Effects of Social Networks on Finding an Apprenticeship in the German Vocational Training System

ABSTRACT

Previous research indicates that social relationships can influence the probability of successfully finding a first job. This paper contributes to this field of research by empirically analysing the rarely studied question, if social networks can help adolescents find an apprenticeship in the strongly company-based vocational training system of Germany. In contrast to most previous studies, not only the youth's networks, but also their mothers' networks are taken into account. In addition to the social, the ethnic composition of the networks is also considered. Event data analyses of German panel data on natives and migrants from the project ‘Young Immigrants in the German and Israeli Educational System’ show that adolescents' networks have no effect on the success of an apprenticeship search. In contrast, the mothers' networks play a crucial role.

Key words:

• social and ethnic network composition
• social capital
• apprenticeship search
• school-to-work transition

Acknowledgements

The article is based on research supported by the Framework Programme for the Promotion of Empirical Educational Research from the German Federal Ministry of Education and Research (Grant: 01JG1053).

Notes

¹ Also in Austria and Switzerland companies are strongly involved in the vocational training (Ashton et al. 2000: 17; Seibert et al. 2009: 598ff.; Hunkler 2010: 214).

² Based on the same data source, analyses of Beicht and Ulrich (2008: 266ff.) show that being a member of associations like sports, singing or cultural clubs has no effect on finding an apprenticeship.

³ Unfortunately, the data does not contain information about the fathers’ networks, although it can be assumed that their networks are more decisive for the search success as they are more often than mothers gainfully employed. Nevertheless, also the networks of mothers should be of importance. Furthermore, previous research has shown that the networks of couples substantially overlap (e.g., Kalmijn 2003: 236ff.), wherefore information about the mother's network can also be seen as a rough proxy variable for the network of both parents. For the interpretation of the results it can be noted that effects of parents’ networks might be probably more pronounced than the estimates for the mothers’ networks.

⁴ While mothers answered the position generator in wave one, for adolescents it was conducted in wave three. Therefore, for adolescents indicators derived from the position generator are not chronologically prior to the dependent variable which questions the causal direction. It cannot be ruled out that the networks of a youth changed during the search process and that it is conditional on the outcome of the search. For example, it could be assumed that those who do an apprenticeship get to know more people who are active at the labour market, wherefore possible effects of the position generator indicators could be overestimated. Nevertheless, information from the position generator is also used for adolescents as it provides the opportunity to use exactly the same indicators of the network composition as for mothers.

⁵ For the exact wording of the position generator and the list of used occupations see appendix.

⁶ For the classification of the occupations the Magnitude Prestige Scale (MPS) from Wegener was used, as it is more appropriate for German particularities than international prestige scales (Wegener 1985; Christoph 2005).

⁷ Empirical results are the same if I use an indicator that measures the proportion of persons with the same ethnic background.

⁸ Parts of the description of the data and the operationalisation are borrowed from Roth and Salikutluk (2012), where one can also find some more information about the position generator.

⁹ Survivor functions are estimated using the Kaplan–Meier product-limit-method.

¹⁰ Additional piecewise constant exponential models with period-specific effects indicate – in accordance with the descriptive presentation of survivor functions in Figure 1 – that the extensity of the network is rather irrelevant at the beginning of the search process but afterwards it has substantial significant effects (results available from the author).

Additional information

Notes on contributors

Tobias Roth

Tobias Roth is research assistant at the Mannheim Centre for European Social Research (MZES) and assistant lecturer at the chair of Sociology, societal comparisons at the University of Mannheim. His main research interests concern social networks as well as social, ethnic and gender inequalities in the education system.