Mapping social capital in vocational education and training: A multi-perspective egocentric social network analysis in a European innovation project

Meyne, Lisa; Siemer, Christine Article Mapping social capital in vocational education and training: A multi-perspective egocentric social network analysis in a European innovation project International Journal for Research in Vocational Education and Training (IJRVET) Provided in Cooperation with: European Research Network in Vocational Education and Training (VETNET), European Educational Research Association Suggested Citation: Meyne, Lisa; Siemer, Christine (2025) : Mapping social capital in vocational education and training: A multi-perspective egocentric social network analysis in a European innovation project, International Journal for Research in Vocational Education and Training (IJRVET), ISSN 2197-8646, European Research Network in Vocational Education and Training (VETNET), European Educational Research Association, Bremen, Vol. 12, Iss. 3, pp. 433-474, https://doi.org/10.13152/IJRVET.12.3.6 This Version is available at: https://hdl.handle.net/10419/323753 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-sa/4.0/ International Journal for Research in Vocational Education and Training (IJRVET) 2025, Vol. 12, Issue 3, 433–474 https://doi.org/10.13152/IJRVET.12.3.6 ISSN: 2197-8646 https://www.ijrvet.net Mapping Social Capital in Vocational Education and Training: A Multi-Perspective Egocentric Social Network Analysis in a European Innovation Project Lisa Meyne*, Christine Siemer University of Bremen, Institute Technology and Education (ITB), Am Fallturm 1, 28359 Bremen, Germany Received: 28 January 2025, Accepted: 23 May 2025 Abstract Context: The importance of the involved stakeholders and their networks in vocational education and training (VET) focussing on international transfer and cooperation is highlighted in various empirical studies. A systematic empirical survey of these by means of social network analysis, however, has hardly been applied to date. This article is concerned with the development of social capital in the course of network formation and its sustainability. The object of investigation is the funded European VET innovation project AI Pioneers within the Erasmus+ program of the European Union. The main objective of the project is to establish and expand an international network in the context of VET in order to support the exchange of expertise on the use of artificial intelligence (AI) in education. Approach: To answer the research questions, the first step was to combine theoretical approaches from a social network perspective from psychology in relation to the analysis of interpersonal trust, sociology regarding the social capital approach and business administration by addressing the roles of actors in innovation processes. Among others, the social network perspective in this study is based on the work of Granovetter as well as Marsden and Campbell. For the data collection, a fully structured interview questionnaire and a semistructured interview guideline were developed based on the theoretical framework of the *Corresponding author: [email protected] 434 Mapping Social Capital in Vocational Education and Training study. In the second step, a multi-perspective egocentric network analysis was carried out: Data on a total of N = 10 egocentric networks were collected from the funded partners in the AI Pioneers project to gain an overall picture of the combined social capital and network structures. For the visualisation of the network data, the type of structured and standardised network maps was used. Findings: Regarding the establishment of social capital in the analysed innovation project AI Pioneers, it can be emphasised that a total of 74 relationships have been recorded in the 10 egocentric networks combined. In line with the project objectives, the education sector is addressed by the majority of the analysed relationships (n = 54), with (technical) vocational schools making up a substantial part of these. Focussing on the sustainability of the surveyed network structures: Most of the analysed relationships already existed before the project start and were consolidated during it (n = 57), while new ones were also established (n = 17). In addition, the continuous development of mutual trust and the need for equal cooperation is emphasised: A relatively high level of mutual trust can be recorded overall in the analysed egocentric networks (n = 55), while a low mutual trust is present in 19 relationships which is described due to e.g. asymmetrical power relations or a lack of commitment. The results show that the relationships analysed primarily contribute their resources in the form of expertise and their networking knowledge to the egocentric networks. Furthermore, a high level of interest and willingness to support the AI Pioneers project can be captured, particularly due to the novelty of the topic and the application of AI in VET. Conclusions: The study makes a significant contribution to VET research and its methodological set by using social network analysis with a combination of qualitative approaches for analysing egocentric networks from multiple perspectives. The importance of allocating resources to the creation of social capital regarding cooperation, network building and the sustainable maintenance of established structures can be emphasised. In this respect the benefits of a network-based approach can be highlighted in the context of the Erasmus+ program and the partnerships for innovation on forward-looking topics. In addition, the development of the two structured survey instruments in this study can be emphasised, which can be further developed on the basis of future research. Further quantitative network analyses would be valuable for VET research, especially against the background of innovation drivers and network formation, such as market and trend-related drivers due to demands and developments in the field of AI in education. Keywords: Social Network Analysis, Egocentric Multi-Perspective, Social Capital, Erasmus+ Innovation Project, Artificial Intelligence, Vocational Education and Training, VET 435 Meyne, Siemer 1 Introduction Educational cooperation has been funded and promoted by the European Union for many years under the Erasmus+ program with innovation1 as the central element: Therefore, the aim of funding is on supporting innovative projects with regard to various priorities, so that, among other areas, cooperation between educational organisations in Europe is promoted under the Partnerships for Innovation in order to support modernisation and novelty on future-oriented topics such as digitalisation and the use of artificial intelligence (AI) in education and training (European Commission, 2022, 2024). Funding is primarily provided in this regard for projects that are made up of a transnational consortium and focus on "mutual learning on forward-looking issues amongst key stakeholders and empowering them to develop innovative solutions and promote the transfer of those solutions in new settings" (European Commission, 2022, p. 277). In addition, other eligible activities are outlined, including 'mapping work' and 'transnational events or networking activities' under the priorities of the so-called forward-looking projects (European Commission, 2022). Therefore, the Erasmus+ program focuses directly on building sustainable international networks as a constant implicit feature (European Commission, 2022, 2024), with some innovation projects even explicitly setting international networking as one of their core objectives. Networking activities like this extend beyond national borders. At the same time, the aspect of networking is increasingly becoming a focus for the use of resources in order to meet the growing challenges and change processes in the education sector (Gruber et al., 2018). As a resource within the networks, the resulting social capital (e.g., trust) in particular can be an indicator of sustainability (Gessler & Siemer, 2020). Therefore, the promotion of network formation and thus social capital, as directly addressed by the European funding program, should be placed in the focus of research interest as well as the output of such funding efforts addressing networking and its sustainability in international contexts. Furthermore, in vocational education and training (VET), the connection between work, technology and education continues to be understood as a dynamic interdependence that must be addressed in a changing world of work (McGrath et al., 2019). The changes resulting from developments in the field of AI should be particularly emphasised in this interdependence, as they result in both fundamental change processes in society and various potentials for teachers, trainers and learners in VET (e.g., to identify individual learning needs; De Witt, 2024; Roppertz, 2021). As the debate surrounding the topic of AI has gained momentum in recent years, the funding and implementation of AI innovation projects was placed 1 Regarding the term innovation under the priority of Partnerships for Innovation focusing on Forward-looking Projects, the Erasmus+ Program refers to "foster[ing] innovation, creativity and participation, as well as social entrepreneurship in different fields of education and training. It will support forward-looking ideas based around key European priorities, (…) and giving input for improving education and training systems, as well as to bring a substantial innovative effect in terms of methods and practices to all types of learning and active participation settings for Europe's social cohesion" (European Commission, 2024, p. 16). 436 Mapping Social Capital in Vocational Education and Training in the foreground of the Erasmus+ program (European Commission, 2024). Accordingly, the importance of AI is reflected in the educational context through policy funding not only at national level (e.g., Digitalisation Master Plan for Lower Saxony in Germany) but also explicitly at international level (e.g., Digital Education Action Plan and Erasmus+ Program for 2021-2027 of the European Commission). Furthermore, under the keyword 'Artificial Intelligence' on the European Commission's project and results platform, there are a total of 689 references to EU-funded projects, whereby 389 projects on AI can already be found for the EU funding period from 2021 to 20272. Appropriately, CEDEFOP and ReferNet (2023, p. 1) emphasise the need "to integrate AI competences in education and training at schools, at the workplace (e.g. apprenticeship training), in teaching and at universities". In order to address the topic of networking and cooperation with regard to the AI application in VET, an Erasmus+ project is analysed as an example for the category of partnerships for innovation and forward-looking projects in this study. The VET innovation project AI Pioneers (funded within the European Union Erasmus+ program, period 2023-2025) promotes the use and teaching of AI in adult and vocational education and training, with a total of 10 project partners from seven different EU countries involved in the project network (Germany, Greece, Portugal, Italy, Spain, Cyprus, Estonia). The focus of the project, besides the development of policy recommendations, AI toolkits, implementation guidelines of AI use cases and ethical and trustworthy use of AI in education, is predominantly on the implementation and establishment of an international network of AI Pioneers. In this regard educators, stakeholders, policy makers and education planners are addressed as reference points in the network for the design and implementation of (future) education projects related to AI (see e.g., Attwell et al., 2023; Deitmer et al., 2024; Grollmann et al., 2024; Meyne & Siemer, 2024). The emerging network will promote the exchange of best practices and national communities of practice (CoP) to increase transparency regarding the application of AI in educational settings. In the current scientific discourse, the importance of the actors involved and the resulting social networks in international VET cooperation is emphasised in a large number of empirical studies (see e.g., Albertz & Pilz, 2025; Gessler, 2019; Peters & Gessler, 2019; Pilz, 2016; Pilz & Zenner‐Höffkes, 2023; Röhrer et al., 2021). However, the state of research in the field of VET (see e.g., Coppe et al., 2023; Schlicht & Moschner, 2018; Messmann et al., 2018) and especially the recording of social capital in international VET research is considerably limited (see Gessler & Siemer, 2020; Siemer & Gessler, 2021). Against this background, the article is based on the assumption that the development of social capital in the course of network formation of funded projects has an influence on the 2 The majority of the projects are located in the Horizon Europe program. If the search terms are used independently of each other, there are 1205 entries with in total 665 projects between 2021 and 2027 (search status: 18th February 2025): https:// ec.europa.eu/info/funding-tenders/opportunities/portal/screen/opportunities/projects-results?order=DESC&pageNumber =1&pageSize=50&sortBy=title&keywords=Artificial%20Intelligence&isExactMatch=false 437 Meyne, Siemer sustainability of those. One approach to identify such outward-facing networking activities comes from innovation research and is based on the promoter model (Gemünden & Hölzle, 2005; Hauschildt & Chakrabarti, 1988; Witte, 1973). This model comprises the following four promoter roles: Power promoter, professional promoter, process promoter and relationship promoter, whereby the latter role is inter-organisationally oriented and mainly characterised by external networking activities (Gemünden et al., 2007; Gemünden & Hölzle, 2005). On the basis of the relationship promoter of the AI Pioneers project partners, this article applies egocentric network analysis and examines the development of social capital (e.g., intensity of the established relationships): Egocentric network data describe the local social environments surrounding individual actors in a network – usually comprising one or more of each focal actor's direct contacts ("alters") and certain qualities of the dyadic relationships between that actor ("ego") and the alters. (Marsden & Campbell, 2012, p. 18) Therefore, the aim of this study is to use egocentric network analysis to examine the development and establishment of (international) networks on forward-looking topics in vocational education and training using the example of the Erasmus+ funded AI Pioneers innovation project. The focus is on the identification of central actors and their development of social capital as well as the intensity of relationships between funded partners in the course of network formation in the form of a multi-perspective egocentric network analysis to gain an overall picture of the combined network structures (see Chapter 2). The analysis is based on network data of 10 relationship promoters (egos) of the project partners funded as part of AI Pioneers and the relationships to the actors (alters) that arise in the project context in their egocentric networks: "Actors [in social networks] are often individuals. But organisations or states can also act as nodes in social networks" (Fuhse, 2018, p. 14; translated by the authors). The present study thus pursues the following main research question (RQ): To what extent has social capital evolved within the network in the international AI Pioneers project? To answer this main question, the following sub research questions arise: –RQ1: Which actors (alters) play a central role in network formation in the AI Pioneers project? –RQ2: How are the relationships within the analysed egocentric networks formed in terms of their intensity? –RQ3: What functions and relevance do these alters bring to the egocentric networks? 438 Mapping Social Capital in Vocational Education and Training To answer these questions, the article is structured as follows: Chapter 2 introduces the theoretical framework and sets out relevant definitions for this study. Chapter 3 presents the research design including the research approach, the data sample, the data collection and analysis. Chapter 4 is dedicated to the results of this study. In chapter 5, the results and the methodological approach are discussed. The article concludes in chapter 6 with the limitations of the study and an outlook for future research. 2 TheoreticalFramework In collecting network data, Hansen et al. (2011) refer to the analysis of overall networks (full networks), sections of these (partial networks) and individual networks of actors (personal/ egocentric networks). According to Hansen et al. full networks are: (…) Often created and available when a single system, such as a social media platform, acts as a hub among a group of connected people or groups (…). A full or complete network contains all the people or entities of interest and the connections among them. All egos are treated equally. (Hansen et al., 2011, p. 36) In partial networks, on the other hand, researchers select certain entities of a superordinate network as the unit of analysis (Hansen et al., 2011). Furthermore, "it is often useful to consider social networks from an individual member's point of view", so that, starting from a single actor (ego), the connections to other persons (alters), and sometimes also the connections between different alters, are analysed (Hansen et al., 2011, p. 36; see Figure 1). The visualisation of different types of network data often occurs in the form of network maps. Hollstein and Pfeffer (2010) also emphasise that studies in which network maps are used often involve comparisons within these maps (e.g., between alters) or between different maps (e.g., different egocentric networks). The analysis of individual egocentric network data without a common context could accordingly be referred to as first-order egocentric analysis, while in the context of the present study, the combination of multiple perspectives of different egocentric networks and their connections to each other within an overall perspective could be thought of as second-order egocentric network analysis (see Figure 2). Based on the Erasmus+ innovation project AI Pioneers, the article adopts the second-order perspective of egocentric network analyses presented here, so that a multi-egocentric network analysis is carried out in the course of the study and thus several egos (N = 10) are surveyed in order to draw conclusions about the developed social capital in the common project context. These two understandings of egocentric network analyses are illustrated below. 439 Meyne, Siemer Figure 1: Visualisation of a "First-Order" Perspective on an Egocentric Network (own compilation in accordance with Hansen et al., 2011, p. 168) Figure 2: Visualisation of a "Second-Order" Perspective on Egocentric Networks (own compilation) 440 Mapping Social Capital in Vocational Education and Training In the following, the theoretical framework of the study is described (see Figure 3), which includes approaches from a social network perspective (see Granovetter, 1973; Marsden & Campbell, 2012) from psychology in relation to the analysis of interpersonal trust (see Schweer, 2008), sociology regarding the social capital approach (see Bourdieu, 1983) and business administration by addressing the roles of actors in innovation processes (see Gessler, 2019; Witte, 1973). Those approaches are subsequently empirically applied to the field of vocational education and training (see Chapter 3 and 4). Figure 3: Concept Map of the Theoretical Framework of the Study (own compilation) To capture the relationships within networks, this paper will draw on the concept of social capital, which is significant for understanding the function of relationships. Among the most prominent authors who shaped the definition of this concept are Pierre Bourdieu (see e.g., 1983) and Robert Putnam (see e.g., 1995). However, both represent different perspectives on the topic: Bourdieu defines social capital as "the totality of actual and potential resources associated with the possession of a durable network of more or less institutionalised relations of mutual knowledge or recognition" (Bourdieu, 1983, p. 191; translated by the authors) and foregrounds social capital as a private good. Although Putnam (2001) increasingly focuses his research on the societal effects of social capital, he states: "There are both public and private faces of social capital" (Putnam, 2001, p. 1). Putnam further characterises social capital as "features of social organisation such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit" (Putnam, 1995, p. 67) and describes the concept of trust as being relevant to social capital. Despite Putnam's view of social capital from the perspective of society as a whole, his considerations on trust are relevant to social 447 Meyne, Siemer them (Häußling, 2009a). Häußling points out that there is a very different understanding of networks in research. It can be assumed that networks themselves are to be understood as boundaries (see e.g., Karafillidis, 2009), that networks have no boundaries at all (see e.g., Mewes, 2009) and/or that networks have blurred boundaries (Häußling, 2009b). Furthermore, Abbott (1995) states: It is wrong to look for boundaries between preexisting social entities. Rather we should start with boundaries and investigate how people create entities by linking those boundaries into units. We should not look for boundaries of things but for things of boundaries. (Abbott, 1995, p. 857) Accordingly, we define the network to be analysed in this study along the thematic focus of the AI Pioneers project. Using multi-perspective egocentric network analysis (see e.g., Perry et al., 2018; McCallister & Fischer 1978; see also Chapter 3.2), the sample is asked about their relevant relationships in the project context, thereby defining the alters of the project partners in terms of their connection and exchange with the AI Pioneers project and thus focusing on a specific number of actors and relationships. 3.2 Phase2:QualitativeSurveyInstruments In the second phase, a fully structured interview questionnaire and a semi-structured interview guideline were developed based on the model by Gessler and Siemer (2020) and on the theoretical framework presented in Chapter 2. The advantage of using fully structured interviews is that they make the interview situation easier for the interviewees by allowing the interviewer to fill in the standardised answers together with the interviewee. This makes it possible to ask questions directly in the event of comprehension problems (Döring, 2023). In this phase, an initial interview was conducted with the interviewees to discuss matters of understanding in advance. The interview situation included an explanation of the categories. The structured interview guideline was then filled out collaboratively. Table 2 in the appendix shows the fully structured interview questionnaire, in accordance with Gessler and Siemer (2020) and enriched by further categories (see Chapter 2) with the corresponding questions/statements and selectable answer options. The selection of the applicable answer option was made by the interviewees themselves (see Table 2 in the appendix). A semi-structured interview guideline was developed on the basis of the fully structured approach (Döring, 2023) to gather further information and contextual knowledge about the relationships. The interview guideline was tested in advance and subsequently adapted (Friebertshäuser & Langer, 2013). The following table shows exemplary in-depth questions along the main categories for the present study (Table 3). 448 Mapping Social Capital in Vocational Education and Training Table 3: Category System With Example Questions Categories Example Questions Information about the person Do you assume generally that a relationship of trust will evolve with new people when you first meet them? Information about the partner organisations Could you please tell us in advance what exactly these organisations are that you have entered in the questionnaire, in other words what they do in their day-to-day business and why you have listed them in general? Intensity of the relationship How do the listed contacts that were described as having a high degree of willingness to mutually exchange information differ from the others? What project-related information is involved here exactly? Function of the organisations Can you give us a few examples regarding the categorised actors in terms of power, expertise, social capital and processes? Relevance of the relationships in the course of the project Can you explain in more detail how the relationships and organisations are classified in terms of their importance for the AI Pioneers project? In this study, the categories listed represent the thematic areas of the fully structured interview questionnaire and the semi-structured interview guideline (see also Table 2 in the appendix). The data was collected by the authors. The data material was analysed on the basis of a previously developed coding system. In a first step, the data material was analysed independently by the authors. In a second step, the results were discussed, which led to a consensus. The data collection was carried out using the video conferencing software Zoom. The interviews were transcribed using the F4x transcription tool. The analysis and coding were initially carried out deductively and subsequently inductively based on qualitative content analysis according to Kuckartz (2018) by using the MaxQDA software to analyse the transcriptions. The methodological reflection of our study is based on the critical appraisal tool for qualitative research by Lockwood et al. (2024). Lockwood et al. (2024, p. 2) pointed out initially: "The purpose of this appraisal is to assess the methodological quality of a study and to determine the extent to which a study has addressed the possibility of bias in its design, conduct and analysis". 3.3 Phase3:ValidationoftheNetworkMaps For the visualisation of the network data, we used the type of structured and standardised network maps (Hollstein & Pfeffer, 2010). The concentric circles in our study express the emotional closeness and distance to the ego. The structuring elements comprise the gradations trust (highest degree of closeness), cooperation, goal-oriented coordination, mutual exchange of information and pure exchange (highest degree of distance) and thus enable comparability of the network maps determined. This means that both the relationships of the alters to the ego within a map (e.g. based on circle and sector assignments) and, in some cases, different maps can be analysed in relation to each other (Hollstein & Pfeffer, 2010). While the use of standardised network maps leads to increased comparability between those, a high degree of standardisation limits the contextual knowledge of the visualised network data. 449 Meyne, Siemer In order to overcome these limitations, we complement the study with qualitative research methods (Hollstein & Pfeffer, 2010). The visualisation of the network maps was created in own compilation by the authors of the study in accordance with the tool VennMaker (Kronenwett & Schönhuth, 2014) based on the data of the egocentric networks collected via the fully structured interview questionnaire. The created network maps were validated in a second appointment via Zoom with the egos (relationship promoters). This procedure was chosen to confirm the validity of the generated network maps on the part of the egos as well as to make changes if these are necessary on-site of ego to represent the relationship to the alters in its visualisation accurately. Finally, two network maps are visualised as examples for this article. 4 Findings In the following, the results of the survey of network structures in the VET Innovation project AI Pioneers are presented, starting by addressing the first research question focused on the analysed relationships (alters) of the multiple egocentric networks (RQ1). Subsequently the topics of the intensity of the surveyed relationships (RQ2) as well as the function of the alters in the network (RQ3) are presented in detail. The presentation of the results does not follow the list in Table 1 in order to ensure the anonymity of the respondents. Two exemplary network maps (see figure 4 and 57) are included in the presentation of the findings to visualise the results of the survey. The identification of the relationships within the egocentric networks (N = 10) of the present study was carried out from the subjective perspective of the egos interviewed. 4.1 WhichActors(Alters)PlayaCentralRoleinNetworkFormationinthe AIPioneersProject? At the time of data collection, a total of 74 relationships were recorded in the 10 egocentric networks combined. The relationships were categorised by the egos as follows with regard to the social functional system of the actors: Politics (n = 4), economy, (n = 3), education (n = 54), science (n = 7), media (n = 0) and others (n = 6; see Table 3). In the area of politics, the interviewees named four ministries and political administrative bodies (E02; E05; E08), while with regard to the economy, one construction company and two AI companies are listed (E01; E07; E10). A more differentiated picture emerges concerning the sectors education, science and others. In line with the project aims, the education sector is addressed by 7 The two exemplary network maps were selected to illustrate the diversity of the structures of the egocentric networks, as they show e.g. the role of an ego with years of experience in the field with a corresponding number of relationships that already existed before the start of the project and were maintained (Figure 4), compared to the network of an ego that could not draw on such an extensive existing network but had established new relationships in the project context (Figure 5). 450 Mapping Social Capital in Vocational Education and Training the majority (n = 54). In this regard, relationships with a total of 24 vocational schools and adult education centres, as well as three secondary schools, were recorded (see Table 4). These primarily include technical vocational schools (TVET), which are for example involved in robotics projects in industry in connection with process and production automation, projects on the use of AI to increase the energy efficiency of buildings, AI for programming in computer science and mechatronics lessons and, in particular, the use of generative AI in teaching (E06). Furthermore, the use of AI in schools is also being considered in the area of hospitality (E10), and in professional education in aesthetics for training courses (E03): The range of potential application scenarios for AI in the field of vocational and adult education is accordingly very diverse (E01; E03; E05; E06; E09; E10). Furthermore, a large number of European and international associations from the field of vocational and adult education are mentioned by the egos (n = 11), which are dedicated to lifelong learning, e-learning and distance education in addition to vocational education in general (E03; E05; E07; E09). The listed relationships of the egos also include a total of six non-profit organisations from the education sector, which deal with topics such as AI in ethics and pedagogy, learning analytics, digital infrastructure, as well as social disadvantage and educational justice (E02; E07). In addition, relationships with three consultancies in the field of AI in education are listed (E07). The remaining relationships are with internal and external research institutions (n = 20) that deal directly or indirectly with the topic of AI in education, whereby one think tank is also listed (E02; E04; E05; E06; E07; E08). Table 4: Characteristics of the Alters Categories n Categories n Characteristics of the Alters Social functional system Location of partner organisation (1) Politics 4 (1) Country of project partner 56 (2) Economy 3 (2) Other EU country 14 (3) Education 54 (3) Country outside EU 4 (4) Science 7 (5) Media 0 (6) Others 6 Intensity of the Relationships Closeness Communication channel (1) No relationship 2 (1) Mainly face to face in presence 3 (2) Pure exchange of information (weak relationship) 14 (2) Mainly face to face online (video call) 17 (3) Mutual exchange (rather weak relationship) 15 (3) Mainly email 31 451 Meyne, Siemer (4) Goal-oriented coordination (promising relationship) 23 (4) Mainly phone 19 (5) Cooperation (rather strong relationship) 13 (5) Mainly chat (e.g., WhatsApp, Telegram, Skype) 4 (6) Trust in each other (strong relationship) 7 Reciprocity/ mutuality of trust Symmetry of the relationship structure (1) No mutual trust 0 (1) Asymmetrical 4 (2) Low mutual trust 19 (2) Rather asymmetrical 17 (3) Rather high mutual trust 43 (3) Rather symmetrical 28 (4) Very high mutual trust 12 (4) Symmetrical 22 (5) Prefer not to answer 3 Years of acquaintance Frequency of contact (1) Half a year or less 6 (1) Less than once a month 50 (2) Between half a year and 1 year 6 (2) Once a month 22 (3) Between 1 and 2 years 16 (3) Once a week 2 (4) More than 2 years 46 (4) More than once a week 0 Relationship initiation before the project (1) Yes 57 (2) No 17 Function of the Alters Power promoter Expertise promoter (1) No power 40 (1) No expertise 6 (2) Some power 25 (2) Some expertise 31 (3) Rather high power 7 (3) Rather high expertise 25 (4) Very high power 2 (4) Very high expertise 12 Relationship promoter Process promoter (1) No relationship capital 5 (1) No process knowledge 11 (2) Some relationship capital 36 (2) Some process knowledge 30 (3) Rather high relationship capital 23 (3) Rather high process knowledge 20 (4) Very high relationship capital 10 (4) Very high process knowledge 13 Relevance of the Alters Attitude towards the project Influence on the project (1) negative (hindering) 0 (1) No influence 34 (2) Rather negative 0 (2) Some influence 29 (3) Neutral (indifferent) 14 (3) Rather high influence 7 (4) Rather positive 53 (4) Very high influence 4 (5) Positive (promoting) 7 Importance for the project Willingness to support the project (1) Not important 6 (1) No willingness 2 (2) Less important 28 (2) Some willingness 35 (3) Rather important 34 (3) Rather high willingness 31 (4) Very important 6 (4) Very high willingness 6 452 Mapping Social Capital in Vocational Education and Training Interest in the project (1) No interest 1 (2) Less interest 15 (3) Rather high interest 51 (4) Very high interest 7 Note. Each category represents the 74 relationships of the 10 egocentric networks combined. With regard to the geographical location of the relationships, it should be noted that 56 contacts originate from the partner countries of the project consortium, while 18 contacts are with actors in other EU countries (n = 14) and countries outside the EU (n = 4; see Table 4). The specific focus on the countries within the project consortium is justified by an ego regarding the role of the partners in the project context: "This is also related to how the project is structured and what is required from us as a partner (…), you're mostly required to add the contacts from this country. (…). I haven't tried contacting VET schools outside [partners country]" (E10, 69). 4.2 HowaretheRelationshipsWithintheAnalysedEgocentricNetworksFormedinTermsofTheirIntensity? Of the relationships mentioned in all egocentric networks, 57 already existed before the start of the project, while 17 were established during the course of the project at the time of data collection. Looking at the results of the closeness of the relationships, a relatively balanced distribution of the relationships can be seen in the middle characteristics of the category, while the two outlying characteristics (no relationship, n = 2; strong relationship, n = 7) were rarely assigned. Therefore, a total of 14 relationships were categorised as pure exchange of information (weak relationship), 15 relationships as mutual exchange but distanced (rather weak relationship), 23 as goal-oriented coordination (promising relationship) and 13 relationships as cooperation (rather strong relationship; see for example Figure 4). 453 Meyne, Siemer Figure 4: Network Structure and Map of Ego 06 (own compilation) The category of pure information exchange (weak relationship) shows that the relationships rated by ego are mainly about "(…) informing them that we have a project. We are working on these results and if you are interested, I can keep you updated" (E05, 36). In contrast to this, the subsequent category is characterised by a mutual exchange (rather weak relationship), but the relationship has the potential to improve in the future: "We have mutual exchange of different information about our relations. At the moment I cannot say that its strong, but could be developed in the future" (E01, 18). Most relationships in the project are located in the category of the goal-oriented coordination (promising relationship; n = 23), so that a purposeful agreement with clear actions to be coordinated is given. Additionally, the egos interviewed increasingly emphasised that a shared interest regarding the topic of communication is important for maintaining contact (E04; E05; E10). The category cooperation (rather strong relationship) goes beyond the exchange within the goal-oriented coordination and is characterised by a certain intertwining of the actors. This category was primarily selected by the egos with regard to relationships that have already been able to gain many years of experience in joint cooperation, for example through the implementation of shared (Erasmus+) projects in the past. Of the 23 relationships in this category, 17 go beyond the project duration and thus already existed before the start of the project (years of 454 Mapping Social Capital in Vocational Education and Training acquaintance). Furthermore, it can be seen that especially those contacts that were categorised as pure exchange of information were mostly newly established during the project period (10 out of 14). With regard to the duration of the relationships in the other subcategories of closeness, there is a clear concentration of longer established relationships (longer than 2 years) focusing on cooperation (rather strong relationship) and trust (very strong relationship). The level of trust is being evaluated during the years. (…) Trust didn't exist in the very beginning. (…) It was much easier for me to contact people and organisations that I already know and there is this mutual trust. (…) The principal of the school, is a very good friend, and I trust him a lot. (…) He assured me that everything will be okay with the school. (E01, 36 - 38) At the same time, it is also evident that personal and friendly contacts are used in the work context, which is associated with certain key positions of these actors in the professional field (E01; E05). Furthermore, some actors who can already look back on a variety of Erasmus+ project experience have larger network structures and can mobilise existing contacts within the AI Pioneers project effectively (E01, E03, E05, E06, E07), although there are also exceptions here in the interplay of network size and professional experience of the egos. Accordingly, egocentric networks are also present, in which extensive expertise is not accompanied by a correspondingly large network in the project context (E02, E08, E09). At the same time, however, it can also be seen that some egos with smaller project-related networks have established more new relationships within the project duration (E02, E08, E10). Ego 6 contradicts these findings, meaning that a certain network size already existed and was mobilised for the analysed project. At the same time, however, 1/3 of the final egocentric network of Ego 6 consists of new contacts that were established within the project duration. With regard to the reciprocity of the relationships, the majority of these are categorised by the egos with a relatively high level of mutual trust (n = 55). However, a low mutual trust is present in 19 relationships, which is mainly due to imbalances with regard to asymmetrical power relations (E02; E05), a lack of commitment or other negative experiences in the working context within the relationship (E05; E07; E10). No relationship was described in terms of an absence of mutual trust. Furthermore, the ongoing development of mutual trust is emphasised, as according to one interviewee, small and medium-sized enterprises in particular must constantly prove their trustworthiness in the work context: For me it has to do with the credibility and these are pictures of the trust, it's an ongoing process. Maybe you are credible today and if you start delivering fake things, you lose your credibility and this is a very dynamic process. (…) It has to do with the size of the organisation and the type, for example, a university. You can trust mostly a university. Even the persons are changing but it's under the university. But for smaller companies like [partner organisation], this is a process that every day we need to prove our credibility in order to build trust. And this is a two-direction process to have trustful partners and to give trust to other partners. (E01, 98 - 101) 455 Meyne, Siemer With regard to the symmetry of the relationship structure, 21 relationships were found to be (rather) asymmetrical, meaning that one partner has more power than the other (asymmetrical, n = 4; rather asymmetrical, n = 17) according to the categorisation of the egos. While one ego did not want to classify the hierarchical potential of three relationships, the majority of the relationships are classified as rather symmetrical (n = 28) and symmetrical (n = 22), which means both parties have the same amount of power. Equal cooperation is increasingly emphasised during the interviews (E01; E03; E06; E08), here personal networking preferences as well as target group-specific aspects become apparent. For example, the desire to balance out the power imbalance is particularly pronounced in the case of an ego in order to establish beneficial work on an equal level: "I always find it completely unpleasant to talk to people where I have the feeling that they are in an exaggeratedly supplicant or somehow inferior position. I don't think anything good comes out of that" (E02, 40). Furthermore, the results show that attracting suitable people from the project's target groups can be a hurdle, which reproduces asymmetrical relationships, as the project is dependent on practitioners, especially VET schools, being interested and wanting to cooperate (E02; E04; E05). In addition, the size of the participating organisations (alters) in terms of available resources and personnel involved in relation to the project partners (egos) in the AI Pioneers consortium is also emphasised, which can also lead to an asymmetrical power imbalance if one side is institutionally significantly larger than the other (E01; E07). In terms of the main communication channels used, there is a clear dominance regarding emails as a central communication medium (n = 31), while the telephone (n = 19) and online video calls (n = 17) are also used. In contrast, face-to-face meetings in person (n = 3) are hardly used as the predominant communication channel, which seems to be partly due to the conditions of the Covid pandemic and the international nature of these projects: "The reality is that since the pandemic, we've (…) cut down grossly the amount of face-to-face contact and I wonder if that affects trust. I think that does affect trust" (E07, 73-87). Finally, chat functions are only used in four relationships as the most frequently used medium for communicating with each other. Although it should be noted that all egos emphasised that they use a variety of communication channels simultaneously, depending on their personal preference and that of their counterpart. While a large number of different channels are used for communication, the frequency of contact via these channels is strongly linked to the respective project phase and the activities that are planned in these in exchange with the project target group. Accordingly, the egos reported an average frequency of contact of less than once a month for 50 relationships. In the case of 22 relationships, communication takes place once a month and in only two relationships once a week. 456 Mapping Social Capital in Vocational Education and Training 4.3 WhatFunctionsandRelevanceDoTheseAltersBringtothe EgocentricNetworks? With regard to the design of the promoter roles and thus the function of the actors in the egocentric networks, the egos interviewed largely agree that the alters in the country-specific networks show low hierarchical potential in the project-related network. The majority of the relationships (n = 40) were categorised regarding the role of the power promoter as having no power, while 25 relationships are marked as having some (see Table 4). Furthermore, one ego highlights: "It's about the research they [alter] do and of course that also has some influence on how we assess the AI applications, the use cases (...), but there is definitely an influence. It is not without influence" (E06, 53). Accordingly, the egos surveyed emphasise that the alters contribute their expertise to the project, but that they nevertheless do not have a certain amount of authority to influence project-related processes (E01; E02; E03; E04; E05; E10). However, there are certain alters that were classified significantly higher in this category either because they occupy a special position as pioneers in the implementation of AI in VET in terms of their technical expertise or because they have a direct link to politics and are therefore involved in discussion processes about AI in VET at policy level (E06; E07; E08). These still represent a minority of the relationships analysed in this survey. Addressing the role of the expertise promoter, the majority of the relationships and actors were categorised as having some expertise (n = 31), rather high expertise (n = 25) and very high expertise (n = 12) in the field of AI in education: "They have a lot of capital all in different areas, they have a lot of experience first, and second a broad network of collaboration. So, they can be multipliers of what we are doing with our project" (E01, 72-73). According to the egos interviewed, such alters with (rather) high expertise possess such expertise due to their many years of professional experience in the implementation of Erasmus+ projects, their practical knowledge in VET, including AI use cases and initiatives with work-based learning approaches, as well as their contribution of strategic perspectives in VET. Furthermore, the academic expertise in the field of AI in education is particularly emphasised among the alters, as well as the technical expertise of AI developers (E01, E02, E04, E05, E06, E07, E08, E10). Only six relationships were marked with no expertise in this regard. Such alters, who have not been identified as having any specialist expertise, are certainly interested in the project results and most importantly have knowledge with regard to the other promoter roles, e.g. high networking skills (E02, E03, E09). Language barriers when it comes to contributing expertise as well as the identification and establishment of new relationships with relevant experts is emphasised as a central hurdle for network formation (E02; E03; E04; E08; E09; E10). Additionally, the distribution of the role of the relationship promoter in the egocentric networks shows that a total of 36 alters have some relationship capital, a further 23 have rather high relationship capital, while 10 alters possess a very high relationship capital related 463 Meyne, Siemer subject-related analyses – as is the case in the use of egocentric network analysis – are to make the methodological procedure transparent and reproducible and to enable verification and falsification of the results. In the following, the use of egocentric social network analysis with regard to the chosen methodological approach (RQ3) is addressed in order to identify findings and obstacles for future VET research. According to Lockwood et al. (2024), the critical evaluation of qualitative research is based on an exact match between the chosen research questions, the theoretical and conceptual foundations used and the resulting data collection instrument. The methodological approach of our study thus allows the following conclusions to be drawn with regard to the quality assurance of the results: – The qualitative approach in this study was systematised in terms of its transparency, comprehensibility and reproducibility by deriving a theory-based research approach. Based on the theoretical framework, the qualitative standardised questionnaire and semi-structured guideline were then developed. This methodological approach makes it possible to increase the comprehensibility of the results and to classify them within a given framework. Based on the theoretical framework, which combines approaches from psychology, sociology and innovation research, the survey instruments could be conceptually and strategically framed, which in turn increases both the transparency of the results and their comprehensibility and enables the results to be compared. – The validity of the methodological approach chosen here receives further empirical evidence through an upstream pre-test and through its application in the AI Pioneers study context. The congruence between research methodology, research question and data collection methods can therefore be regarded as given (Lockwood et al., 2024). The validity of the identified network maps also receives further evidence through the review and approval of the egos. Nevertheless, it should be noted that the survey instruments are not validated up to now. – Against the background of the objectivity of the results, it should be taken into account that the assessments of the egocentric networks are based on the subjective perception of the egos. Nevertheless, the combination of the fully and semi-structured survey instruments proved to be useful in the course of the data collection during the interview settings in order to create comparability of the data and to capture complex and in-depth information about the subject of investigation. – According to the accuracy of the research questions, the theory, the survey instruments and the case selection, the field access and the unit of analysis must also be 464 Mapping Social Capital in Vocational Education and Training chosen appropriately so that the research design is coherent (Lockwood et al., 2024). In this respect, the present study referred to the funded project partners within the AI Pioneers project, so that direct field access was already given by the joint funding context. In accordance with the survey method of multi-perspective egocentric network analysis, the sample was selected on the basis of the funded project partners in the network and thus on their relationship promoters to get an in-depth view of the project-related network on the basis of the personal network structures combined. – The application of the promoter model proved to be useful in the context of this survey, so that the egocentric networks could be analysed from the perspective of the person who summarises the contacts per partner organisation. On the basis of this, the network can be narrowed down to the individual assessment of the interviewees. The definition of network boundaries is essential for the data collection of egocentric networks (see Chapter 3). The natural boundaries defined by the funding framework of the AI Pioneers project made it possible to narrow down the relevant relationships for this empirical study. – The perspective of the alters was not recorded, so that a partial view of the project network could be provided from multiple perspectives. Accordingly, the empirical results do not contain an objectifiable measure, as would be the case in whole network structures. Consequently, the network maps are an individual reflection of the social capital of the egos interviewed. The connections to alters shown in the network maps provide insights into the individual social structures of the AI Pioneers project partners. At the time of the survey, they provided important insights into personal resources and enable a description of the intensity and function of these from a subjective perspective. – By using social network analysis, it was possible to identify predominantly strong ties between ego and alters. However, with the methodological approach used here, it remains unclear to what extent the alters mentioned assess the bond with the egos and how the alters are interconnected with each other with regard to the exchange of content on artificial intelligence in vocational education and training. It should also be pointed out that further relevant institutions (alters) might be included in the further course of the project and that the network maps shown in the article do not provide a complete picture of the network structure of the AI Pioneers project as would be the case in the recording of whole network structures. 465 Meyne, Siemer Despite the efforts listed here to design the methodological approach used for this study in a comprehensible and transparent manner, our study has limitations. This paper concludes with the limitations of the study and an outlook for future research. 6 Conclusion,LimitationsandFutureResearch In this study, the methodological approach of a multi-perspective egocentric social network analysis was used to analyse the network structures from the perspective of individual actors (personal networks). This has been addressed on the basis of an innovation project with the central goal of network formation about AI in VET in an international context by collecting data on ego-alter relationships. Furthermore, the combination of egocentric data as a multiple-perspective egocentric network analysis is beneficial for an overarching view of a common context (e.g., project or programme) in order to gain an initial insight into network structures that go beyond the perspective of a single ego, but at the same time do not correspond to the undertaking of recording overall network structures. In a first step, the two (standardised) survey instruments were created for the purpose of the study, so that in a second step, they were applied in the course of a qualitative data collection of 10 egocentric networks (partners of the Erasmus+ AI Pioneers project as egos). Subsequently the data were used to generate network maps to visualise exemplary egocentric networks of the study. One advantage of the chosen methodological approach is that in the context of the project-related network development of AI Pioneers, a detailed analysis of a person's social relationships (ego) and their direct connections (alters) was possible. Furthermore, the study not only makes a significant contribution to VET research and its methodological set, but also to the Erasmus+ funding landscape by highlighting personal resources, social support and individual perceptions of project-related networks. On the basis of the knowledge gained in this study, it would be possible in a next step to take measures to expand the project network structures regarding important groups of stakeholders that are not yet sufficiently reached (e.g., from politics and media) to use the results of the study as a foundation for future project steering decisions. Accordingly, the methodological approach pursued here offers the possibility of making transfer and network activities of project consortia visible in order to draw attention to the necessary resources to build and maintain those. These findings can be taken into account when announcing further funding programmes with a similar focus. With a view to future programmes, the extent to which emergent phenomena influence the network formation of the egos during the funding period could also be investigated, e.g. to potentially identify network patterns in terms of the response to the project environment of the actors involved. To this end, a process-orientated analysis designed as a 466 Mapping Social Capital in Vocational Education and Training longitudinal study with different time horizons up to the end of the project, could provide information on the background against which further relationships (e.g., new social capital due to expertise) are entered into, if existing relationships are terminated prematurely (e.g., due to misuse of trust) or in how far topic related trends affect the motivation to join new groups and networks. This allows both the dynamics of network formation to be recognised and the sustainability of the networks after the funding period to be determined. Future research could also investigate the extent to which network size and previous professional practice affect each other in detail. Moreover, the role of forward-looking actors and innovations that arise within networks could also be investigated my means of future research projects. The following research questions could guide future research as a continuation of the present study: –To what extent do the egocentric networks change over the course of the implementation of the project? –What factors influence the intensity of the relationships between ego and alter? –To what extent is there a relation between network size and professional experience from previous similar project contexts? –What forms of project consortia are particularly beneficial for the creation of innovations within networks? Corresponding data could be collected in the form of partial networks and/or an extension of alter-alter relationships in (multi-perspective) egocentric networks. For the present study of the AI Pioneers project, the relationships between the 10 egos would also be beneficial to analyse, as well as the resulting conclusions for internal relationship management within a consortium. In accordance, for example the connections between Ego 1, Ego 2 and their shared relationships (alters) could be analysed. Such a methodical approach would make it possible to identify changes in the subjectively perceived relevance of the relationships that might have a long-term positive effect on the design of the egocentric networks and consequently on the success of the project. Furthermore, the findings of this study are based on a small sample size which nevertheless produced an extensive amount of qualitative data. Statements about a complete survey of all relationships within a social system would mean recording the entire network. Following on from this article, the entire network of AI Pioneers could be analysed in a further empirical study. The following question could guide this: How extensive are the whole network structures of the AI Pioneers project and the interconnections between alters? 467 Meyne, Siemer This qualitative approach could be expanded in a follow-up study with a quantitative approach. Accordingly, the next step could be to develop a quantitative questionnaire based on the theoretical framework used here, which draws on comprehensive scales in order to be able to make valid statements regarding the overall network of AI Pioneers. Such a study could provide information on the extent of cooperation between stakeholders or institutions in the establishment and expansion of an innovative international network in the field of vocational education and training. In this context, with regard to the social capital, it would also be interesting to analyse strong and weak relationships in the network in more detail, for example by measuring centrality. It would also be interesting to see whether and how certain characteristics (nodes or edges) are related to each other. Such an approach would make it possible to identify relationship behaviour more comprehensively. In line with the study's approach, it should be emphasised that the approaches of recording egocentric networks and whole networks are not mutually exclusive. Both methodological approaches to survey social network structures can complement each other. For example, researchers can start with egocentric network studies in order to understand central actors and their relationships with each other, and then follow up with an analysis of the entire network or partial sections that go beyond the analysis of individual egocentric networks. In any case, the delimitation of the network is of central importance, which is particularly useful with regard to the investigation of funded collaborative projects and relationships that have arisen or are maintained in the project context due to the clear possibility of defining network boundaries. Due to the specific framework conditions of the funded AI Pioneers project, it is assumed that the relationships identified initially only apply to the multiple unique egocentric networks. It is unclear to what extent the relevance and intensity of the relationships can be transferred to other projects. The study also does not allow any conclusions to be drawn about the connection between the identified egocentric networks and the probability of success of the AI Pioneers project as a whole. In conclusion, it can be pointed out that despite the limitations described, the methodological approach is fruitful due to the combination of qualitative approaches for analysing the institutionally related egocentric networks from multiple perspectives and the description of social capital for the European project landscape. Further qualitative and quantitative network analyses would be an enrichment for VET research, especially against the background of innovation drivers and network formation. Innovation drivers such as market-related and trend-related drivers for example regarding the technical development of AI or related training programmes offered by vocational training providers could have an impact on the content dimension of network formation. Increased demand for the integration and ethical use of AI in educational settings may lead to an upswing in the willingness of individuals and organisations to reach out to others 468 Mapping Social Capital in Vocational Education and Training and participate in topic-specific networking activities. In accordance, further analysis of forward-looking and topic-specific network formation processes with a view to the internationalisation of vocational education and training in order to learn from each other across national borders would be an enrichment for international comparative vocational education and training research. EthicsStatement The research presented here was carried out according to the principles for research involving the participation of human subjects based on the guidelines described in IJRVET's ethical statement. The interviewees consented to the collection, recording and analysis of the interviews and the anonymous presentation of the findings. Acknowledgement This article is based on the joint Erasmus+ project "AI and the future of Education" (acronym: AI Pioneers; Project number: 101087261). The project is funded as part of the Erasmus+ program under the "Forward-Looking-Projects". We would particularly like to thank the AI Pioneers project partners for their willingness to participate in this survey. References Abbott, A. (1995). Things of boundaries – Defining the boundaries of social Inquiry. Social Research, 62(4), 857–882. https://www.jstor.org/stable/40971127 Albertz, A., & Pilz, M. (2025). Green alignment, green vocational education and training, green skills and related subjects: A literature review on actors, contents and regional contexts. International Journal of Training and Development, 29(2), 243–254. https://doi.org/10.1111/ijtd.12359 Attwell, G., Deitmer, L., & Bekiaridis, G. (2023). AI Pioneers: Developing a community of practice for artificial intelligence (AI) and vocational education and training. In V. Tūtlys, L. Vaitkutė & C. Nägele (Eds.), Vocational Education and Training Transformations for Digital, Sustainable and Socially Fair Future. Proceedings of the 5th Crossing Boundaries Conference in Vocational Education and Training, Kaunas, 25. – 26. May (pp. 30–37). European Research Network on Vocational Education and Training, VETNET, Vytautas Magnus University Education Academy, Institute of Educational Science. https://doi.org/10.5281/zenodo.7808076 Bourdieu, P. (1983). Ökonomisches Kapital, kulturelles Kapital, soziales Kapital. In R. Kreckel (Ed.), Soziale Ungleichheiten (pp. 183–198). Schwartz. Burt, R. S. (1992). Structural holes: The social structure of competition. Harvard University Press. Cedefop., & ReferNet. (2023). Strategy for artificial intelligence in education and training: Austria. Timeline of VET policies in Europe [online tool]. Cedefop, ReferNet. Cedefop/ReferNet. https://www.cedefop.europa.eu/en/tools/timeline-vet-policies-europe/search/29855 469 Meyne, Siemer Coppe, T., März, V., & Raemdonck, I. (2023). Second career teachers' work socialization process in TVET: A mixed-method social network perspective. Teaching and Teacher Education, 121, 103914. https://doi.org/10.1016/j.tate.2022.103914 Deitmer, L., Dersch, I., Meyne, L., Siemer, C. (2024). How do technical vocational school teachers deal with artificial intelligence in the classroom? An attempt to analyze AI use at German VET schools. Ubiquity Proceedings, 4(1), 21. https://doi.org/10.5334/uproc.143 Döring, N. (2023). Forschungsmethoden und Evaluation in den Sozialund Humanwissenschaften (6. Auflage). Springer. https://doi.org/10.1007/978-3-662-64762-2 De Witt, C. (2024). Künstliche Intelligenz in der Berufsbildung. Technologische Entwicklungen, didaktische Potenziale und notwendige ethische Standards. BWP, 53(1), 8–12. https://www.bwpzeitschrift.de/dienst/publikationen/de/19416 European Commission. (2022). Erasmus+ Program Guide (Version 2: 26-01-2022). European Commission. https://erasmus-plus.ec.europa.eu/sites/default/files/2022-01/2022-erasmusplus-programme-guide-v2_en_0.pdf European Commission. (2024). Erasmus+ Program Guide (Version 1: 28-11-2023). European Commission. https://erasmus-plus.ec.europa.eu/sites/default/files/2023-11/2024-Erasmus%2BProgrammeGuide_EN.pdf Friebertshäuser, B., & Langer, A. (2013). Interviewformen und Interviewpraxis. In B. Friebertshäuser, A. Langer & A. Prengel (Eds.), Handbuch qualitative Forschungsmethoden in der Erziehungswissenschaft (pp. 437–455). Beltz Juventa. Fuhse, J. A. (2018). Soziale Netzwerke, Konzepte und Forschungsmethoden (2. edition). utb. Gemünden, H.-G., & Hölzle, K. (2005). Schlüsselpersonen der Innovation. Champions und Promotoren. In S. Albers & O. Gassmann (Hrsg.), Handbuch Technologie und Innovationsmanagement (pp. 457–474). Springer. https://doi.org/10.1007/978-3-322-90786-8 Gemünden, H. G., Salomo, S., & Hölzle, K. (2007). Role models for radical innovations in times of open innovation. Creativity and Innovation Management, 6(4), 408–421. https://doi.org/10.1111/j.1467-8691.2007.00451.x Gemünden, H. G., & Walter, A. (1998). Beziehungspromotoren: Schlüsselpersonen für zwischenbetriebliche Innovationsprozesse. In J. Hauschildt & H. G. Gemünden (Eds.), Promotoren: Champions der Innovation (pp. 113–132). Springer Gabler. Gessler, M. (2019). Promotoren der Innovation im transnationalen Berufsbildungstransfer: Eine Fallstudie. In M. Gessler, M. Fuchs & M. Pilz (Eds.), Konzepte und Wirkungen des Transfers dualer Berufsbildung (pp. 231–279). Springer Fachmedien. https://doi.org/10.1007/978-3-658-23185-9 Gessler, M., & Siemer, C. (2020). Nachhaltigkeit internationaler Berufsbildungszusammenarbeit: Erfassung des sozialen Kapitals mittels personaler Netzwerkanalysen. In Berufsbildung International – Nachhaltigkeit (pp. 44–47). Bundesministerium für Bildung und Forschung (BMBF). https://www.berufsbildung-international.de/wp-content/uploads/2022/08/ibb_publikation_ nachhaltigkeit_final.pdf Granovetter, M. S. (1973). The strength of weak ties. American Journal of Sociology, 78(6), 1360–1380. https://doi.org/10.1086/225469 Grollmann, P., Meyne, L., & Neff, D. (2024). Ausbildungspersonal und Künstliche Intelligenz – KI Pioniere: Ein Blick aus der Perspektive internationaler Berufsbildungsexperten. berufsbildung, Zeitschrift für Theorie-Praxis-Dialog, 78(3), 26–28. https://doi.org/10.3278/BB2403W 470 Mapping Social Capital in Vocational Education and Training Gruber, H., Hirschmann, M., & Rehrl, M. (2018). Bildungsbezogene Netzwerkforschung. In R. Tippelt & B. Schmidt-Hertha (Hrsg.), Handbuch Bildungsforschung (4. ed., pp. 1339–1356). Springer Fachmedien. https://doi.org/10.1007/978-3-531-19981-8_60 Hansen, D. L., Shneiderman, B., & Smith M. A. (2011). Analyzing social media networks with NodeXL. Insights from a connected world. Elsevier. https://doi.org/10.1016/C2009-0-64028-9 Hauschild, J., & Chakrabarti, A. K. (1988). Arbeitsteilung im Innovationsmanagement – Forschungsergebnisse, Kriterien und Modelle. Zeitschrift Führung und Organisation, 57(6), 378–389. Häußling, R. (2009a). Einleitung. In R. Häußling (Ed.), Grenzen von Netzwerken (pp. 7–14). VS Verlag für Sozialwissenschaften. https://doi.org/10.1007/978-3-531-91856-3 Häußling, R. (2009b). Zur Bedeutung von Emotionen für soziale Beziehungen. Möglichkeiten und Grenzen der Netzwerkforschung. In R. Häußling (Ed.), Grenzen von Netzwerken (pp. 81–103). VS Verlag für Sozialwissenschaften. https://doi.org/10.1007/978-3-531-91856-3 Hollstein, B., & Pfeffer, J. (2010). Netzwerkkarten als Instrument zur Erhebung egozentrierter Netzwerke. In H. G. Soeffner (Eds.), Unsichere Zeiten: Herausforderungen gesellschaflicher Transformationen. Verhandlungen des 34. Kongress der Deutschen Gesellschaf für Soziologie in Jena. VS Verlag für Sozialwissenschaften. http://www.pfeffer.at/egonet/Hollstein%20Pfeffer.pdf Hosmer, L. T. (1995). Trust: The connecting link between organizational theory and philosophical ethics. The Academy of Management Review, 20(2), 379–403. https://doi.org/10.2307/258851 Jecker, C., & Spachmann, K. (2023). Vertrauen in der internen Kommunikation von Organisationen. In J. Basel & P. Henrizi (Eds.), Psychologie von Risiko und Vertrauen Wahrnehmung, Verhalten und Kommunikation (pp. 225–248). Springer. https://doi.org/10.1007/978-3-662-65575-7 Karafillidis, A. (2009). Entkopplung und Kopplung - Wie die Netzwerktheorie zur Bestimmung sozialer Grenzen beitragen kann. In R. Häußling (Ed.), Grenzen von Netzwerken (pp. 105–131). VS Verlag für Sozialwissenschaften. http://doi.org/10.1007/978-3-531-91856-3_6 Kronenwett, M., & Schönhuth, M. (2014). VennMaker 1.4.0 Anwenderhandbuch. VennMaker. https:// www.vennmaker.com/files/VennMaker_1_4_0_Anwenderhandbuch.pdf Kuckartz, U. (Ed.) (2018). Qualitative inhaltsanalyse, Methoden, Praxis, Computerunterstützung (4th ed.). Beltz Juventa. Lockwood, C., Porritt, K., Munn, Z., Rittenmeyer, L., Salmond, S., Bjerrum, M., Loveday, H., Carrier, J., & Stannard, D. (2024). Systematic reviews of qualitative evidence. In E. Aromataris, C. Lockwood, K. Porritt, B. Pilla & Z. Jordan (Eds.), JBI manual for evidence synthesis. JBI. https://doi.org/10.46658/JBIMES-24-02 Marsden, P. V., & Campbell, K. E. (1984). Measuring tie strength. Social Forces, 63(2), 482–501. https://doi.org/10.2307/2579058 Marsden, P. V., & Campbell, K. E. (2012). Reflections on conceptualizing and measuring tie strength. Social Forces, 91(1), 17–23. https://doi.org/10.1093/sf/sos112 McCallister, L., & Fischer, C. S. (1978). A procedure for surveying personal networks. Sociological Methods & Research, 7(2), 131–148. https://doi.org/10.1177/004912417800700202 McGrath, S., Mulder, M., Papier, J., & Suart, R. (2019). Handbook of vocational education and training: Developments in the changing world of work (1st ed.). Springer Nature. https://doi.org/10.1007/978-3-319-94532-3 Merton, R. K. (1968). The Matthew Effect in science. Science, 159(3810), 56–63. https://doi.org/10.1126/science.159.3810.56 471 Meyne, Siemer Messmann, G., Mulder, R. H., & Palonen, T. (2018). Vocational education teachers' personal network at school as a resource for innovative work behaviour. Journal of Workplace Learning, 30(3), 174– 185. https://doi.org/10.1108/JWL-08-2017-0069 Mewes, J. (2009). Die räumlichen Grenzen persönlicher Netzwerke. In R. Häußling (Ed.), Grenzen von Netzwerken (pp. 33–53). VS Verlag für Sozialwissenschaften. https://doi.org/10.1007/978-3-531-91856-3 Meyne, L., & Siemer, C. (2024). Social capital of actors in VET: An egocentric case study based on the AI Pioneers project. In C. Nägele, B. Esmond, N. Kersh & B. E. Stalder (Eds.), Trends in vocational education and training research, vol. VII. Proceedings of the European Conference on Educational Research (ECER), Vocational Education and Training Network (VETNET) (pp. 75–81). VETNET. https://doi.org/10.5281/zenodo.13236742 Perry, B. L., Pescosolio, B. A., & Borgatti, S. P. (2018). Egocentric network analysis: Foundations, methods, and models. Cambridge University Press. Peters, S., & Gessler, M. (2019). Die B edeutung von Promotoren im Berufsbildungstransfer. In DLR Projektträger (Eds.), Geschäftsmodellentwicklung. Berufsbildung International (pp. 26–28). DLR Projektträger. https://www.berufsbildung-international.de/wp-content/uploads/2022/08/ibb_publikation_01-19_geschaeftsmodellentwicklung.pdf Petzold, H. G. (1993). Integrative Therapie. Modelle, Theorien und Methoden für eine schulenübergreifende Psychotherapie. 1. Klinische Philosophie: Transversale Diskurse. Junfermann Verlag. Pilz, M. (2016). Typologies in comparative vocational education: Existing models and a new approach. Vocations and Learning, 9(3), 295–314. https://link.springer.com/article/10.1007/s12186-016-9154-7 Pilz, M., & Zenner‐Höffkes, L. (2023). "There's always someone else to blame": A comparison of the political discourse surrounding 'vocational maturity' in Germany and 'Employability' in the UK. Journal of Vocational Education & Training, 77(2), 392–420. https://doi.org/10.1080/13636820.2023.2229796. Platzköster, M. (1990). Vertrauen. Theorie und Analyse interpersoneller, politischer und betrieblicher Implikationen. Beleke. Project Management Institute. [PMI] (2017). A guide to the project management body of knowledge (PMBOK® Guide; 6th ed.). Newton Square, Project Management Institute. Putnam, R. D. (2001). Social Capital: Measurement and Consequences. Isuma: Canadian Journal of Policy Research, 2, 41–51. Putnam, R. D. (1995). Bowling Alone: America's Declining Social Capital. Journal of Democracy, 6(1), 65–78. Roppertz, S. (2021). Die Rolle und Bedeutung von Künstlicher Intelligenz in der Berufsausbildung – Implikationen für angehende Berufsund Wirtschaftspädagog*innen. bwp@ Berufsund Wirtschaftspädagogik – online, 40, 1–23. https://www.bwpat.de/ausgabe40/roppertz_bwpat40.pdf Röhrer, N., Vogelsang, B., & Fuchs, M. (2021). The region matters—for whom? The regional actor network for vocational education and training in tourism of Cancún (Mexico). International Journal of Training and Development, 25(3), 244–258. https://doi.org/10.1111/ijtd.12217 Schenk, M. (2010). Medienforschung. In C. Stegbauer & R. Häußling (Eds.), Handbuch Netzwerkforschung (pp. 773–884). VS Verlag für Sozialwissenschaften. https://doi.org/10.1007/978-3-531-92575-2 472 Mapping Social Capital in Vocational Education and Training Schlicht, J., & Moschner, U. (2018). Kommunizieren und kooperieren, um erneuerbare Energien zu nutzen Befunde und Perspektiven für eine berufliche Bildung für Nachhaltigkeit. In J. Schlicht & U. Moschner (Eds.), Berufliche Bildung an der Grenze zwischen Wirtschaft und Pädagogik (pp. 91–113). Springer VS. https://doi.org/10.1007/978-3-658-18548-0_6 Schweer, M. (2008). Vertrauen und soziales Handeln - Eine differentialpsychologische Perspektive. In E. Jammal (Ed.). Vertrauen im interkulturellen Kontext. Perspectives of the other: Studies on intercultural communication (pp. 13–26). VS Research. http://doi.org/10.1007/978-3-531-91038-3_2 Schweer, M. (2006). Vertrauen. In D. H. Rost (Ed.), Handwörterbuch pädagogische Psychologie (3rd ed., pp. 848–852). Beltz PVU. Schweer, M. (2022). Vertrauen und Misstrauen im Kontext aktueller gesellschaftlicher Herausforderungen – zwei zentrale psychologische Regulatoren. In M. Schweer (Ed.), Facetten des Vertrauens und Misstrauens Herausforderungen für das soziale Miteinander (pp. 1–16). Springer VS. Siemer, C. (2023). Studienpartner im BMBF-Programm "Internationalisierung der Berufsbildung": Eine qualitative Analyse zur Untersuchung der Selbstverständnisse wissenschaftlicher Begleitung auf Projektebene. Zeitschrift für Berufsund Wirtschaftspädagogik (ZBW), 119(2), 236–254. https://doi.org/10.25162/zbw-2023-0010 Siemer, C., & Gessler, M. (2021). The role of research partners in funded model projects in the context of the internationalisation of VET: Research partners as promoters. In C. Nägele, N. Kersh, & B. E. Stalder (Eds.), Trends in vocational education and training research, Vol. IV. Proceedings of the European Conference on Educational Research (ECER), Vocational Education and Training Network (VETNET) (pp. 270–278). https://doi.org/10.5281/zenodo.5172437 Witte, E. (1973). Organisation für Innovationsentscheidungen: Das Promotoren-Modell. Verlag Otto Schwartz & Co. Zaltman, G., & Moorman, C. (1988). The importance of personal trust in the use of research. Journal of Advertising Research, 28(5), 16–24. BiographicalNotes Lisa Meyne, M.A., is a doctorate candidate and a research associate at the Institute Technology and Education (ITB) at the University of Bremen in Germany. Her research interests include the international transfer of vocational education and training (VET) and the complexity and innovation of VET services. She also focuses on the analysis of social networks in international VET contexts. Dr Christine Siemer is a postdoctoral researcher at the Institute Technology and Education (ITB) at the University of Bremen in Germany. Her research focuses on internationalisation and artificial intelligence in vocational education and training. The current themes of her research are technology-enhanced teaching, and instructional design of learning environments as well as social network analyses in the context of international VET cooperations.