This study is an empirical contribution to the explanation of class-specific educational inequality in continuing vocational education and training (CVET). Inequalities in CVET build up or down over the life course. It is in relation to this life course dependency that I want to examine educational behavior in CVET (O'Rand 2006). Class-differentiating educational decisions are related to class-specific differences in the cost-benefit trade-off. Those decisions then - mediated by the selection and allocation function of the educational system and the resources of the parental home - lead to social inequality of educational opportunities. An important conceptual distinction of social educational inequalities is that between primary and secondary effects (Boudon 1974). The latter is aimed in particular at social inequalities that arise outside the education system and already before entry into the (pre-)school education system - i.e., primarily within the family of origin. Inequalities during a person’s life course shape their attitudes towards CVET (Loeng 2020). 

In the public and academic debate, a close and direct connection between educational systems and educational decisions is often assumed (Van de Werfhorst and Mijs 2010). Such a connection is of interest mainly because educational institutions are among the factors that are in principle open to political control and can thus be shaped. In contrast to early pre-vocational education, two characteristics become important in the case of CVET. First, educational decisions are no longer made within the framework of institutionally predetermined decision-making latitude, i.e., under specifications of the respective decision alternatives and access criteria. According to educational research, a large part of adult learning processes takes place outside of educational organizations (Livingstone 1999, Livingstone 2001, Holland 2019). The decision for or against CVET and the knowledge of relevant and necessary educational offers are thus even more dependent on the individual and his or her decision-making and subjective and objective knowledge about educational offers. Second, the dimension of standardization ceases to apply. An institutional anchoring of education with predefined (humanistic) educational ideals and determined standards is increasingly being replaced by the practice- and application-oriented perspective on education. Education becomes economically exploitable employability, which is also discussed under the term “subjectification of education” (Ryökkynen, Maunu et al. 2022). In this sense, the term "vocational education" is becoming more diffuse as it is applied to an increasingly wide range of different learning processes and as CVET becomes less and less definable by institutional or content-related criteria. In particular, the specification of forms of "organized learning" (Bildungskommission 1970: 197) can no longer be convincing today in view of the increasing importance of informal learning (Eraut* 2004, Holmgren and Sjöberg 2022). 

The paper first provides a brief overview of empirical findings on social inequalities in the (German) CVET landscape in order to clarify key comparative dimensions of the observed educational differences. Then, based on a number of problems identified, the paper outlines the main features of a research program that combines the analysis of CVET with that of an individual decision-making behavior and educational history. The research question is Q: What impact do (parental) educational decisions during school time, bounded rationality/norms and framing of education and the learning environment during school time have on educational decisions on CVET?

I draw data from the German National Educational Panel Study Starting Cohort 4 (SC4) (Blossfeld and Roßbach 2021, NEPS-Netzwerk 2021) to study a cohort of students starting from Grade 9 into work life, when the transition from school to work life and early tracking of CVET is possible. Based on the research question derived from the theory, there are five main variables relevant for this study: The participation in different CVET programs (outcome variable); (parental) educational decisions during school time (predictor), bounded rationality/norms and framing of education (predictor), learning environment during school time (predictor), individual educational history (predictor) and, additionally, social economic status (moderating variables). I use structural equation modeling (SEM) to investigate the influence of educational background to CVET decision-making. Structural equation models are well-suited for analyzing panel data in the field of education (Voelkle, Oud et al. 2012). The ability to examine changes in relationships over time makes SEM a powerful tool for studying educational outcomes. Firstly, SEM can be used to model the impact of family background, peer effects, bounded rationality/norms and framing of education and the learning environment on student achievement. Secondly, SEM can handle both within-individual and between-individual effects, which is important in the analysis of panel data in vocational education. This allows for the examination of changes in individuals' educational behavior over time, as well as differences in behavior across individuals. This can provide valuable insights into the factors that drive vocational educational choices and the mechanisms through which they impact outcomes. Thirdly, SEM can control for unobserved heterogeneity, such as individual abilities and preferences, which may affect both educational behavior and outcomes. This helps in reducing bias and improving the accuracy of the estimates. With the rational choice concepts of costs and benefits and status preservation I will trace educational decisions of respondents and their parents throughout respondents’ school education. Additionally, bounded rationality/norms and framing of education serves as concepts to measure attitudes and aspirations towards education, both of respondents’ and their parents/familiar background.

CVET is becoming even more crucial for workers’ employability as the rise of AI technologies shape and change more and more jobs. It is important to better understand inequalities in vocational education, where they come from and how they function, in order to minimize them. Educational trajectories are characterized by episodes and transitions in the individual life course. The social inequality in education changes during the life course due to cumulative selection processes (Mare 1980). The educational status observed in a person at a particular point in time cannot necessarily be explained by current conditions. The decisive factor is often the individual's previous history, and this must also be taken into account when analyzing educational inequalities in CVET. Correlations result from the individual or parental decision-making behavior during the corresponding transition. This behavior is linked to the development of preferences, but also to the individual performance development of the child or adolescent. Therefore, the importance of longitudinal performance measurement becomes apparent. Only this way can it be decided at which levels social selections 'ultimately' take place and to what extent the further development of competencies, the acquisition of certificates and the genesis of educational decisions tend to be mere consequences of previous selection processes. Along with relative risk aversion theory (Boudon 1974, Breen and Goldthorpe 1997) I assume that expectation of employability has a strong impact on CVET choices. Lower educated workers should see more benefits in directly applicable knowledge. Status maintenance considerations should cause higher educated workers to enter courses that facilitate access to higher status positions. Additionally, they should be better able to navigate through the diffusion of informal educational offers and, hence, use more diverse and informal learning programs. In sum, these decisions factors should contribute to socially selective decision behavior in the choice of CVET programs.

Bildungskommission, D. B. (1970). Strukturplan für das Bildungswesen: verabschiedet auf der 27. Sitzung der Bildungskommission am 13. Februar 1970, Bundesdruckerei. Blossfeld, H.-P. and H.-G. Roßbach (2021). Education as a lifelong process: The german national educational panel study (NEPS). Wiesbaden, Springer. Boudon, R. (1974). Education, opportunity, and social inequality: Changing prospects in western society. New York, Wiley. Breen, R. and J. H. Goldthorpe (1997). "Explaining educational differentials: Towards a formal rational action theory." Rationality and society 9(3): 275-305. https://doi.org/10.1177/104346397009003002. Eraut*, M. (2004). "Informal learning in the workplace." Studies in Continuing Education 26(2): 247-273. https://doi.org/10.1080/158037042000225245. Holland, A. A. (2019). "Effective principles of informal online learning design: A theory-building metasynthesis of qualitative research." Computers & Education 128: 214-226. https://doi.org/10.1016/j.compedu.2018.09.026. https://www.sciencedirect.com/science/article/pii/S0360131518302719. Holmgren, R. and D. Sjöberg (2022). "The value of informal workplace learning for police education teachers’ professional development." Journal of Workplace Learning 34(7): 593-608. 10.1108/JWL-04-2021-0040. https://doi.org/10.1108/JWL-04-2021-0040. Livingstone, D. W. (1999). "Exploring the icebergs of adult learning: Findings of the first Canadian survey of informal learning practices." https://eric.ed.gov/?id=ED436651. Livingstone, D. W. (2001). "Adults' informal learning: Definitions, findings, gaps and future research." Centre for the Study of Education and Work. https://hdl.handle.net/1807/2735. Loeng, S. (2020). "Self-directed learning: A core concept in adult education." Education Research International 2020(Article ID: 3816132). https://doi.org/10.1155/2020/3816132. Mare, R. D. (1980). "Social background and school continuation decisions." Journal of the American Statistical Association 75(370): 295-305. 10.1080/01621459.1980.10477466. https://www.tandfonline.com/doi/abs/10.1080/01621459.1980.10477466. NEPS-Netzwerk (2021). Nationales Bildungspanel, Scientific Use File der Startkohorte Klasse 9. Bamberg, Leibniz-Institut für Bildungsverläufe (LIfBi). https://doi.org/10.5157/NEPS:SC4:12.0.0. O'Rand, A. M. (2006). Nine - Stratification and the Life Course: Life Course Capital, Life Course Risks, and Social Inequality. Handbook of Aging and the Social Sciences (Sixth Edition). R. H. Binstock, L. K. George, S. J. Cutler, J. Hendricks and J. H. Schulz. Burlington, Academic Press: 145-162. https://doi.org/10.1016/B978-012088388-2/50012-2. Ryökkynen, S., A. Maunu, R. Pirttimaa and E. K. Kontu (2022). "Learning about students’ receiving special educational support experiences of qualification, socialization and subjectification in finnish vocational education and training: A narrative approach." Education Sciences 12(2). 10.3390/educsci12020066. Van de Werfhorst, H. G. and J. J. B. Mijs (2010). "Achievement Inequality and the Institutional Structure of Educational Systems: A Comparative Perspective." Annual Review of Sociology 36(1): 407-428. 10.1146/annurev.soc.012809.102538. https://doi.org/10.1146/annurev.soc.012809.102538. Voelkle, M. C., J. H. L. Oud, E. Davidov and P. Schmidt (2012). "An SEM approach to continuous time modeling of panel data: Relating authoritarianism and anomia." Psychological Methods 17: 176-192. 10.1037/a0027543.