From access to agency: addressing the conversion gap in agripreneurship education for young migrant and NEET women

Entrepreneurial participation has risen, yet disparities persist – particularly for young women who are Not in Education, Employment or Training (NEET) and those with migrant backgrounds (Galvão et al., 2025b; Santos-Ortega et al., 2021). In contexts of constrained employment and livelihood insecurity, entrepreneurship can offer a route to economic participation. However, intersecting barriers – unequal education access, stereotypes and discriminatory institutional practices – restrict both labour-market and entrepreneurial engagement (Abamosa, 2023; Berry et al., 2024). In the European Union (EU), the NEET rate among women aged 15–29 is 14.4% (10.7% for men), often linked to caregiving responsibilities (Santos-Ortega et al., 2021). Among youth born outside the EU, NEET prevalence is higher (21% vs 12.5% for natives), especially among migrant girls and young women (25.9%), and 25% of foreign-born women reported obstacles in finding suitable work in 2021 (Orav, 2023; Azmat, 2013).

Within this context, agripreneurship – entrepreneurship in the agricultural sector – is often framed as a locally grounded and comparatively accessible livelihood pathway, linked to self-employment, rural revitalisation and sustainable development (Aggarwal and Johal, 2021). A recent systematic review further shows that migrant entrepreneurship often takes sustainable and social forms, aligning support with green and inclusive growth agendas (Raimi and Mahmoud, 2025). Yet gendered constraints remain entrenched, including unequal access to land, finance, networks and markets (Dias et al., 2019; Lourenço et al., 2014). Programmes also often overlook caregiving and related time and mobility constraints, reproducing marginalisation (Kawarazuka et al., 2023).

Against this backdrop, entrepreneurship education is often framed as inherently empowering, but it frequently fails to address the intersecting inequalities shaping marginalised women’s engagement (Gidley et al., 2010; Kawarazuka et al., 2023). Anchored in social inclusion theory – as a progression from access to participation to success through empowerment (Gidley et al., 2010) – this study examines how higher education (HE) and vocational education and training (VET) operationalise inclusion in agripreneurship. Despite the expansion of entrepreneurship education across HE and VET, prevailing approaches often overlook intersectional identities and barriers, reducing relevance for marginalised groups (Lahooti, 2025; Mahajan et al., 2025). Evidence of migrant women’s strategic navigation of exclusion points to the limits of individual skill-building in the absence of systemic change (Lahooti, 2025). Consistent with this, migrants’ contributions to job creation and innovation are strongest when financial support and favourable policies are paired with genuinely inclusive environments (Raimi and Mahmoud, 2025). Accordingly, the first objective is to examine how agripreneurship is integrated within HE and VET and to identify delivery and structural features – present or absent – that shape access and the quality of participation for young women in NEET situations and with migrant backgrounds.

Even when access exists, curricula often prioritise individual competencies while sidelining structural inequalities such as gender bias, discriminatory policies and exclusion from male-dominated networks (Hughes et al., 2012; Kawarazuka et al., 2023). Standardised designs can ignore intersectional realities – caregiving, economic insecurity and cultural barriers – limiting participation and slowing progression towards success through empowerment (Pliakoura et al., 2020). Feminist critiques therefore stress gender-responsive, intersectionally informed pedagogies that accommodate women’s dual economic and domestic roles (Ekinsmyth, 2014; Gustavsson, 2021). Without explicit integration of such approaches, entrepreneurship education can reinforce rather than challenge systemic inequalities (Mahajan et al., 2025). The second objective is therefore to identify, from educators’ perspectives, the specific training needs of these groups to inform gender-responsive, contextually inclusive pedagogies.

Advancing empowerment requires reforms that support navigation of discriminatory practices, socio-economic barriers and market exclusion, rather than a narrow focus on competencies (Azmat, 2013; Kyrgidou and Petridou, 2013). Many initiatives still neglect gendered power relations and social reproduction, limiting sustainable outcomes (Aggarwal and Johal, 2021; Dias et al., 2019; Hughes et al., 2012; Kawarazuka et al., 2023). In a low-density context, most immigrant and return-migrant entrepreneurs received no start-up support mainly because they lacked information about support sources (Galvão et al., 2025b). Thus, the third objective is to assess educators’ views on programme effects, persistent barriers, market-readiness challenges, and whether enabling supports (e.g. finance or buyer brokerage) are embedded to translate participation into success through empowerment.

Guided by qualitative evidence from HE professors and VET trainers across 11 European countries, and focusing on young women in NEET situations and with migrant backgrounds, the study addresses the following research question:

The analysis addresses three linked dimensions aligned with staged inclusion:

1. integration and pedagogy shaping access and participation;

2. training needs and how constraints are framed (skills gaps vs structural barriers); and

3. assessed effects, market-readiness challenges and whether enabling supports (e.g. finance or buyer brokerage) are reported to translate participation gains into success through empowerment.

This research contributes in three ways. Firstly, it extends agripreneurship education through staged social inclusion by specifying a conversion gap that interrupts movement from access and participation to success through empowerment for young women in NEET situations and with migrant backgrounds (Kawarazuka et al., 2023; Marlow, 2020). Secondly, drawing on educator accounts across HE and VET, it shows how skills-first provision can recode intersectional structural constraints as individual deficits, highlighting the need for gender-responsive redesign (Dias et al., 2019). Thirdly, it specifies a feminist pedagogy conversion mechanism – recognition, redistribution and relationality – and links it to institutional supports and market brokerage that may help convert participation gains into empowerment trajectories (Mahajan et al., 2025).

Gendered constraints restrict women’s access to land, finance, networks and markets in agripreneurship (Dias et al., 2019). Feminist scholarship links these constraints to patriarchal norms and unequal power relations that normalise male authority in agriculture and entrepreneurial ecosystems, shaping recognition and valuing of women’s work (Hughes et al., 2012; Kawarazuka et al., 2023). Masculinised framings of agriculture and rural enterprise constrain entry, networking and legitimacy, while care responsibilities and time poverty shape venture choices and limit growth trajectories (Gawel et al., 2024; Gustavsson, 2021). These dynamics matter for education because they affect feasibility of participation and whether learning translates into agency and market traction.

A core debate concerns what explains unequal outcomes: skills and individual agency, or structural conditions and institutional gatekeeping. Mainstream accounts prioritise competence development, with structural inequalities treated as background (Lourenço et al., 2014). In contrast, gender and migration scholarship emphasises layered exclusion – discrimination, bureaucratic and legal hurdles, cultural bias and thin networks – that shapes what skills can convert into opportunities (Abamosa, 2023; Azmat, 2013; Berry et al., 2024; Lahooti, 2025). For migrant women, exclusion is intensified by language barriers, unfamiliar institutional frameworks and racialised/ethnicised positioning, and by the invisibility of women’s agricultural contributions in masculinised discourses (Gustavsson, 2021; Webster, 2017). Evidence from a low-density context further differentiates barriers: return migrants foreground bureaucracy, while immigrants additionally stress language constraints, discrimination, limited local-market knowledge and lack of financial support (Galvão et al., 2025b). The unresolved tension is not whether skills matter, but whether skills-first designs suffice when structural barriers shape who participates and what participation yields.

Social inclusion theory conceptualises inclusion as progression through access, participation and success through empowerment (Gidley et al., 2010). Access concerns entry and availability; participation concerns engagement quality and supports; and success through empowerment concerns capabilities, agency and influence beyond completion (Gidley et al., 2010). Empowerment is the realised capacity to make strategic choices when resources enable agency and achievements (Kabeer, 1999). Analytically, empowerment can be distinguished as individual capability and confidence, relational voice and networks, and structural access to enabling institutions, productive assets and market linkages (Huis et al., 2017). This staged lens shifts attention from whether programmes exist to whether conditions support movement from entry to meaningful participation to durable empowerment.

In entrepreneurship education, inclusion is often equated with participation, leaving underexamined the conversion conditions that enable participation to translate into success through empowerment (Kawarazuka et al., 2023). Even where access is formal, curricula can neglect intersectional realities – care obligations, mobility constraints, language barriers, digital exclusion and discrimination – reducing sustained participation and relevance (Ekinsmyth, 2014; Hughes et al., 2012; Kilpatrick and Johns, 2014; Pliakoura et al., 2020). Delivery formats and community-embedded approaches may increase feasibility (Simões and Brito do Rio, 2020), yet institutional and market barriers can still block progression when supports and linkages remain discretionary rather than embedded (Azmat, 2013; Kyrgidou and Petridou, 2013). Migrant women’s “translocal practices” may further shape strategies in ways generic curricula rarely address (Webster, 2017).

Overall, the literature suggests that programmes may widen access and participation while leaving intact the constraints that determine whether skills and confidence convert into agency and market power. Less explicit is how HE and VET provision operationalises staged inclusion in agripreneurship and where educator-perceived frictions interrupt progression for young women in NEET situations and with migrant backgrounds.

A second debate concerns how entrepreneurship education should respond to gendered and intersectional inequality. Skills-centric approaches treat empowerment as a likely outcome of competence development, while feminist critiques argue that such designs can recode structural exclusion as individual deficit unless curricula, relationships and supports address power relations and social reproduction (Dias et al., 2019; Hughes et al., 2012; Kawarazuka et al., 2023). From this perspective, the issue is not only what is taught, but whether programmes recognise lived constraints, redistribute enabling resources and build relationships and brokerage needed to navigate gendered institutions and markets (Ekinsmyth, 2014; Gustavsson, 2021). This implies a conversion gap: learners may enter and participate, but empowerment remains partial when the conditions enabling application and advancement are absent or optional.

Within agripreneurship education, feminist pedagogy can be treated as a conversion mechanism because it targets the conditions through which participation becomes empowerment: making exclusion mechanisms actionable, valuing care and time constraints as design parameters, and organising learning as relational and mentored (Ekinsmyth, 2014; Petridou and Glaveli, 2008). It aligns with calls to build agency while confronting structural barriers rather than assuming competencies will suffice (Azmat, 2013; Kyrgidou and Petridou, 2013). Gender-neutral “one-size” designs often fail to address combined effects of gender, migration status and socio-economic precarity, limiting inclusive outcomes (Dias et al., 2019; Mahajan et al., 2025).

Taken together, staged inclusion identifies where progression can stall, while feminist pedagogy specifies how designs might alter conversion conditions. This framing positions the study to (i) examine integration and delivery across HE and VET, (ii) analyse how educators frame needs and constraints, and (iii) identify where educator-perceived frictions interrupt movement from access and participation to success through empowerment.

A qualitative design elicited educators’ accounts of agripreneurship provision and inclusivity for young women in NEET situations and with migrant backgrounds. Purposive maximum-variation sampling combined convenience access across countries, sectors and roles (Patton, 1990). Eligible participants were HE professors and VET trainers involved in agriculture/agrifood entrepreneurship education (Robinson, 2014). Inclusion criteria were current/recent involvement and English for in-depth interviewing; exclusion criteria were administrative-only roles, generic entrepreneurship without an agriculture/agrifood focus, and lapsed involvement (Robinson, 2014). Recruitment used a desk-mapped, publicly sourced contact frame (programme webpages and institutional listings) supplemented by professional networks; outreach was iteratively adjusted to maximise variation (Patton, 1990). The final sample comprised 90 interviews (49 HE, 41 VET) across 11 European countries (Table 1). Given uneven country cell sizes and non-stratified sampling, systematic cross-country comparison was not pursued; national examples are illustrative rather than tests of welfare-, education-, gender-norm or rural labour-market regime effects. Educators are treated as institutional informants on design, delivery constraints and perceived effects rather than learner outcomes.

Semi-structured interviews (25–60 min) were conducted in English, face-to-face or online, audio-recorded with consent, and transcribed verbatim. Participation was voluntary and uncompensated; transcripts were de-identified prior to analysis. A common interview guide supported cross-country and cross-sector comparability while allowing flexibility to follow participants’ meanings. Open-ended questions covered:

• integration of agripreneurship into curricula/training;

• pedagogical approaches and learning activities;

• participant profiles and participation patterns;

• needs and barriers for young women in NEET situations and/or with migrant backgrounds; and

• outcomes, feedback and gaps as assessed by educators.

Core prompts included: “How is agripreneurship addressed in your current training provision?”, “What are the current training needs and trends in the field of agripreneurship?”, and “What training approaches do you use, which are most effective, and why?”. Follow-up probes sought clarification and examples, including: “What are the training needs and characteristics of young women in NEET situations and/or minority groups who have participated?”. Data integrity was supported through secure file handling, systematic de-identification, and an audit trail of recruitment, data preparation and analytic decisions.

Complementary desk research of HE and VET websites contextualised educator accounts and enabled limited triangulation across data sources (Ghauri et al., 2020; Schreier, 2012). Public programme descriptions were retained only when they explicitly referenced agriculture or agrifood entrepreneurship; level, audience, delivery format and thematic coverage were recorded. Duplicates were removed, and offerings were classified as HE, VET, lifelong learning or “Other” provision outside formal HE/VET. The mapping constitutes a study-constructed data set compiled from publicly accessible descriptions and is used to contextualise interview evidence rather than provide an exhaustive registry. Because the mapping relies on online descriptions, the apparent distribution across countries is shaped by differential web visibility and documentation practices; country counts should therefore be interpreted as indicative rather than national totals.

Interview and desk research materials were analysed using reflexive thematic analysis (RTA) to develop themes as patterns of shared meaning organised around central concepts rather than topic summaries (Braun and Clarke, 2022). RTA is theoretically flexible and interpretive: themes are generated through iterative engagement with the data, prioritising reflexivity and analytic transparency over coding consensus, inter-rater reliability or a fixed codebook (Braun and Clarke, 2022). This approach suited the aim of examining how educators make sense of inclusion, pedagogy and empowerment within institutional contexts while identifying theoretically informed patterns across heterogeneous multi-country settings.

Analysis followed six iterative phases: familiarisation; flexible, theory-informed coding; generation of candidate themes; review against coded data and the data set; definition and naming; and analytic writing. Coding was sensitised by staged inclusion and feminist pedagogy; empowerment talk was interpreted as individual (confidence/self-efficacy), relational (voice, peer/mentoring ties) and structural (access to finance, land/assets and market/buyer linkages). Quality was pursued through conceptual coherence, memoed decisions and rich extracts; consistent with RTA, inter-rater reliability and codebook standardisation were not sought (Braun and Clarke, 2022). Positionality was considered during coding and refinement. Quotations were organised as anchor, workhorse and supporting extracts to make evidentiary warrant explicit (Rockmann and Vough, 2024). These roles are indicated in  Appendix. Figure 1 summarises the research design and analytic process from desk mapping through sampling, interviews, preparation and RTA to the study outputs.

Using RTA (Braun and Clarke, 2022) and a claim-centred presentation (Rockmann and Vough, 2024), findings synthesise educator accounts into three themes aligned with staged social inclusion (access, participation, success through empowerment) and address integration, needs and perceived impacts. Desk research identified 210 agripreneurship programmes across HE, VET, lifelong learning, and “Other” provision outside formal HE/VET (Table 3), contextualising interview evidence. Table 2 summarises themes and subthemes; detailed claim tables are in  Appendix.

This theme examines, drawing on educator accounts, how agripreneurship is integrated into HE and VET and how pedagogical choices shape access and meaningful participation for young women in NEET situations and with migrant backgrounds.

Educator accounts trace a continuum – from comprehensive embedding (e.g. Slovenia), through partial, dispersed coverage (e.g. Cyprus), to gaps within specific institutions (e.g. some Polish HE contexts). Desk mapping also mirrored this variation: Slovenia listed 34 programmes – the highest in the data set (Table 3):

At VET level, integration is often pragmatic, threaded through broader agribusiness modules:

Across settings, content concentrates on business and production fundamentals. For example, educators highlighted:

Taken together, these accounts indicate uneven curricular integration across contexts; explicit references to gender-responsive curricular design were uncommon in educators’ descriptions of integration, suggesting that formal access alone may not secure meaningful participation for young women in NEET situations and with migrant backgrounds.

Pedagogies are described as modernising (problem-based, experiential) and delivery modes differentiate (face-to-face, distance, blended), but explicit discussion of intersectional participation constraints is uncommon in these accounts. HE educators pointed to expanded reach and practice orientation:

VET trainers reported consistently participatory methods:

Educators largely described pedagogy in generic terms (e.g. lectures/team projects, role-play, blended delivery), with only occasional references to participation supports such as language mediation or device access (e.g. translation and laptop lending in Subtheme 2.1). Childcare/care credits and other gender-responsive scaffolds were rarely mentioned.

This theme examines how educators identify training needs and participation constraints for young women in NEET situations and with migrant backgrounds, and how these are framed (skills gaps vs structural barriers).

VET trainers emphasised employability-oriented foundations, complemented by practical access supports:

HE educators prioritised the development of soft skills, enterprise and market-readiness competencies:

Across interviews, needs were most often framed as skill development; even where structural barriers were acknowledged, educators tended to translate them into individualised competency priorities, narrowing the pathway from participation to success through empowerment.

This theme examines how educators assess programme effects and market-readiness challenges, and the extent to which enabling supports (e.g. finance or buyer brokerage) are reported to translate participation gains into success through empowerment. Programmes generated proximal gains, yet educator accounts rarely described structured post-training market linkages (e.g. finance or buyer connections) as programme-embedded features.

Educators reported consistently positive proximal outcomes – confidence, peer connection, and tailored guidance:

Across educator accounts and the evidentiary quotes in  Appendix (Table A3), post-training linkages were rarely described as programme-embedded obligations; impacts were framed primarily as confidence gains and peer networks.

At the same time, educators identified persistent constraints that blunt the translation of these gains into durable market outcomes:

Across these accounts, programmes are described as building readiness – confidence, literacy and networks – while educators focus mainly on skills and market-literacy constraints rather than on programme-embedded changes to institutional resources or market access mechanisms. As a result, success through empowerment is described as partial and contingent in educator accounts.

Findings on integration and pedagogy reveal substantial heterogeneity across HE and VET, consistent with prior literature on policy–practice disjunctures in entrepreneurship education (Fitz-Koch et al., 2018). Desk mapping reinforces this picture, with dense provision in some countries and thinner ecosystems elsewhere (Table 3). Pedagogies have modernised (problem-based, experiential), yet delivery seldom engages gendered and migration-related constraints shaping participation; business-centric and ostensibly neutral curricula risk reproducing exclusion when unequal access to land, finance, networks and time is ignored (Dias et al., 2019; Hughes et al., 2012; Kabeer, 2015; Kawarazuka et al., 2023). This aligns with evidence that top-down standardisation rarely guarantees equity and that technocratic logics can crowd out participatory, place-based approaches that surface the needs of marginalised learners (Azmat, 2013; Gidley et al., 2010), but contrasts with research arguing that community-embedded and mentored designs can make participation materially more inclusive by embedding relational and redistributive supports (Ekinsmyth, 2014; Seuneke and Bock, 2015). Even as programmes integrate digital tools and industry engagement, pedagogy remains infrequently gender-responsive, leaving the access–participation link fragile for young women in NEET situations and with migrant backgrounds. Prior studies similarly note gender-sensitivity gaps in entrepreneurship training and the absence of inclusion-by-design safeguards – patterns echoed in the interviews here (Fradejas-García et al., 2024; Martinho, 2020). Across the educator accounts, narratives infrequently referenced explicit, institutionally resourced finance or buyer brokerage, which may help explain why gains in confidence and literacy did not reliably convert to market traction.

Educator interpretations that prioritise skills requirements reveal a second pattern. VET providers emphasised foundational digital and language skills and employability supports; HE prioritised ideation, soft skills market analysis and firm management. This focus on individual deficits resonates with a dominant capability narrative in entrepreneurship education but underplays intersectional and institutional barriers that circumscribe agency for young women in NEET situations and with migrant backgrounds, including legal precarity, discriminatory norms and thin networks (Azmat, 2013; Dias et al., 2019; Lüküslü and Çelik, 2022). Feminist analyses caution that competence-only solutions risk recoding structural exclusion as personal shortfall, narrowing what participation can achieve (Kabeer, 2015; Marlow, 2020). The findings reinforce calls to recognise identity work and social navigation as central in masculinised sectors and to position mentoring, peer affirmation and community-embedded practice as integral to capability formation rather than add-ons (Ekinsmyth, 2014; Seuneke and Bock, 2015). Consistent with pandemic-era studies of uneven digital access and care burdens, the online/blended formats described by educators were rarely paired with institutionally resourced care scaffolds and only occasionally included language or device supports (Kawarazuka et al., 2023). In short, the skills-first framing evident across providers helps explain stalled movement towards success through empowerment; this contrasts with feminist and intersectional scholarship that treats empowerment as contingent on resourced supports (care/time, language/digital access) and brokerage into finance and markets, not skills acquisition alone (Kabeer, 2015; Kawarazuka et al., 2023; Marlow, 2020).

Evidence on impacts and persistent constraints completes the picture. Educators consistently reported proximal gains in confidence, conceptual grasp and peer ties, aligning with work showing that experiential learning and cohort effects can catalyse entrepreneurial self-efficacy and aspiration (Dias et al., 2019; Kyrgidou and Petridou, 2013). Yet the same accounts pointed to gaps in sector linkages, applied opportunities and market literacy, and to the enduring weight of gendered and institutional hurdles. Complementary evidence from a place-based agrifood sector suggests that strategic academic training and experience outside the enterprise can introduce modernising perspectives without breaking the link to inherited values (Galvão et al., 2025a). This combination reflects a well-documented conversion problem where capabilities built in classrooms do not reliably translate into market power without changes to the rules, resources and relationships that govern entry and growth in agrifood systems (Boldureanu et al., 2020; Seuneke and Bock, 2015). In contrast to studies that emphasise ecosystem conditions – inclusive finance, buyer/supply-chain brokerage, and enabling institutional frameworks – as central to translating training into market outcomes, educator accounts seldom described these as programme-embedded mechanisms (Dias et al., 2019; Kawarazuka et al., 2023). In addition, prior work notes informal strategies (e.g. translocal networks) that were largely absent from educator accounts, reducing practical resonance and limiting post-training traction (Gustavsson, 2021; Webster, 2017). The result, in educator accounts, is a pattern where confidence and networks rise but success through empowerment remains partial and contingent – consistent with broader evidence that durable change hinges on addressing structural constraints alongside pedagogy and skills development (Kabeer, 2015; Pliakoura et al., 2020).

Across HE and VET, educator accounts indicate that agripreneurship provision is unevenly embedded and largely business-centric; participation supports (e.g. language mediation, care-related scaffolds, digital access) and post-training market brokerage (e.g. inclusive finance and buyer linkages) are rarely standardised, so impacts are assessed mainly as proximal gains (confidence, peer ties) rather than durable market participation and agency. These patterns suggest that agripreneurship education may benefit from staged, equity-by-design architectures that treat inclusion as a pathway from access to participation to success through empowerment, recognition–redistribution–relationality as core design logics, and evaluation based on empowerment-sensitive outcomes beyond completion. Accordingly, an inclusion-to-empowerment conversion framework is advanced to explain where progression stalls and how feminist pedagogy may help close the conversion gap under specified boundary conditions (Gidley et al., 2010; Hughes et al., 2012; Kabeer, 2015; Kawarazuka et al., 2023; Marlow, 2020). A theoretical model is advanced (Figure 2) integrating antecedent conditions, staged inclusion, conversion mechanisms and feedback loops for institutional change.

Firstly, the framework reframes inclusion in agripreneurship education as a staged pathway with stage-specific failure points rather than a single access event. Educator accounts and programme mapping suggest that access is primarily shaped by programme availability and institutional offer, whereas participation is shaped by delivery design and intersectional everyday-life constraints (care, mobility, language, digital access). The conversion gap manifests differently across settings: in HE, frictions cluster around dispersed curricular integration and evaluative logics privileging completion and skills signals; in VET, frictions cluster around short, employability-oriented formats where accommodations are discretionary and post-training brokerage is rarely embedded. As a result, policy-aligned provision can widen participation without shifting perceived agency, market traction or resilience unless conversion mechanisms and enabling conditions are in place (Azmat, 2013; Dias et al., 2019; Lahooti, 2025; Mahajan et al., 2025). This reconceptualisation extends social inclusion theory from description to explanatory mechanism by treating success through empowerment as an achieved state contingent on conversion conditions, not an automatic sequel to participation. Here, success through empowerment denotes realised empowerment; in future evaluations it may be proxied by venture establishment, revenue stability and asset access (Kabeer, 1999).

Secondly, the model theorises feminist pedagogy as a conversion mechanism – specified as a triad of recognition, redistribution and relationality – that may transform participation into success through empowerment. Recognition aligns content and pedagogy with intersectional realities – racialised/ethnicised positioning, socio-economic precarity, rural/peripheral geographies and mobility constraints, caregiving responsibilities, migration status, language and asset constraints – so learning validates lived experience and identities (Ekinsmyth, 2014; Gustavsson, 2021). Redistribution embeds enabling resources into programme architecture, including flexible scheduling, childcare or care credits, digital access, stipends and gender-responsive financial literacy and market brokerage (Aggarwal and Johal, 2021; Dias et al., 2019; Kawarazuka et al., 2023). Gender-responsive finance denotes credit and investment instruments designed to mitigate gendered barriers in access, terms and collateral. Relationality reorganises learning as mentored, peer-networked and community-based, so participants accrue social capital, role models and post-training navigation support (Kabeer, 2015; Seuneke and Bock, 2015). Success through empowerment is more likely when these three logics operate together and are institutionally resourced, rather than when experiential techniques or digital tools are added to otherwise business-centric, not explicitly gender-responsive curricula (Fradejas-García et al., 2024).

Thirdly, the framework introduces boundary conditions and feedback effects that make the conversion mechanism durable. Boundary conditions include educator capability for feminist practice, programme linkages to inclusive finance and markets, and governance metrics valuing empowerment outcomes beyond enrolment and completion. Where these conditions are weak, programmes produce confidence and networks without structural mobility; where they are strong, success through empowerment feeds back to reshape institutional logics and future curricula, gradually reducing the conversion gap in subsequent cohorts (Pliakoura et al., 2020; Webster, 2017). The model explains mixed impacts across HE and VET by showing that similar pedagogical forms generate divergent outcomes depending on the presence of feminist conversion conditions within the classroom, the institution and the surrounding ecosystem.

Finally, the framework yields theoretically grounded propositions for future testing and refinement: conversion to success through empowerment increases when recognition, redistribution and relationality are explicitly designed and resourced; conversion is dampened when policy and funding logics privilege market outputs over equity; and conversion gains are sustained when post-programme ecosystems – mentorship, finance and buyer brokerage – are embedded as programme obligations rather than optional add-ons (Aggarwal and Johal, 2021; Dias et al., 2019; Kabeer, 2015; Kawarazuka et al., 2023). By specifying how and where to intervene, the model moves agripreneurship education scholarship beyond skills-centric accounts towards a structural pedagogy of empowerment for marginalised women.

Implementing the inclusion-to-empowerment conversion framework in HE and VET depends on institutional capacity, resources and buy-in; redesign will be uneven across settings and may face resistance. For access, institutions can reduce non-essential entry barriers and use targeted outreach via migrant/NEET intermediaries, coupled with micro-pathways that recognise prior learning and enable re-entry after interrupted trajectories (Azmat, 2013; Fitz-Koch et al., 2018; Lüküslü and Çelik, 2022; Pliakoura et al., 2020). For participation, providers can embed recognition- and redistribution-oriented conditions as standard delivery architecture – flexible scheduling, childcare or care credits, language mediation and device/connectivity and mobility supports – alongside feminist, participatory teaching and assessment that rehearses navigation of gendered markets and legal regimes (Kyrgidou and Petridou, 2013; Mahajan et al., 2025; Marlow, 2020; Seuneke and Bock, 2015).

For success through empowerment, institutions and policymakers can make relationality and market access programme obligations through structured links to inclusive finance, secure land tenure/leases, incubation/extension support and buyer/supply-chain brokerage, strengthening post-training progression beyond participation and reducing reliance on ad hoc networks (Aggarwal and Johal, 2021; Dias et al., 2019; Petridou and Glaveli, 2008). This aligns with evidence from rural family-owned wineries where continuity is sustained through intergenerational transmission of values and tacit knowledge, early involvement in daily tasks, and family support in strategic decision-making, alongside incremental innovation (Galvão et al., 2025a). Governance can reward empowerment outcomes beyond enrolment and completion by tracking progression to venture establishment/market participation, revenue stability, asset security and durable networks, with monitoring disaggregated by intersecting positions (e.g. migrant status, socio-economic status, place-based constraints) to make the conversion gap visible in accountability systems, and by prioritising women-led graduate ventures in procurement and ecosystem services (Gidley et al., 2010; Kabeer, 2015). To retain conceptual consistency, indicator sets can align with the individual–relational–structural empowerment definition by combining measures of self-efficacy and strategic-choice confidence (individual), network durability and voice/mentoring ties (relational), and market participation and asset security (structural). Where recognition–redistribution–relationality elements are introduced without resourcing or institutional buy-in (e.g. care, language and market-brokerage supports), redesign may remain symbolic and shift additional navigation burdens onto learners rather than reducing structural frictions (Kabeer, 2015; Marlow, 2020).

The evidence base relies on English-language educator accounts and European programme mapping, creating epistemic boundaries. Educator-based data describe programme design and perceived proximal change but cannot validly assess lived empowerment beyond training settings (e.g. exercised agency in real market decisions, market integration, financial independence, intra-household bargaining power, asset security and long-term venture sustainability). Although empowerment is specified earlier across individual, relational and structural dimensions (Huis et al., 2017), the evidence mainly captures the first two, while structural outcomes beyond educators’ view cannot be verified. Claims are therefore limited to perceived conversion conditions and plausible trajectories rather than demonstrated effects or causal attributions from training to outcomes. Reporting may also be shaped by institutional and professional incentives (e.g. skills-deficit framings, social-desirability bias, and under-reporting discrimination or implementation resistance), and the cross-sectional design using self-report and documentary sources constrains triangulation and obscures post-training trajectories and within-programme heterogeneity (Abamosa, 2023; Kilpatrick and Johns, 2014).

Future research should centre young women as co-researchers to surface situated strategies and design preferences that institutional accounts miss, strengthening contextual validity and ethical robustness (Lahooti, 2025; Mahajan et al., 2025). Longitudinal mixed-methods studies should test the framework by tracking whether recognition–redistribution–relationality designs sustain movement along the access–participation–success through empowerment pathway and produce durable outcomes (venture establishment/market participation, revenue stability, asset security and network durability). Realist evaluations and comparative case studies can specify for whom and under what policy and finance conditions feminist pedagogies and enabling ecosystems close the conversion gap (Gidley et al., 2010; Kabeer, 2015). Design-based implementation trials comparing standard business-centric modules with co-designed feminist alternatives can identify scalable configurations, alongside studies of educator professional-learning models and the mediating roles of inclusive finance, buyer brokerage and cooperative membership. Equity-centred evaluations of digital and green agritech should assess not only uptake but also shifts in bargaining power and risk exposure for marginalised women, ensuring technological adoption translates into empowerment rather than new exclusions (Fradejas-García et al., 2024; Kawarazuka et al., 2023).