academic journals vocational education

Advertisement

Curricular integration of academic and vocational education: a theory-based empirical typology of dual study programmes in Germany

• Open access
• Published: 15 July 2022
• Volume 85 , pages 1257–1279, ( 2023 )

Cite this article

You have full access to this open access article

• Lisa Mordhorst   ORCID: orcid.org/0000-0001-5749-8469 1 &
• Tobias Jenert   ORCID: orcid.org/0000-0001-9262-5646 1  

4199 Accesses

4 Altmetric

Explore all metrics

In Germany, dual learning programmes are increasingly offered by higher education institutions. These programmes’ main characteristic and greatest challenge is their integration of academic and vocational learning. So far, this challenge has frequently been stated without specifying its exact nature and consequences for learners. The present study addresses this pedagogical research gap and examines the extent of variation in the degree of integration among dual study programmes . With reference to curriculum theory, the study develops an empirical typology of curricular integration in dual programmes. The data sample consists of 152 programmes at (dual) universities and universities of applied sciences. Data is analysed using hierarchical cluster analysis. Results indicate that the currently prevailing forms of curricular integration should best be differentiated according to five types. The five overlapping types of integration are located on a continuum ranging from parallelism through organisational linking to full curricular integration targeted immediately at students’ personal integration. The analysis confirms that there are problems with complying with integration standards set on the policy level. Above all, the study offers new insights on what marks the diverse integration landscape of dual study programmes. It proves that approaches to integration are more differentiated than previous research has shown.

Similar content being viewed by others

Professional Education Between School and Practice Settings: The German Dual System as an Example

Integrating Mathematics into Science: Design, Development and Evaluation of a Curriculum Model

Curriculum Integration and the Semicentennial of Basil Bernstein’s Classification and Framing of Educational Knowledge

Avoid common mistakes on your manuscript.

Introduction

The main objective of dual study programmes in higher education (HE) is to offer students the benefits that result from integrating academic and vocational learning. While these dual academic programmes are formally part of tertiary education, the vocational elements are typically located at the secondary level. According to the German Science Council (GSC, 2013 ), dual bachelor studies aim at systematically linking these two realms of learning. An increasing number of students in Germany favour dual studies (Authoring Group Educational Reporting, 2020 ), with universities of applied sciences (UAS)—the main provider of dual programmes—reporting stable growth rates. In 2005, the share of dual students at UAS was 1.4%; in 2017 it was 9.7% (Mordhorst & Nickel, 2019 ). According to the national student survey at universities and UAS, every second student can imagine studying dually (Multrus et al., 2017 ). Dual vocational programmes at the secondary level have a long tradition in Germany (Wolter, 2017 ). The recent expansion of “hybrid programmes” (Graf, 2017 ) at the intersection of the HE and the vocational system, in turn, is an innovation, since traditionally the two systems have been strictly separated (Baethge, 2006 ). The growth of dual programmes in tertiary education relates to a crisis of the dual vocational system and to the international trend of massification in postsecondary education: an increasing number of professions is moving out of the vocational and into the HE system.

Research on their structure suggests that the degree to which academic and vocational learning are integrated varies greatly between programmes (Krone, 2019 ; Langfeldt, 2018 ). Often the students themselves are challenged to connect their distinct learning experiences (Faßhauer & Anselmann, 2021 ; Kupfer et al., 2014 ). Different degrees of integration are generally possible if the programmes have coherent profiles. Parallelism of the two realms of learning, however, is not in line with the normative frame of these programmes. A lower degree of integration usually asks a higher transfer capacity of students and should be made transparent (GSC, 2013 ). Transfer capacity refers to a student’s ability to make sense of and link vocational and academic learning experiences. As research on these programmes in general and on pedagogical aspects in particular remains scarce, the challenge of integration is frequently discussed without specifying degrees, types or consequences for learners. Until now, the few studies on the educational quality of dual programmes mostly reported students’ perceptions (e.g. Krone, 2019 ; Nickel et al., 2018 ) and are often short of representative data (Langfeldt, 2018 ; Weiß, 2016 ). Commonly, research lacks pedagogical concepts of what integrating academic and vocational learning implies for the actual educational design of dual programmes (Faßhauer & Anselmann, 2021 ; Meyer, 2019 ).

In this paper, we investigate the differences in integration of academic and vocational learning. We focus on the pedagogical implications of different curricular characteristics, exploring the extent of variation in the degree of integration among dual study programmes. We contribute to the ongoing discourse on the design and the quality of dual programmes by first introducing a pedagogical perspective, where hitherto theory-based empirical contributions have focused on system-related aspects or structural issues (Graf, 2017 , 2018 ; Krone et al., 2019 ; Schiller & Leišytė, 2020 ). Second, we contribute by identifying different types of curricular integration in dual programmes. To this end, we propose indicators grounded in curriculum theory (e.g. Kelly, 2009 ) and apply them in a cluster analysis. To understand typical curricular types of design, we examine dual bachelor programmes in the most common subject areas: engineering and economics (Hofmann et al., 2020 ). We draw on a representative data set from the German Centre for Higher Education (CHE) on dual undergraduate programmes. This makes the study the most complete empirical contribution in this field. As a result, we present a curricular typology of integration for dual bachelor programmes. With our paper, we show how curricular design elements support learners in integrating academic and vocational learning experiences. By adding a pedagogical perspective on the state of integration in dual programmes, it provides a differentiated picture of the status quo in Germany and an analytical framework for international policy makers, researchers and dual programme designers.

Literature review

In the German education system, dual educational programmes have a long tradition in initial vocational training at the secondary level (Wolter, 2017 ). Situated in “the dual system of vocational education and training (VET)” (Baethge & Wolter, 2015 , p. 97), these vocational programmes combine “training on-the-job” and “school-based learning in vocational colleges” (Sloane, 2014 , p. 402). For many years, VET has been the backbone of the German educational system with entrants well above those to HE. This distribution changed over the last decades and in recent years entrants to HE have been higher than entrants to VET (Authoring Group Educational Reporting, 2020 ). Tertiary dual programmes must be perceived in the context of HE massification and the simultaneous “apprenticeship crisis” (Deißinger, 2006 , p. 181) in VET. One of the reasons for the crisis is companies’ declining interest in the VET system (Deißinger, 2006 ). While learners and companies increasingly ask for academic degrees, many still acknowledge the benefits of VET (such as immediate access to a job or trainees’ workforce), making dual programmes at the tertiary level a relevant phenomenon.

To grasp the variation in the degree of curricular integration among dual programmes, we have to consider the dual character of these programmes not only structurally but also in pedagogical terms. For several years now, scholars have noted that the discourse on how to design dual academic programmes lacks a focus on pedagogy (Faßhauer & Anselmann, 2021 ; Meyer, 2019 ; Mordhorst & Gössling, 2020 ), i.e. the relationship between structural decisions and the individual learner’s development. Until now, there has been no comprehensive theoretical framework in HE which focuses specifically on duality, i.e. the question of how to design the integration of academic and vocational learning. Such a framework, however, is necessary to delineate what duality or integration mean and which characteristics of these concepts are present in the curricular designs as well as the pedagogical practices of dual programmes. To develop theoretical categories which allow for a pedagogical understanding of duality and integration, first, we expound on the German dual VET system. Second, we refer to curriculum theory, deducing general curricular dimensions relevant for shaping educational experiences in dual programmes. Third, these dimensions are further qualified with reference to policy recommendations. Fourth, we refer to the state of research on curricular integration in dual programmes.

Duality in vocational and higher education

The German dual system of VET is highly regulated and complex. The 16 federal states are each responsible for designing syllabi for vocational schools, while training regulations for companies are the responsibility of the federal government (Sloane, 2014 ). Conceptually, the term dual system suggests that two systems collaborate to form a whole, which has been emphasised since the 1960s (Euler et al., 1999 ). Duality goes beyond governance or the organisational combination of two venues of learning and extends to the teaching and learning activities (Sloane, 2014 ). While the legal framework for VET emphasises this extensive duality, the degree to which it is implemented varies (Euler et al., 1999 ). Zabeck ( 1996 ) distinguishes between dualism and duality . Where the former suggests parallel activities of schools and companies with hardly any cooperation, the latter implies intense collaboration.

The VET-related discourse on duality provides basic categories for understanding degrees of integration between academic and vocational learning. At the same time, the transferability to dual study programmes has its limits (Faßhauer & Anselmann, 2021 ). This is largely due to the different regulatory settings of HE and VET and their consequences.

The German HE system is binary, comprising universities and UAS. The system is further characterised by academic freedom of teaching and research as well as relatively autonomous higher education institutions (HEIs). While the federal states exercise oversight and accreditation provides some quality control of study programmes (Kehm, 2018 ), there is no direct regulation regarding educational content and pedagogies.

From a regulatory perspective, there are two types of dual academic programmes, which differ in regard to the contracts students sign with corporate partners and the awarded degrees. In both programmes, students gain a bachelor’s degree and enrol in a HEI. In so-called training-integrating programmes students sign a contract with a company for a legally regulated profession and obtain an additional VET degree. In practice-integrating programmes students only enrol for the university degree. Besides their status as HE students, they have a contract for extended work placements with a corporate partner (GSC, 2013 ).

The German Standing Conference of the Ministers of Education and Cultural Affairs (GCMEC, 2017 ) defined the criteria for the accreditation of dual programmes in an exemplary statutory order. This statutory order (GCMEC, 2017 ) defines the following aspects as crucial dimensions for carrying the label dual: Besides the academic principle, the core element characterising dual studies is the combination of at least two venues of learning. It further indicates that duality in the sense of organisational linkages between the academic and the vocational venue of learning provides the basis for an integration that should extend to learning content . As opposed to VET on the secondary level the organisational and the curricular side of integration are defined as two connected, but separate elements which are not standardised.

Thus, in the context of dual academic programmes, the term duality refers to organisational linkages between HE and VET, remaining rather unspecific about content and pedagogies. It requires adequate concepts to investigate how learners are supported in linking and integrating the learning experiences in dual programmes. Curriculum theory can provide these concepts. The need for such an approach is underpinned by the fact that most dual programmes encompass some degree of organisational linking, while problems occur with the systematic integration of learning content (German Accreditation Council (GAC), 2020 ). The principles of academic freedom and autonomy of HEIs make a greater variety of approaches to integration more likely. Additionally, dual study programmes were never strongly supported by political initiatives (Nickel et al., 2018 ). It is up to each HEI to spell out how to accomplish duality and integration. To sum up, HEIs are granted much more autonomy in how to run their dual programmes than the VET institutions (Graf, 2017 ).

Curricular integration in higher education

Curriculum theory is concerned with how to develop broad educational objectives into a programme which provides experiences that help learners to achieve those objectives as best as possible (Kelly, 2009 ). Generally, the curriculum provides a structural framework (i.e. modules, workload, assessment modalities) as well as an institutional framework (i.e. expectations towards learners and educators) in which educational activities take place.

The literature defines common elements of curriculum to include goals, content, sequence, integration, and assessment. Goals define the outcomes expected from an educational programme (Diamond, 2008 ). Content refers to the subject matter selected to foster the goals (Lattuca & Stark, 2009 ). Sequence deals with the vertical relationships in the curriculum and the systematic organisation of content over time (Goodland & Su, 1992 ), for instance, over the entire length of a bachelor’s programme. Integration intends to evolve a curriculum in a way that its elements mutually contribute to a holistic learning experience (Goodland & Su, 1992 ). In the case of dual programmes, integration refers essentially to the integration of academic and vocational learning experiences. While the “ultimate integration is in the learner […] the process is aided presumably by the way in which the curriculum components are organized” (Goodland & Su, 1992 , p. 330). Thus, we can differentiate between personal integration and curricular integration . Assessment refers “to judgements about student learning” (Harlen, 2016 , p. 693).

These categories provide a frame of reference for analysing programmes. A pedagogical analysis of dual programmes would investigate how both the academic and the vocational dimensions become visible in the curricular design. Integration is a core category for investigating dual programmes as it refers to all the dimensions mentioned above and qualifies them in regard to the relationship of academic and vocational components. Here, the core question is (how) do curricular design elements support learners in integrating the academic and the vocational learning experiences?

Characteristics of curricular integration in dual study programmes

While curriculum theory provides overall categories for a curricular analysis of dual programmes, these dimensions need to be further qualified concerning the pedagogical implications of the dual character . The GSC ( 2013 ) suggests measures to systematise dual studies ranging from governance aspects via organisational measures to curricular elements. In Table 1 , we show those measures relevant for designating curricular integration and duality (organisational linking). We group the measures by the dimensions from curriculum theory. In addition, we differentiate between organisational and curricular integration—inherent in curriculum theory as well as in the policy discourse (Kelly, 2009 ; GSC, 2013 ).

Regarding organisational linking , the measures mentioned below are emphasised (GSC, 2013 ):

Stable contractual relations with corporate partners: cooperation is regulated through contracts with the corporate partners who train the students

Exchange in cooperation committees with actors from HEI and vocational institutions: regular meetings in committees can foster organisational and learning-related cooperation

Cooperation in admission processes: cooperation committees can support a joint admission which is otherwise often exclusively carried out by the corporate partner for the training part; in return, HEIs need to discuss their admission criteria with the corporate partners; the final decision on the admission must remain with the HEIs

With respect to curricular integration , the GSC ( 2013 ) provides the following measures:

Coordination of curricula and learning goals: this measure also encompasses decisions on sequencing and the integration of training by the corporate partner in the study plan

Comprehensive learning concepts, which encompass a regular exchange between actors in the academic and vocational field: such concepts entail, for instance, joint study projects offered by HEI and corporate partners

Special classes for dual students, if conventional and dual students study together in a programme: dual students must be supported through tailored modules in integrating the learning experience at the corporate partner

Recognition of phases at corporate partners as study experience, formalised through credit points: recognition must be anchored in the study plan

The introduced characteristics of curricular integration inform the analysis of dual programmes, as they qualify the meaning of the terms duality and integration .

Empirical evidence on curricular integration in dual study programmes

In this section, we review empirical research on dual programmes in Germany using Table 1 as a frame of reference to provide an overview of the state of research on characteristics relevant for our analysis. Theoretical concepts on dual studies are not reviewed as there are hardly any (Gerstung & Deuer, 2021 ; Mordhorst & Gössling, 2020 ; Schiller & Leišytė, 2020 ) and none refer to curriculum theory.

As far as we know, there is no empirical study investigating curricular integration as a core theme. Extant research often focuses on structural issues within systems and on an institutional level (Graf, 2017 , 2018 ; Krone et al., 2019 ; Schiller & Leišytė, 2020 ). While the measures in Table 1 are occasionally part of such studies, they have not been investigated from the viewpoint of curriculum theory.

Organisational issues with the linking of different institutions are frequently addressed (Kupfer et al., 2014 ; Langfeldt, 2018 ). A small-scale study on dual students ( n  = 14) found that around two thirds report a tight cooperation between workplace and HEI (Nickel et al., 2018 ). In a study on dual engineering and economics programmes, 30% of the companies interviewed ( n  = 292) reported to be part of at least one committee for cooperation. Using document analysis, the same study revealed that all dual programmes have some sort of cooperation committee (Langfeldt, 2018 ). Qualitative empirical evidence from case studies at UAS describes cooperation to be loose in many dual programmes. HEIs and corporate partners report cooperation structures to have a low degree of formalisation based on contracts, committees, etc. (Kupfer et al., 2014 ). Comparing the results of these two studies reveals that the question whether formal instruments such as committees are in common use has not been fully resolved yet. However, the publication dates of the studies suggest that there might have been some shift in recent years towards a wider use of committees.

An investigation on the potential of dual science, technology, engineering and mathematics programmes reported in a qualitative part that the level of combination of academic and vocational learning ranges from no coordination between HEI and corporate partner, via non-committal recommendations by the HEI to reflection modules for phases at the partner company. This description hierarchically arranges organisational linking and curricular integration, reducing integration to the structural aspect of modules. Staff from HEIs in this study pointed out that integration is especially hard to establish in those programmes which combine regular and dual studies (Wolter et al., 2014 ).

Kupfer et al. ( 2014 ), Langfeldt ( 2018 ) and Krone et al. ( 2019 ) found qualitative and quantitative evidence that dual programmes often lack curricular integration . Only a quarter of the students who participated in the online survey ( n  = 4125) by Langfeldt ( 2018 ) judged the combination of learning content in their study programme to be ‘good’ or ‘very good’. Likewise, Krone et al. ( 2019 ) report students ( n  = 9285) to be only partially content with the integration of academic and vocational learning in their dual programmes. 79% of the companies asked ( n  = 280) in an online survey by Kupfer et al. ( 2014 ) reported HEIs to be solely in charge of academic learning. 91% agreed that corporate partners have no support from HEIs when it comes to the vocational education. Thus, the empirical studies seem to confirm that curricular integration is a particularly difficult challenge (cf. GAC 2020 ).

To sum up, extant research lacks representative empirical data on dual programmes (Langfeldt, 2018 ; Weiß, 2016 ). Krone et al. ( 2019 ) seem to be an exception but do not address the topic of curricular integration as a core theme. Until now, there has been no systematic survey investigating the characteristics mentioned in Table 1 . Thus, a quantitative survey exploring the issue of curricular integration in dual programmes based on pedagogical criteria is a gap in the research landscape with important consequences, which has yet to be filled. Currently, challenges with integration are lamented, but their specific nature and consequences for learners are not specified. This paper analyses the differences in integration from an educational perspective. We investigate the following research question: How (and to what extent) does the degree and nature of integration among dual study programmes differ? Thus, we focus on curricular integration and organisational linking, as they are a prerequisites for personal integration. We propose indicators grounded in curriculum theory and apply them in a cluster analysis. As a result, we intend to develop a typology of curricular integration for dual bachelor programmes. Advancing the analysis of these programmes is relevant because more and more students enrol in these programmes (Mordhorst & Nickel, 2019 ). Their integrative approach is an innovation, since traditionally the two systems of HE and VET have been strictly separated (Baethge, 2006 ).

Sample and procedure

The research was conducted as a cross-sectional study on the integration of vocational and academic learning in dual programmes at German (dual) universities and UAS. The sample covers programmes in engineering and economics as these are the most common subjects of dual studies (Hofmann et al., 2020 ). The data was provided by the CHE. All university departments of dual programmes in engineering and economics registered in the database for study programmes by the German Rectors’ Conference (GRC) were asked to participate. Programmes in this database are “state-approved and/or accredited” (GRC, 2020 ). Accredited programmes not listed there were added manually to obtain a complete sample. The data was collected from May 2018 to January 2020 by the CHE. The response rate was 72.3%, equalling 219 questionnaires. Such a response rate is high for online surveys among university staff (Kaplowitz et al., 2012 ). In total, 66 cases missed data in single variables. We had to exclude these cases from the sample due to data requirements in SPSS for the method chosen in the final data analysis. One case was excluded during data preparation. Hence, a total of 152 cases were included in the final sample (see Table 2 ). The distribution by type of governing body and HEI reflects the general characteristics of the dual programmes in engineering and economics; also, the ratio of engineering and economics fits the real distribution (GRC, 2020 ). There is no specific data on the relative proportion of practice- versus training-integrating programmes in engineering and economics. However, data on this distribution is available for all subject areas together, which shows that regarding this criterion our data set is representative (Hofmann et al., 2020 ). Overall, we regard the sample as representative of dual programmes. The data was checked twice for plausibility. After the first plausibility check, the university departments were asked by the CHE to review their data and add missing variables. We carried out the second plausibility check after the survey was conducted, before proceeding to analyse the data. Additionally, we verified some variables manually comparing open-ended text items with the closed questions used for analysis.

In total, eleven variables were included in the analysis (see Table 3 ), all of which are binary. While some items were binary in the questionnaires already, others had to be downscaled to allow for the intended cluster analysis. The nominal characteristics were transformed to binary items using a binary split. We applied a median split to transform the continuous variable ‘number of students in study programme’ as the box plot for the variable showed many outliers. Thereby, we optimised discrimination without losing too much information or having to expect distortion when analysing our data, as the nominal items were ideal for a binary split.

The variables operationalise the characteristics of organisational linking and curricular integration. Most of the variables cover curricular dimensions due to the paper’s focus on curricular integration. All items were derived from the literature on dual programmes. Two organisational variables on duality are also included (sample item: ‘Is there a written agreement on the institutional exchange between HEI and the corporate partner?’). Nine items refer to curricular integration (sample item: ‘Does the study plan identify study content to be provided by the corporate partner?’).

We chose hierarchical cluster analysis (HCA) to group our cases as it fits binary, medium-size data sets and groups cases in a stage process without a fixed number of clusters. Cluster analysis is commonly used for typology studies in higher education (e.g. Bahr, 2010 ; Inkelas et al., 2008 ) and allows for a sample-driven approach. Providing the degree of distance among cases, cluster analysis groups similar cases into the same clusters. We decided against using two-step cluster analysis or k-means, as the algorithm for the two-step method is hardly documented and k-means asks to define the number of clusters beforehand. The survey data were analysed using SPSS Statistics 27.

We considered other methods for typology studies used in HE research such as latent class analysis (LCA; e.g. Custer & Akaeze, 2019 ; Denson & Ing, 2014 ) and factor mixture modelling (FMM; e.g. Hatch & Bohlig, 2016 ). While LCA is applicable with categorial data, it ignores heterogenous facets of data and is not sample- but model-dependent (Denson & Ing, 2014 ) and therefore not fit for our purpose. Similarly, FMM is a model-based approach (Hatch & Bohlig, 2016 ), and thus no option for our analysis either.

As cluster analysis is an exploratory approach, we tested different cluster solutions before finding a robust grouping. For different options, we changed the proximity measure and the merging algorithm. To test for outliers, we carried out an HCA twice, once using the Simple Matching Coefficient and then the Euclidian Distance as the proximity measures, applying both times the agglomerative merging algorithm Single Linkage. Two different proximity measures were used to yield robust results. Single Linkage was used as the merging algorithm, because it can identify outliers. We excluded one case from the sample based on the results. Before doing so, however, we carried out a manual plausibility check as well as a cluster analysis with the outlier (Rapkin & Luke, 1993 ), resulting in a clear decision to exclude it. For identifying clusters with mutually related variables in the final HCA we applied the agglomerative merging algorithm Complete Linkage. The Simple Matching Coefficient was chosen as the proximity measure, as it allows to simultaneously consider all unanimous variables whether an item applies or not. In a first step, the number of clusters was not determined beforehand, as we had no criteria for estimating how many clusters would make sense. We checked for plausibility and considered not only the dendrogram, but also the scree plot (see Appendix Fig. 2 ) to deduce a statistically meaningful number of clusters, identifying five clusters as the best number of clusters. In a second step, we processed the data once again with the same algorithm and proximity measure and the fixed number of five clusters resulting in a robust cluster solution with a high homogeneity within the clusters and high heterogeneity between the clusters. We compared the variables used for clustering between the resulting clusters to investigate the cluster properties. This investigation is like a validity check of the cluster solution. We used crosstabs as descriptive presentation, Chi-Square tests for checking significant difference and Cramer’s V as an effect size measure. We found out that all variables used for the clustering show statistically significant differences between the clusters (cf. Appendix Table 5 ). Additionally, we did further comparisons between the groups concerning relevant variables not included in the cluster analysis. Again, we used the procedure described above. With this analysis, we characterised the groups further and checked for possible differences between the clusters. This helped us to better understand the different groups and their defining characteristics.

In this section, we present the five-cluster solution. Each of the 152 programmes was assigned to one of five discrete clusters. Table 4 gives an overview of the clusters. It indicates the percentage of programmes which are fully dual (no regular students enrolled). The main integration characteristics refer to the attributes identified in the cluster analysis. A cluster’s degree of integration draws on its overall scoring in the analysis, referring to both categories of organisational linking and curricular integration.

Cluster A ‘Incoherent curricular integration in programmes mixed with non-dual programmes’:

Regarding organisational linking, this cluster is characterised by a relatively low share of contractual relations on institutional exchange between HEI and the corporate partner (60.0%). However, it has the highest share of collaborative admission of HEI and corporate partner (26.7%) of all clusters. In terms of curricular integration , e.g. regarding collaboration for integration, the cluster lacks transparency. For instance, the study plan does not provide content to be studied at the corporate partners. Yet, in more than half of the programmes, staff from the partners regularly teach in HEI and in 46.7% of the programmes students earn credit points for learning activities at the partner company. The cluster comprises only programmes where dual and regular students study together. In 93.3% of the programmes, dual students study partly in separate classes without regular students to foster curricular integration of vocational and academic content . Programmes in this cluster are mostly smaller with no more than 58 students (80.0%). Regarding the regulatory classification , most of the programmes (40.0%) can be studied both as practice- and training-integrating programmes. These programmes are called mixed programmes in the following. A typical programme in this cluster would be a management programme at a private UAS with some curricular integration elements, such as recognition of training at the corporate partner formalised through credit points. However, the cluster is incoherent because often the study plan does not indicate the recognised content.

Cluster B ‘Assessment-based curricular integration in programmes mixed with non-dual programmes’:

The cluster is characterised by a low degree of organisational linking , e.g. 19.0% of the programmes have contracts with the corporate partners on institutional exchange regarding their cooperation. Most variables also illustrate a lack of curricular integration . For instance, 9.5% of the programmes rely on agreements related to goals and sequence determining how to coordinate study content, classes and assessments. In the dimension content , only 9.5% of the programmes rely on special classes for dual students. This is unexpected, as in all programmes in this cluster dual and regular students study together and the number of dual students per programme is rather high. In 57.1% of the programmes, between 59 and 1270 students are enrolled. 90.5% of the dual programmes regulate the assessments of study content provided by the corporate partners and 85.7% coordinate assessments with them. Thus, integration in this cluster is essentially assessment-based. With respect to regulatory classification , programmes in this cluster primarily lead to a VET (71.4%) next to the academic degree. An example for a typical programme in this cluster would be a public UAS programme in mechanical engineering integrating a regulated VET degree in industrial mechanics mainly through assessment practice.

Cluster C ‘Clearly regulated, coherent curricular integration in fully dual programmes’:

Regarding organisational linking , dual programmes in this cluster are characterised by stable contractual relations (all programmes). The degree of curricular integration is also high: Related to goals and sequence , in 92.5% of programmes, agreements regulate how to coordinate study content, classes and assessments. Regarding collaboration for integration, content in the study plan is explicitly assigned to the corporate partners in 67.2% of the cases and 71.6% of the programmes have accredited modules at the partners. In no other cluster company staff teaches more often at the HEI (89.6%). Regarding content, most of the programmes are fully dual (77.6%). This cluster consists primarily of larger programmes with at least 59 students (70.1%). All programmes determine how to assess content provided by the corporate partners and coordinate assessment s with them. In 74.6% of the programmes, students earn credit points for learning activities at the partner corporation. Considering the regulatory classification , most of the programmes are practice-integrating (70.1%). Typical for this cluster would be a large, fully dual and coherent practice-integrating programme in business management at a public dual university or public UAS with a high degree of regulation on the organisational and curricular level.

Cluster D ‘Holistic, transparent curricular integration in programmes mixed with non-dual programmes’:

Programmes in this cluster have a relatively high degree of organisational linking , e.g. HEI and corporate partners collaborate in 78.6% of cases based on a written agreement. Variables reveal holistic curricular integration patterns. For instance, regarding collaboration for integration, in 85.7% of cases content is directly assigned to the corporate partners in the study plan and 78.6% report to have accredited modules at the partners’, both the highest scores in comparison. Regarding the content category, in 7.1% of the cases, students study in fully dual programmes. In 50.0% of programmes, students learn in separate classes without regular students. Regarding student numbers, the cluster is in the middle range: in 78.5% of programmes, there are 13 to 139 students. Coordination of assessments (96.4%) is common. It is the only cluster with a higher rate of coordination of assessments than guidelines (85.7%). Regarding regulatory classification , 39.3% of the programmes are mixed and another 35.7% lead to a vocational degree in addition to the academic degree. Typical for this cluster would be a public engineering programme designed as a dual training- (and practice-) integrating and as a regular programme with a transparent and holistic curricular integration. It could be provided by a UAS or a university.

Cluster E ‘Mainly organisational linking in programmes mixed with non-dual programmes’:

Regarding organisational linking , institutional cooperation is aided by written agreements in 66.7% of dual programmes and 19.0% of the programmes practice admission collaboratively–both intermediate scores. Results depict a low degree of curricular integration in this cluster. For instance, looking at the content , 14.3% of the programmes in this cluster are fully dual and 9.5% have separate classes, which is low considering the share of programmes combined with regular programmes. The cluster properties provide 76.2% smaller programmes (0 to 58 students). 23.8% have guidelines on assessing study content provided by the corporate partners, but none coordinate assessments with them (the overall lowest scores). With respect to the regulatory classification , 47.6% of programmes in this cluster lead to a vocational in addition to the academic degree. Typical for this cluster would be a public or private UAS or university programme in engineering offered as a regular and as a dual study path, lacking support for students with integration.

To sum up, the analysis yielded two clusters with a high degree of curricular integration: clusters C and D. Programmes in cluster A show some elements of organisational linking and curricular integration but lack transparency. Clusters B and E have a low degree of curricular integration.

Curricular integration of academic and vocational learning is both a core characteristic and a great challenge for designing dual programmes (Mordhorst & Gössling, 2020 ; Nickel & Püttmann, 2015 ). This study demonstrates how curricular design elements support learners in integrating academic and vocational learning experiences.

Overall, the typology developed in this study confirms that the degree of integration varies significantly between study programmes labelled as dual (cf. Krone, 2019 ; Langfeldt, 2018 ). Our analysis found different forms of curricular integration . As discussed, the concepts from VET are only partly applicable to dual academic programmes. The different institutional set-ups of dual academic programmes and VET have direct effects on the curriculum level. Duality in VET goes beyond the organisational combination of two venues of learning and extends to the learning activities through an extensive legal framework. Due to the principles of academic freedom and institutional autonomy of HEIs, legal regulations on duality in dual study programmes remain rather unspecific about content and pedagogies, asking for a differentiated perspective on organisational linking (duality) and curricular integration.

Therefore, the divergent types of curricular integration we identified could be placed on a continuum ranging from parallelism through duality to full curricular integration targeting students’ personal integration directly (see Fig.  1 ).

Continuum of curricular integration

Duality as the core category for organisational linking refers to the cooperation of HEI and corporate partners realised through measures such as committees. These measures are important to support strong structural links between the actors and usually have no direct impact on students. Yet, some of these measures affect students directly, such as admission procedures. Our analysis reveals that all programmes rely on some sort of organisational linking for the cooperation of HEI and corporate partners (cf. GAC 2020 ; see Fig.  1 ). Additionally, our results confirm the problem that there are programmes which hardly go beyond duality (cf. GAC, 2020 ; see cluster E), lacking curricular integration elements. A lack of curricular integration conflicts with the core promise of dual learning (GAC, 2020 ). In turn, programmes with a high degree of curricular integration also have stable contractual relations with cooperation partners (clusters C and D). This supports the assumption that organisational duality is the basis for curricular integration (cf. GSC, 2013 ). However, organisational duality does not guarantee curricular integration.

Regarding curricular integration aiming at personal integration, results indicate that establishing a systematic integration is more challenging in programmes combining regular and dual studies (cf. Wolter et al., 2014 ). Programmes mixing both—different types of dual programmes as well as dual and regular programmes—are especially difficult to design. While such a differentiation asks for a curricular response with measures such as specifically tailored modules for dual students, this is not common in all the clusters. Some programmes rather seem to follow a very economical approach, relying on similar study plans for regular and dual as well as for practice- and training-integrating programmes. A small number of modules are offered in different programmes, resulting in quality constraints and blurry programme profiles. For instance, purely assessment-based integration (cluster B) is not a form of systematic curricular integration (similar degree of integration as cluster E). Integrated assessment does not support students in the process of integrating their learning experiences. It expresses, however, an expectation for such intellectual integration. Additionally, measures such as separate classes for dual students, credit points for learning content provided by corporate partners or having corporate staff teach at the HEI is not necessarily a sign of quality, if students do not know which content is provided by the partner. Especially practice-integrating programmes without clearly regulated work placements are more likely to align their goals and content with company-specific rather than profession-specific skill demands (Graf, 2018 ). Hence, transparency of learning content and concepts is especially important in these programmes. However, there are programmes with an active role of corporate partners and no such transparency (see cluster A). In contrast, the high share of programme-specific agreements in cluster C with 70.1% practice-integrating programmes might be a positive sign, indicating that HEIs take on their responsibility in those programmes where the work-based part is not based on regulations in VET.

The results also illustrate that training-integrating programmes have a lower degree of curricular integration than practice-integrating programmes (cf. Langfeldt, 2018 ). However, there is no clear tendency for an overall low degree of integration, as all formal types of dual programmes can be found in all the clusters. Yet, the nature of integration varies according to the regulatory classification (see clusters C and D). Whether the regulatory classification grants much autonomy in designing the work-based part of the programme (practice-integrating programmes) or whether there are legal regulations standardising the vocational part of the programme (training-integrating programmes) affects the integration design.

A systematic integration—as conceptually defined for dual programmes—may be achieved through curricular measures on different levels, e.g. by designing programmes which are exclusively for dual students (cluster C) or by establishing programme structures in mixed setups that support the learning experience through additional elements specifically designed for dual students (cluster D). We assume that coherent curricular integration fosters personal integration. However, it does not guarantee an integrated learning experience. How students perceive learning in those programmes, which are formally characterised by a holistic curricular integration, is another research gap, yet to be addressed.

Limitations and future research

In the following section, we discuss the limitations of the study, its implications for future research and the generalisability of the findings. The study is limited in three ways. First, as the study is based on secondary data, after deducing criteria for analysis, we assigned fitting variables from our data set to the categories. Most of the items cover the criteria derived from the literature very well. However, as extant databases on dual programmes are not pedagogical, so are some of the items. This was problematic in one case: While our category from curriculum theory was goals and sequencing, as there was no better fit, the item we selected refers to content, classes and assessments, leaving goals and sequencing aside. Educational goals of a programme, however, are essential, as they often serve as a starting point to design sequencing and content. Therefore, our criteria derived from the literature review could inform future quantitative studies on curricular integration of dual programmes.

Second, the quantitative data says little on the concrete realisation of curricular integration including cultural aspects of teaching and learning (cf. James, 2014 ). Aiming at advancing integration, it could be interesting to address questions such as: What do comprehensive learning concepts look like? Which role do goals and sequencing play? How do institutional and programme cultures enable the cooperation between HEIs and partners? These questions can lead to the overall research question: What are design principles fostering curricular integration in dual programmes? To resolve the question, case studies are necessary.

Third, our study is limited regarding subjects and groups of persons questioned. While economics and engineering programmes make up most dual programmes and are thus representative, future research could show whether the typology fits for other subjects too. Computer science and health care programmes would be interesting to investigate, as they are other relevant subject groups in dual studies (Hofmann et al., 2020 ). While the study focusses reasonably on HEIs, it would be interesting to look at learners’ experiences regarding the insights from our analysis.

Despite these limitations, the study contributes to our understanding of curricular integration in dual programmes. Altogether, the study is valuable, because it yields a robust and differentiated typology on the integration of academic and vocational learning. It is the first pedagogical empirical study on German dual programmes in HE. By adding a differentiated view on the state of integration in dual programmes it provides new information for policy makers, researchers and dual programme designers alike. The insights are necessary for the ongoing process of institutionalisation (Graf, 2018 ) of dual programmes becoming a special type of many educational programmes in a differentiated HE system. The analytical framework can inform quality monitoring and policy making. Results are also relevant for students’ orientation. On an international level, the findings are fruitful for any dual programme regardless of the HE system, e.g. a professional degree programme on the masters’ level integrating job experience. Yet, those findings targeting organisational aspects depend on governance constellations and are therefore not easy to transfer.

Data availability

Data was provided by the CHE. The data is provided free of charge and on request for any independent research purpose by the CHE (contact: [email protected]).

Authoring Group Educational Reporting (2020). Education in Germany 2020 . Retrieved May 10, 2021, from https://www.bildungsbericht.de/de/resolveuid/6301fccbdac5499fa9454b6338c697d0

Baethge, M. (2006). Das deutsche Bildungsschisma: Welche Probleme ein vorindustrielles Bildungssystem in einer nachindustriellen Gesellschaft hat. SOFI-Mitteilungen, 34 , 13–27.

Google Scholar  

Baethge, M., & Wolter, A. (2015). The German skill formation model in transition: From dual system of VET to higher education? Journal for Labour Market Research, 48 (2), 97–112. https://doi.org/10.1007/s12651-015-0181-x

Article   Google Scholar  

Bahr, P. R. (2010). The Bird’s Eye View of Community Colleges: A Behavioral Typology of First-Time Students Based on Cluster Analytic Classification. Research in Higher Education, 51 (8), 724–749. https://doi.org/10.1007/s11162-010-9180-5

Custer, B. D., & Akaeze, H. O. (2019). A Typology of State Financial Aid Grant Programs Using Latent Class Analysis. Research in Higher Education, 62 (2), 175–205. https://doi.org/10.1007/s11162-019-09585-5

Deißinger, T. (2006). The apprenticeship crisis in Germany: The national debate and implications for full-time vocational education and training. In L. Mjelde & R. Daly (Eds.), Studien zur Berufs- und Weiterbildung: Vol. 3. Working knowledge in a globalizing world: From work to learning, from learning to work (pp. 181–196). Lang.

Denson, N., & Ing, M. (2014). Latent Class Analysis in Higher Education: An Illustrative Example of Pluralistic Orientation. Research in Higher Education, 55 (5), 508–526. https://doi.org/10.1007/s11162-013-9324-5

Diamond, R. M. (2008). Designing and Assessing Courses and Curricula: A Practical Guide (3rd ed.). Jossey-Bass.

Euler, D., Berger, K., Hertel, H.‑D., Höpke, I., Krafczyk, T., Walden, G., & Weber, H. (1999). Kooperation der Lernorte im dualen System der Berufsbildung: Bericht über eine Auswertung von Modellversuchen (Materialien zur Bildungsplanung und zur Forschungsförderung No. 73). Bund-Länder-Kommission für Bildungsplanung und Forschungsförderung.

Faßhauer, U., & Anselmann, S. (2021). Akademische Beruflichkeit in dualen Studiengängen - Entwicklungspfade aus der Berufsbildung. In J. Seifried, G. Sailmann, & A. Brodsky (Eds.), Praxispotentiale im Dualen Studium: Lernen am Arbeitsplatz als Element akademischer Qualifizierung (pp. 11–27). wbv Media.

German Accreditation Council (2020). Zwischenbilanz: Zwei Jahre neues Akkreditierungssystem (2018–2019). Ausblick bis Ende 2021 . Retrieved September 14, 2020, from https://www.akkreditierungsrat.de/sites/default/files/download/2020/AR%20Zwischenbilanz%202020.pdf

German Rectors‘ Conference (2020). Higher education compass . https://www.hochschulkompass.de/en/study-in-germany.html . Accessed 11 May 2021

German Science Council (2013). Empfehlungen zur Entwicklung des dualen Studiums: Positionspapier . Retrieved September 12, 2020, from https://www.wissenschaftsrat.de/download/archiv/3479–13.pdf

German Standing Conference of the Ministers of Education and Cultural Affairs (2017). Musterrechtsverordnung gemäß Artikel 4 Absätze 1 – 4 Studienakkreditierungsstaatsvertrag: Beschluss der Kultusministerkonferenz vom 07.12.2017 . Retrieved May 11, 2021, from https://www.akkreditierungsrat.de/sites/default/files/downloads/2019/Musterrechtsverordnung.pdf

Gerstung, V., & Deuer, E. (2021). Theorie-Praxis-Verzahnung im dualen Studium: Ein konzeptioneller Forschungsbeitrag. Zeitschrift Für Hochschulentwicklung, 16 (2), 195–213. https://doi.org/10.3217/zfhe-16-02/14

Goodland, J. I., & Su, Z. (1992). Organisation of the Curriculum. In P. W. Jackson (Ed.), Handbook of Research on Curriculum: A Project of the American Educational Research Association (pp. 327–344). Macmillan.

Graf, L. (2017). Work-based higher education programmes in Germany and the US: Comparing multi-actor corporatist governance in higher education. Policy and Society, 36 (1), 89–108. https://doi.org/10.1080/14494035.2017.1278872

Graf, L. (2018). Combined modes of gradual change: The case of academic upgrading and declining collectivism in German skill formation. Socio-Economic Review, 16 (1), 185–205. https://doi.org/10.1093/ser/mww044

Harlen, W. (2016). Assessment and the curriculum. In D. Wyse, L. Hayward, & J. Pandya (Eds.), The SAGE Handbook of Curriculum, Pedagogy and Assessment (Vol. 2, pp. 693–709). SAGE Publications.

Hatch, D. K., & Bohlig, E. M. (2016). An Empirical Typology of the Latent Programmatic Structure of Community College Student Success Programs. Research in Higher Education, 57 (1), 72–98. https://doi.org/10.1007/s11162-015-9379-6

Hofmann, S., Hemkes, B., Leo-Joyce, S., König, M., & Kutzner, P. (2020). AusbildungPlus − Duales Studium in Zahlen 2019: Trends und Analysen. Bundesinstitut für Berufsbildung. Retrieved June 3, 2020, from https://www.bibb.de/dokumente/pdf/06072020_AiZ_dualesStudium-2019.pdf

Inkelas, K. K., Soldner, M., Longerbeam, S. D., & Leonard, J. B. (2008). Differences in Student Outcomes by Types of Living-Learning Programs: The Development of an Empirical Typology. Research in Higher Education, 49 (6), 495–512. https://doi.org/10.1007/s11162-008-9087-6

James, D. (2014). Investigating the curriculum through assessment practice in higher education: the value of a ‘learning cultures’ approach. Higher Education, 67 (2), 155–169. https://doi.org/10.1007/s10734-013-9652-6

Kaplowitz, M. D., Lupi, F., Couper, M. P., & Thorp, L. (2012). The Effect of Invitation Design on Web Survey Response Rates. Social Science Computer Review, 30 (3), 339–349. https://doi.org/10.1177/0894439311419084

Kehm, B. M. (2018). Higher Education Systems and Institutions, Germany. In P. Teixeira & J. C. Shin (Eds.), Encyclopedia of International Higher Education Systems and Institutions 30 , 1–10. Springer. https://doi.org/10.1007/978-94-017-9553-1_369-1

Kelly, A. V. (2009). The Curriculum: Theory and Practice (6th ed.). Sage Publications.

Krone, S. (2019). Duales Studium aus der Perspektive der Studierenden. In B. Hemkes, K. Wilbers, & M. Heister (Eds.), Berichte zur beruflichen Bildung. Durchlässigkeit zwischen beruflicher und hochschulischer Bildung (pp. 462–478). Budrich.

Krone, S., Nieding, I., & Ratermann-Busse, M. (2019). Dual studieren − und dann? Eine empirische Studie zum Übergangsprozess Studium − Beruf dualer Studienabsolvent/inn/en. Study: Vol. 413 . Hans-Böckler-Stiftung.

Kupfer, F., Köhlmann-Eckel, C., & Kolter, C. (2014). Duale Studiengänge – Praxisnahes Erfolgsmodell mit Potenzial? Abschlussbericht zum Entwicklungsprojekt: Analyse und Systematisierung dualer Studiengänge an Hochschulen. Wissenschaftliche Diskussionspapiere: Vol. 152 . Retrieved May 11, 2021, from https://www.bibb.de/dienst/veroeffentlichungen/de/publication/download/7368

Langfeldt, B. (2018). Lernortkooperation im dualen Studium – zu viel oder zu wenig Einfluss der Hochschulen auf die betrieblichen Praxisphasen? Bwp@ Berufs- und Wirtschaftspädagogik (34), 1–20. Retrieved July 6, 2020, from http://www.bwpat.de/ausgabe34/langfeldt_bwpat34.pdf

Lattuca, L. R., & Stark, J. S. (2009). Shaping the college curriculum: Academic plans in context (2nd ed.). Jossey-Bass.

Meyer, R. (2019). Integration von Berufs- und Hochschulbildung − betriebliche und hochschulische Lernkulturen. In B. Hemkes, K. Wilbers, & M. Heister (Eds.), Berichte zur beruflichen Bildung. Durchlässigkeit zwischen beruflicher und hochschulischer Bildung (pp. 423–438). Budrich.

Mordhorst, L., & Gössling, B. (2020). Dual study programmes as a design challenge: Identifying areas for improvement as a starting point for interventions. EDeR. Educational Design Research , 4 (1). https://doi.org/10.15460/eder.4.1.1482

Mordhorst, L., & Nickel, S. (2019). Grenzenloses Wachstum? Entwicklung des dualen Studiums in den Bundesländern. (Arbeitspapier 212, CHE) Retrieved May 11, 2020, from http://www.che.de/downloads/CHE_AP_212_duales_Studium_Bundeslaendervergleich.pdf

Multrus, F., Majer, S., Bargel, T., & Schmidt, M. (2017). Studiensituation und studentische Orientierungen: Zusammenfassung zum 13. Studierendensurvey an Universitäten und Fachhochschulen . Retrieved June 4, 2020, from https://www.bmbf.de/upload_filestore/pub/Studierendensurvey_Ausgabe_13_Zusammenfassung.pdf

Nickel, S., & Püttmann, V. (2015). Qualitätsentwicklung im dualen Studium. Ein Handbuch für die Praxis . Stifterverband für die deutsche Wissenschaft.

Nickel, S., Püttmann, V., & Schulz, N. (2018). Trends im berufsbegleitenden und dualen Studium: Vergleichende Analyse zur Lernsituation von Studierenden und Studiengangsgestaltung . Study: 396. Hans-Böckler-Stiftung.

Rapkin, B. D., & Luke, D. A. (1993). Cluster analysis in community research: Epistemology and practice. American Journal of Community Psychology, 21 (2), 247–277. https://doi.org/10.1007/BF00941623

Schiller, B., & Leišytė, L. (2020). Study program innovation in the Triple Helix context: The case of cooperative study programs at a German University of Applied Sciences. Triple Helix, 7 (2–3), 160–188. https://doi.org/10.1163/21971927-bja10002

Sloane, P. F. (2014). Professional education between school and practice settings: The German dual system as an example: Chapter 15. In S. Billett, C. Harteis, & H. Gruber (Eds.), Springer International Handbooks of Education: Vol. 1. International Handbook of Research in Professional and Practice-based Learning (pp. 397–425). Springer.

Weiß, R. (2016). Duale Studiengänge - Verzahnung beruflicher und akademischer Bildung. In U. Faßhauer & E. Severing (Eds.), Berichte zur beruflichen Bildung: Vol. 16. Verzahnung beruflicher und akademischer Bildung: Duale Studiengänge in Theorie und Praxis (pp. 21–38). W. Bertelsmann.

Wolter, A. (2017). The expansion and structural change of postsecondary education in Germany. In P. G. Altbach, L. Reisberg, & H. de Wit (Eds.), Responding to massification: Differentiation in postsecondary education worldwide (pp. 100–109). Körber-Stiftung.

Wolter, A., Kamm, C., Lenz, K., Renger, P., & Spexard, A. (2014). Potenziale des dualen Studiums in den MINT-Fächern: Eine empirische Untersuchung (acatech Studie). Retrieved May 11, 2021, from https://www.acatech.de/publikation/potenziale-des-dualen-studiums-in-den-mint-faechern-eine-empirische-untersuchung/download-pdf?lang=de

Zabeck, J. (1996). Die dualistische deutsche Berufsausbildung als wissenschaftliche Herausforderung. In W. Seyd & R. Witt (Eds.), Materialien zur Berufsbildung: Vol. 6. Situation, Handlung, Persönlichkeit: Kategorien wirtschaftspädagogischen Denkens. Festschrift für Lothar Reetz (pp. 71–86). Feldhaus.

Download references

Open Access funding enabled and organized by Projekt DEAL. No funding was received for conducting this study.

Author information

Authors and affiliations.

Department of Business and Human Resource Education, University of Paderborn, Warburger Straße 100, Paderborn, Germany

Lisa Mordhorst & Tobias Jenert

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Tobias Jenert .

Ethics declarations

Conflicts of interest.

Financial or Non-financial interests: All authors certify that they have no affiliations with or involvement in any organisation or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript.

Employment: Lisa Mordhorst is employed by the non-profit CHE Centre for Higher Education in Gütersloh/Germany (which provided the data). The paper was essentially written during a sabbatical for research purposes at the University of Paderborn.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Scree plot of heterogeneity measure

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Mordhorst, L., Jenert, T. Curricular integration of academic and vocational education: a theory-based empirical typology of dual study programmes in Germany. High Educ 85 , 1257–1279 (2023). https://doi.org/10.1007/s10734-022-00889-7

Download citation

Accepted : 17 June 2022

Published : 15 July 2022

Issue Date : June 2023

DOI : https://doi.org/10.1007/s10734-022-00889-7

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Dual study programmes
• Curricular integration
• Curriculum theory
• Empirical typology
• Cluster analysis
• Find a journal
• Publish with us
• Track your research

Vocational education : Journals

• Career & Technical Education Research "Publishes articles dealing with research and research-related topics in vocational education."
• Career development and transition for exceptional individuals "Career Development and Transition for Exceptional Individuals (CDTEI) specializes in the fields of secondary education, transition, and career development for persons with documented disabilities and special needs. CDTEI focuses on the life roles of individuals as students, workers, consumers, family members, and citizens. Articles cover qualitative and quantitative research, scholarly reviews, and program descriptions and evaluations."
• Empirical research in vocational education and training "The main focus of this journal is to provide a platform for original empirical investigations in the field of professional, vocational and technical education, comparing the effectiveness, efficiency and equity of different vocational education systems at the school, company and systemic level. The journal fills a gap in the existing literature focusing on empirically-oriented academic research and stimulating the interest in strengthening the vocational part of the educational system, both at the basic and higher education level."
• Journal of counseling and development : JCD "Journal of Counseling & Development publishes practice, theory, and research articles across 18 different specialty areas and work settings. Sections include research, assessment and diagnosis, theory and practice, and trends."
• The Journal of human resources "The Journal of Human Resources is among the leading journals in empirical microeconomics. Intended for scholars, policy makers, and practitioners, each issue examines research in a variety of fields including labor economics, development economics, health economics, and the economics of education, discrimination, and retirement. Founded in 1965, the Journal of Human Resources features articles that make scientific contributions in research relevant to public policy practitioners."
• Techniques: Connecting Education & Careers "Provides news about legislation affecting career & technical education, in-depth features on issues & programs, profiles of educators & other newsmakers, notices of new products, and ACTE events."
• Vocations and learning "Vocations and Learning: Studies in Vocational and Professional Education provides an international forum for papers on the broad field of vocational learning, across a range of settings: vocational colleges, schools, universities, workplaces, domestic environments, voluntary bodies, and more. Coverage includes such topics as curriculum and pedagogy practices for vocational learning; the role and nature of knowledge in vocational learning; the relationship between context and learning in vocational settings; analyses of instructional practice and policy in vocational learning and education; studies of teaching and learning in vocational education; and the relationships between vocational learning and economic imperatives, and the practices and policies of national and trans-national agencies."
• << Previous: Recent print books
• Next: Databases >>
• Background information
• Recent e-books
• Recent print books
• Connect to Stanford e-resources
• Last Updated: May 26, 2022 10:57 AM
• URL: https://guides.library.stanford.edu/vocational_ed

• Search Menu
• Advance articles
• Author Guidelines
• Submission Site
• Open Access
• About European Sociological Review
• Editorial Board
• Advertising and Corporate Services
• Journals Career Network
• Self-Archiving Policy
• Dispatch Dates
• Journals on Oxford Academic
• Books on Oxford Academic

• < Previous

Vocational education, tertiary education, and skill use across career stages

• Article contents
• Figures & tables
• Supplementary Data

Wiebke Schulz, Heike Solga, Reinhard Pollak, Vocational education, tertiary education, and skill use across career stages, European Sociological Review , Volume 39, Issue 5, October 2023, Pages 741–758, https://doi.org/10.1093/esr/jcac074

• Permissions Icon Permissions

Vocational education enhances smooth transitions into the labour market. However, this initial advantage might vanish over the career and eventually turn into a disadvantage because the skills of vocationally trained workers become outdated faster. So far, research has examined this potential vocational trade-off by assessing labour market outcomes such as employment and income. This study uses a different approach, it directly examines how different types of skills used at work change over the career of vocationally trained workers compared to tertiary-educated workers, and how career events shape skill-use changes. With data from the German National Education Study (NEPS), we examine five skills use dimensions based on job-tasks measures: analytical, creative, managerial, interactive, and manual skills. We find that skill-use differentials between vocational and tertiary-educated workers are only small to modest. The clearest differences relate to analytical and manual skills. Looking across career stages, the observed skill-use differentials remain rather stable across career stages—thus, the vocational skill trade-off thesis is only partially supported. Occupational mobility and unemployment contribute to observable changes, whereas job-related further training does not. Our results challenge skill-based explanations of a vocational trade-off.

Email alerts

Citing articles via.

• Recommend to your Library

Affiliations

• Online ISSN 1468-2672
• Print ISSN 0266-7215
• Copyright © 2024 Oxford University Press
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

• Architecture and Design
• Asian and Pacific Studies
• Business and Economics
• Classical and Ancient Near Eastern Studies
• Computer Sciences
• Cultural Studies
• Engineering
• General Interest
• Geosciences
• Industrial Chemistry
• Islamic and Middle Eastern Studies
• Jewish Studies
• Library and Information Science, Book Studies
• Life Sciences
• Linguistics and Semiotics
• Literary Studies
• Materials Sciences
• Mathematics
• Social Sciences
• Sports and Recreation
• Theology and Religion
• Publish your article
• The role of authors
• Promoting your article
• Abstracting & indexing
• Publishing Ethics
• Why publish with De Gruyter
• How to publish with De Gruyter
• Our book series
• Our subject areas
• Your digital product at De Gruyter
• Contribute to our reference works
• Product information
• Tools & resources
• Product Information
• Promotional Materials
• Orders and Inquiries
• FAQ for Library Suppliers and Book Sellers
• Repository Policy
• Free access policy
• Open Access agreements
• Database portals
• For Authors
• Customer service
• People + Culture
• Journal Management
• How to join us
• Working at De Gruyter
• Mission & Vision
• De Gruyter Foundation
• De Gruyter Ebound
• Our Responsibility
• Partner publishers

Your purchase has been completed. Your documents are now available to view.

International Journal of Vocational Education Studies

• Type: Journal
• Language: English
• Publisher: transcript Verlag
• First published: March 1, 2024

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 18 January 2024

Dynamics of returns to vocational education in China: 2010–2017

• Jie Chen 1 &
• Francesco Pastore 2  

Humanities and Social Sciences Communications volume  11 , Article number:  118 ( 2024 ) Cite this article

991 Accesses

Metrics details

In this paper, we use the Chinese General Social Survey data (2010–2017) to analyze the dynamics of returns to different education qualifications. We find increasing returns for all types of education, with returns to vocational and academic education increasing roughly at the same speed. We additionally compare the returns to vocational education with returns to academic education. Compared to those who only complete compulsory education, upper secondary graduates earn about 20% more, vocational college graduates earn 50%, and academic university graduates earn 75% more. At tertiary level, academic education pays better than vocational education, although the difference shrinks over the years. At upper secondary level, the evidence is indeterminate, depending on different econometric techniques (i.e., OLS, IV, Lewbel method, or PSM). These findings add to the limited quantitative evidence on returns to vocational education. The dynamics emerged from the findings echo the discussion on labor market mismatch and overeducation in China, which has important policy implications.

Similar content being viewed by others

Worldwide divergence of values

Bayesian statistics and modelling

A national experiment reveals where a growth mindset improves achievement

Introduction.

The vocational education and training sector has been at the heart of recent economic reforms all over the world (Chappell 2003 ; Tran 2021 ). As the economy develops, more high value-added industries will substitute low value-added industries (Loyalka et al. 2016 ). Consequently, the demand for high-skilled workers grows. The positive relationship between economic development and demand for skilled workers has been observed in many developing countries in recent years. To better prepare future graduates for the job market, a country’s education system has to evolve accordingly. Vocational education thus plays a vital role in a country’s education system (Tran 2021 ; Patrinos et al. 2021 ).

As one of the world’s largest and most vibrant economic entities, China is no exception in upgrading its vocational education and training system. In China, vocational education and vocational training are different. Vocational education is school-based and earns a credential whereas vocational training is work-based and may or may not earn a credential. Our focus is on the former, which is a kind of education qualifications. Footnote 1

China’s rapid economic growth sees a dramatic increase in the demand of high-skilled workers (Heckman and Yi 2014 ). As a response, the State Council has enacted a series of policies (e.g., Decisions on Accelerating Development of Modern Vocational Education in 2014) aiming to improve the social status of skilled workers. However, as a result of a long-standing tradition of degrading vocational graduates, lower secondary graduates are reluctant to pursue a vocational credential unless their academic performance is too poor for an academic degree (Hansen and Woronov 2013 ). Figure 1 presents the share of individuals with different education qualifications in China over the period 2010–2017. From Fig. 1a , we see that the largest proportion of individuals only completed lower secondary education; around 14% completed academic upper secondary education while only around 9% completed vocational upper secondary education; from 2010–2013 to 2015–2017, there is a shift from lower education qualifications to higher education qualifications, reflecting an overall improvement in the quality of labor supply. Figure 1b shows that the story is similar when we zoom into the young cohort. Notably though, in 2015–2017, the share of academic university graduates surpasses the share of lower secondary graduates. The difference shows a trend that more well-educated employees are needed for China’s booming economy.

a All working age people, b Young cohort (age 24–35).

Despite the upsurge in the supply of academic university graduates, employers in China still face a supply shortage of skilled labors. This is for two reasons. On the one hand, curricula of academic universities are theory-oriented and do not really prepare students for skill-intensive jobs. On the other hand, academic university graduates prefer “white-collar” jobs and are reluctant to do “blue-collar” jobs. Consequently, the supply and demand of labor constantly mismatch. To alleviate the shortage of supply of skilled workers, the Chinese government has been actively promoting vocational education since the 1990s. Yet, skilled workers are still under-supplied in the labor market (Fudan University, Tsinghua University 2016 ). Two reasons likely have contributed to this perpetuating supply shortage. One reason is that the culture of degrading vocational graduates persists, discouraging parents from sending their child to vocational institutes in the first place. The other reason is that the returns to vocational education in the labor market is not high enough to attract sufficient students. In this paper, we speak to the second reason by looking at returns to different education qualifications, especially vocational education ones, from 2011 to 2017 in China. In education-related literature, dynamics means the changes of returns to education over a time period (Arias and McMahon 2001 ; Söderbom et al. 2006 ; Crespo Cuaresma and Raggl 2016 ). Specifically, we aim to answer the following research questions: What are the dynamics of returns to vocational education in China? What implications do the dynamics have on education policy and practice? On the one hand, we see high-quality vocational workers are in short supply. On the other hand, we see high-quality vocational workers are in great demand. Thus, understanding the changes of returns to vocational education in China has important and urgent policy and practical implications.

The research makes contribution to the dynamics of returns to vocational education in China – a rarely addressed issue in the following aspects. Along with this investigation, we provide an updated and more holistic picture of the returns to different education qualifications in China during 2010–2017, focusing on vocational education. Second, to cope with the notorious endogeneity issue in estimating returns to education (Card 1999 ; Woodridge 2010 ; Castellano et al. 2014 ), we adopt rigorous econometric techniques (i.e., conventional IV, heteroskedasticity-based Lewbel (Lewbel, 2012 ) IV) to compare the returns to vocational education and academic education, both at the upper secondary level and at the tertiary level. In doing this, we not only come up with a novel instrument (i.e., the proportion of university graduates relative to the entire population per year but also replicate the enrollment ratio IV as is used in Dai and Martins ( 2020 ). Although we are able to replicate Dai and Martins ( 2020 ) using the same IV and find similar “wage premium” enjoyed by vocational graduates as opposed to academic graduates using the same instrument, the “premium” disappeared when we try with the novel instrument or various econometric techniques as robustness checks. The fact that some findings are not robust to alternative econometric specifications serves as a cautionary reminder to exercise greater diligence in the selection of estimation techniques before drawing definitive conclusions. Additionally, we find evidence of increasing gender inequality in subsample analysis. Last but not least, we tie the discussion of the dynamics in the returns to different education qualifications with the ongoing vocational education reforms in China, providing international readers with an enriched picture of vocational education in China.

This paper is structured as follows. In Section “Literature review”, we explain the research context of the education system in China, as well as summarizing the related literature. In Section “Theoretical framework”, we provide a theoretical framework as a roadmap guiding the empirical analysis in this paper. In Section “Methodology”, we describe the data and discuss the methodology. Section “Findings and discussions” presents the results and discussion. We conclude in Section “Conclusion”.

Literature review

Vocational education has attracted much attention in recent decades, especially after the massification of higher education in developed countries and recently in some developing countries including China (Meng 2012 ; Mok and Marginson 2021 ). In this section, we first provide contextual explanation of the current education system in China and its potential implications on the changes of returns to vocational education. Then we summarize the literature studying the returns to education in China and position our paper in this specific strand of literature.

The education system in China

The current education system in China mainly consists of two stages. The first stage is compulsory and lasts 9 years - 6-year primary education plus 3-year lower secondary education. Footnote 2 The second stage is competitive and thus noncompulsory, including upper secondary level education and those above. Figure 2 illustrates the major progression paths in China’s education system. As can be seen, a student normally sits in the High-school Entrance Exam (HSEE) after completing the compulsory education (Davey et al. 2007 ). Depending on his/her performance in the exam, he/she may enter an academic upper secondary school or a vocational upper secondary school. Some scholars have argued the HSEE to be “a form of class sorting” in modern China as the majority of those who fail the HSEE are “destined to a life in the working class” (Woronov 2016 ). The upper secondary education normally lasts three years, after which students compete to enter tertiary level education or directly enter the job market. The path to tertiary education is distinct for academic students and vocational students, which is different from the education system in some other countries (Brunello and Rocco 2017 ). An academic upper secondary graduate has to attend the College Entrance Examination to enter tertiary education. Depending on his/her performance, he/she may end up in an academic university or a vocational college. Meanwhile, a vocational upper secondary graduate only needs to attend a transitional exam, which is normally carried out by individual vocational colleges, to progress to a vocational college. Meanwhile, a vocational upper secondary graduate may still choose to take the National College Entrance Examination if he or she prefers not to take the transitional exam. Although the latter path to tertiary education seems easier, it is less attractive to students for two reasons. First, the latter path eliminates the possibility of directly entering an academic university which is more highly valued in the current job market. Second, the latter path makes it hard to continue pursuing postgraduate degrees. Until 2019, a vocational college graduate cannot directly attend the Graduate Entrance Exam, a necessary path for general undergraduates to progress to postgraduate education. Clearly, vocational education was designed as a suboptimal option compared to academic education.

Education system in China.

Meanwhile, as the Chinese economy expand and grow, the Chinese government and China’s industries demand specialized, technically adept workers, meaning that vocational graduates face favorable policies (Schulte 2013 ). In 1996, the Chinese government enacted the Vocational Education Law of the People’s Republic of China, setting out official standards for vocational education in China. Since then, vocational education has received increased attention and resources from the government (Li et al. 2021 ). In 2005, the Chinese government announced Decision of the State Council on Vigorously Developing Vocational Education, which stimulated a rapid growth period of vocational education. For example, from 2005 to 2017 the national financial funds for vocational education reached 2.7 trillion yuan. As of 2017, the total investment in vocational education was about 434 billion yuan, nearly five times that of 93.9 billion yuan in 2005, with an average annual growth rate of 14%. In 2019, the State Council released the National Vocational Education Reform Implementation Plan, which further clarified the types and statuses of vocational education, and at the same time carried out pilot vocational education at the undergraduate level. In 2021, Opinions on Promoting the High-quality Development of Modern Vocational Education was initiated by the central government. In 2022, the Vocational Education Law of the People’s Republic of China was amended to formally recognize vocational education at the undergraduate and postgraduate level.

However, the effort to improve the situation of vocational education is counteracted by two main forces. First, China has a long tradition of discriminating vocational graduates (Schulte 2013 ). Even parents in rural areas are reluctant to send a child to a vocational institute unless getting into an academic upper secondary school is impossible (Hansen and Woronov 2013 ). As a result of the tradition, good students try to avoid vocational education if they can, which further lowers the quality of vocational graduates and perpetuates the vicious cycle (Woronov 2016 ). Second, the massification of higher education since 1999 escalates the supply of university/college graduates, both academic ones and vocational ones (Dai and Martins 2020 ). Both kinds of university graduates then have to compete in the same job market. Yet, until September 2019, vocational college graduates can only obtain diplomas or occupational qualification certificates but not degrees (Xiong 2011 ; Li et al. 2022 ), being disadvantaged even further. As the increasing rates of university graduates are much higher than those of positions requiring skilled workers, job market competition has soared, and academic graduates have had to compete with vocational graduates.

Returns to (vocational) education in China

In the existing education literature in China (and neighboring countries such as Japan and Korea), the focus has been largely on academic stream education but not vocational stream education (Hansen and Woronov 2013 ). More specifically, only a handful of studies quantitatively explore the returns to vocational education, especially vocational upper secondary education, in developing countries including China. The lack of relevant evidence is partly due to data unavailability (Guo and Wang 2020 ; Woronov 2016 ).

Among existing studies on this topic, the evidence is mixed. Using the 2003–2015 Chinese General Social Survey data on all workers, Dai and Martins ( 2020 ) compare returns to upper secondary vocational education versus returns to upper secondary academic education in China. The authors find that, although secondary vocational graduates do not enjoy wage premia over secondary vocational graduates on average in OLS regressions, secondary vocational education does result in more than 30% wage premia for individuals with average earnings potential when estimated by instrumental variable quantile regressions. Using CGSS survey data from 2008, Guo and Wang ( 2020 ) take propensity score approach and find that vocational upper secondary education attracts significantly higher private returns, especially for academically low-performing students, compared to academic upper secondary education. The overall premium of attending vocational education is around 8–8.4%. Loyalka et al. ( 2016 ) carried out a two-wave longitudinal study involving 10,071 first-year computer-majored students among 259 upper secondary schools in two provinces in China. They find that attending vocational upper secondary school negatively affects a student’s general skill by 0.30–0.44 SDs (as measured by math score) and does not significantly improve a student’s specific skills (as measured by computer skills). If anything, attending a vocational high school seems to detract a student from accumulating human capital. In the same vein, the authors also find that students in vocational schools are significantly more likely to drop out compared to students in academic schools, with a particularly high drop-out rate among low-income and low-ability students. A similar paper with a two-wave longitudinal study involving more than 12,000 students in one Chinese province has similar findings Yi et al. ( 2018 ). Using data of twins in urban China, Li et al. ( 2012 ) find that returns to upper secondary vocational education is around 22%, whereas return to academic upper secondary education is not statistically different from return of compulsory education. Their finding is in drastic contrast with the most recent findings from Dai and Martins ( 2020 ).

The mixed evidence in these studies can be attributed to several reasons. First, different datasets are utilized in various studies. For example, the data in Yi et al. ( 2018 ) are collected within one province in China whereas the data in Li et al. ( 2012 ) are based on urban twins, making their conclusions less generalizable. Wang and Wang ( 2023 ) only include post-secondary graduates from the CFPS dataset in their analysis, leaving out the individuals with secondary or below level of education. The data in Guo and Wang ( 2020 ) are the 2008 wave CGSS, which are collected more than 15 years ago and may not represent the current situation. By contrast, our study utilizes the more recent waves of the CGSS data and covers the years from 2010 to 2017. We also include individuals with all levels of education. In this sense, our study provides a more updated and holistic picture of the returns to education, especially vocational education, in China. Additionally, given the relatively long span of the data, we are able to gauge the dynamics of the educational returns over the years. Even if we assume the results based on the nationally representative samples (e.g., CGSS, CFPS) are largely comparable, the different estimated returns to education signal a potential shift in the supply and demand of vocational graduates due to the massive expansion and promotion of vocational education in the recent decades. All the above factors call upon further investigation regarding returns to vocational education, a gap that our research aims to fill in.

Theoretical framework

This paper is motivated by supply and demand theory from labor economics, with analysis guided by determinants of returns to education from the economics of education and discussions enriched by vocationalism-related theories. As was pointed out in several seminal works (Tinbergen 1975 ; Psacharopoulos 1989 ), the changes in returns to education are results of the race between technology and education. Specifically, technology development shifts the demand curve for educated labor to the right, although educational expansion shifts the supply curve of educated labor to the same direction as well, leaving the equilibrium returns to education undetermined and “an empirical matter” (Psacharopoulos 1989 ). In empirical studies, the returns to education are found to relate to a list of factors (more details are provided later in this section). Vocational education exhibits its uniqueness by virtue of its pronounced result-orientation nature and its proclivity to closely mirror economic fluctuations. In subsequent paragraphs, we discuss each of the three strands in greater details, linking to China’s specific context.

In labor economics, a competitive labor market will reach its equilibrium wage and equilibrium employment when labor supply equals labor demand (Borjas and Van Ours 2010 ). However, due to endless shocks in a modern industrialized economy, the labor market is unlikely to reach a stable equilibrium. Rather, wages and employment tend to be reaching new equilibria. As can be seen from Fig. 3 , labor supply and labor demand jointly affect the returns to education while supply and demand are both constantly affected by other factors. Labor supply is affected by the educational background of the labor force, whether the education system is vocationalism in its policies and practices, the education policies implemented by the government, and the technological advancement in the economy. For example, although the demand of skilled workers has upsurged in China after the Reform and Opening-up policy in 1978, the returns to education may not have risen accordingly in recent decades due to the supply boost of highly educated labor force by the higher education massification. Labor demand is more affected by the latter two factors, i.e., policy intervention of the government and technological advancement, because the educational background of individual workers and the vocationalism practice of educational institutes is unlikely to directly affect the demand side.

Theoretical framework.

Returns to investment in education can be affected by various factors. Early works in the economics of education focus on the relationship between education investment and labor outcomes. Controlling for individual demographics is of vital importance in estimating the effects of education (McCaffrey et al. 2004 ; Castellano et al. 2014 ). Such factors include gender, social economic status, marital status, union membership etc. (Angrist and Krueger 1992 ; Woodridge 2010 ; Castellano et al. 2014 ). In addition to these generic demographics, returns to education are also affected by heterogeneous factors that vary across contexts. For example, hukou is an influencing factor which is unique to China. Thus, we need to take hukou into account when estimating the returns to education.

Vocationalism is particularly suitable when discussing the dynamics of returns to education to China, because the changes in returns to education in China largely resonate the country’s tremendous economic development and industrial upgrading. Vocationalism means that the design and operation of the educational system reacts to the operation of the economic system (Bills et al. 2016 ; Xiong 2011 ; Xiong 2013 ). More specifically, vocationalism involves an array of practices and policies aiming to maximize the occupational value of schooling, by meeting different skill requirements of employers. It encompasses, but is not limited to, vocational education. In the context of this paper, all different levels of education speak to the demand of the employers in some way. For the older generations who were born when education was not compulsory, they may enter the job market with very limited (or even no) education. At that time, China was rather underdeveloped with recovering agriculture and early-stage industry, both of which were labor-intensive and did not require much professional knowledge of the labor force. By contrast, those who were born after the one-child policy may not enter the job market until their late 20 or 30 s due to overeducation and heavy investment into the single child by the entire family (Li et al. 2008 ). The overeducation situation was exaggerated by the higher education expansion around 2000, adding uncertainties to the returns to education. At this stage, China’s industry has seen huge growth and is becoming a leading power in many sectors worldwide, attracting talents from all over the globe. Consequently, educational requirement of the labor force has drastically increased while the returns to education are becoming more dispersed.

In sum, the dynamics of returns to education, especially vocational education, in China is a multi-disciplinary matter that does not induce a definitive answer. Theories across various disciplines are considered and discussed to inform us on the various forces that contribute to the dynamics. Figure 3 provides a holistic theoretical framework which depicts the interplay among the different theories that eventually affect the returns to education. The holistic theoretical framework in this section can be used as a guide for reading the empirical evidence.

Methodology

Data description.

This paper utilizes quantitative data from the Chinese General Social Survey (CGSS), one of the nationally representative surveys in China. It is a repeated cross-sectional survey project that was launched by the Renmin University of China in 2003. The survey adopts a multi-stage stratified sampling method, covering households from all 31 mainland provinces. Footnote 3 Starting in 2010, a modified sampling method is adopted to collect a more representative sample nationwide. In this paper, we analyze six annual survey datasets collected from 2010 to 2017 (except for 2014 and 2016, due to data unavailability). The 2017 data was released in 2020 and is the latest dataset available. After dropping observations without annual wage entries, we are left with a total of 27,545 observations. Table 1 summarizes the main variables of interest. The average hourly wage across all six years is 15.391 yuan, consistent with existing literature (Asadullah and Xiao 2020 ; Dai and Martins 2020 ).

In the summary statistics, the years of education is constructed by mapping the Chinese education system onto the International Standard Classification of Education Footnote 4 . Note that because it typically takes three years to complete vocational education at tertiary level and four years to complete academic education at tertiary level in China, the years corresponding to vocational education are one year shorter than those of academic education. Following the convention in the literature (Kang et al. 2021 ), vocational tertiary education is referred to as vocational college and academic tertiary education is referred to as academic university throughout this paper. In our econometric analysis, we treat the education qualifications as dummies. The actual years of education are irrelevant when we focus on the two types of upper secondary education, both of which typically last three years. The years of education is controlled when we investigate other education qualifications.

The average years of work experience for a waged worker are 29.08. Female composes about 41.2% of the sample. Ethnically, only less than 9% are from a minority group. 60.3% of the sample have an agricultural hukou . 15.0% of the sample are union members, whereas 11.6% are communist party members. The average years of education are 5.72 for father and 3.97 for mothers. 82.6% of the sampled are married and 10.5% remain single. As for education qualifications, more than 56.3% of the sample only attended compulsory or less education. 8.0% finished vocational upper secondary education, whereas 13.1% finished academic upper secondary education. Roughly the same proportion - around 10.7% - of vocational college graduates and academic university graduates are observed in the sample. Only 1.3 out of 100 completed postgraduate and above level of education. In terms of occupation sector, more than 33.9% of the sampled subjects are in the agricultural sector. More than 38.3% work in privately owned enterprises. 20.2% work in state owned enterprises, followed by collectively owned enterprises, foreign funded enterprises, and Hong Kong, Macau or Taiwan funded enterprises. Location-wise, 37.6% are from the eastern provinces, 26.5% are from the western ones, 23.1% are from the central, and 12.8% are from the northeast provinces. Footnote 5

Figure 4 shows the relationship between change in employment rate and the share of individuals with certain credential. Each dot corresponds to an industrial sector categorized following ISCO2008 two-digit codes. Footnote 6 In all four figures, the vertical axis represents the sector-wise change in employment from 2010 to 2017. The horizontal axis captures the share of tertiary graduates (in figures a and c) or upper secondary graduates (in figures b and d). Figure 4a , b are for all the working age subjects, whereas c and d are for the young cohort aged 24–35. Comparing the figures horizontally, we see that the majority of sectors cluster around the left end in a and the distribution spreads more evenly in b. In other words, in most of the sectors, upper-secondary credential holders are relatively common whereas university degree holders are rare (mostly occupying less than 10% of the population in a sector). Comparing the figures vertically, the share of tertiary graduates is less clustered for the young cohort as opposed to the entire working age group, indicating that workers with tertiary degrees are more common in most sectors; the share of upper secondary graduates are more clustered on the right end for the young cohort as opposed to the entire working age group, indicating upper-secondary credential holders also become more common. Additionally, the figures manifest one constantly expanding sector (i.e., 52 - Sales Workers) and one constantly shrinking sector (i.e., 1 - rural agriculture). The expansion of the sales sector echoes the massive growth of Chinese market, especially the housing market, in the past decade. The compression of the rural agriculture sector corroborates the well-known rural-urban migration in China.

a All working age subjects - tertiary graduates, b All working age subjects - upper secondary graduates, c Young cohort (age 24-35) - tertiary graduates, d Young cohort (age 24–35) - upper secondary graduates.

Baseline estimation

To estimate the returns to different education qualifications across different years, we apply baseline estimation strategy in this section. Because we are also interested in comparing the returns to same-level academic and vocational education, we apply various other techniques to address endogeneity problem and identify causality in the comparison, which simultaneously serve as robustness checks of the baseline comparison. These additional techniques are explained in Section “Robustness check techniques”.

Our baseline estimation strategy is the extended Mincerian equation (Mincer 1974 ; Duraisamy 2002 ; Psacharopoulos 1994 ). The classical semi-logarithmic functional form is as follows.

where Y is the natural logarithm of hourly working income. edu i is a vector of dummies corresponding to different education qualifications, with the baseline group being those who have completed the 9-year compulsory education at most. X is a list of controlled covariates including years of work experience, experience-squared, gender, marital status, ethnicity, hukou status, union membership, party membership, the occupation sector, and the province/location in which one completed the survey. Specifically, province fixed effect is controlled in the OLS regression. In all later regressions, provinces are grouped into four regions to avoid over-controlling dummies. We also control for the interaction between marital status and gender, as existing literature has demonstrated marital status affects men and women’s earnings differently (Antonovics and Town 2004 ; Juhn and McCue 2017 ; Chen and Pastore 2021 ). We additionally control for parents’ years of education, as Card ( 1999 ) noted that family background measures such as parental education typically positively affect earnings.

When education qualifications are controlled as dummies, scholars typically utilize the following formula in calculating the yearly returns corresponding to a specific education qualification (Duraisamy 2002 ).

where r i represents the yearly return rate to education qualification edu i . β i and β i −1 are the returns to education qualification at the i th and i  − 1th level. t and t  − 1 are the years of schooling corresponding to their respective qualifications. In our case, both the vocational and the academic upper secondary education typically last three years. Both types of upper secondary education also share the same baseline group (i.e., compulsory and below level of education). Thus, comparing the statistical difference between r i and r i −1 is equivalent to comparing β i and β i −1 directly. Clearly, if one is to obtain the yearly returns, one still needs to convert β i to r i .

Robustness check techniques

As robustness checks of the baseline estimation, three different econometric techniques are applied: the instrumental variable method, the Lewbel method, and the propensity score matching method Footnote 7 . Each method adds insights to our baseline results.

The Instrumental Variable (IV) method is helpful since the Durbin and Wu–Hausman test weakly rejects the hypothesis that the choice of vocational upper secondary education against academic upper secondary education is exogenous. IV method is often susceptible to the choice of IVs. To provide robust IV estimates, we try several seemingly feasible IVs in this paper and compare their estimates. The first IV is the proportion of university (including colleges) graduates over the total population by year. Figure 5 plots the proportion of university graduates over the total population from year 1978 to 2019. We see that the proportion of university graduates increased slowly before 2000 and surged after that. The dramatic change is due to the higher education expansion in 1999 (Meng 2012 ; Meng et al. 2013 ; Zhong 2011 ). This variable is exogenous in that any individual decision cannot affect the proportions over the years. This is particularly true if we think about the year in which the expansion policy is announced: the change in proportions is mainly out of a policy shock. The variable is relevant in that the changes in the proportion of university graduates will relate to the changes in upper secondary graduates. Hence, the proportion of university graduates over the total population is a valid instrument.

Proportion of university graduates relative to total population.

The second instrument is the upper secondary enrollment ratio, the number of vocational upper secondary students enrolled over the number of lower secondary graduates, by province. Enrollment ratio has been used as instrument in Dai and Martins ( 2020 ) and Chen and Pastore ( 2021 ). To construct the enrollment ratio, we first extract the number of lower secondary graduates and the number of students enrolled in vocational upper secondary schools from the Educational Statistics Yearbook of China from 1987 to 2015. Then we calculate Eratio vus , which is the average of province-level enrollment ratios each year. A province-level enrollment raio is defined as the number of students enrolled in vocational upper secondary schools in year t in a province divided by the number of lower secondary graduates in year t in the same province. A graphical representation of the Eratio vus can be found in Fig. 6 . As a robustness check, we have also replicated Fig. 7 in Dai and Martins ( 2020 ), which can be found in Fig. 7 in the Appendix. The quota is relevant in that it can predict the probability of a student entering the vocational upper secondary education in a specific province in a given year. The quota is exogenous because there are strong reasons to believe that enrollment ratio does not affect one’s income several years later (Chen and Pastore 2021 ).

Enrollment ratio.

Enrollment ratio (replication of Fig. 7 in Dai and Martins 2020 ).

In the third IV specification, we apply the conventionally used parental education and compare the results across three specifications. The purpose of doing this is to benchmark our results to more conventional specifications and facilitate understanding of our results. Similarly, the four IV is a dummy which equals one if the enrollment year to upper secondary education is after 1985 and zero otherwise. This IV is identical to that in Guo and Wang ( 2020 ). The rational of this IV is that a policy shock that supposedly boosts enrollment to vocational upper secondary education is announced in 1985. The IV is included for comparison purpose as well.

In addition to the conventional IV method, we also apply a heteroskedasticity-based instrumental method introduced in Lewbel ( 2012 ); Baum and Lewbel ( 2019 ). This method can utilize the heteroskedasticity in the error term of the reduced form equation to construct instruments when no external instruments are available. It can also serve as a robustness check and validity test of external instruments (Baum and Lewbel 2019 ).

When comparing the returns to vocational upper secondary and academic upper secondary education, one challenge is that students from those two tracks may differ in various characteristics before they enter the upper secondary education. For example, boys may prefer vocational education while girls may prefer academic education, or rural students may prefer vocational education while urban students prefer academic education. To maximize the possibility of comparing the like with the like, we also apply propensity score matching (PSM) method. The PSM method assigns observations into the treatment group and the control group, then it estimates propensity scores for each observation based on pre-treatment characteristics, and comparisons are done between observations with closely matched propensity scores. In this paper, nearest neighbor matching is used, meaning for each treatment observation, one control observation with the closest propensity score is chosen for comparison. In our case, subjects are matched based on experience, gender, ethnicity, hukou , parental education, union membership, and party membership.

Findings and discussions

Dynamics of returns to different education qualifications.

First, we find that the overall dynamic is that the returns to all types of education, both vocational and academic ones, have an increasing trend in the period from 2010 to 2017. Table 2 presents the OLS estimates of the extended Mincerian equation. Noting that we have controlled for a rich set of variables, we perform the multicollinearity test using the Variance Inflation Factor. No multicollinearity is detected. Overall, compared to the baseline group who only complete compulsory education, those who complete vocational upper secondary education earns 24.5% more, those who complete academic upper secondary education earns 19.5% more, those who complete vocational college education earns 51.4% more, those who complete academic university earns 75.3% more, and those who complete postgraduate and above level of education earns 108.7%. Clearly, the returns to education increase with the level of education qualifications. From 2010 to 2017, returns to all types of education are increasing. Yet, returns to upper secondary education increase more substantially (vocational: 25.1% Footnote 8 , academic: 77.5%) than returns to tertiary and above levels of education (vocational: 24.1%, academic: 10.0%, postgraduate and above: 16.0%).

Second, we compare the dynamics of returns for different types of education. We find that, at the upper secondary level, returns to vocational education increase more slowly than returns to academic education; at the tertiary level, vocational education and academic education increase roughly at the same speed. Comparing vocational education with academic education, we see that, returns to vocational upper secondary education (24.2%) are higher than returns to academic upper secondary education (19.5%). As for whether the difference is statistically significant, we relegate the comparison to Section “Comparison of the dynamics of returns to vocational andacademic education”. Looking through the six years, returns to academic upper secondary education seem to increase faster than returns to vocational upper secondary education.

Third and to uncover further nuances, we explore the dynamics of returns in subgroups with different gender, union status, hukou status, ethnic group, and occupation. From the coefficients corresponding to marital status, we see there is almost a “wage transfer” from married females to married males. Specifically, if one is married and male then he gets a 28% increase in wages compared to a single male, whereas a married female gets a 21% decrease in wages compared to a married male. From 2010 to 2017, males seem to enjoy an increasing marriage premium from 23.2% to 30.9%, grown by more than a third. This implies that males are potentially even more highly valued in the Chinese labor market nowadays than ten years ago, a strong signal of workplace gender inequality. This finding is in line with a recent meta-analysis on gender wage gap in China (Iwasaki and Ma 2020 ), which states that in recent years the gender wage gap in China has grown rapidly.

Looking further down the table, we find that union membership no longer brings additional returns starting from 2015. Those with agricultural hukou tend to suffer a certain level of wage penalty. The minority ethnic groups are paid comparably to the Han ethnic group, indicating no ethnical discrimination. Party membership is related to a lower wage premium of 7% only when all the years are pooled together. Returns are generally positively related to parental education, in line with the mainstream literature. Compared to the baseline agricultural jobs, all five types of enterprises pay substantially higher wages across all the years, implying that the payment situation in agriculture has not improved much over the years. Probably, this is also why the Chinese government initiated the “rural revitalization ( xiangcunzhenxing )” strategy in late 2017.

Technically, because our OLS estimation is essentially a log-linear transformation of an exponential function, we can interpret the coefficients as semi-elasticity. In other words, the coefficients measure the percentage change in the hourly wage in response to any unit change in the independent variable (Pastore and Verashchagina, 2006 ). Because the education qualifications are dummies, the semi-elasticity interpretation can be more precise when transformed via ( e β  − 1) × 100 (Halvorsen et al. 1980 ). By its nature, the difference between the original estimates and the transformed coefficients gets larger with the coefficients increasing. The transformed coefficients are reported in Table 3 . By and large, all the values based on the original semi-elasticity interpretation hold. However, we note the imprecision in the original estimates. As expected, the larger the coefficients, the more imprecise the semi-elastic measures are. For example, the returns to postgraduate and above level education is 108.7% in Table 2 but almost doubled to 196.6% in Table 3 , underlying the importance of noting this caveat in future research.

Comparison of the dynamics of returns to vocational and academic education

As mentioned in Section “Dynamics of returns to different education qualifications”, simple OLS regression may suffer from endogeneity issue, when estimating the returns to education. Thus, we apply various econometrics techniques as robustness check to the main findings.

Specifically, we do pairwise comparison of the returns to vocational education and those to academic education, both at the upper secondary level and at the tertiary level. Table 4 presents the coefficient difference test results of the transformed OLS coefficients, based on delta method. The coefficient difference is calculated as follows.

From Panel A, we see that returns to vocational upper secondary education are 6.2% higher than returns to academic upper secondary education. The positive difference is in line with Dai and Martins ( 2020 ) in which a 20% wage premium is captured among vocational upper secondary graduates as opposed to academic graduates. Yet our estimated difference is much smaller in magnitude (6%) than that in Dai and Martins ( 2020 ). This is partly due to the fact that we use transformed coefficients for the education qualification dummies, which presumably generates a more precise measure of the difference. Since both vocational and academic upper secondary education typically last three years, the yearly difference is around 2.1%. Panel B shows that the difference in returns to vocational college and academic university is much more substantial, vocational college graduates earning 45.3% less than academic university graduates. As explained before, most vocational college programs last three years and academic university programs last four years. When comparing the yearly difference, different denominators need to be used. The yearly difference is around 6% for the entire sample, comparable in magnitude to the benefit of obtaining an additional vocational upper secondary credential. Because the difference is both economically and statistically strongly significant, it is unsurprising that the pattern remains stable across the years. Notably though, the yearly return difference between vocational college education and academic university education keeps shrinking from 2010 to 2017. This again matches the higher education massification over the past two decades and the labor market mismatch in China - academic university degree is losing its additional charm slowly in the labor market and vocational college credential is catching up.

Meanwhile, comparisons are done for the difference between the two types of upper secondary education when all the years are pooled together. Panel A of Table 5 presents the IV estimates using the four sets of instrument variables discussed in Section “Methodology”. Because both types of upper secondary education programs last three years, we redefine the main treatment variable as a dummy which equals zero for academic graduates and equals one for vocational graduates. The first instrument is the proportion of university graduates by year, which is novelly constructed by the authors. The second instrument is the enrollment ratio, similar to the one in Dai and Martins ( 2020 ). The third set of instruments are parental education, which is conventionally used in the returns to education literature. The fourth set of instruments is identical with the one in Guo and Wang ( 2020 ), which is included for comparison purpose as well. Footnote 9 Odd-numbered columns provide the F statistic of the first stage. Even-numbered columns present the coefficient of interest estimated from the second stage. As a rule-of-thumb, an acceptable F statistic of the first stage should be greater than 10 (Staiger and Stock 1997 ). As can be seen, only our novel instrument has an F statistic larger than 22, well above the conventional threshold of 10 for a qualified instrument. Interestingly, with the novel instrument, we find that vocational graduates suffer a wage penalty of close to 70% compared to academic graduates. With the other three sets of instruments, the estimated coefficients are positive. However, the weak first stage or failing to pass the Durbin test cast doubt on the validity of these positive estimates. In sum, the IV estimation instrumented by the proportion of university graduates provides evidence that vocational graduates are actually paid less well compared to academic graduates.

To further check the robustness of our IV estimates, we utilize the Lewbel method and the PSM method. The definition of the treatment variable and the external instruments in the Lewbel model are the same as those in the corresponding IV specifications. Note that the Lewbel method additionally incorporates all the remaining control variables as internal instruments. Results from the Lewbel regressions are reported in Panel B of Table 5 . We see that all four specifications pass the Breusch-Pagan test with strong statistical significance, validating the usage of the Lewbel method. As expected, the results instrumented by the proportion of university graduates (Model 1) remains negative and statistically significant. Results in the other three specifications are insignificant, raising further concern towards instrumenting vocational upper secondary graduates using its enrollment ratio.

Turning to the PSM estimates in Panel C of Table 5 , there is a lack of statistical significance which may again result from the dramatic reduction in the number of observations. This in itself has interesting implications, suggesting that the overall samples are very different from each other. In other words, the difference in returns to education between vocational and academic education is essentially a matter of the different samples. This is in line with the idea that unfortunately people going into vocational education are the worst in terms of skills and talent. This adds to the story, is not in contrast with the story. Meanwhile, the PSM result is interesting when compared with the positive difference which was found in the OLS estimates. One possibility is that the alternative to vocational education for academically low performing students is academic upper secondary education, but they perform even worse in this case by moving to academic education. Without matching, we may have underestimated the returns to academic upper secondary education because a subgroup of academic students are the would-be low performing vocational students. With matching, we have a like for like comparison. Now that the three different econometric techniques generate three drastically different answers, we conclude that the return difference between vocational and academic upper secondary education cannot be determined as far as the current paper goes. On a related note, readers will benefit from reading this paper jointly with the paper by Dai and Martins ( 2020 ) for the following reasons. First, both studies use the same CGSS dataset, although ours additionally includes a new round. Second, the second IV specification in the current study is almost identical to the one in Dai and Martins ( 2020 ) (see Fig. 7 for a reproduction of Fig. 7 in Dai and Martins ( 2020 )). Third, the OLS estimates in both studies show that vocational upper secondary graduates earn a wage premium over academic upper secondary graduates. Fourth, Dai and Martins ( 2020 ) find no difference via IV estimations whereas we find a negative difference using a stronger IV. Last but not least, we attempt Lewbel and PSM methods to further investigate the overall difference (albeit without obtaining a determinant answer) whereas Dai and Martins ( 2020 ) switch their focus onto local differences utilizing quantile regressions.

To sum up, from the comparisons of vocational and academic education, we have the following main findings. First, academic university education degree pays around 45% better than vocational college education credential, meaning each additional year of academic university education earns 6% more than each additional year of vocational college education. This indicates the higher level of higher education can get higher returns than higher education at a lower level since vocational college education is located at junior college (zhuanke) level. Second, from 2010 to 2017, the payment premium of academic university education shrinks constantly. After 2015, yearly returns to academic university education no longer strongly differ from those to vocational college education. It seems that, with increasingly strong policy support, higher vocational education has gained major development and has started to gain social credibility and acceptance. The indication is that vocational education is gaining higher social status. Meanwhile, although OLS estimates show that vocational upper secondary graduates enjoy a 6% wage premium over academic upper secondary graduates, the difference does not survive alternative specifications (IV, Lewbel, PSM). In subsample analysis, we also find evidence that married men enjoy increasing wage premium compared to married women during 2010–2017. This signal of enlarging gender inequality merits further investigation, which is beyond the scope of the current paper.

In this paper, we investigate the dynamics of returns to different education qualifications in China. We also compare the returns to vocational education and returns to academic education, both at the upper secondary level and at the tertiary level.

We have the following main findings. First, compared to the baseline group with compulsory education and below levels of education, upper secondary graduates enjoy a wage premium of around 20%, vocational college graduates enjoy around 50% premium, academic university graduates enjoy 75.3% premium, and those who complete postgraduate and above levels of education more than double their returns. Second, compared to academic upper secondary education, vocational education earns a wage premium of around 6% according to OLS estimates, although not robust to alternative specificaitons. Third, compared to academic university education, vocational college education pays 45.3% less in returns, both economically and statistically strongly significant. As mentioned before, this is evidence that higher levels of higher education pays better than lower levels of higher education. Hence, a system extending to undergraduate level may help improve returns of higher vocational education. In fact, since 2014, China has started to expand higher vocational education to undergraduate levels. Fourth, during the period from 2010 to 2017, returns on investment in education at all levels exhibited an upward trend. Notably, returns to upper secondary education witnessed a more substantial increase compared to those of tertiary and higher levels, relative to their respective baselines, signaling the after-effect of higher education massification after the millennium. Last but not least, we find that males enjoy an increasing marriage premium over the study period.

Alongside the new findings, the novel econometric techniques we adopt in this paper will hopefully enrich the toolbox for researchers quantitatively studying vocational education. First, a novel instrument (i.e., the proportion of university graduates relative the entire population over the years) is introduced to account for endogeneity in the choice of vocational education against academic education at the upper secondary level. Evidence shows that this instrument outperforms the conventional parental education instruments, the enrollment ratio instrument which was used in Dai and Martins ( 2020 ), and the policy shock dummy in Guo and Wang ( 2020 ). Second, heteroskedasticity-based Lewbel method is used to fully utilize both the externally constructed instruments and internal instruments (i.e., the full set of control variables or a subset). Third, propensity score matching method is applied to match vocational students with academic students who have similar characteristics, so that self-selection bias can be partialled out to a certain extent. Nonetheless, the paper has its limitations. One is our sample captures people across all the working age. People in different generations may face completely different social and economic conditions. Readers need to bear this caveat in mind when interpreting the results. Another is that our sample size is not so large relative to the 1.4 billion population in China.

Although we do not reach a definitive answer when comparing vocational and academic upper secondary education, it is undoubtful that vocational college graduates are paid significantly lower than academic university graduates. Meanwhile, we observe that vocational education graduates are increasingly getting better paid in recent years. On the one hand, the payment disadvantage facing vocational graduates can be explained by the specific context in China. The Chinese government’s early effort in encouraging vocational education is countered by the country’s long-standing tradition of degrading vocational graduates. This conflict puts vocational graduates into a seemingly vulnerable position compared to academic graduates. On the other hand, the increasing social credibility, and consequently returns, of vocational education demonstrates the achievement of the increasing policy support by the Chinese government. Footnote 10 In the foreseeable future, the Chinese education sector will continue paying more attention to vocational education and training. Understanding how the returns to vocational education change as China transits its industry can better prepare other countries that face a similar situation.

Data availability

The survey data in this paper is from the research project “Chinese General Social Survey (CGSS)” of the National Survey Research Center (NSRC), Renmin University of China. Accessible via the link http://cgss.ruc.edu.cn/English/Home.htm .

The Chinese government has specific requirements regarding the design of vocational education programs. First, a total of 19 areas (corresponding to 1349 majors), which cover all sectors of the economy, are set by the Professional Catalogue of Vocational Education issued by the Ministry of Education. Design of vocational majors should be within the catalogue. Second, individual curricula should be designed following the national guidelines. Third, practical training should be no less than 50% of the total credit hours of the vocational education credential.

In some places, the primary education takes 5 years and the lower secondary education takes 4 years, still summing up to 9 years in total. For more detailed discussion of China’s education system, please refer to the working paper version of this paper at:

A detailed description of the sampling method used can be found here.

The mapping follows Chen and Pastore ( 2021 ). Specifically, compulsory education = 9 years, upper secondary education = 12 years, vocational college = 15 years, academic university = 16 years, master’s education and above = 19 years.

The geographical classification of locations is according to the National Bureau of Statistics in China.

except sector number 1, which correspond to rural agriculture , distinct from the commonly understood agriculture, as a result of the Chinese hukou system. hukou is a registration system in China that separates its population into rural and urban populations, restricting the former’s mobility and access to state-sponsored benefits and welfare.

For detailed discussion of the econometric techniques, please refer to the working paper version at this link:

As an example, the ratio is calculated as the following: \(\frac{{0.289 - 0.231}}{{0.231}} = 25.1\%\) .

Actually, we tried another instrument based on the 2015 policy which aimed at developing vocational education vigorously. Yet, because that policy is rather recent and only a small proportion of the sample is affected, the estimates were too imprecise to be informative. Hence, the fifth IV is dropped.

China gradually started its transition from low value-added industry to high value-added industry in recent years. In 2015, China’s President Xi initiated the “supply-side reform”, encouraging the industry to improve the quality of production and optimize its efficiency. Shortly after this, the “Made in China 2025” strategy was announced by Premier Li Keqiang in his Government Work Report. Traditionally, “Made in China” tends to be associated with cheap and low-quality goods. The “Made in China 2025” aims at transforming the Chinese industry so that products of China will be linked to technology-intensive and high-quality goods. The introduction of the “supply-side reform” and the “Made in China 2025” strategy has seen an increasing demand of high-quality skilled workers, followed by heavy investment and extensive media coverage on vocational education institutions.

Angrist JD, Krueger AB (1992) The effect of age at school entry on educational attainment: an application of instrumental variables with moments from two samples. J Am Stat Assoc 87(418):328–336

Article   Google Scholar  

Antonovics K, Town R (2004) Are all the good men married? uncovering the sources of the marital wage premium. Am Econ Rev 94(2):317–321

Arias O, McMahon WW (2001) Dynamic rates of return to education in the US. Econ Educ Rev 20(2):121–138

Asadullah MN, Xiao S (2020) The changing pattern of wage returns to education in post-reform China. Struct Change Econ Dyn 53:137–148

Baum CF, Lewbel A (2019) Advice on using heteroskedasticity-based identification. Stata J 19(4):757–767

Bills DB, Graf L, Protsch P (2016) Vocationalism. In: A. Furlong (ed.) Routledge handbook of youth and young adulthood (2nd edn.). New York, NY: Routhledge, pp 129–136

Borjas GJ, Van Ours JC (2010) Labor economics. McGraw-Hill/Irwin, Boston, p 144

Google Scholar  

Brunello G, Rocco L (2017) The effects of vocational education on adult skills, employment and wages: What can we learn from piaac? SERIEs 8(4):315–343

Card, D (1999) The Causal Effect of Education on Earnings. In: O. Ashenfelter, D. Card (eds) Handbook of Labor Economics, Volume 3a. Amsterdam; New York and Oxford: Elsevier Science, North-Holland, pp 1801–1863

Castellano KE, Rabe-Hesketh S, Skrondal A (2014) Composition, context, and endogeneity in school and teacher comparisons. J Educ Behav Stat 39(5):333–367

Chappell C (2003) Researching vocational education and training: where to from here? J Vocat Educ Train 55(1):21–32

Chen J, Pastore F (2021) “Study hard and make progress every day”: Updates on returns to education in China. GLO Discussion Paper Series No. 787. Global Labor Organization (GLO)

Crespo Cuaresma J, Raggl A (2016) The dynamics of returns to education in Uganda: National and subnational trends. Dev Policy Rev 34(3):385–422

Dai L, Martins PS (2020) Does vocational education pay off in China? instrumental-variable quantile-regression evidence. GLO Discussion Paper Series No. 495. Global Labor Organization (GLO)

Davey G, De Lian C, Higgins L (2007) The university entrance examination system in China. J Furth High Educ 31(4):385–396

Duraisamy P (2002) Changes in returns to education in india, 1983–94: by gender, agecohort and location. Econ Educ Rev 21(6):609–622

Fudan University, Tsinghua University (2016) New skills at work. Available at: https://news.fudan.edu.cn/_upload/article/files/41/1f/0abd108e4bb592d339bda3dcaf80/a1f429e3-9191-4194-bc5f-cbacbf7b0313.pdf

Guo D, Wang A (2020) Is vocational education a good alternative to low-performing students in China. Int J Educ Dev 75:102187

Halvorsen R, Palmquist R et al. (1980) The interpretation of dummy variables in semilogarithmic equations. Am Econ Rev 70(3):474–475

Hansen MH, Woronov TE (2013) Demanding and resisting vocational education: a comparative study of schools in rural and urban China. Comp Educ 49(2):242–259

Heckman JJ, Yi J (2014) Human Capital, Economic Growth, and Inequality in China. In The Oxford Companion to the Economics of China, edited by Shenggen Fan et al., p 459– 64. Oxford: Oxford University Press

Iwasaki I, Ma X (2020) Gender wage gap in China: a large meta-analysis. J Labour Mark Res 54(1):1–19

Juhn C, McCue K (2017) Specialization then and now: Marriage, children, and the gender earnings gap across cohorts. J Econ Perspect 31(1):183–204

Kang L, Peng F, Zhu Y (2021) Returns to higher education subjects and tiers in China: evidence from the China Family Panel Studies. Stud High Educ 46(8):1682–1695

Lewbel A (2012) Using heteroscedasticity to identify and estimate mismeasured and endogenous regressor models. J Bus Econ Stat 30(1):67–80

Article   MathSciNet   Google Scholar  

Li B, Li Z, Tang L (2022) 职业本科学位设置: 类型要义、主要争议及路径选择[Vocational undergraduate degree setting: type essence, main disputes and path selection]. 现代教育管理 [Mod Educ Manag] 383(02):119–128

Li F, Morgan WJ, Ding X (2008) The expansion of higher education, employment and over-education in China. Int J Educ Dev 28(6):687–697

Li H, Liu PW, Zhang J (2012) Estimating returns to education using twins in urban China. J Dev Econ 97(2):494–504

Li Z, Zheng J, Xiong J (2021) Examining project-based governance of higher vocational education in China: a case study. Higher Educ Policy 36:1–20. https://doi.org/10.1057/S41307-021-00251-Z

Loyalka P, Huang X, Zhang L, Wei J, Yi H, Song Y, Shi Y, Chu J (2016) The impact of vocational schooling on human capital development in developing countries: evidence from China. World Bank Econ Rev 30(1):143–170

McCaffrey DF, Lockwood JR, Koretz D, Louis TA, Hamilton L (2004) Models for value-added modeling of teacher effects. J Educ Behav Stat 29(1):67–101

Article   PubMed   PubMed Central   Google Scholar  

Meng X (2012) Labor market outcomes and reforms in China. J Econ Perspect 26(4):75–102

Meng X, Shen K, Xue S (2013) Economic reform, education expansion, and earnings inequality for urban males in China, 1988–2009. J Comp Econ 41(1):227–244

Mincer J (1974) Schooling, experience, and earnings. New York: Columbia University Press

Mok KH, Marginson S (2021) Massification, diversification and internationalisation of higher education in China: critical reflections of developments in the last two decades. Int J Educ Dev 84:102405

Pastore F, Verashchagina A (2006) Private returns to human capital over transition: a case study of belarus. Econ Educ Rev 25(1):91–107

Patrinos HA, Psacharopoulos G, Tansel A (2021) Private and social returns to investment in education: the case of turkey with alternative methods. Appl Econ 53(14):1638–1658

Psacharopoulos G (1989) Time trends of the returns to education: cross-national evidence. Econ Educ Rev 8(3):225–231

Psacharopoulos G (1994) Returns to investment in education: a global update. World Dev 22(9):1325–1343

Schulte B (2013) Unwelcome stranger to the system: Vvocational education in early twentieth-century China. Comp Educ 49(2):226–241

Söderbom M, Teal F, Wambugu A, Kahyarara G (2006) The dynamics of returns to education in Kenyan and Tanzanian manufacturing. Oxf Bull Econ Stat 68(3):261–288

Staiger D, Stock JH (1997) Instrumental variables regression with weak instruments. Econometrica , 65:557–586

Tinbergen J (1975) Income distribution analysis and policies. North-Holland, Amsterdam

Tran C-TTD (2021) Efficiency of the teaching-industry linkage in the australian vocational education and training. Empir Res Vocat Educ Train 13(1):1–28

Wang G, Wang Z (2023) Vocational education: a poor second choice? A comparison of the labour market outcomes of academic and vocational graduates in China. Oxf Rev Educ 49(3):408–427

Wooldridge JM (2010) Econometric analysis of cross section and panel data, 2nd edn. Cambridge, MA: MIT Press

Woronov TE (2016) The high school entrance exam and/as class sorter: working class youth and the HSEE in contemporary China. Y. Guo (ed) Handbook on class and social stratification in China, Edward Elgar Publishing, pp. 178–196, https://doi.org/10.4337/9781783470648.00020

Xiong J (2011) Understanding higher vocational education in China: vocationalism vs confucianism. Front Educ China 6(4):495–520

Xiong J (2013) Institutionalization of higher vocational education in China: a neoinstitutionalist perspective. Front Educ China 8(2):239–265

Yi H, Li G, Li L, Loyalka P, Zhang L, Xu J, Kardanova E, Shi H, Chu J (2018) Assessing the quality of upper-secondary vocational education and training: evidence from China. Comp Educ Rev 62(2):199–230

Zhong H (2011) Returns to higher education in China: what is the role of college quality? China Econ Rev 22(2):260–275

Download references

Acknowledgements

This paper is part of the Global Labor Organization (GLO) Virtual Young Scholars Program. Special thanks go to Klaus F. Zimmermann and Almas Heshmati for their suggestions that enriched this paper substantially. We also thank the program director, Olena Nizalova, and all the advisors and participants of the 2020/2021 GLO VirtYS program for their suggestions to the development of the paper. The survey data in this paper is from the research project “Chinese General Social Survey (CGSS)” of the National Survey Research Center (NSRC), Renmin University of China. This work is supported by the following projects/grants: (1) Senior Talent Foundation of Jiangsu University, Project No. 22JDG001; (2) the 14th Five-Year Plan Project of Educational Science in Jiangsu Province (Project title: Evaluation of the Effect of Jiangsu Vocational Education Reform Policy under the Background of Talent Power Strategy in the New Era, Project No. C-c/2021/01/68); and (3) Jiangsu Province University Philosophy and Social Sciences Research Project (Project title: Research on the effect evaluation and optimization path of vocational education policies in Jiangsu Province, Project No. 2022SJYB2199).

Author information

Authors and affiliations.

School of Finance & Economics, Jiangsu University, Zhenjiang, China

Department of Economics, University of Campania “Luigi Vanvitelli”, Capua (Caserta), Italy

Francesco Pastore

You can also search for this author in PubMed   Google Scholar

Contributions

JC conceived of the original idea, conducted all the formal analysis, acquired the funding, and wrote the original draft. FP supervised JC and provided insightful feedback throughout the process. Both authors discussed the results and contributed to the final manuscript.

Corresponding author

Correspondence to Jie Chen .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Ethical approval

Ethical approval was not required. This article does not contain any studies with human participants performed by any of the authors.

Informed consent

This article does not contain any studies with human participants performed by any of the authors.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Chen, J., Pastore, F. Dynamics of returns to vocational education in China: 2010–2017. Humanit Soc Sci Commun 11 , 118 (2024). https://doi.org/10.1057/s41599-024-02616-2

Download citation

Received : 24 January 2023

Accepted : 08 January 2024

Published : 18 January 2024

DOI : https://doi.org/10.1057/s41599-024-02616-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

• Skip to main content
• Keyboard shortcuts for audio player

Many in Gen Z ditch colleges for trade schools. Meet the 'toolbelt generation'

Windsor Johnston

Sy Kirby dreaded the thought of going to college after graduating from high school. He says a four-year degree just wasn't in the cards for him or his bank account.

"I was facing a lot of pressure for a guy that knew for a fact that he wasn't going to college," Kirby says. "I knew I wasn't going to sit in a classroom, especially since I knew I wasn't going to pay for it."

Instead, at the age of 19, Kirby took a job at a local water department in southern Arkansas. He said the position helped him to develop the skills that helped him start his own construction company.

Sy Kirby, who runs his own construction company, says a four-year degree just wasn't in the cards for him or his bank account. Will Anderson hide caption

Sy Kirby, who runs his own construction company, says a four-year degree just wasn't in the cards for him or his bank account.

Now at age 32, Kirby finds himself mentoring many of his employees, who also opted to learn a skilled trade rather than shelling out tens of thousands of dollars to pursue a degree that they wouldn't use after graduating.

Kirby says blue-collar work is lucrative and allows him to "call the shots" in his life. But, he says the job also comes with a downside, mainly because of the stigma attached to the industry.

"I think there's a big problem with moms and dads coming home from quote-unquote 'dirty' jobs. Coming home with dirty clothes and sweating. You had a hard day's work and sometimes that's looked down upon," he says.

High-paying jobs that don't need a college degree? Thousands of them sit empty

Kirby is among the growing number of young people who have chosen to swap college for vocational schools that offer paid, on-the-job training.

Skilled trades make a comeback

Lisa Countryman-Quiroz is the CEO of JVS, or Jewish Vocational Service, a nonprofit in San Francisco that provides career training for unemployed workers to find jobs, including in skilled trades. She says that over the years there has been a shift — with skilled trade making a comeback, especially among members of Generation Z.

"Folks have really prioritized a college education as a path to the middle class and a path to a cushy office job." But, Countryman-Quiroz says, "over the last 10 to 15 years, we are seeing a trend among young people opting out of universities. Just the crushing debt of college is becoming a barrier in and of itself."

More than half of Gen Zers say it's possible to get a well-paying job with only a high school diploma, provided one acquires other skills. That's according to a survey by New America, a Washington Think Tank that focuses on a range of public policy issues, including technology, education and the economy.

The driver of the big rig one lane over might soon be one of these teenagers

The high cost of college prompts a change in career paths.

In addition, the Education Data Initiative says the average cost of college in the United States has more than doubled in the 21st century.

With that price tag increasing, many Gen Zers say they've been left with no choice but to leave the college path. Many say living with their parents until they can pay off their college debt isn't an option.

The Indicator from Planet Money

Do i need a four-year degree.

Nitzan Pelman is founder of Climb Hire, a company that helps low-income and overlooked people break into new careers. She says many young people say graduating from college with a six-figure debt is a non-starter.

"It's not a secret that the cost of college has gone up so dramatically in the last decade that it's really cost prohibitive at this point," she says.

Pelman says pursuing skilled trades can also help "level the playing field," especially for young people from less-privileged backgrounds and for people of color.

Construction boom helps fuel job gains in March

"We don't see a lot of Black men in construction, but more Latino men in construction and you don't see many women in construction. Social capital is a really big gatekeeper and a door-opener for accessing high-quality jobs and helping people break into certain industries," she says.

In 2021, President Biden signed a $1.2 trillion bipartisan infrastructure bill. Since then, he's been traveling the country promoting the law, which he says will open up thousands of new jobs in trades.

Comparing college costs to the amount a student expects to earn after graduation

"you can expect to get your hands dirty and that's ok".

The high cost of college isn't the only factor driving many young people toward skilled trades. With the use of artificial intelligence on the rise, many Gen Zers see manual labor as less vulnerable to the emerging technology than white-collar alternatives. They also say vocational schools are a straight path to well-paying jobs.

Pelman says increasing salaries and new technologies in fields such as welding, plumbing and machine tooling are giving trade professions a face-lift, making them more appealing to the younger crowd.

"There are a lot of vocational jobs out there that are pretty attractive — HVAC repair and installation, electricians, solar panel installer — there's so much demand for wind turbine installers who, in many cases, make more than $100,000 a year — so there's a lot of demand for manual labor," she stresses.

Diego Aguilar works at a trade center at East Bay Municipal Utility District in Oakland, Calif. Marla Aufmuth/JVS hide caption

Diego Aguilar works at a trade center at East Bay Municipal Utility District in Oakland, Calif.

That was the case for 25-year-old Diego Aguilar, who says a traditional desk job was out of the question for him. Aguilar now works full time at a trade center at East Bay Municipal Utility District in Oakland, Calif., after going through the JVS training program.

"When I went into a trade program I learned how much money I could make performing a very specific kind of work. You need mechanics, you need machinists, you need carpenters, operators you need painters. You can expect to get your hands dirty and that's OK," Aguilar says.

Jobs Friday: Why apprenticeships could make a comeback

Data from the National Student Clearinghouse Research Center shows the number of students enrolled in vocational-focused community colleges increased 16% from 2022 to 2023.

As for Kirby, he says his mission is to keep raising awareness about what he calls the "toolbelt generation."

"Where they can walk out of the school of hard knocks, pick an industry, work your 10 years, take your punches, take your licks and hopefully you're bringing jobs and careers back to the community," he says.

When asked if he regrets his decision to go into skilled trades, Kirby chuckles. "Not for a second," he says.

Millions in funding for Kansas higher education leveraged against DEI practices

Diversity, equity and inclusion in higher education could see serious changes in the coming academic year.

As Democratic Gov. Laura Kelly weighed what to do with anti-DEI legislation passed by the Republican-led Legislature, the Kansas Board of Regents voted earlier this month to limit DEI practices at state universities.

Kansas lawmakers passed two anti-DEI bills

Lawmakers passed House Bill 2105, which prohibits postsecondary institutions from making hiring, application or student aid decisions based on support or opposition statements about DEI or "any political ideology or movement."

The bill doesn't prevent universities from upholding anti-discrimination laws or infringe on the educational freedoms of faculty to conduct research on DEI. If an institution violates the law, they will owe up to $10,000 per violation and will pay all the investigation fees.

Additionally, lawmakers added a contingency plan to the state budget.

They inserted a budget proviso into Senate Bill 28 that would require university presidents to stand before legislative leaders, confirming their universities will uphold the requirements of the bill. As part of the provision, if university leaders failed to do so, $35.7 million in funding would be cut from the state budget for higher education.

But, the budget proviso in SB 28 wouldn't apply if the legislation in HB 2105 became law.

Regents changed diversity policy as lawmakers targeted DEI

As Kelly weighed what to do with both pieces of legislation, the Regents finalized its own action on diversity, equity and inclusion.

KBOR voted earlier this month to implement an amendment to the board policy statement on diversity and multiculturalism. The board's change aligns its DEI policy with the legislative intent in HB 2105.

When KBOR voted on the amendment, board chair Jon Rolph said work had been going on for the last year or so and taking this vote would simply formalize the policy. He also said it would be congruent with the state Legislature's plans.

"This is, again, our good faith effort in trying to listen to the Legislature as they listen to the people," Rolph said. "This concern they brought up is not something central to our practices around wanting student success and trying to fulfill our promises to people when we invite them onto our campuses."

While KBOR could change its policy on DEI again in the future, the change will now be more permanent because HB 2105 is now law.

Gov. Laura Kelly allowed anti-DEI bill to become law

While HB 2105 and SB 28 were on her desk awaiting her action, Kelly told reporters last week that she hadn't decided what to do on the DEI legislation.

"I have been focused a lot on taxes lately," Kelly said. "So, I haven't looked at it and I really don't know the details. No, I haven't seen that one yet, so, I'll have to look and see the ramifications of it. Sometimes those bills, you know, they really don't do much and the universities can continue to function the way they need to function.

"So, I need to take a real good look at how impactful that bill is."

Then on Friday, Kelly allowed the bill to become law without her signature.

"While I have concerns about this legislation, I don't believe that the conduct targeted in this legislation occurs in our universities," Kelly said. "We need to move forward and focus our efforts on making college more affordable and providing students from all backgrounds with the tools they need to succeed.

"I am focused on advancing policies that drive economic growth and develop tomorrow's workforce. For that reason, I will allow the bill to become law without my signature."

Kelly hasn't acted on the budget bill — she has until Thursday to do so — but HB 2105 becoming law should negate the provisions that would strip funding from state universities because of DEI.

New state law will affect Washburn University

While Washburn University isn't technically a Regents institution, if the state bill or budget proviso are enacted, Washburn will be required to comply or receive less funding from the state.

Because of the state's efforts to change DEI policies, Washburn University has stated it would comply with the state's requests.

"This bill includes all public postsecondary educational institutions which includes Washburn University and Washburn Tech," the university said in a statement. "Washburn works diligently to remain in compliance with Kansas Law as we carry out our mission to create educational pathways to success for everyone."

Kansas Legislature had lively debate

On April 5, the Senate approved the bill 28-11 and the budget 26-12. The House passed the bill 81-40 and the budget 78-44.

The House was in a ruckus during the April 5 discussion of the bill and the budget. Speaker Pro Tem Rep. Blake Carpenter, R-Derby had to bang his gavel and call out misconduct multiple times and even addressed the entire House on its misconduct.

He told representative body to contain themselves because they were hissing, moaning and groaning at speakers who were given the floor.

While the budget proviso and bill passed in the House, many speakers held opposition to the bill.

Rep. Mike Amyx, D-Lawrence, said he wouldn't support the bill because of how it impacts DEI.

"I do believe that we are just not making this bill better," Amyx said. "In fact, we might be making it tougher on our post-secondary schools."

Rep Susan Ruiz, D-Shawnee, said she didn't appreciate having funding held back against them if legislators and higher education leaders chose not to comply.

"It feels like blackmail or like we are being held hostage or something," Ruiz said. "If we don't pass this bill, they're going to cut $35 million. Is that how the bill is written?"

The carrier of the budget bill, Rep. Steven Howe, R-Salina said the legislation doesn't tell universities they can't use DEI training or teaching but prevents requiring staff from using DEI.

"I wanted to do something that was responsible and respected academic freedom and the first amendment," Howe said. "But the Senate took a different position and offered a budget proviso."

Ruiz said she appreciated his efforts to be responsible but doesn't agree with his execution.

"This is a mess," Ruiz said. "It still feels like the House Bill is being held hostage but the other thing is, if we didn't have his House bill, we wouldn't be in this position in the first place. So, we need to be careful about bills we construct and I appreciate the representative wanting to do it in a responsible way, but it feels like it's backfired on us."

Rep. KC Ohaebosim, D-Wichita, agreed with Ruiz.

"With respect for this bill, I cannot support it at all," Ohaebosim said. "Eliminating DEI, I don't understand why we are doing it. Kind of like my colleague from Johnson County said, looks like we are holding universities hostage with money."

Rep. Barbara Ballard, D-Lawrence, said it's important to emphasis DEI is diversity, equity and inclusion because people find it easier to undermine its importance when minimizing it to an acronym.

"Diversity is what is in this room," Ballard said. "We are all diverse. Big towns, small towns and all kinds of things. We are diverse. So, I cannot see what is wrong with diverse. So, if we would put it all together in terms of diversity, equity and inclusion, I cannot see anything negative about those words."

After multiple speakers came forward asking questions to Howe, he refused to answer questions for Rep. Jason Probst, D-Hutchison.

Probst said this is a dangerous precedent to set in the chamber.

"The floor is an opportunity to have a discussion and a debate for the rest of us who haven't had a chance to hear a bill in committee," he said. "So, if we're going to get into the business of not answering questions on the floor, we're not actually having a full debate."

• Open access
• Published: 16 November 2020

Open access in vocational education and training research: results from four structured group discussions

• Laura Getz 1 ,
• Karin Langenkamp 1 ,
• Bodo Rödel 1 ,
• Kerstin Taufenbach 1 &
• Meike Weiland 1  

Empirical Research in Vocational Education and Training volume  12 , Article number:  15 ( 2020 ) Cite this article

3439 Accesses

11 Altmetric

Metrics details

Open Access fosters the exchange of academic research information by making publications free of charge and, wherever possible, available through open licences and without any technical barriers. Although the Open Access publication model is already well established in the natural sciences, there seems to be more resistance towards Open Access in the social sciences, including the field of vocational education and training research. The research project “Open Access in Vocational Education and Training Research” aims to uncover the conditions influencing the acceptance, dissemination and use of Open Access in vocational education and training research. The project is grounded in a sociology of knowledge approach and in media theory. It comprises of two parts. First, four structured group discussions are conducted as focus groups and analysed using a qualitative content analysis. This paper focusses on this first part of the research project, the implementation and analysis of the group discussions as well as the results thereof. The second part of the research project will be based on an online questionnaire built upon these results. The questionnaire will be sent out to authors involved in vocational education and training research in the second part of the research project.

The analysis of the group discussions reveals several thematic clusters. According to group discussion participants, the scope of their publications as well as transparent quality assurance procedures in publishing are particularly relevant. The reputation of their chosen publication medium is another central aspect. It also becomes clear that in some cases an information deficit regarding the financing of Open Access publications or accompanying licensing models exists. Finally, participants discuss changing literature research strategies and changes of work and reading practices. The latter being clearly discernible in an increasingly digitalised daily work routine of vocational education and training researchers, while academic research communication is also an important topic discussed.

Introduction–objectives and structure of the research project

The research project “Open Access in Vocational Education and Training Research” aims to uncover the following research question: Which technical and structural, policy-related and normative conditions, as well as conditions inherent in the academic research system influence the acceptance, dissemination and use of Open Access (OA)? The research project approaches this question from the perspective of authors working in the field of vocational education and training (VET) research, thereby uncovering attitudes, opinions and restraint of these researchers, and to derive recommendations. The project also looks into the differing roles authorship and readership of academic publications. Generally, authors are also users of academic research publications, and this means that their interests in these two capacities may diverge.

Since VET research is an interdisciplinary research field, combining different academic disciplines (Sloane 2006 , p. 610, Weiß, 2008 , p. 79), results of the research project may be partially transferable to the humanities and social sciences. Results will therefore help to provide more insight into OA in these research fields. In turn, habits of related academic research fields within the social sciences and humanities may influence the use and perception of OA in VET research (Taubert 2009 , p. 658).

What is Open Access?

There is no standardized concept to describe OA. Within the research project, however, we define OA with regard to three essential aspects:

Access to academic literature is free of charge: Readers do not need to pay for electronic access to academic literature. No usage or licensing fees are charged.

However, since editorial processing is indispensable and causes costs to cover production and layout of manuscripts in the run-up to publication, these costs are usually covered by so-called article processing charges (APCs): The author or institution pays for the article to become an OA publication. Fees in the form of so-called Book Processing Charges (BPCs) are charged in the same way for the publication of monographs.

Licensing is as open as possible: Legal protection of OA publications often takes place through a form of licensing fostering the dissemination of academic literature, rather than by means of copyright law, which–at least in Germany–is highly author-centric. Creative Commons Licences (CC Licences) are one example of such a common licensing model. Licensing likely plays an important role in German VET research and will therefore be taken into account in the research project.

OA publications should be easily findable: OA publications should be easily searchable and accessible, unhindered by technical restrictions. A standardised meta data structure needs to be in place. Repositories and similar infrastructures are needed, in order to store literature permanently. At the same time, access to academic publications should be free from technical obstacles. Documents should be downloadable and come in suitable file formats.

The publication market in VET research

The subject area of VET research is characterized by its great interdisciplinary diversity. It is therefore a challenge to describe the publication landscape in this field. Linten et al. ( 2019 ) differentiate between journals that are dedicated to the core area of VET research and those that address the wider field of the subject area. A distinction can also be made between refereed and non-refereed journals as well as between journals that follow the OA publication model and those that are subscription-based see Table  1 .

Established research institutes in VET research also publish their research and work results on OA websites, but this is not always the case. Efforts by publishers to convert their business models into OA are still in their infancy (e.g. the crowdfunding model “wbv OpenLibrary” from wbv Media Verlag).

In related disciplines of VET research, academic repositories make publications available in OA: PEDOCS and ERIC (educational sciences), SSOAR (social sciences) and EconStor (economics). The VET Repository, on the other hand, covers the core areas of VET research. These specialist repositories are used for initial publications in the form of grey literature, but also for secondary publications of articles that have been previously published in subscription journals. Such repositories increase the visibility of specialist literature and make it accessible in one central online space.

A large number of results from VET research are published as journal articles, but monographs are still a common publication format. According to Seifried (cf. 2020, p. 17–18), a trend indicating a decreasing relevance of monographs and an increasing popularity of journal articles can be found in educational sciences. However, this still needs to be proven for the field of vocational education and training research (ibid.). In 2014 Söll, Reinisch & Klusmeyer published research results from a survey on reading and publication behavior among academics from the field of professional and business education. With regard to the academic appointment process and the acquisition of third-party funding for research, subjects, who were all members of the Vocational and Business Education Section of the German Educational Research Association (GERA), assigned the highest ranking in terms of reputation to academic journals. This was also reflected in the perceived high future importance subjects assigned to academic journals (Söll et al. 2014 , pp. 511–513). It should be noted, that only members of the section for Vocational and Business Education participated in the survey and that academic journals were distinguished not only from non-academic practical journals, but also from online journals. Accordingly, results indicate a trend towards increased publication activity in academic journals.

In terms of academic reputation in VET research, the same principles as in the natural and engineering sciences cannot be applied. In a study of the publications contained in the German Education Index ( FIS: Fachinformations - System Bildung ), Klusemeyer et al. (cf. 2011 , p. 340) found that almost 95% of the journal articles were written in German. Accordingly, in their follow-up research Söll, Reinisch and Klusmeyer 2014 found that an increase in academic reputation in VET research is not the product of publishing in highly ranked academic journals. In fact, section members ascribed a great deal of importance to journals, which had a peer-review process in place, while the “Impact Factor […] only received below average attention” (Söll et al. 2014 , p. 525). This may be due to the fact that none of the German-language journals of vocational education and training research are ranked in the Social Science Citation Index (SSCI). In the international area of VET research, however, there are ranked journals, such as the Empirical Research in Vocational Education and Training (ERVET), Journal of Vocational Education and Training (JVET) or the International Journal for Research in Vocational Education and Training (IJRVET) (cf. SCImago Journal & Country Rank. Retrieved 2020).

Current state of research and feature space of the research project

The project team examined the current state of research on OA in the German-speaking social sciences thoroughly at the beginning of the research project (Herb 2015 , 2017 ; Bambey 2016 , Dallmeier-Tiessen et al. 2011 ; Pampel 2019 and, for a summary, see Langenkamp et al. 2018 , Getz et al. 2019 ). The analysis of the relevant literature on OA indicates that technical and structural, policy-related and normative as well as conditions inherent in the academic research system may influence the acceptance, dissemination and use of OA.

Technical and structural conditions include factors, such as storage, archiving, distribution and findability of OA publications. Repositories serving as a location for the organised storage of documents are one aspect of these conditions. The financing of OA publications, e.g. through publication funds, represents another aspect.

Policy-related and normative conditions mainly concern legal foundations of OA. These range from Article 5 (1) of the Basic Law of the Federal Republic of Germany and its implications for transparency and the democratic decision-making process, to regulations contained within the Federal Act Governing Access to Information (IFG). They also include the Copyright Act (Rödel 2017 , pp. 4 ff., Linten et al. 2019 , pp. 9 ff.). Publishing in OA through alternative licensing models, such as Creative Commons Licences, is an additional aspect.

Conditions inherent within the academic research system include quality assurance procedures such as peer review and the academic reputation system. Our assumption is that quality assurance in the OA publication model is a crucial factor for the acceptance, dissemination and use of OA. Against the backdrop of the prevailing pressure to publish in research (“publish or perish”), we need to consider how quality assurance and publication pressure relate to each other with regard to OA.

The matrix below illustrates the possible feature space examined in the research project and summarises possible conditions for the acceptance, dissemination and use of OA. Acceptance means that authors understand, approve of and support the OA publication model by publishing in OA. Dissemination refers to the various models used for OA publications (e.g. green OA, gold OA). Use means that authors use OA publications for their own academic research (even if their opinion on OA is a critical one) (Table  2 ).

Methods–theoretical and methodological approach

The theoretical foundation of the research project is based on a sociology of knowledge and media theory approach, in order to describe and reflect on developments in the field of OA within a broader framework. The project aims to create an increased understanding of the economic relevance of knowledge and academic research and is set to display processes of science communication and publication systems varying across academic disciplines. The project mainly refers to the works of Wilke ( 1998 ), who describes the transformation of the work and industrial society into a knowledge society and depicts knowledge as a production factor. Lyotard ( 2015 ) describes the value of knowledge as a commodity, meaning that the economisation of knowledge affects the academic research system and the publication and communication structure within (Taubert and Weingart 2016 ). Alongside the economisation of knowledge, digitalisation also causes a shift in the formal communication of academic research, for which Taubert and Weingart (cf. 2016 , p. 5) deem peer review essential, in order to verify research results. Digitalisation has brought about a change in both mass media and academic research communication (cf. Taubert and Weingart 2010 , pp. 5 ff.).

Finally, the reciprocity between the academic publication system and the academic reputation system needs to be taken into account (Taubert and Weingart 2010 ). Given the abundance of academic publications, Luhmann ( 1971 ) argues that quality assurance procedures of academic texts need to take place prior to the reading process because readers cannot carry out sufficient quality assurance. Generally, readers trust in quality assurance procedures used in the academic research system, which include the widely acknowledged peer review by reviewers with a high reputation in their respective research discipline. It remains debatable whether these quality assurance procedures serve their desired purpose. In any case, together with the “journal impact factor”, up-and-coming researchers in particular tend to regard them as gatekeepers (for information on this debate, cf. e.g. Rödel 2020 ; Roberts 2017 ; Schekman 2013 ; Callaway 2016 ; Fanelli 2012 ; Ioannidis 2005 ).

Structured group discussions as empirical data

In order to explore the research question, which has been subject to very little investigation in research so far, structured group discussions were carried out in the form of focus groups (cf. Krueger and Casey 2014 ). In line with the first qualitative part of the research project, group discussions followed a set structure with open questions. This enabled participants to bring their perspectives and experiences into the discussion while ensuring that important aspects were covered.

Four group discussions, each involving five to eight participants, were conducted in the second quarter of 2019. All participants had an academic background in VET research. Participants were of different age groups and genders, and occupied various status levels within the academic system (academic researcher, post doc, lecturer, and professor). This selection of participants ensured room for varying attitudes, preferences, experiences and user behaviours with regard to OA.

Prior assumptions and possible hypotheses in preparation for the group discussions

In order to set up a structured guide for the group discussions with suitable questions for participants, the research team compiled a collection of prior assumptions and possible hypotheses divided into (a) Influencing factors and (b) Aspects . The order of items does not represent an evaluation.

Influencing factors: Influencing factors relate to institutions or individuals and will be part of the second part of the research project as independent variables (Table  3 ).

Aspects: Aspects describe (framework) conditions for the acceptance, dissemination and use of OA. The table below illustrates the possible features and characteristics of these conditions (Table  4 ).

The project team operationalised the above-mentioned six aspects and allocated each of them individually to the technical and structural, policy-related and normative, as well as to conditions inherent within the academic research system. This produced six variations of the RLTW Matrix, which visualised the possible feature space and characteristics of the acceptance, use and dissemination of OA. The six matrices served as a basis for the group discussions (Tables  5 , 6 , 7 , 8 , 9 and 10 ).

Methodological approach to structuring the group discussions

In order to set up a structure for the group discussions, the six matrices above were further compacted. For this purpose, the project team generated three individual matrices for acceptance , dissemination and use respectively. Each of these matrices contained all six aspects, assigned to technical and structural, policy-related and normative conditions, and conditions inherent in the academic research system. Based on the matrices, the project team carried out an evaluation of the individual characteristics of the conditions displayed. This way the project team identified the most relevant topics to be included in the group discussions, assigning points and using a scoring system.

One point = “not particularly important”.

Two points = “quite important”.

Three points = “very important”.

These points were added up for each feature and rated according to their relevance.

*** = 12, 11, 10 points (very important).

** = 9, 8, 7 points (quite important).

* = 6, 5, 4 points (not particularly important).

For all aspects of the respective matrix, average values for all possible characteristics were calculated.

From each matrix, the three aspects with the highest average values were selected as relevant topics for the group discussion (Table  11 ).

In addition to including those three aspects from the three matrices with the highest average values, individual characteristics with an assessment of three stars (***) were also identified as particularly relevant conditions by the project team.

Following this, the project team formulated questions for all chosen characteristics and rephrased them into open questions, structured along the three thematic areas of use , acceptance and dissemination of OA. Footnote 1

In the first part on the use of OA, participants were asked about the search criteria they apply when in need of academic literature as readers and to what extent they specifically search for OA publications in this process. Another question focussed on green OA and gold OA and on whether participants know what these terms stand for. Here, we asked how participants handle restricted access to publications prior to the end of the embargo period. The first part closed with a question on whether participants had noticed an increase of the use of OA in their academic communities and what they appreciate about the digital format of OA publications in general. Moving on to the second part of the group discussion on the acceptance of OA, we placed an emphasis on the participants’ perspectives as authors of academic literature. The guiding questions revolved around participants’ priorities and preferences when publishing their own research in the OA format. We also asked them how the appeal of OA could be increased, in order to encourage more researchers to publish in OA. At the end of the second part, we asked participants to elaborate on their view on requirements to publish in OA in the framework of third-party funded research, emphasizing their role as academic authors. Two main guiding questions structured the final thematic part of the group discussion on the dissemination of OA. We asked participants how the dissemination of OA in academia could be advanced in general. Following this, we then asked participants to elaborate on their view on requirements to publish OA within third-party funded research projects, this time emphasizing their role as readers who need access to research results in a timely manner, as opposed to their role as authors.

Group discussions started with a warm-up session, in which participants introduced themselves, and closed with a final brainstorming on OA. While the moderator of the group discussions created sufficient space in the conversation for participants to engage in spontaneous exchanges, she also ensured that the group returned to the set structure on a regular basis. This made possible the comparability of the group discussions, which lasted for about 90 min each.

Organisation of the group discussions and selection of participants

Based on desk research, the project team were able to identify a sufficient number of potential participants at universities and research institutions in Germany. In three locations, professorships and/or research institutes, as well as researchers with a strong focus on VET research, were found. Six to eight researchers at different stages of their academic careers were invited to each group discussion. A trial group discussion was conducted at the Federal Institute for Vocational Education and Training (BIBB) prior to group discussions in Hamburg, Cologne and Paderborn, Germany. Overall, the project team aimed to invite VET researchers, who had already published in OA, came from different professional academic cultures, and represented different status levels, ages and genders, to ensure an even distribution thereof.

The participant acquisition process began in April 2019. Potential participants were initially contacted by telephone and/or received an invitation by email. Participants did not receive any information on the research project except for a link to the project website, a two-page project description and a letter inviting them to the group discussion. Participants were also informed that no preparation for the group discussion would be necessary on their part, since the project team was interested in their personal view on and experiences with OA.

Composition of group discussion participants

See Figs.  1 , 2 , 3 and 4 .

Participants in the group discussions by gender, n = 26

Participants in the group discussions by age, n = 26

Participants by status of information on OA (self-assessment), n = 26

Participants by OA publication experience (self-assessment), n = 26

Results and discussion

In accordance with the principles of Open Science, the anonymised German transcripts of the group discussions are available for subsequent use on the homepage of the research project: www.bibb.de/oabbf .

The analysis of the group discussions followed the qualitative content analysis by Mayring ( 2015 ), complemented with Kuckartz ( 2018 ). This approach allowed for the exploration of various thematic areas and layers in the data obtained. In addition, this methodological approach enabled the project team to uncover primary content, as well as latent content through interpretation. Content analysis also permits quantification, for example with regard to the frequency of certain topics brought up by participants. In order to structure the analysis, prominent topics were clustered into the thematic areas described below.

Outreach of publications and addressing target groups

Two initial topics emerge from the analysis of the group discussions: the outreach of publications on the one hand and the targeted addressing of readers on the other. Both these topics concern the participants’ point of view as authors and do not concern OA specifically, but rather the publication process as a whole. Participants explain for example that they wish to reach an expert audience that is interested in the topic of their publication. D8 for instance stresses the desire to make publications available to readers respectively: “It is important for me to actually reach my intended target group.” Participants also state that the way in which they address readers when they publish their research may vary depending on its thematic focus and alignment. In their role as academic authors, participants distinguish between academic research projects and practically oriented projects, as they state this influences their writing style and choice of publication medium. In this regard, D6 reports to adjust the writing style depending on the target audience: “We write differently when we are […] writing for a financed public audience [or] when we are writing for the professional academic community.”

Peer review and transparent quality assurance procedures in the VET research community

Participants view quality assurance procedures as an integral part of the academic research system and as deeply rooted in academic culture. From an author’s point of view, quality assurance procedures in the publication process are largely described by participants as being indispensable. The peer review process in particular is regarded as a central instrument for quality assurance within the academic publication system in general. Authors also perceive the feedback they receive through peer review processes as particularly useful for their own academic development: “[…] [through peer review] academic research quality, evidence […] is being ensured […] possibly even secured for the first time” A3 remarks. On the other hand, some participants express a critical view on peer review and state that they would prefer more transparency and comprehensibility when it comes to decisions made by reviewers. D7 points out this perceived lack of transparency in the review process: “I find that very often there are review procedures in which […] feedback […] is not very comprehensible.”

With regard to the financing of quality assurance procedures, participants reflect critically on the approach adopted by publishing houses not to pay researchers for peer review, may it be for reviewing OA or non-OA publications: “It is something which is simply seen as part of your job, if you are an academic researcher […],” says C4. In addition to discussing the payment of reviewers critically, participants debate a stronger shift of quality assurance procedures away from large publishing houses and more towards the VET research community, e.g. through an open peer review on online platforms. Participants thus appear to consider the dissemination of OA and the quality assurance thereof partly as a question of financial and human resources.

With regard to OA and based on their trust in institutions and networks of the research community, participants express the wish for the VET research community to look after OA media and platforms, in order to make OA more appealing to researchers and increase the dissemination thereof. Participants state that as both readers and authors of academic publications they appreciate transparent quality assurance provided by the academic community. They wish for trustworthy platforms run by well-known institutions. B4, for example, expresses a desire for “[…] a sort of network of persons to be organised [and] which would attend to this.” Accordingly, participants suggest that BIBB should set up an infrastructure to facilitate OA publications for VET research. C3, for instance, can “[…] well imagine that BIBB actually would be a good place to organise this process of publishing through Open Access […].” Some participants can also imagine uploading literature online without any prior review like in an “academic research Wikipedia”. On the other hand, they fear a “Wikipedia effect” , as it may lever out systematic review procedures and reliable quality assurance. D1 expresses a certain concern regarding the removal of quality assurance procedures prior to publication: “ […] if [OA] takes off this way and everyone is able to publish something anywhere [they like], then we will have a Wikipedia effect.”

Reputation and ranking of publication media, persons, and OA

With regard to their own publishing activity as authors, participants prefer high-ranking publication media that are recognised in the academic community. They express that, from an author’s perspective, it is very important whether a publication is perceived as high-ranking and whether it is actually or presumably held in a certain esteem, as C6 explains: “[…] it’s important that [journals] have a good reputation, in any case they should be peer reviewed […] and ideally they have an impact factor, too.” This is especially true for academic researchers who are at the beginning of their career and who are building up a reputation. Participants also state that the prestige of a publication medium may also affect the reputation of an author. Concerning the recognition of publication media within the academic community, OA is viewed as lower ranking by some participants and is often associated with “grey literature”, according to D8: “ Open Access still has this slight taste of grey literature […].” Accordingly, authors employ publication strategies to publish in highly ranked and often closed access publications. This approach to publishing in OA appears to be a contradiction to participants’ views on publication practices within third-party funded research, in which it often is a requirement to make research results available through OA. In all of the four group discussions participants state that taxpayer-financed research and results from third party-funded projects must be publicly available. D3 explains for example that taxpayers ought to have access to the research they have ultimately paid for: “[…] if we are permitted to carry out research using taxpayer funds, […] then […] the public has a right to be informed about the results.” In their capacity as authors who wish to build a reputation in the academic community participants face a conflict when it comes to their own publications. They wish for research results to be publicly available through OA. However, OA is perceived to provide less of a reputation. Therefore, it may be difficult for authors to publish in high-ranking publications enhancing their reputation and to publish in OA in the framework of third-party funded projects at the same time. Consequently, authors tend to prefer conventional and prestigious closed access formats and print publications over OA, even though the latter would be freely accessible to the public.

With regard to gaining an academic reputation, participants explain that they feel there is a need to publish research results as quickly as possible. This need is further amplified by the ubiquitous and immediate access to literature. The pressure to read and publish quickly, which one participant describes as “fast food” , is explicitly linked with OA in one of the group discussion. D1 expresses concern over the immediate accessibility of an abundance of literature online: “ I have immediate access, I can take a quick look, I can process things quickly […] [but] all this time I am wondering […] how sustainable this really is.” Participants assume that immediate access to OA may foster a fast and unreflecting consumption of literature. They are concerned that such fast consumption may leave only little room for critical discourses and reflections among academic researchers. Consequently, they fear, the quality of academic literature may suffer from an increasing pressure to publish.

Generally, participants state that the involvement of prestigious authors and publishers, “certain […] opinion leaders […] persons, luminaries […]” as A2 puts it, could increase the popularity and dissemination of OA. Participants indicate that OA would be more appreciated, if prestigious individuals were involved in various steps of the publishing process, thereby increasing trust in OA amongst academic researchers. Participants further describe those publication media as prestigious and trustworthy, that have become established in the VET research community.

Literature research strategies and the invisibility of OA

With regard to research and the acquisition of academic research literature, group discussions show that OA is not a selection or search criterion for participants. From the point of view as readers and users of academic literature, other criteria such as access opportunities (e.g. through libraries), research topics, specific authors and the current validity of a publication are more crucial to participants. In addition, journal subscriptions, which are available to participants through their institutional connections, render the topic of OA invisible when they search for literature. Participants even state that it is often not clear to them whether a publication is OA or not. C7 explains in this regard that institutional access to publications complicates the identification of a resource as OA: “ When I sit at my desk, logged into the university network, I am not aware of whether I can open a certain PDF file because the university has a licence or because the document is Open Access.” In general, when it comes to literature search, content matters more to participants than the way literature can be accessed.

In case, it is not possible for participants to access academic literature immediately, they resort to alternative strategies for literature retrieval. These may be accessing social media platforms such as ResearchGate and the online platform Google Books, or interlending literature through the university library. Participants also state that it is important for them to remain up to date in their academic field through subscriptions to specific newsletters.

Participants mention an increasing flood of information as being one of the drawbacks of the digital availability of literature. However, they do not mention OA as the root cause of this. Instead, they explain that literature research is becoming more complex in general and that they often accumulate documents more or less indiscriminately or may not even read them at all. D8 states to be overwhelmed by the abundance of online article at times: “Sometimes I am unable to […] find articles [on my computer] again […] and then I end up downloading them two or three times.” One participant compares the research of a certain topic with putting together the pieces of a puzzle. The individual “pieces” of the puzzle are collected by conducting research through various databases and platforms as well as by attending conferences, and networking with other researchers until a topic has been adequately researched, the participant explains.

Since participants are able to obtain closed access literature through their respective institutional affiliations, OA is of secondary importance to them. Instead, the convenient retrieval of literature is what matters most to participants, regardless of whether this concerns their own publications or those of others. An improved availability of digital literature could, for example, be ensured through thematic repositories and stable URLs, as participants state. With regard to the digitalisation of literature, participants discuss the scarcity of digital versions of older literature. When being pressed for time, they tend to choose literature based on immediate online availability, even if this means that some literature will fall through, C2 explains: “If a text cannot be […] found online, […] I no longer take it into account [in my research].”

Work practices and changes to academic research communication

A recurring topic in the group discussions concerns changing work practices brought about by the digitalisation of literature within the publication system in general. Participants state that they appreciate the immediate availability and flexibility in terms of the location they work from when handling the digital format (of OA and non-OA texts). They also emphasise that digital texts can be conveniently highlighted, cited, and worked with through search and commentary functions. Participants frequently explain that they highly appreciate working with digital texts and prefer them over printed documents. A1 explains: “ I prefer reading on a screen, on a tablet or on something similar; [the text] just needs to be digitally available.” The digital format of literature is generally valued by participants because it is convenient to retrieve and easy to handle in everyday working life. This applies to authors, as well as readers of academic literature.

Although a large number of participants perceive working with digital texts to be convenient and up to date, some express a high appreciation for books and reading on paper. They state that printouts are especially convenient on the commute to work or when needing to get an overview over various publications, e.g. in a bookshelf. They particularly highlight the tactile experience of handling paper and a perceived safe space to slow down when reading on paper. D7 explains a deep appreciation for books in this regard: “ There’s just something special about a book, it’s something that really matters to me, reading on paper.”

Participants also indicate that changes in science communication and digitalised reading and research practices are becoming increasingly important in everyday working life. In addition, remote conference participation is becoming more common, participants explain. In terms of OA in particular, some express a desire to actively participate in the changes that OA brings about and wish to contribute to science communication in the face of digitalisation.

Financing of OA

Participants explain that, as authors they are often unsure how to finance OA publications. Hence, they suggest that OA should be included in research project plans and budgets from the start, in order for employers or third parties to defray APCs. “Often the decision [to publish in OA] is based on whether the employer will pay for it or not” A4 explains in this regard. Participants also mention crowdfunding as a possible alternative to conventional financing opportunities for research publications.

Licensing models and legal conditions

Regarding the legal conditions within the publication process, participants frequently state that they do not feel well informed about the various licensing models in OA. This applies to both readers and authors of OA publications and their differing usage of publications. Participants call for the better protection of author rights in the publication process and demand for policy-makers to improve the legal protection of authors in the face of digitalisation.

Conclusions

The analysis of the group discussions shows that OA has become an established topic discussed among VET researchers. However, there may be an information deficit within the VET community regarding individual aspects of OA. Quality assurance is a central issue discussed and participants view OA critically in this context. This is surprising, since quality assurance procedures of journals using the OA model are not fundamentally different to those of non-OA journals. Participants consider peer review to be a crucial aspect of academic quality assurance. Nevertheless, they discuss the framework conditions for reviewers and the transparency of review procedures rather critically. Some participants even suggest a shift of review procedures from traditional peer review to new forms of an open review and OA infrastructures provided by the VET community. Generally, participants wish to disseminate their research widely and effectively. They aim to reach their respective target audiences by adjusting their writing style and the choice of their publication medium for each publication they work on. Frequently, they mention that results from third-party funded research should be made accessible to the public through OA and that an OA budget should be included in project plans from the start. Nevertheless, they often choose prestigious non-OA journals when disseminating their research. Authors seem to find themselves in conflict between enabling access to their publication and building a reputation at the same time. This ultimately leads to dissonance between the choice to publish or not to publish in OA. Participants’ remarks regarding the digital format of literature also stand out. They are critical of the fast pace of digital communication and the steady increase of publications in general, resulting in an abundance of literature that is challenging for researchers to keep track of. Participants wish for a deceleration of reading and work practices, while at the same time they appreciate the flexibility and convenience of working with digital literature. Digital and paper-based work practices seem to complement each other.

In the further course of the research project, the research team will design an online questionnaire based on the analysis of the group discussions. The questionnaire will be sent out to approximately 5000 authors working in VET research in 2020. This will allow for a further exploration of the technical and structural, policy-related and normative conditions, as well as conditions inherent in academic research system influencing the acceptance, use and dissemination of OA in VET research.

Availability of data and materials

The datasets generated and analysed during the current study are available in the website https://www.bibb.de/de/98091.php .

An overview of the structure and individual questions of the group discussions (in German) can be obtained from the project website: www.bibb.de/oabbf .

Bambey, D. (2016). Fachliche Publikationskulturen und Open Access. Fächerübergreifende Entwicklungstendenzen und Spezifika der Erziehungswissenschaft und Bildungsforschung [Specialist publication cultures and open access. Cross-disciplinary tendencies and specific characteristics of the educational sciences and education research]. Retrieved from https://tuprints.ulb.tu - darmstadt.de/5603/1/Bambey_Dissertation_2016_01_19_Ver%C3%B6ff_final.pdf

Callaway E (2016) Beat it, impact factor! Publishing elite turns against controversial metric. Nature 535:210–211. https://doi.org/10.1038/nature.2016.20224

Article   Google Scholar  

Dallmeier-Tiessen, S., Darby, R., Goerner, B., Hyppoelae, J., Igo-Kemenes, P., Kahn, D., Van der Stelt, W. (2011). Highlights from the SOAP project survery. What scientists think about open access publishing. Retrieved from https://arxiv.org/ftp/arxiv/papers/1101/1101.5260.pdf

European Commission (n.d.) Horizon 2020 online manual. https://ec.europa.eu/research/participants/docs/h2020-funding-guide/cross-cutting-issues/open-access-data-management/open-access_en.htm . Retrieved 11 Nov 2020

Fanelli D (2012) Negative results are disappearing from most disciplines and countries. Scientometrics 90(3):891–904. https://doi.org/10.1007/s11192-011-0494-7

Gantert K (2016) Bibliothekarisches Grundwissen [Basic library knowledge]. De Gruyter Saur, Berlin

Book   Google Scholar  

Getz, L., Langenkamp, K., Rödel, B., Taufenbach, K., & Weiland, M. (2019). Open Access in Vocational Education and Training Research–A Research Project at BIBB. In B. E. Stalder & C. Nägele (Eds.), Trends in Vocational Education and Training Research. Volume 2, Proceedings of the European Conference on Educational Research (ECER), Vocational Education and Training Network (VETNET). Hamburg, Germany, 3 to 6 September 2019 (pp. 131–139). https://doi.org/10.5281/zenodo.3371461

Herb, U. (2015). Open Science in der Soziologie. [Open Science in sociology] Eine interdisziplinäre Bestandsaufnahme zur offenen Wissenschaft und eine Untersuchung ihrer Verbreitung in der Soziologie [Open Science in sociology. An interdisciplinary analysis of open science and its dissemination in sociology]. https://doi.org/10.22028/d291-23737

Herb, U. (2017). Sozialwissenschaften [Social sciences]. In K. Söllner & B. Mittermaier (Eds.), Praxishandbuch Open Access [Practical Handbook of Open Access] (pp. 254–260). https://doi.org/10.1515/9783110494068-029

Ioannidis, J. P. A. (2005). Why most published research findings are false. PLoS Med, 2(8, e124), 696–701. https://doi.org/10.1371/journal.pmed.0020124

Klusemeyer J, Reinisch H, Söll M (2011) Wo publizieren Berufs- und Wirtschaftspädagogen? Eine Vorstudie zur Erfassung des Publikationsverhaltens in der Berufs- und Wirtschaftspädagogik [Where do professional and business educators publish? A preliminary study to record the publication behaviour in professional and business education]. Zeitschrift für Berufs- und Wirtschaftpädagogik 107(3):328–350

Google Scholar  

Kuckartz U (2018) Qualitative Inhaltsanalyse. Methoden, Praxis, Computerunterstützung. [Qualitative Content Analysis. Methods, Practices, Computerized Support], 4th edn. Beltz, Weinheim

Krueger RA, Casey MA (2014) Focus Groups. A practical guide for applied research, 5th edn. Sage Publications, Thousand Oaks, CA

Langenkamp K, Rödel B, Taufenbach K, Weiland M (2018) Open Access in Vocational Education and Training Research. Publications 6(29):1–29. https://doi.org/10.3390/publications6030029

Linten, M., Rödel, B., Taufenbach, K. & Woll. C. (2019). Wissenschaftliches Publizieren in Zeitschriften der Berufsbildungsforschung [Academic Research Publishing in Vocational Education and Training Research Journals] (2. ed.). Retrieved from https://www.bibb.de/veroeffentlichungen/de/publication/download/9801

Luhmann, N. (1971). Selbststeuerung der Wissenschaft [Self-direction of academic research]. In N. Luhmann (Ed.), Soziologische Aufklärung. Aufsätze zur Theorie sozialer Systeme [Sociological Enligtenment. Essays on the Theory of Social Systems] (2. ed., pp. 232–252). Opladen, Germany: Westdeutscher Verlag

Lyotard J-F (2015) Das postmoderne Wissen. Ein Bericht [Post-Modern Knowledge. A Report], 8th edn. Passagen-Verlag, Wien

Mayring P (2015) Qualitative Inhaltsanalyse. Grundlagen und Techniken [Qualitative Content Analysis. Theoretical Foundations and Basic Procedures], 12th edn. Beltz, Weinheim

Pampel, H. (2019). Open Access an wissenschaftlichen Einrichtungen in Deutschland. Ergebnisse einer Erhebung im Jahr 2018. Bericht [Open Access at Scientific Institutions in Germany. Results of a Survey in 2018. Report]. https://doi.org/10.2312/os.helmholtz.005

Roberts RJ (2017) Bibliometrics. An obituary for the impact factor. Nature 546:600. https://doi.org/10.1038/546600e

Rödel, B. (2017). Open Access in der Berufsbildungsforschung. Status quo und Perspektiven [Open access in vocational education and training research. Status quo and perspectives]. Retrieved from https://nbn-resolving.org/urn:nbn:de:0035-0651-9

Rödel, B. (2018). Lass dich nicht ausrauben!. Anmerkungen zur Diskussion um Predatory Journals [Don’t get robbed! Remarks on the discussion about Predatory Journals]. Berufsbildung. Zeitschrift für Theorie-Praxis-Dialog [Vocational Education and Training. Journal for Theory-Practice Dialogue]. 72(173), 48–50sch

Rödel, B. (2020). Peer Review. Entstehung – Verfahren – Kritik [emergence – method – criticism]. Retrieved from https://www.bibb.de/veroeffentlichungen/de/publication/show/10827

Schekman, R. (2013, December 9). How journals like Nature, Cell and Science are damaging science. The Guardian. Retrieved from https://www.theguardian.com/commentisfree/2013/dec/09/how-journals-nature-science-cell-damage-science

Sloane P (2006) Berufsbildungsforschung [Vocational education and training research]. In: Arnold R, Lipsmeier A (eds) Handbuch der Berufsbildung [Handbook of vocational education and training], 2nd edn. VS Verlag für Sozialwissenschaften, Wiesbaden, pp 610–627

Söll M, Reinisch H, Klusmeyer J (2014) Publikation und Reputation: eine Studie zum Rezeptions- und Publikationsverhalten von wissenschaftlich tätigen Berufs- und Wirtschaftspädagoginnen und-pädagogen [Publication and reputation: a study on literary reception and publishing behaviour of vocational and economic education researchers]. Zeitschrift für Berufs- und Wirtschaftspädagogik [J Vocat Econ Educ Res] 110(4):505-528

Taubert NC (2009) Eine Frage der Fächerkultur? Akzeptanz, Rahmenbedingungen und Adaption von Open Access in den Disziplinen [A question of the specialist culture? Acceptance, general conditions and adaption of open access in the disciplines]. Forschung Lehre 16(9):564–566

Taubert NC, Weingart P (2010) Open Access-Wandel des wissenschaftlichen Publikationssystems [Open access-change in the academic research publication system]. In: Sutter T, Mehler A (eds) Medienwandel als Wandel von Interaktionsformen [Media shift as a change of forms of interaction]. VS-Verlag, Wiesbaden, pp 159–181

Chapter   Google Scholar  

Taubert N, Weingart P (2016) Wandel des wissenschaftlichen Publizierens-eine Heuristik zur Analyse rezenter Wandlungsprozesse [Shift in academic research publishing-A heuristic for the analysis of recent change processes]. In: Weingart P, Taubert N (eds) Wissenschaftliches Publizieren [Academic research publishing]. De Gruyter, Berlin, pp 3–38

Weiß, R. (2008). Nach der Evaluation ist vor der Evaluation. Zur Berufsbildungsforschung im BIBB [Evaluation is always followed by another evaluation. On vocational education and training research at BIBB]. In D. Euler, J. Howaldt, G. Reinmann & R. Weiß (Eds.), Neue Forschungsverständnisse in den Sozialwissenschaften. Konsequenzen für die Berufsbildungsforschung im Bundesinstitut für Berufsbildung [New research understandings in the social sciences. Consequences for vocational education and training research at the Federal Institute for Vocational Education and Training] (pp. 75–93). Retrieved from http://www.bibb.de/dokumente/pdf/WDP_94_Screen.pdf

Willke, H. (1998). Systemisches Wissensmanagement [Systematic Knowledge Management]

Söll, M., Reinisch, H., Klusmeyer, J. (2014). Publikation und Reputation: eine Studie zum Rezeptions- und Publikationsverhalten von wissenschaftlich tätigen Berufs- und Wirtschaftspädagoginnen und-pädagogen [Publication and reputation: a study on literary reception and publishing behavior of vocational and economic education researchers]. Zeitschrift für Berufs- und Wirtschaftspädagogik [Journal for Vocational and Economic Education Research] 110 (4), 505-528

Download references

Acknowledgements

Not applicable.

The research project is funded by the Federal Institute for Vocational Education and Training (BIBB).

Author information

Authors and affiliations.

Federal Institute for Vocational Education and Training (BIBB), Robert-Schuman-Platz 3, 53175, Bonn, Germany

Laura Getz, Karin Langenkamp, Bodo Rödel, Kerstin Taufenbach & Meike Weiland

You can also search for this author in PubMed   Google Scholar

Contributions

LG coding, evaluation, KL Open Access specialist, BR head of project, KT Open Access specialist, MW methodological concept. All authors read and approved the final manuscript.

Authors’ information

Laura Getz, is a research associate at the Strategic Office “Publications and Scientific Information Services” at BIBB. She is responsible for the project implementation, methodology and dissemination of the research project “Open Access in Vocational Education and Training Research”. She studied Cultural Anthropology and English-Speaking Cultures, and holds an M.A. in Transcultural Studies. She has been working at BIBB since 2018.

Karin Langenkamp, is a research associate at the Strategic Office “Publications and Scientific Information Services” at BIBB and is part of the research project team “Open Access in Vocational Education and Training Research”. She ist also responsible for the Content of the VET Repository and the documentations services at BIBB. She holds an M. A. in Library and Information Science. She has been working at BIBB since 2017. Before that, she was head of the compulsory department of the Baden state library in Karlsruhe.

Bodo Rödel, is head of the Strategic Office “Publications and Scientific Information Services” at BIBB. He has been working for BIBB since 2012 and is responsible for publications here. Before he was working in a publishing house. Bodo Rödel holds a Diploma in Pedagogy and a PhD in Education Science.

Kerstin Taufenbach, is a research associate at the Strategic Office “Publications and Scientific Information Services” at BIBB and is part of the research project team “Open Access in Vocational Education and Training Research”. She is also responsible for the library and documentation services at BIBB. Kerstin Taufenbach holds a Diploma in Library and Information Science.

Meike Weiland, is a research associate at the Strategic Office for “Online Communications and Knowledge Management” at BIBB. She is responsible for the methodological concept of the research project “Open Access in Vocational Education and Training Research”. Meike Weiland holds an M.A. in Educational Science and Sociology, as well as an MBA in Educational Management.

Corresponding author

Correspondence to Bodo Rödel .

Ethics declarations

Competing interests.

The authors declare that they have no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Getz, L., Langenkamp, K., Rödel, B. et al. Open access in vocational education and training research: results from four structured group discussions. Empirical Res Voc Ed Train 12 , 15 (2020). https://doi.org/10.1186/s40461-020-00101-z

Download citation

Received : 05 July 2020

Accepted : 05 November 2020

Published : 16 November 2020

DOI : https://doi.org/10.1186/s40461-020-00101-z

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Open Access
• Vocational education and training (VET)
• Academic research communication
• Group discussions
• Qualitative content analysis
• Science communication

• Education News

TS Inter 1st, 2nd Year Result 2024: Check district-wise and gender-wise pass percentages

District-wise & Gender-wise pass percentage - General First Year - IPE March 2024

District-wise & gender-wise pass percentage - vocational first year - ipe march 2024, district-wise & gender-wise pass percentage - general second year (regular) - ipe march 2024, district-wise & gender-wise pass percentage - vocational second year (regular) - ipe march 2024, visual stories.