Education pays off. Many international studies show that more education also means better labour market opportunities (e.g. Psacharopoulos and Patrinos, 2004). However, some findings suggest that, given the same level of education, it also depends on which education is invested in (Birkelund et al., 2021). This is particularly relevant for Germany, where the educational system is an “archetype” (Schindler, 2017) of a strongly stratified system (as well as Austria and Czechia) and the assignment to different educational tracks is conducted at an early age. In order to correct school type decisions once they have been made, there is a wide range of ‘alternative pathways’ at upper secondary level. Catching up on school certificates, mainly higher education entrance qualifications (HEEQ), is particularly possible at vocationally oriented schools (Schuchart, 2013). On average, pupils in these school types perform worse compared to similar pupils from general schools (Trautwein, 2007). Nevertheless, alternative pathways typically provide vocational knowledge in a certain field (e.g. ‘Health’ or ‘Economy’), although they do not award vocational degrees. For instance, a lower secondary level graduate with an intermediate certificate (IC) can catch up on a HEEQ at upper secondary level and meanwhile obtain some basic knowledge in a vocational field.

Our general research question is connected to labour market opportunities that arise with these qualifications: Does the attainment of school-leaving certificates via vocational schools lead to similar opportunities as their acquisition in the general school system? Recent research has shown that general education pathways to school-leaving certificates seem to lead to higher labour market returns than alternative pathways (Heckman et al., 2006; Schuchart and Schimke, 2019). However, this research has at least two weaknesses: First, it is unclear whether these findings are due to self-selection or selection by employers. To address this point, we focus on the demand side of labour and analyse potential selection. We are interested in hiring processes at entry level positions on the labour market. Although, we focus on Germany the results are at least transferable to countries with dual vocational training systems (e.g. Austria, Denmark and Switzerland), where employers decide on who to hire for apprenticeships (Poulsen and Eberhardt, 2016).

Second, previous research has not explicitly taken into account the significance of acquiring basic professional knowledge with the attainment of a school-leaving certificate. Since a general school-leaving certificate is also acquired at vocationally oriented schools, graduates are free to apply for vocational training programs that are within the scope of the basic vocational knowledge they have already attained or not. In the former case, this could mean a great advantage also over graduates with the same school-leaving certificate but from general education schools, in the latter this would not be the case. To develop more precise assumptions, we draw on screening (Arrow, 1973; Stiglitz, 1975) and queuing theory (Thurow, 1975). According to these theories, different paths to the same educational degree could be associated with differing signalling effects. Hiring is always an uncertain investment and therefore employers are most likely to decide for applicants who are expected to have the lowest training costs. Based on these arguments, we derive the following hypotheses:

H₁: Graduates with a HEEQ from vocationally oriented schools who obtain basic knowledge in a similar vocational field to the training occupation are more likely to be invited to vocational training interviews than general school graduates with the same school-leaving certificates.

H₂: General school graduates with a HEEQ are more likely to be invited to vocational training interviews than graduates from vocationally oriented schools with the same school-leaving certificate but without specific vocational knowledge.

Our study is based on a discrete choice experiment (DCE) (Street and Burgess, 2007). The DCE was conducted in January 2023 with German HR professionals who hire applicants for apprenticeships in at least one of 10 occupations. These occupations were selected to represent 1.) professions, where apprentices typically have HEEQ and 2.) are quantitatively important on the German labour market. The sample was generated by collecting HR professional’s contact information from real job advertisements posted on the employment agency’s website between January 2022 and 2023. Each respondent was confronted with three choice decisions. They were asked to select the most promising out of three candidates (opt-out alternative was also presented) for a vocational training interview invitation in the profession they have hiring expertise. Different studies on determinants of hiring have shown the validity of this measurement (e.g. Humburg and van der Velden, 2015) and the methodology of in-lab experiments is also known to represent real world behaviour (Petzold and Wolbring, 2019; Hainmueller et al., 2015). Information on each applicant was experimentally varied and included gender, age, migration background, educational attainments (IC and HEEQ, results for IC are not reported) combined with educational pathways focusing on different general (Gymnasium or comprehensive schools) and vocationally-oriented schools (‘Fachoberschule’ or ‘Berufliches Gymnasium’) with different vocational fields as well as final grades. In order to take the specific characteristics of the German federal educational system adequately into account, we had to ensure that each respondent is assigned to profiles with correct designations of school types and available vocational fields at the vocationally oriented schools based on the federal state of the HR professional’s workplace. The final sample consists of 1,329 respondents who made 3,839 decisions on 11,517 profiles. The experiment is structurally identical for all 10 occupations. The DCE profiles was built on a fraction of vignettes from a population of 1092 profiles (2^2*3*7*13) (DB-error: 0.219). The final DCE contains 270 profiles that were experimentally selected and also experimentally assigned to 90 choice sets, which subsequently have been split into 30 blocks. Each respondent is randomly assigned to one block and the three choice sets of a block as well as the three profiles within a choice set are presented in a random order. The design was constructed using the R package ‘idefix’ (Traets et al., 2020). Analyses are conducted using Stata 17 and multilevel logit models with random intercepts.

Main findings for the DCE match theoretical considerations. Graduates from vocationally oriented schools that lead to general HEEQ (‘Berufliches Gymnasium’) and provide basic vocational knowledge in a field that is related to the vocational training program, are most often invited to interviews. Compared to graduates from general schools with the same educational degree, these profiles significantly stand out by 17.9 percentage points against graduates from Gymnasium and 24.9 p. p against those from comprehensive schools. Analogous results can be stated for graduates from ‘Fachoberschulen’ which provide specific HEEQ for universities of applied sciences. Both findings support H1. The results to test H2 illustrate the value of basic vocational knowledge regarding hiring of apprentices. Compared to vignettes with HEEQ from the general educational system, the same degree from vocationally oriented schools providing unrelated vocational knowledge is strongly devalued. The average marginal effect for invitation is reduced by 25.3 p. p. compared to the Gymnasium and 18.3 p. p. in contrast to graduates from comprehensive schools. The above mentioned analogies for graduates from ‘Fachoberschulen’ are also reproduced and support H2. All findings are also in line with results that show the importance of occupation-specific human capital for hiring indicating lower training costs (e.g. Humburg and van der Velden, 2015). Simultaneously, the results shed light on open research questions regarding educational inequalities and labor market opportunities. Pupils in vocationally oriented schools perform worse compared to similar pupils from general schools and these educational paths are more often followed by children from educationally disadvantaged social groups (Trautwein, 2007). In addition to school-leaving certificates, school types of both the general and vocationally oriented fraction of the educational system seem to play a further role for educational returns. These findings are particularly relevant for educational counselling of pupils and their parents, trainees and students.

Arrow, K. J. (1973). Higher education as a filter. Journal of Public Economics, 2, 193–216. Birkelund, J. F., Capsada-Munsech, Q., Boliver, V. and Karlson, K. B. (2021). Lives on track? Long-term earnings returns to selective school placement in England and Denmark. The British Journal of Sociology, 72, 672–692. Hainmueller, J., Hangartner, D. and Yamamoto, T. (2015). Validating vignette and conjoint survey experiments against real-world behavior. Proceedings of the National Academy of Sciences of the United States of America, 112, 2395–2400. Heckman, J. J., Stixrud, J. and Urzua, S. (2006). The effects of cognitive and noncognitive abilities on labor market outcomes and social behavior. Journal of Labor Economics, 24, 411–482. Humburg, M. and van der Velden, R. (2015). Skills and the graduate recruitment process: Evidence from two discrete choice experiments. Economics of Education Review, 49, 24–41. Petzold, K. and Wolbring, T. (2019). What can we learn from factorial surveys about human behavior? Methodology, 15, 19–30. Poulsen, S. B. and Eberhardt, C. (2016). Approaching Apprenticeship Systems from a European Perspective. Bonn. Psacharopoulos, G. and Patrinos, H. A. (2004). Returns to investment in education: a further update. Education Economics, 12, 111–134. Schindler, S. (2017). School tracking, educational mobility and inequality in German secondary education. Developments across cohorts. European Societes, 19, 28–48. Schuchart, C. (2013). Kein Abschluss ohne Anschluss? Durchlässigkeit und Vergleichbarkeit in der Sekundarstufe II. Die Deutsche Schule, 105, 345–363. Schuchart, C. and Schimke, B. (2019). Lohnt sich das Nachholen eines Schulabschlusses? Alternative Wege zur Hochschulreife und ihre Arbeitsmarkterträge. Kölner Zeitschrift für Soziologie und Sozialpsychologie, 71, 237–273. Stiglitz, J. E. (1975). The theory of "screening," education, and the distribution of income. American Economic Review, 65, 283–300. Street, D. J. and Burgess, L. (2007). The Construcution of Optimal Stated Choice Experiments. Theory and methods. New Jersey: John Wiley & Sons. Thurow, L. C. (1975). Generating Inequality. Mechanisms of Distribution in the U.S. Economy. New York: Basic Books. Traets, F., Sanchez, D. G. and Vandebroeck, M. (2020). Generating Optimal Designs for Discrete Choice Experiments in R: The idefix Package. Journal of Statistical Software, 96, 1–41. Trautwein, U. (Ed.) (2007). Schulleistungen von Abiturienten. Regionale, schulformbezogene und soziale Disparitäten. Münster: Waxmann.