Keywords

1 Introduction

Many people mobilize their social networks during the job search and application process, because social contacts can provide useful information and support on the labour market (Kramarz & Skans, 2014). Whereas several studies have investigated differences in network mobilization between natives and migrants, few studies have examined these differences at the transition from school to work. In particular, there is a lack of empirical research based on data that provide detailed prospective and retrospective information on the mobilization of several kinds of social contacts for various purposes.

Against this background, we provide an in-depth analysis of differences between students with and without a migration background in Germany when it comes to their mobilization of social capital during the transition to vocational education and training (VET) after lower secondary education. In Germany, the school-to-work transition for most youths without a general university entrance qualification (Abitur) is characterized by entry into the strongly company-based VET system. At the end of general schooling, adolescents usually do not yet have sufficient labour market experience and information, and are thus likely to depend heavily on support from their social contacts when searching for an apprenticeship (Roth, 2014a, 2018). Among these ties, parents play a key role in providing support and information during their children’s vocational orientation and apprenticeship search (e.g. Beicht & Granato, 2010; Hoenig, 2019; Roth, 2014a). However, parents with a migration background are likely to lack knowledge about the vocational training system and about open VET positions (e.g. Kretschmer, 2019; Roth, 2014a). Hence, they might not be able to provide help in the same way as native parents. Due to their important role, we will pay special attention to ethnic differences in the mobilization of information and support from parents.

Using data from the first five waves of Starting Cohort 4 (Grade 9) of the National Educational Panel Study (NEPS), we can overcome several shortcomings of previous research and add to the existing literature in several ways: in contrast to most other data used in previous studies, NEPS data provide nationwide, longitudinal information on a large sample of 9th graders attending regular schools. Students were subsequently interviewed annually or bi-annually, covering the transition period from lower secondary to upper secondary general education or vocational training. Hence, NEPS data allow us to analyse not only retrospective information from adolescents after having left general schooling but also (hypothetical) prospective information from students at the end of general lower secondary education. This is an important improvement, because the problem with retrospective information is that the actual mobilization of social contacts depends not only on social capital resources but also on the need for support—for instance, due to unsuccessful experiences in searching formally for an apprenticeship. Moreover, NEPS data provide not only longitudinal but also uniquely detailed social capital information, and this allows us to investigate commonalities and differences in network mobilization between adolescents with and without a migration background in more detail depending on the type of social contact and the kind of support considered. Finally, whereas previous results on network mobilization have been mainly descriptive, we additionally run multivariate analyses in order to investigate whether differences in the mobilization of parents between adolescents with and without a migration background are in fact caused by ethnicity or rather by socio-economic background. In our multivariate analyses, we also divide adolescents with a migration background into those who were foreign born themselves or whose parents were both foreign born and those with at least one German-born parent.

This chapter is structured as follows: we first give a brief overview of the German education and vocational training system followed by a discussion of the theoretical considerations and previous research. After describing the data and analytical strategy, we present our empirical results. The chapter concludes by summarizing and discussing our findings.

2 The German Education and Vocational Training System

In the highly stratified German education system (Müller, 2005), students who attend the most demanding track (Gymnasium) in lower secondary education usually continue in the same school to general upper secondary education, whereas those attending the less demanding tracks mostly start VET after lower secondary education. Those who neither continue with general education nor start VET participate in pre-vocational measures in order to increase their chances of finding a training position later on. Most vocational training programmes take place in the form of dual training (Bundesministerium für Bildung und Forschung, 2016) that combines practical training in a company with theoretical education at a vocational school. In order to obtain a dual apprenticeship, adolescents have to apply directly to companies offering such positions. Hence, the search and application processes are similar to those for regular jobs (Glauser & Becker, 2016; Roth, 2014b), and almost 70% of the dual VET trainees are hired by companies after training (Bundesministerium für Bildung und Forschung, 2018). A smaller number of VET programmes are pursued as school-based training in which practical phases are often also part of the training (Solga et al., 2014). For adolescents without a university entrance qualification, formal vocational qualifications are of key importance for a smooth school-to-work transition and a stable professional career. Because most adolescents with a migration background do not succeed in obtaining a higher education entrance qualification, VET qualifications are particularly important for the successful integration of ethnic minorities into the German labour market (Granato & Kalter, 2001; Hunkler, 2010). At the same time, however, adolescents with a migration background make the transition to vocational training less often than native adolescents (Bundesministerium für Bildung und Forschung, 2018).

3 Theoretical Background

Social networks are an important source of information and support during the transition to VET, because school leavers themselves usually lack labour market experience and thus have to rely on others to receive assistance. Social networks can provide general support or counselling during the adolescents’ job-choice and career planning. They can also offer more specific, instrumental assistance during the VET search—for example, by providing information on apprenticeship vacancies or about potential training companies and their hiring practices. Social contacts can also support apprenticeship seekers—for instance, by helping to write applications or prepare for interviews. Lastly, they can act as referrals and put in a good word for the applicant with potential employers (Burt, 1992; Granovetter, 1995; Kogan, 2011; Roth, 2014b).

Young school leavers usually start from scratch when it comes to establishing labour-market-relevant social ties themselves. Because their friends usually go to school as well, we can assume that they likewise mostly lack substantive work experience and knowledge about the labour market. This also applies to their siblings, who are often of similar age. Consequently, adolescents at the end of lower secondary education are likely to rely mainly on the support of ‘older’ social contacts who already have labour market and/or VET experience such as parents or other older relatives and acquaintances. It can be assumed that parents are particularly important in this respect, because teenagers usually still live with them in the same household, and they are therefore easily available to them. In addition to this spatial proximity, their emotional proximity makes parents highly motivated to assist their children during the transition from school to work (Moerbeek & Flap, 2008; Roth, 2018).

The contacts mentioned above are usually acquired via birth or by private means in everyday interaction. However, ties acquired through institutional channels might also help school leavers find an apprenticeship. Obviously, schools are the most significant institutions in this respect, and, consequently, teachers are expected to be the most important institutional social contacts for the provision of information and support. Teachers might, for example, initiate partnerships and joint projects with companies, act as referrals, visit vocational guidance events with their students, or practice how to write job applications with them.

Overall, we assume that young people often draw on their institutional and non-institutional social contacts during their search for a training place, with parents being of major importance. Based on these general theoretical observations about the mobilization of social ties during the apprenticeship search, we shall now take a closer look at potential differences in the mobilization of different categories of social ties between adolescents with and without a migration background.

Concerning non-institutional social contacts, there are no obvious theoretical reasons to assume that the motivation of these contacts to provide support differs between ethnic groups. Consequently, the extent of involvement and general information on vocational orientation and career planning available from non-institutional social contacts should be similar for adolescents with and without a migration background. In contrast, providing more specific, instrumental assistance during the apprenticeship search such as giving information about attractive vacancies or helping with an application depends not only on the willingness of the social ties to support but also on their actual ability to do so. This again depends on their language skills, knowledge about the German education and vocational training system, and their information about and connections into the labour market. Because these resources are receiving-country-specific, there are good reasons to assume that native contacts are more competent in providing instrumental information and support than migrant contacts (Alba, 2008; Alba & Nee, 2003; Esser, 2004; Kalter & Kogan, 2014; Lancee, 2012; Lancee & Hartung, 2012). This should apply especially to first-generation migrant contacts who were not fully socialized and educated in the receiving country, whereas, in contrast, the differences between native contacts and migrant contacts born in the receiving country (second generation) should be less pronounced. Consequently, adolescents with a migration background whose parents are both foreign born should substantially less often receive instrumental information and support from their parents in their transition to vocational training than native adolescents.

Adolescents with a migration background with at least one native-born parent are expected to be less disadvantaged because, on average, their parents should have more receiving-country-specific information and resources needed at the transition from general schooling to vocational training than parents who were both foreign born. Given the disadvantages of ethnic minority youths at the transition from school to work in Germany and the tendency towards ethnic homophily in friendship networks (Leszczensky & Pink, 2015), we expect that the siblings and friends of children with a migration background are less able to provide specific information and support compared to those of native youths. However, we do not expect these ethnic differences to be exceptionally large for two reasons: first, both siblings and friends normally are of similar age as the respondents, and most of them should thus also have attended German schools. Consequently, their individual knowledge about the education and VET system should not differ much from that of natives. Second, the utility of siblings and friends of native adolescents should also be limited, because many of them have not yet gained experience in the VET system. Additionally, negative ethnic effects regarding siblings might be counteracted by the fact that migrant families have, on average, more children (Bundesministerium für Familie, Senioren, Frauen und Jugend, 2013), leading to a larger pool of contacts who can provide information and support. We also expect ethnic differences in the support from relatives outside the nuclear family. Because these relatives are more likely to be older and foreign born than siblings and friends among adolescents with a migration background, the extent of ethnic differences in support from relatives outside the nuclear family should be more similar to that of parental support.

Institutional social contacts, such as teachers, should usually provide both general and specific, instrumental information and support equally to native adolescents and adolescents with a migration background. However, if migrant children expect their parents or other non-institutional social contacts to be less helpful, they might turn more often to institutional social contacts for information and assistance than natives. In addition, teachers might systematically give more support to students who need special assistance during the apprenticeship search, such as socially disadvantaged or low-performing students among whom migrants are over-represented. On the other hand, if ethnic discrimination on behalf of teachers exists, native adolescents might receive more assistance from teachers than adolescents with a migration background.

In summary, our expectations for ethnic differences in the amount of information and support received from institutional social contacts are not clear. For non-institutional social contacts, we expect that there are no differences concerning general involvement and obtaining unspecific information from social network contacts between adolescents with and without a migration background. In contrast, we expect that friends and siblings of ethnic minority youths less often provide specific, instrumental information and support compared to friends and siblings of native youths, although differences should be comparatively small. Furthermore, we expect adolescents with a migration background to less often receive specific, instrumental information and assistance from other relatives, and particularly from parents, than native adolescents. The latter should be true especially for adolescents whose parents were both born abroad, whereas differences to natives should be less pronounced when at least one of the parents was born in Germany.

4 Previous Empirical Findings

Previous research has shown that the majority of apprenticeship seekers in Germany receive information and help from relatives, friends, or acquaintances during their job search, with parents clearly being the most important source (e.g. Beicht & Granato, 2010; Bundesinstitut für Berufsbildung, 2017; Eberhard, 2012; Eberhard et al., 2018; Hoenig, 2019; Roth, 2014a, 2018; Ulrich et al., 2018). In addition, parents’ involvement and transmission of information during the transition to VET seem to increase the likelihood of obtaining an apprenticeship (Beicht, 2011; Hoenig, 2019; Lindemann & Gangl, 2019; Ulrich, 2013).

Quantitative empirical findings on ethnic differences in network mobilization in Germany are based mainly on data from the BA/BIBB Applicant Survey which includes adolescents who registered in the Federal Employment Agency (BA) as training applicants (Beicht, 2011; Bundesinstitut für Berufsbildung, 2014; Eberhard et al., 2018). Retrospective accounts from VET applicants indicate that, in general, adolescents with and without a migration background tried to mobilize their social contacts during their career choice and apprenticeship search to a similar extent. Concerning actual mobilization, results show that native adolescents have more often talked to their parents during their career orientation and apprenticeship search than adolescents with a migration background, whereas no differences are found with respect to friends and acquaintances. With data from the BIBB-Transitional Study, Beicht and Granato (2010) showed that adolescents with a migration background less often received help from their social network in making contact with companies. However, the authors did not distinguish between different kinds of social ties. The most differentiated information on ethnic differences in the network mobilization can be found in Roth (2014a). Based on a small, regionally limited sample, he showed that natives more often received information relevant for planning their VET and assistance in finding an apprenticeship from their social contacts. Ethnic differences were most pronounced with respect to instrumental support received from parents and from parents’ friends and acquaintances, whereas there were no clear differences with respect to information and help received from adolescents’ friends and acquaintances.

Whereas the overall results to date are consistent with our theoretical expectations, previous studies have several, already described shortcomings that limit their explanatory power and generalizability.

5 Data and Analytical Strategy

We base our empirical analyses on data from Starting Cohort 4 (SC4) of the National Educational Panel Study (NEPS). This starting cohort provides longitudinal information on the educational and occupational career of ninth graders selected via a Germany-wide school-based sampling approach who have been interviewed annually or bi-annually since then. We use the first five waves that cover a period of about 2 years in which most of the students make their transition from lower secondary education to upper secondary general education or to vocational education and training. We do not consider students attending the Gymnasium in Grade 9 in our analyses, because nearly all of them continue to general upper secondary education. This would make it impossible to compare their answers to questions about entering VET to those of students attending the less demanding tracks. We also exclude students from special needs schools, because they did not receive the same questions as students from general schools.

Indicators of Network Mobilization

We derive our indicators of network mobilization from seven different questions: two referring to general and five to specific goal-oriented information and support from social contacts. The two general and three of the specific goal-oriented indicators comprise prospective information from students still attending lower secondary education (Waves 1 to 3). In addition, two of the specific goal-oriented indicators comprise retrospective information from adolescents after having left general schooling (Waves 3 and 5). For each question, students were presented with a list containing several groups of persons who might be helpful for certain challenges. In the analyses, we consider teachers (institutional social contacts) as well as parents, siblings, other relatives, and friends (non-institutional social contacts). This rich source of information on social capital mobilization is clearly an advantage of the NEPS SC4 data over other sources. The downside, however, is that we are faced with differing samples of analyses for the different indicators, with each sample representing a selection of the student population. Furthermore, the response categories and the social contacts considered differ slightly between the questions (a full overview of the wording of all questions can be found in Tables 17.A4, 17.A5 and 17.A6 in the appendix). Consequently, findings for the different indicators are only partially comparable.

Two prospective questions capture the importance of different social contacts in providing general information for job choice and during the apprenticeship search. Respondents were asked to rate their importance on a 4-point scale that we recoded to dummy variables (1: ‘rather important’ or ‘very important’ versus 0: ‘very unimportant’ or ‘rather unimportant’). The first question was given to students at the beginning of Grade 9 (Wave 1) who planned to enter an apprenticeship the following year (Sample 1). The second question was given to students at the end of Grade 9 (Wave 2) or Grade 10 (Wave 3) who planned to apply for an apprenticeship during the respective school year (Sample 2). For this question, we considered the last information available before respondents leave general education.

In contrast to this information on general involvement and assistance, the remaining three prospective questions measure specific goal-oriented assistance during the apprenticeship search. All students in general education in Wave 2 or 3 were asked to indicate on a 4-point scale ranging from 1 (very unlikely) to 4 (very likely) how likely it was that their social contacts would inform them about interesting apprenticeships, make an effort towards getting an apprenticeship for them, or help them write an application. In a second step, students who answered rather likely or very likely were asked to name all relevant sources for the respective type of support from a list of possible sources of information.Footnote 1 As above, we use information at the time of students’ last year of general lower secondary schooling (Sample 3). Information on help with writing an application is available only for Wave 3 (i.e. at the end of Grade 10; Sample 4).

In addition to this prospective information, students who left the general education system provided information about which social contacts actually told them about interesting open training positions or helped them to find an apprenticeship. For these retrospective questions, we consider the first information available after respondents left general education (either Wave 3 or Wave 5; Sample 5). Most of these respondents pursue VET, whereas a smaller share attends a transition programme of vocational preparation and very few do something else (e.g. working without a vocational qualification or being unemployed).

Migration Background

We define adolescents as having a migration background if they themselves, at least one of their parents, or at least two of their grandparents were not born in Germany. All other respondents are considered natives. In the multivariate analyses, we further divide adolescents with a migration background into those whose parents were both foreign born and those with at least one German-born parent.

Analytical Strategy

In a first step, we provide a comprehensive descriptive overview on the extent of social capital mobilization of natives and migrants with respect to different groups of social contacts and types of support. To test for statistically significant differences between natives and migrants, we apply bivariate linear probability models. Because Starting Cohort 4 of the NEPS survey applied a stratified sampling approach by which certain school types were oversampled (Hauptschule, Integrierte Gesamtschule, Freie Waldorfschule, and Förderschule; Steinhauer & Zinn, 2016), we use design weights that account for the different inclusion probabilities of the students.Footnote 2 We apply cluster-robust standard errors that take into consideration the correlation of observations on the school level.

In a second step, we run multivariate linear probability models to check whether differences between adolescents with and without a migration background in the mobilization of their parents are in fact due to ethnic rather than socio-economic differences. In these models, we control for students’ socio-economic background by considering the highest educational attainment (lower secondary degree or less, intermediate secondary degree, at least Abitur) and the highest occupational status (ISEI-08) of their parentsFootnote 3 as well as the number of books in the household. In addition, our models include the adolescents’ gender, year of birth (before 1995, 1995, after 1995), and whether they live in a single-parent household. Lastly, we also include Grade 9 school dummies. We thus estimate school fixed-effects regression models by which we control for all school characteristics such as school type or social or ethnic composition of the student population as well as for regional variation of VET demand and supply. In our analyses, we allow the number of cases to vary across the different measures of social capital mobilization to obtain the most informative sample possible for each of the measures. In additional analyses, we hold the number of cases constant across several measures to make results comparable. Table 17.1 gives an overview of the number of cases and the distribution of all control variables and school types in Grade 9 after listwise deletion for the five different samples and for all cases that are part of any of these samples (last column).

Table 17.1 Descriptive overview over the variables used in the multivariate analyses and school types in grade 9 (differentiated by sample of analyses)
Full size table

6 Empirical Results

6.1 Descriptive Results

We start the empirical analyses with a descriptive overview of our prospective measures of general support from social network contacts. Figure 17.1 shows that both institutional and non-institutional social contacts are important sources of general information for the job choice. Each of the four kinds of social contacts is considered important by at least 70 per cent of adolescents who indicated at the beginning of Grade 9 that they wanted to search for an apprenticeship during the school year (Sample 1). At about 85 per cent, parents are mentioned most often by both adolescents with and without a migration background. There are also no ethnic differences in the importance of relatives and friends. In contrast, significantly more adolescents with a migration background (difference of around 9 percentage points) regard teachers as important sources of information.

Fig. 17.1
A horizontal double bar graph estimates the percentage of prospective support received by native and migratory adolescents from their teachers, parents, other relatives, and friends or acquaintances. The support from parents is the highest for both groups of adolescents, followed by the teachers.

Prospective support: information regarding job choice (Sample 1)

Source: National Educational Panel Study (NEPS): Starting Cohort 4, authors’ own calculations

Significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001; N = 3333; cluster-robust standard errors, design-weighted

Bars indicate per cent of adolescents who answered that the respective source is ‘rather important’ or ‘very important’

Full size image

Concerning the importance of receiving general information from social contacts during the apprenticeship search shown in Fig. 17.2 (Sample 2), we find similar patterns to those for the first indicator. All four kinds of social contacts are considered important; however, on a lower level than for general information regarding the job choice. Again, parents are most often rated as important. Ethnic differences are small (around 5 percentage points) yet significant on the 5% level. We again do not find any substantive ethnic differences for other relatives and friends, whereas teachers are significantly more often rated important by adolescents with a migration background (difference of around 10 percentage points).

Fig. 17.2
A horizontal double bar graph estimates the percentage of prospective support received by native and migratory adolescents from their teachers, parents, other relatives, and friends during their apprenticeship. The support from parents is the highest for both groups of adolescents, followed by the teachers in the case of migratory adolescents and relatives in the case of native adolescents.

Prospective support: information during apprenticeship search (Sample 2)

Source: National Educational Panel Study (NEPS): Starting Cohort 4, authors’ own calculations

Significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001; N = 2954; cluster-robust standard errors, design-weighted

Bars indicate per cent of adolescents who answered that the respective source is ‘rather important’ or ‘very important’

Full size image

Thus far, we conclude that the majority of apprenticeship seekers regard social contacts as an important source of general information during their transition to vocational training. In this regard, parents are most often considered important by both adolescents with and without a migration background, and ethnic differences are rather small. This supports our notion that, when it comes to general support during this transition phase, parents born in Germany and parents born abroad are equally involved in their children’s job choice and apprenticeship search. However, teachers seem to be important as well, particularly for adolescents with a migration background. This could be due to the fact that ethnic minorities more often need guidance in their job choice and encounter problems when entering the apprenticeship system.

In the next step, we show descriptive results for our prospective measures on specific, instrumental information and support. These refer to students at the end of lower secondary schooling (Figs. 17.3, 17.4 and 17.5; Samples 3 and 4). Overall, we find that many adolescents consider it likely to receive specific information and support from their social contacts. At the same time however, the various types of contacts are named by considerably fewer respondents than was the case for general information and support. Thus, it seems that the amount of possible support embedded in young adolescents’ social networks is lower in the form of specific assistance compared to that regarding general assistance. Again, and in line with results on general information, we find that parents are seen as the most helpful source followed by relatives, friends, and siblings. The distance between parents and the other kinds of social ties is much more pronounced for specific, instrumental assistance than for general assistance. Moreover, in contrast to the results on general information, teachers do not seem to play an important role when it comes to specific assistance.

Fig. 17.3
A horizontal double bar graph estimates the percentage of prospective support received by native and migratory adolescents from their teachers, parents, other relatives, and friends after the apprenticeship. The support from parents is the highest for both groups of adolescents, followed by relatives.

Prospective support: information about interesting open apprenticeship positions (Sample 3)

Source: National Educational Panel Study (NEPS): Starting Cohort 4, authors’ own calculations

Significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001; N = 6468; cluster-robust standard errors, design-weighted

Bars indicate per cent of adolescents who answered that information or support from the respective source is ‘rather likely’ or ‘very likely’

Full size image
Fig. 17.4
A horizontal double bar graph estimates the percentage of prospective support received by native and migratory adolescents from their teachers, parents, siblings, other relatives, and friends. The support from parents is the highest for both groups of adolescents, followed by relatives.

Prospective support: efforts towards getting an apprenticeship (Sample 3)

Source: National Educational Panel Study (NEPS): Starting Cohort 4, authors’ own calculations

Significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001; N = 6468; cluster-robust standard errors, design-weighted

Bars indicate per cent of adolescents who answered that information or support from the respective source is ‘rather likely’ or ‘very likely’

Full size image
Fig. 17.5
A horizontal double bar graph estimates the percentage of prospective support received by native and migratory adolescents from their teachers, parents, siblings, other relatives, and friends for writing an application. The support from parents is the highest for both groups of adolescents, followed by siblings in the case of native and friends in the case of migratory adolescents.

Prospective support: help with writing an application (Sample 4)

Source: National Educational Panel Study (NEPS): Starting Cohort 4, authors’ own calculations

Significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001; N = 4771; cluster-robust standard errors, design-weighted

Bars indicate per cent of adolescents who answered that information or support from the respective source is ‘rather likely’ or ‘very likely’

Full size image

Concerning differences between adolescents with and without a migration background, we do not find any significant ethnic differences regarding teachers for any of the three measures of specific, instrumental assistance. Ethnic minority youth report less often than natives that their siblings and friends are likely to provide specific information and support. However, as expected, these ethnic differences are rather small and statistically significant for only one (friends) or two (siblings) of the measures.Footnote 4 In contrast, substantively more native adolescents than adolescents with a migration background expect their parents to provide information about interesting apprenticeship vacancies (difference of around 13 percentage points) or to help them obtain such a position (difference of around 14 percentage points). Ethnic differences are even more pronounced with regard to the expected support from parents when writing an application for an apprenticeship (difference of around 27 percentage points). This underlines the importance of receiving-country-specific resources, especially for parents who want to provide specific instrumental information and support to their children: whereas parents need knowledge about the German vocational training system as well as information about and connections into the labour market in order to inform their children about interesting vacancies and help them find an apprenticeship, assistance in writing an application requires language proficiency that can be expected to be considerably lower among migrants compared to natives.

Concerning the first two prospective measures on specific, instrumental information and support, we also find clear ethnic differences with respect to assistance from relatives. However, this is less pronounced than for parents (difference of around 12 percentage points for information and around 8 percentage points for efforts towards getting an apprenticeship). This further supports the notion that contacts who were socialized and educated abroad might lack receiving-country-specific resources and are thus less helpful in providing specific support on the labour market. Contrary to our expectations, however, we find no differences between adolescents with and without a migration background when looking at their expectation concerning received help from relatives with writing an application for an apprenticeship (Figs. 17.3, 17.4 and 17.5).

In sum, our prospective results show that adolescents see especially their parents as a key source of information and support during their transition from general education to vocational training. Whereas this is equally true for adolescents with and without a migration background with regard to general information, the possibilities that migrant parents have to provide specific, instrumental assistance seem to be clearly limited. Although less pronounced, this also seems to be the case for relatives outside the nuclear family. A contrasting picture emerges for the role of teachers in this transition process. Whereas students with and without a migration background do not differ in how they perceive the potential of teachers to provide specific support, students with a migration background more often expect to need assistance from teachers during job choice and general information during apprenticeship search. In order to exclude the possibility that comparisons of the results on general and specific prospective information are biased by different analysis samples, we run additional analyses in which we restrict the sample to cases with information on all five measures. General results remain unchanged (cf. Table 17.A1 in the appendix).

In a last step of our descriptive analyses, we investigate the two retrospective measures of specific, instrumental information and support from adolescents who left the general education system (see Figs. 17.6 and 17.7). As in the prospective measures, parents are the source most often considered important by both adolescents with and without a migration background. Whereas adolescents with a migration background substantively less often state that their parents told them about interesting apprenticeship or training vacancies (ethnic gap of around 12 percentage points), differences with regard to help in finding an apprenticeship are only small and not significant (difference of around 3 percentage points). Unexpectedly, adolescents with a migration background received information about VET vacancies from teachers significantly more often than native adolescents (difference of around 10 percentage points) and more often experienced that teachers helped them obtain an apprenticeship (difference of around 8 percentage points). This indicates that they try to compensate their lack of support from their parents by mobilizing institutional social contacts. Teachers might either assist them in their search efforts by actively connecting them with training companies or vocational schools or by informing them about other institutional opportunities such as employment agencies. Ethnic differences with regard to other kinds of social ties are comparatively small. Compared to natives, ethnic minority youths somewhat more often report having received information about interesting VET vacancies from siblings and friends. Lastly, in line with our theoretical arguments, native youths more often receive information about open positions from their relatives compared to ethnic minority youths (difference of around 4 percentage points).

Fig. 17.6
A horizontal double bar graph estimates the percentage of retrospective support received by native and migratory adolescents from their teachers, parents, siblings, other relatives, and friends. The support from parents is the highest for both groups of adolescents, followed by the teachers in the case of migratory adolescents and other relatives in the case of native adolescents.

Retrospective support: information about interesting open apprenticeship positions (Sample 5)

Source: National Educational Panel Study (NEPS): Starting Cohort 4, authors’ own calculations

Significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001; N = 4593; cluster-robust standard errors, design-weighted

Bars indicate per cent of adolescents who answered that they received information/support from the respective source

Full size image
Fig. 17.7
A horizontal double bar graph estimates the percentage of retrospective support received by native and migratory adolescents from their teachers, parents, siblings, other relatives, and friends in their efforts towards getting VET positions. The support from parents is the highest for both groups of adolescents, followed by the teachers in the case of migratory adolescents and other relatives in the case of native adolescents.

Retrospective support: efforts towards getting a VET positions (Sample 5)

Source: National Educational Panel Study (NEPS): Starting Cohort 4, authors’ own calculations

Significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001; N = 4593; cluster-robust standard errors, design-weighted

Bars indicate per cent of adolescents who answered that they received information/support from the respective source

Full size image

Interestingly, whereas adolescents more often mention their parents as an important source of information about interesting VET vacancies retrospectively rather than prospectively, the opposite is true in the case of providing actual help to find an apprenticeship. Figure 17.8 shows that this pattern also holds when we restrict our analysis sample to those adolescents with information on all four measures—that is, to those who left general education after Grade 9 or 10.

Fig. 17.8
A horizontal double-bar graph compares the percentage of prospective and retrospective effort and information received by native and migratory adolescents. The prospective effort and retrospective information received by both groups of adolescents are high.

Comparison of prospective and retrospective measures on information and support from parents

Source: National Educational Panel Study (NEPS): Starting Cohort 4, authors’ own calculations

Significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001; N = 4343; cluster-robust standard errors, design-weighted

Prospective information: Bars indicate per cent of adolescents who answered that information or support from the respective source is ‘rather likely’ or ‘very likely’

Retrospective information: Bars indicate per cent of adolescents who answered that they received information/support from the respective source

Full size image

Our descriptive analyses show that all social ties considered are important sources of information and support for adolescents in their school-to-work transition, with parents being by far most important. The results for non-institutional social contacts are mostly in line with our expectations. Adolescents with a migration background less often receive specific help and information from their parents than native adolescents, and they also anticipate this. In contrast, this is not the case with regard to obtaining general information, for which receiving-country-specific resources should not be decisive. Whereas results for relatives outside the nuclear family are not completely conclusive, they tend to take the same direction as those for parents. In contrast, we find no clear ethnic differences with respect to friends; and whereas natives somewhat more often expect their siblings to be helpful, ethnic differences do not emerge retrospectively. Concerning institutional social contacts, results indicate that adolescents with a migration background more often rely on the assistance of teachers than native adolescents. This rather speaks against ethnic discrimination by teachers and supports the argument that adolescents with a migration background are more dependent on support from teachers, who, in turn, are willing to provide it. In addition to general job application training in class or referring students to career counselling, this support can also include individual assistance with writing applications, finding suitable apprenticeship vacancies, and even organizing partnerships with regional companies and arranging internships.

6.2 Multivariate Results

As shown above, parents play a key role in providing information and support to young school leavers during the transition from school to work. This is in line with results from previous studies on the role of social capital during the apprenticeship search. However, parents of adolescents with a migration background seem to be substantively less helpful during this transition than parents of native adolescents. This might be due to a lack of migrant parents’ receiving-country-specific capital such as language resources and knowledge about the VET market or to a lack of native contacts in the social network. However, it is also possible that lower socio-economic status or contextual factors, such as regional availability of training opportunities, are responsible for this ethnic gap. Therefore, in the following, we shall further investigate the role of adolescents’ parents by running multivariate analyses in which we control for parents’ socio-economic status. In order to also control for regional variations and context characteristics on the school level, we run school fixed-effects regressions. Moreover, we differentiate between adolescents with a migration background whose parents are both first-generation migrants and adolescents with a migration background of whom at least one parent was born in the receiving country. In the first set of models, we simply replicate the descriptive findings for each indicator with migration background differentiated by migrant generation. The second set of models presents ethnic differences in multivariate analyses including all control variables and school fixed effects as described above (coefficients for the control variables can be found in Tables 17.A2 and 17.A3 in the appendix).

Table 17.2 shows multivariate analyses for the two indicators referring to rather general information and support. Model 1a and Model 1b overall replicate the descriptive results from Figs. 17.1 and 17.2. Ethnic differences are statistically significant only for the first and second generation regarding general information received during the apprenticeship search (Model 1b). However, these differences are reduced and become non-significant once we control for socio-economic background and school characteristics in Model 2b. This further corroborates our assumption that migrant and native parents are very similar with respect to providing general support.

Table 17.2 Coefficients from linear probability regressions: general information
Full size table

Table 17.3 shows multivariate results referring to indicators of instrumental support—both prospectively for students at the end of lower secondary schooling and retrospectively for those who have left general education after Grade 9 or Grade 10. Regarding prospective information and support, Model 1a, Model 1b, and Model 1c show substantive ethnic differences, as already seen in Figs. 17.3, 17.4 and 17.5. In addition, the models show that ethnic disadvantages are especially pronounced for adolescents with a migration background whose parents are both born abroad, whereas they are far smaller (and in case of Model 1a not significant) for those with at least one German-born parent. The inclusion of our control variables reduces ethnic penalties, but the results remain substantively the same as in the bivariate models. Regarding retrospective information and support, results are similar. Although somewhat less pronounced than for the prospective measures, we again find statistically and substantively significant differences in bivariate and in multivariate models between native adolescents and adolescents whose parents were both born abroad. For school leavers with one German-born parent, differences to natives are small and statistically significant only with regard to information about open positions when socio-economic background is not taken into account (Model 1d).

Table 17.3 Coefficients from linear probability regressions: specific, instrumental information and support
Full size table

We can conclude that, whereas ethnic differences regarding the provision of specific, instrumental information and support from parents are particularly pronounced between native adolescents and adolescents who themselves or whose parents were both born abroad, differences to natives are weak or even non-existent when at least one parent of the adolescents with a migration background was born in Germany. In addition, substantive results remain largely unchanged when we control for socio-economic background and context characteristics in our multivariate analyses. This is in line with our theoretical assumption that ethnic differences are mostly not due to differences in socio-economic status or structural factors. Instead, they are likely to result from migrant parents’ insufficient receiving-country-specific knowledge, resources, and contacts due to their shorter length of stay in the receiving country and their lack of own experience with the German education and vocational training system.

7 Summary and Conclusion

Based on refined information from NEPS Starting Cohort 4, we gave a comprehensive overview of (potential) social capital mobilization among adolescents with and without a migration background during the transition to vocational training at the end of lower secondary education. In addition, we provided in-depth insights into the role parents play in this process. In line with previous research, we find that students rely heavily on information and support from their social network, with parents playing the most important role. Besides this replication of earlier results on the basis of a nationwide sample of ninth graders in regular schools, our analyses provide new information concerning ethnic differences in the mobilization of social ties at the transition to VET that enhance current knowledge in several ways. Our main results can be summarized as follows: first, concerning institutional social contacts, there are indications that adolescents with a migration background more often rely on information and support from teachers than native adolescents do. Second, for non-institutional social contacts, we see hardly any ethnic differences in the expected network mobilization with regard to general information and support. Third, we also find no strong ethnic differences in the expected specific, instrumental assistance from siblings and friends, whereas adolescents with a migration background tend to less often receive specific assistance from relatives outside the nuclear family and substantively less often specific assistance from parents. Fourth, ethnic differences in actually received specific, instrumental assistance widely mirror the findings of our prospective measures for parents, but this is not the case for other sources of information. Fifth, our multivariate analyses show that ethnic differences in the provision of specific assistance from parents are especially pronounced for adolescents who themselves or whose parents were both born abroad, and that these differences cannot be traced back to socio-economic background or context factors.

Overall and in line with our theoretical expectations, results suggest that the general motivation of non-institutional social contacts to provide support at the transition to VET does not differ between natives and migrants, but that the ability of these ties to provide more specific, instrumental assistance depends on their receiving-country-specific resources and thus on their migration history. Consequently, adolescents whose parents were born abroad are clearly disadvantaged, because they are likely to less often receive specific, instrumental assistance from their parents. The finding that adolescents with a migration background are more likely to rely on information and help from teachers suggests that institutional social contacts might be used to counteract these disadvantages. This may, however, not be enough to fully compensate for the disadvantages caused by the lack of assistance from parents. These findings are relevant in the German context, because, as we have seen, parents are by far the most important source of information and support at the transition to VET. Moreover, previous research indicates that parental assistance actually increases the likelihood of obtaining an apprenticeship (Beicht, 2011; Hoenig, 2019; Lindemann & Gangl, 2019; Ulrich, 2013). Finding an apprenticeship is an important first step in acquiring a vocational qualification; and for adolescents who do not enter tertiary education, such a qualification is indispensable to establish a successful career in the German labour market in which occupation-specific credentials are highly valued (Bills et al., 2017; Müller, 2005). Because comparatively few young adults with a migration background possess a vocational degree and about one third of them have neither a university degree nor a VET qualification (Autorengruppe Bildungsberichterstattung, 2018), a successful transition into the VET system plays a key role in the structural integration of adolescents with a migration background in Germany. Thus, our results suggest that it is important to assist adolescents with a migration background at the transition to VET in order to foster a smooth school-to-work transition, for example through the organization of internships or information campaigns, thereby compensating for the limited possibilities of their parents to support them.

On a more general level, we have shown that differentiated analyses are important to better understand ethnic differences in network mobilization in the school-to-work transition. In this respect, it seems especially important for future research to distinguish between different kinds of institutional and non-institutional social contacts as well as between general and specific assistance. Furthermore, future research should investigate more closely why it is specifically migrant parents who are less likely to provide specific assistance during this transition. Our theoretical argumentation and empirical findings suggest that the reason lies in a lack of receiving-country-specific resources, but an explicit empirical investigation of the exact mechanisms is beyond the scope of this chapter and is therefore left to future research. Finally, future research should further explore how social capital mobilization and apprenticeship search success are related. The refined information in the NEPS offers opportunities for a differentiated analysis that can further enhance existing knowledge from previous research.