southern Europe fit? In C. Inglessi, A. Lyberaki, & H. Vermeulen (Eds.), Immigration and integration in Northern versus Southern Europe (pp. 2739). Athens, Greece: Netherlands Institute at Athens.
de Vries, M. (1990). Roddel nader beschouwd [A closer view on gossip]. Leiden, Netherlands: COMT.
Waldinger, R. (2001). Strangers at the gates: New immigrants in urban America. Berkeley: University of California Press.
Werner, H. (1994). Integration of foreign workers into the labour market: France, Germany, the Netherlands and Sweden. Geneva, Switzerland: International Labour Organisation.
Worbs, S. (2003). The second generation in Germany: Between school and labor market. International Migration Review, 37, 10111038.