Stability by Design: How STEM Ecosystems Support Pathways for Underrepresented Youth

Abstract

Across the United States, equity-oriented STEM initiatives are frequently launched through short-term grants, yet few persist once external funding ends, particularly in economically marginalized communities where institutional capacity is constrained. This longitudinal qualitative study investigates how an out-of-school-time (OST) robotics initiative developed the relational and organizational capacity to transition from a time-limited grant project into a functioning STEM ecosystem that has persisted for a decade. Drawing upon eight years of focus groups and field notes analyzed through integrated deductive and inductive approaches, the study traces how STEM ecosystem tenets were enacted, adapted, and reinforced as partners navigated resource constraints. Findings identify four mutually reinforcing mechanisms that stabilized the ecosystem beyond the grant period: relational infrastructure coordinating work across students, educators, families, university partners, and district leaders; community recognition and collective pride conferring legitimacy and mobilizing local support; parental validation and logistical advocacy; and youth identity development and near-peer leadership renewing commitment and circulating expertise. Together, these mechanisms converted initial grant-funded inputs into durable capacity by reducing coordination costs, strengthening shared responsibility, and embedding STEM participation within community meaning-making. The study contributes to STEM ecosystem research by advancing a theory-building, process-oriented explanation of how equity-focused initiatives achieve durability.

1. Introduction

Recent policy shifts have intensified concern about the rollback of efforts to expand educational pathways into science, technology, engineering and mathematics (STEM) for historically underrepresented groups, yet this volatility is not new. STEM initiatives have long been governed by cycles of political enthusiasm and withdrawal, producing an educational landscape in which opportunities for participation, particularly for racialized learners, are fragile, contingent, and unevenly sustained across time and institutions (Bullock, 2017; Judson, 2025; Vakil & Ayers, 2019). Because federal and state laws prohibit discrimination but do not require institutions to implement specific STEM diversification programs, these initiatives are often treated as discretionary rather than mandatory (National Academies of Sciences, Engineering, and Medicine [NASEM], 2025; U.S. Government Accountability Office, 2024). As a result, they are commonly positioned as add-on or short-term projects whose continuity depends on shifting political priorities and unstable funding rather than on any guaranteed legal obligation to provide them. In response to White House educational initiatives such as the President’s Council of Advisors on Science and Technology (2010, 2012), national efforts to expand out of school time (OST) STEM learning were pursued primarily through short term grant funded or pilot oriented initiatives, broad policy rhetoric promoting federal investment for “STEM for All” (Johnson, 2012) and targeted research focused on the science of broadening STEM participation (McNeely et al., 2018). In K–12, STEM equity efforts are largely advanced through discretionary, competitive grants and changing federal priorities rather than standing legal mandates, which contributes to cycles of short-term expansion and retrenchment and to uneven access that affects students’ preparation for postsecondary STEM (National Academies of Sciences, Engineering, and Medicine [NASEM], 2025; U.S. Government Accountability Office, 2024). This pattern of short term expansion followed by retrenchment has and will continue to contribute to fragmented preparation and uneven access and transitions into college-level and higher education STEM programs. As grant funding cycles conclude and or federal investments are constrained, STEM ecosystems offer a durable alternative to fragmented career pathway efforts by cultivating sustained, cross-sector commitments among schools, communities, postsecondary institutions, and industry (Allen et al., 2020a).

STEM ecosystems are place-based, cross sector collaborations that intentionally connect schools, out of school providers, community organizations, higher education, industry, and other local assets to coordinate STEM learning opportunities across various settings and over time, with equity outcomes shaped by local histories, governance, resource flows, and power relations among partners (Timko et al., 2023; Vance et al., 2016). Timko et al.’s (2023) case study shows that STEM ecosystems build durable, community-anchored networks that mobilize local assets, strengthen cross-sector collaboration, and create continuity for STEM learning even when external funding fluctuates or ends. Additional empirical studies of regional organizational structures and STEM ecosystems find that distributed leadership and shared governance structures help embed equity goals across schools, community organizations, and industry partners, which in turn have the potential to reduce reliance on any single federal incentives, thereby buffering against political volatility (Marshall & Galey-Horn, 2024; Traphagen & Traill, 2014). Rather than guaranteeing equitable outcomes, the literature suggests that STEM ecosystems create the structural conditions under which equity-oriented collaborations become possible, contingent on local power dynamics and institutional commitments.

Research on out-of-school-time (OST) STEM ecosystems increasingly emphasizes the value of cross-sector coordination; however, much of the literature pays limited attention to the relational infrastructures through which commitments are sustained, roles are negotiated, and formal structures are enacted across partners. This omission is especially consequential in communities experiencing long-term structural disadvantage and chronic underinvestment, where formal governance arrangements and funding mechanisms are necessary but insufficient on their own to stabilize ecosystems over time (Ostman, 2024; Timko et al., 2023). In these contexts, relational infrastructures that support shared governance shape how formal structures are taken up, maintained, or adapted, influencing not only the distribution of resources but also the durability of collaboration and the enactment of equity-oriented practices. Thus, the aim of this study is to qualitatively examine how a STEM ecosystem established during a time-limited grant cycle developed the relational and organizational capacity needed for sustaining a pre-engineering afterschool robotics program serving African American youth in a rural, economically disadvantaged community. In doing so the study seeks to identify which STEM ecosystem tenets were most consequential for shaping program outcomes and long term sustainability; to analyze how cross sector interactions among students, university partners, district leaders, teachers, parents, and industry partners supported collaboration and influenced the evolution of a STEM ecosystem; and to explain how program outcomes reinforced stakeholder commitment and collective responsibility during and beyond the grant period. By tracing these dynamics longitudinally, the study addresses a persistent gap in STEM ecosystem research, which has documented STEM ecosystem formation but offered limited empirical evidence on how such ecosystems in under-resourced, racially marginalized communities sustain programming after external funding ends. This work contributes to the field by illuminating the mechanism through which relational infrastructures and cross-sector partnerships support durable STEM opportunities for historically underserved youth.

The following literature review establishes why underrepresentation in STEM derives from cumulative, structural inequities that limit rigorous K–12 preparation and early STEM identity formation in economically marginalized communities. It then examines out-of-school-time (OST) STEM programs as equity-centered learning environments that can counter opportunity gaps by fostering engagement, motivation, and identity, while noting recurring challenges of funding instability, organizational capacity, and program longevity. Building on these strands, the review foregrounds STEM ecosystems due to partnership quality, resource alignment, and community participation as conditions that influence whether effective programs endure beyond initial grant cycles. Building on these insights, the study focuses on a pre-engineering robotics after-school program situated in a severely economically disenfranchised community with limited access to STEM opportunities. The program continued for an additional five years beyond its initial five-year federal funding period and is now operating in its tenth year. This extended lifespan provides a valuable context for examining the interdependent ecosystem dynamics that contributed to the program’s sustainability and enabled it to persist in an environment where comparable initiatives often struggle to endure. This longevity is significant because it demonstrates the program’s capacity to consistently advance equitable STEM pathways for students whose identities have been historically marginalized.

2. Literature Review

Underrepresentation in STEM is not incidental; it reflects deeply entrenched structural inequities that take root in early childhood and intensify across educational trajectories, including higher education. These inequities shape the quality of early learning environments, restrict access to enrichment and advanced learning opportunities, and limit the institutional capacity required to provide rigorous and sustained STEM preparation throughout the K–12 system (McGee & Robinson, 2020; McNeely et al., 2018; Prime, 2019). Districts experiencing concentrated economic disadvantage face systemic barriers to offering high-quality STEM learning opportunities, which makes them especially vulnerable to policy retrenchment (Knight, 2017). Consistent with these patterns, schools serving high concentrations of low-income students provide significantly fewer advanced courses in mathematics, science, computer science, and engineering, producing opportunity gaps that begin early and widen across students’ educational trajectories (King et al., 2021; Hypolite & Rogers, 2023). These gaps are further exacerbated by chronic shortages of qualified STEM teachers in economically marginalized districts, which constrain schools’ ability to sustain advanced coursework and specialized STEM programs at scale (Hansen et al., 2024). Beyond formal schooling, students in economically disadvantaged communities often have limited access to informal STEM activities such as STEM summer camps, enrichment activities or research experiences, and project-based learning opportunities, all of which are significant for developing STEM identities and supporting college and career-focused STEM pathways (Means et al., 2017). Holzman et al. (2024) show that areas identified as “STEM deserts,” defined by limited or nonexistent opportunities to explore STEM pathways or learning experiences, produce geographically patterned inequities that restrict students’ participation in advanced STEM learning based on where they live, even when they possess the academic preparation to succeed.

For more than a decade, OST STEM programs have served as a critical equity-oriented reform strategy by broadening access to high-quality learning experiences and fostering the motivational and identity-related dispositions that sustain participation in STEM pathways, particularly for students who have been historically marginalized (Allen et al., 2020b). OST STEM programs are organized, voluntary learning experiences in STEM that occur outside compulsory school hours, typically in afterschool, weekend, or summer formats, and are offered by schools, universities, museums, libraries, youth-serving nonprofits, and community organizations (Shah et al., 2018). Additionally, OST STEM programs promote youth engagement and agency by centering collaborative, hands-on inquiry, design, and reflective practices that simultaneously advance disciplinary learning and cultivate affective and identity-related outcomes, including interest in STEM, persistence, and career knowledge (Shah et al., 2018). Allen et al. (2019) conducted a national survey of 1599 youth in 158 afterschool STEM programs and found that 65–85% of participants reported meaningful gains in STEM engagement, identity, career interest, and knowledge, as well as in critical thinking, perseverance, and positive relationships. From this perspective, OST STEM programs are essential as they function as a powerful antidote to the documented decline in students’ interest in STEM subjects such as science as they progress through K–12 and into postsecondary pathways (Osborne, 2003). Although numerous studies emphasize the need for longitudinal research to illuminate the programmatic influences on persistence-related dispositions that support STEM pathways, the limited operational timespan of many interventions significantly constrains the feasibility of such long-term designs.

A growing body of research highlights that while OST STEM programs play a critical role in broadening participation and enriching STEM learning, maintaining these programs over time is difficult due to structural, financial, organizational, and contextual barriers. OST STEM programs frequently depend on short-term grants, philanthropic funding cycles, or competitive external awards, a reliance that leaves them structurally vulnerable when these limited funding streams expire. The destabilizing of funding often has a ripple effect on staffing capacities, curriculum design with supporting resources, and organizational leadership. Koch and Penuel (2015) note that many high-quality informal and afterschool STEM projects do not persist beyond the grants that fund their development, highlighting the importance of planning for scale and sustainability from the outset. This persistent pattern has prompted researchers to develop plans, models and frameworks for scaling and sustainability to ensure that effective programs can endure beyond short-term funding cycles. One of their central recommendations is the cultivation of strategic partnerships that not only distribute capacity and resources across organizations but also foster shared ownership of the program’s long term success. Shepherd et al. (2020) contend that although STEM outreach programs face persistent threats from funding volatility, institutional constraints, leadership turnover, and unforeseen disruptions, those that cultivate robust partnerships, diversify funding streams, communicate impact strategically, and operate through adaptive, cost-efficient models are substantially more likely to achieve long-term sustainability and continued impact. The recurring emphasis on cross-sector partnerships highlights the importance of studying the interdependent dynamics that sustain STEM ecosystems and enable them to function as systems-level supports for the broadening participation work historically carried out by OST programs.

3. Conceptual Framework

The STEM ecosystems framework informs this study by examining how interdependent dynamic such as interconnected relationships, shared governance structures, distributed resources, and programmatic factors, shape stakeholder experiences and sustain the development and maintenance of the system. The STEM ecosystems framework used in this study emerged from a collection of studies that have done foundational work on STEM ecosystems. The combined contributions of the Noyce Foundation’s executive summary, How Cross-Sector Collaborations are Advancing STEM Learning (Traphagen & Traill, 2014), Partnerships to Transform STEM Learning (Allen et al., 2020b), University of San Diego’s Critical Factors Report regarding Design for Success: Developing a STEM Ecosystem report (Vance et al., 2016), and the Timko et al. (2023) RAIN rural ecosystems study offer a comprehensive and coherent knowledge base for understanding STEM learning ecosystems. The Noyce report establishes the original definition, attributes, and field-wide strategies that anchor the concept of STEM ecosystems across formal, informal, and community settings. Allen et al.’s (2020b) Tulsa case then demonstrates how these strategies operate in practice, providing evidence of implementation, partnership evolution, program quality, and student outcomes in a mature ecosystem. The University of San Diego’s Critical Factors Report regarding Design for Success: Developing a STEM Ecosystem translates the broader vision into a codified framework of eight essential factors and developmental stages, enabling researchers and practitioners to assess and strengthen ecosystem effectiveness. Finally, Timko et al.’s (2023) case study extends this knowledge into rural, place-based contexts, documenting sustainability processes, community identity, and the adaptations required for long-term viability in under-resourced regions. Together, these studies provide the field with conceptual for examining context-sensitive sustainability, making them an exceptionally cohesive and foundational set of publications for understanding how STEM ecosystems are designed, implemented, and sustained across diverse communities.

STEM learning ecosystems across all four studies share seven tightly aligned tenets that together form a coherent foundation for how ecosystems function and improve. For clarity, the core tenets are italicized to facilitate identification. Cross-sector collaboration serves as the organizing structure, bringing together schools, OST programs, STEM-rich institutions, higher education, industry, families, and community partners to create learning opportunities no single sector can provide. Strong, intentional leadership acts as the enabling driver, coordinating partners, aligning goals, and sustaining momentum across initiatives. Ecosystems also design connected learning pathways, ensuring experiences across home, school, and community build progressively toward STEM interest, identity, and careers. High-quality implementation requires educator capacity-building across sectors, supported by shared tools and professional learning that strengthen instruction in both formal and informal settings. A commitment to equity and community engagement guides design, broadening participation and rooting programs in local assets and cultural contexts. To sustain improvement, ecosystems depend on data and continuous learning, using common measures and feedback loops to refine practice and assess outcomes across settings. Finally, all four studies reveal the essential role of relational infrastructure, the trust, shared language, shared governance routines, and intermediary functions that allow partners to coordinate effectively and maintain collaboration over time, which significantly impacts student outcomes within the STEM ecosystem.

Designing STEM initiatives as ecosystems from the outset, rather than expecting isolated projects to evolve into them over time, establishes the social, organizational, and instructional conditions necessary for equitable outcomes that endure beyond a single grant cycle. Achieving this durability requires embedding the seven ecosystem tenets at inception. Co-design with cross-sector partners builds the relational infrastructure required for sustained collaboration, including trust, shared language, boundary-spanning routines, and intermediary roles that enable joint work across cultural and institutional contexts and maintain coherence through leadership transitions (Penuel et al., 2015; Coburn & Penuel, 2016). Early planning for connected learning pathways further ensures that learning experiences are intentionally aligned across home, school, and community settings. Research from the National Academies links such cross-setting alignment to gains in student interest, identity development, and achievement when connections are deliberately designed and assessed (Honey et al., 2014; National Research Council, 2015). Building educator capacity and systems for quality measurement from the beginning is equally consequential, as national evidence indicates that programs using shared frameworks and continuous feedback processes produce stronger youth outcomes and support early, system-level improvement (Allen et al., 2019; Shah et al., 2018). Finally, if broadening participation is a central aim, equity must be treated as a front-end design requirement rather than a later-stage enhancement since longitudinal and interview studies consistently show that early and connected STEM exposure predicts later coursework, persistence, and degree attainment. Collectively, these studies indicate that the strongest student outcomes occur when programs are designed from the start with ecosystem tenets guiding decisions about partnerships, pathways, and learning quality (Allen et al., 2019; National Research Council, 2015; Penuel et al., 2015; Traphagen & Traill, 2014).

4. Methods

This longitudinal qualitative case study examines a pre-engineering out-of-school-time robotics program operating within a STEM ecosystem in an economically marginalized rural community over ten years, drawing on eight years of annual focus group interviews and longitudinal field notes and using an integrated deductive and inductive thematic analysis guided by seven STEM ecosystem tenets to examine sustainability processes under conditions of chronic resource constraint. The qualitative sample included 19 stakeholders representing six roles within the ecosystem: 11 students, 2 teachers, 1 university partner, 2 parents, 1 school district administrator, and 2 industry partners, all selected for their sustained engagement and ability to illuminate relational, organizational, and identity-based dynamics shaping long-term ecosystem durability.

Conducted in an economically disadvantaged region with limited STEM infrastructure, the study focused on which tenets most effectively fostered partner collaboration, supported persistence beyond the original grant cycle, and enabled the program’s continuation as an operational STEM ecosystem serving historically underrepresented groups. A longitudinal qualitative approach was essential for capturing these dynamics as they unfolded, documenting relationships, practices, and resource flows within an adaptive system rather than as isolated events. As Rees and Ottrey (2026) note, this approach is well suited for examining processes over extended periods, allowing researchers to trace how design features and interdependent interactions are enacted, reinforced, and reconfigured over time. This temporal sensitivity aligns with STEM ecosystems, which operate as interconnected systems shaped by sustained interactions among educators, institutions, and community partners. From an ecological and systems perspective, educational outcomes emerge from multilevel interdependencies across families, schools, communities, and broader contexts. In STEM education research, STEM ecosystems that facilitate STEM opportunities are conceptualized as complex, dynamic systems of interconnected partners and environments that shape participation and learning over time (Allen et al., 2020a). In regions with limited STEM infrastructure, a longitudinal design also captures how cross-sector collaborations form, organizational capacity develops gradually, and stakeholders navigate persistent resource constraints to expand opportunities (Timko et al., 2023). By following historically underrepresented groups engaged in emerging STEM pathways, this design reveals changes in participation, identity, and perceived opportunity, reflecting the lived experience of systemic change (Yao et al., 2023). These factors demonstrate that a longitudinal qualitative study is necessary for understanding the evolving interdependencies that support the development and sustainability of a STEM ecosystem in a resource-constrained region.

The study examined how embedding a STEM ecosystem design at inception influenced interdependent dynamics, program outcomes, and long-term sustainability. The research questions are as follows:

• Which core STEM ecosystem tenets emerged as most consequential in shaping program outcomes, interdependent dynamics, and long-term sustainability?
• How did cross-sector interactions among stakeholders support collaboration, influence program outcomes, and contribute to the establishment and sustainability of the STEM ecosystem?
• How did program outcomes reinforce stakeholder commitment and collective responsibility for sustaining the STEM ecosystem both during and beyond the initial grant period?

4.1. Study Context

In alignment with former calls to broaden participation in STEM for historically underrepresented groups (Committee on Equal Opportunities in Science and Engineering, 2019; National Science Foundation, 2017), particularly African American students, the pre-engineering robotics afterschool program was developed to address persistent inequities that restrict access to meaningful STEM opportunities. Structural barriers in high-poverty schools, including limited access to advanced mathematics and science courses, continue to depress STEM participation. National analyses have shown that as poverty levels rise, access to advanced STEM coursework sharply declines (King et al., 2021; U.S. Government Accountability Office, 2024). Within the economically disadvantaged region where the program operates, these inequities intersect with the limited visibility of STEM professionals who share students’ racial and gender identities, which can significantly constrain their ability to imagine themselves entering STEM fields (Lancaster & Xu, 2017).

The initiative was grounded in Bourdieu’s theory of capital, which argues that unequal access to institutional, cultural, and academic resources reproduces existing social inequalities rather than expanding opportunity (Bourdieu, 1983). Regional readiness data reinforced the need for intervention. ACT science and mathematics results indicated that African American students in the region rarely show the preparation levels typically associated with persistence in STEM majors (ACT, 2016). In response, the four-year mixed methods robotics program was designed to strengthen engineering self-efficacy, cultivate science identity, and support long-term STEM aspirations through structured pre-engineering and career-focused activities. Quantitative trends revealed increases in engineering self-efficacy and science identity, and these shifts were echoed in focus group data that described developing confidence, belonging, and a growing sense of possibility within STEM fields (Taylor & Brand, 2021).

Although the original grant funded only four years of implementation, the program’s impact was substantial enough that the participating school and community partners chose to continue the initiative after the grant ended. The robotics program remained active for an additional five years without external funding and is now entering its tenth year of continuous operation. This persistence in a region with limited STEM infrastructure led researchers to begin a new line of inquiry focused on understanding the factors and interdependent dynamics that have enabled the initiative not only to survive but to remain a stable and meaningful STEM pathway. Emerging questions center on how relationships among educators, families, community partners, district leaders, university partners and students have supported the program’s long-term sustainability, how local practices have evolved to reinforce STEM participation, and how these intertwined processes have allowed a time-limited intervention to become a lasting component of the community’s educational ecosystem.

4.2. Program Structure

The program is designed for students to enter in the ninth grade and continue through graduation in twelfth grade. Each year, the program unfolds in three structured phases, providing students with sequential experiences that build technical knowledge, problem-solving skills, and professional readiness. During the fall semester, students are introduced to engineering design principles through hands-on challenges rooted in real-world issues within their local community. These projects emphasize the intersection of STEM with societal needs, encouraging students to consider broader social, environmental, and economic implications of technological solutions. Students are introduced to robotics and guided through an engineering design challenge in which they develop a prototype robot using either programmable EV3 robots or drone platforms to address a specific local problem. For instance, one year, students designed agricultural drones capable of distributing seeds along predetermined paths in support of local farming initiatives. To accomplish these projects, students acquire foundational knowledge in electronics, computer-aided design (CAD), and programming, learning from both classroom teachers and university partners who provide specialized instruction and mentorship. At the conclusion of the fall semester, students showcase their prototypes during an annual STEM Night, presenting their solutions to parents, school administrators, teachers, and local industry representatives. This event highlights not only their technical accomplishments but also their ability to communicate complex STEM concepts to a broader audience.

During the spring semester, students build on the knowledge and skills acquired in the fall to design and construct competition-level robots intended for national-level contests. These projects require students to apply advanced engineering, programming, and collaborative problem-solving skills while adhering to strict technical and performance criteria. Throughout this process, students develop strategies for iterative design, testing, and optimization, simulating professional engineering environments and fostering teamwork under realistic constraints. In the summer, students participate in college and career readiness workshops that provide guidance on standardized tests such as the ACT and SAT, preparation for advanced STEM coursework, and essential academic strategies, including study skills, time management, and professional communication. Additionally, students take on leadership roles in a student-led summer robotics program for upper elementary and middle school students. This component serves as both a mentorship opportunity and a community pipeline, nurturing younger learners’ interest in STEM while reinforcing the upper-level students’ knowledge and leadership abilities. Through this multi-year, phased approach, the program integrates technical skill development, real-world problem solving, and professional preparation, creating a comprehensive STEM learning pathway that promotes sustained engagement and fosters the next generation of STEM innovators.

4.3. Study Participants

The study involved six stakeholder groups whose perspectives were essential for understanding how the robotics initiative developed the capacity to evolve into a sustained STEM ecosystem: students, teachers, university partners, parents, administrators, and industry partners. Across the ten-year lifespan of the program, approximately 170 students participated, with annual cohorts of 15 to 25 ninth through twelfth graders. Eleven students are represented in this study. All student participants were African American youth from an economically disadvantaged region with limited access to advanced STEM coursework. Students were included because their experiences illuminated identity development, engagement, and the relational mechanisms that supported persistence.

Five teachers contributed across the program’s duration, bringing backgrounds in mathematics, science, and career and technical education. Two of these teachers were included in the study because they had the longest and most sustained interactions with students throughout the duration of the program. Their long-term instructional and mentoring roles positioned them to speak to program continuity, student development, and organizational routines. University partners included seven engineering faculty and undergraduate mentors who provided specialized instruction in electronics, programming, and computer-aided design. One of the university partners is represented in this story for his longevity with the program. Their involvement connected secondary and postsecondary STEM environments and supported students’ exposure to engineering pathways. Parents were included because of their critical role in sustaining student engagement and supporting the program after external funding ended.

Approximately 80 parents consistently contributed through chaperoning, assisting with STEM Night, supporting fundraising, and providing logistical help during competitions and community events. Two parents are represented in this study who attended a STEM Night in later years. Administrators consisted of two school district leaders who offered institutional support, public recognition, and long-term commitments to sustaining the initiative. One administrator is represented in this study due to his close interfacing with the program. Their perspectives were essential for understanding policy-level decisions and resource allocations. Industry partners included five local and regional organizations that provided sponsorships, technical demonstrations, and career-focused engagement opportunities. Two industry partners are represented in this study. Their participation strengthened program visibility and diversified resources.

Student recruitment occurred through coordinated school and community outreach, including middle-school demonstrations, flyers, morning announcements, pep-rally presentations, and two major annual events: the STEM summer camp and community STEM Night. All procedures followed ethical guidelines approved by the University Institutional Review Board. Parents received consent materials during online informational sessions, and assent and consent were completed with students during the first program session each year. Participation in interviews and observations was voluntary for all stakeholder groups.

4.4. Researcher Positionality

Reflexive thematic analysis requires researchers to consider how their identities and relationships shape data collection and interpretation (Braun & Clarke, 2006). Because this study examines a long-term STEM ecosystem grounded in relational work, the research team engaged in ongoing reflexive discussions throughout the project. The first researcher is an African American woman and early faculty STEM educator and has developed longstanding relationships in the community where the robotics program operates. Her deep familiarity with the program and its participants supported trust and open dialogue during interviews and observations. She remained attentive to her relational status and commitments to equity-shaped interpretation and engaged in reflexive memoing to examine these influences. The second researcher is a STEM education scholar who is from the broader region and has extensive experience working with local schools and families. Although she is regionally rooted, the specific community served by the robotics program is a tightly knit environment with its own histories and social boundaries. She therefore occupied a position of partial insider familiarity combined with awareness that she was not fully embedded in the everyday social fabric of the community. Her regional expertise enriched the interpretation of partnership dynamics, while her relative distance from the program required ongoing reflexive attention. Through analytic memos, she systematically documented how her familiarity with the context and her professional authority might influence interpretive judgments and shape the analytic process. The third researcher is a South Asian male doctoral student with training in engineering and human-centered design. As a newcomer to both the community and qualitative research, he approached data collection with curiosity and openness. His technical background supported attention to engineering aspects of the program, and his outsider status prompted the team to articulate assumptions and clarify analytic categories. Together, the team treated positionality as an analytic resource. Reflexive dialogue across differing identities, levels of community embeddedness, and disciplinary backgrounds supported transparency and rigor in coding, theme development, and interpretation.

4.5. Data Collection

Although program data were collected throughout the full duration of the initiative, the present study focuses specifically on the first eight years. This delimitation was chosen to ensure analytic coherence, as these years represent the period during which the program’s foundational structures, stakeholder relationships, and ecosystem dynamics were most actively developing and most extensively documented. Concentrating on this formative phase allows for a more precise examination of the mechanisms that were critical to sustainability. Data collection for this study drew upon two primary qualitative sources: focus group semi-structured interviews with students and teachers, and field notes documenting all other stakeholder groups, including parents, administrators, university partners, and industry partners. This design aligned with longitudinal qualitative approaches that examine evolving relationships, practices, and contextual conditions within adaptive systems (Rees & Ottrey, 2026; Saldaña, 2003). Interviews with students and teachers were conducted annually at the end of each program year. Students and teachers were selected for interviews because they interacted with the program on a daily basis and were positioned to provide detailed accounts of instructional practices, identity development, collaboration, and the relational mechanisms that shaped persistence. Interviews were semi-structured to balance consistency across years with flexibility for participants to elaborate on emerging experiences, consistent with qualitative approaches that foreground depth, voice, and participant meaning making (Braun & Clarke, 2006; Morgan, 1997).

Field notes served as the primary data source for all other stakeholder groups whose engagement was episodic, event-based, or embedded within broader community activities. Field notes were collected during program sessions, competitions, planning meetings, STEM Night, summer camps, and informal interactions with parents, administrators, university partners, and industry partners. This approach allowed the research team to document cross-sector interactions, relational dynamics, decision-making processes, and emergent practices that stakeholders might not explicitly articulate in interviews (Emerson et al., 2011). Using field notes for these groups was intentional, as their contributions were often distributed across events rather than concentrated in sustained instructional roles. Integrating interviews and field notes strengthened the credibility and depth of the study through methodological triangulation (Denzin, 1978; Creswell & Poth, 2018). The longitudinal design enabled systematic tracking of changes in stakeholder experiences and ecosystem dynamics over time, which is essential for understanding how collaboration, shared governance, and relational infrastructure develop and persist in STEM ecosystems serving historically underrepresented youth (Coburn & Penuel, 2016; Yao et al., 2023).

4.6. Data Analysis

Data analysis in this longitudinal qualitative study employed a combined deductive and inductive analytic approach to examine how specific STEM ecosystem tenets embedded at program inception shaped interdependent dynamics and sustained outcomes related to STEM participation over time. Deductive analysis was guided by the seven STEM learning ecosystem tenets articulated in the conceptual framework and presented in

Table 1. STEM Ecosystems Tenets.

STEM Ecosystem Tenet	Explanation
Cross-Sector Collaboration	Serves as the organizing structure by bringing together schools, OST programs, STEM-rich institutions, higher education, industry, families, and community partners to create learning opportunities that no single sector could provide alone.
Intentional Leadership	Acts as the enabling driver, coordinating partners, aligning goals, and sustaining momentum across initiatives to ensure ecosystem coherence and effectiveness.
Connected Learning Pathways	Designs progressive experiences across home, school, and community that build toward STEM interest, identity, and career readiness, providing continuity and relevance in learning.
Educator Capacity Building	Strengthens instruction in both formal and informal settings through professional learning, shared tools, and cross-sector support, ensuring high-quality program implementation.
Equity and Community Engagement	Guides program design to broaden participation and embed initiatives within local assets, cultural contexts, and community priorities, ensuring inclusive access to STEM opportunities.
Data and Continuous Learning	Uses shared measures, feedback loops, and ongoing assessment to refine practice, monitor outcomes, and sustain continuous improvement across the ecosystem.
Relational Infrastructure	Establishes trust, shared language, governance routines, and intermediary functions that enable effective coordination and long-term collaboration among partners, positively impacting student outcomes.

: cross-sector collaboration, intentional leadership, connected learning pathways, educator capacity building, equity and community engagement, data and continuous learning, and relational infrastructure. These tenets served as a priori coding framework applied to focus group interview transcripts and field notes. Deductive coding enabled systematic examination of how stakeholder dialogue within focus group interviews reflected the enactment of these tenets and how related patterns of partner collaboration, post-grant persistence, and ecosystem continuity developed across the study period. Focus group data were particularly valuable for illuminating shared meanings, collective sense-making, and relational processes that are central to ecosystem functioning. The use of a predefined coding framework is well aligned with qualitative approaches that examine theoretically grounded constructs across time and settings while maintaining analytic rigor (Miles et al., 2014; Fereday & Muir-Cochrane, 2006).

In parallel with deductive coding, we conducted inductive thematic analysis of focus group interviews and field notes to capture emergent patterns, processes, and meanings not fully specified in the initial tenet framework. Inductive analysis was essential for understanding contextualized experiences of historically underrepresented participants, unanticipated partnership mechanisms, and shifts in relationships or resource flows as the program transitioned from grant-funded support to an operational ecosystem. We followed established thematic analysis guidelines emphasizing iterative engagement with the data, allowing codes and themes to arise from participants’ words while remaining grounded in the longitudinal narrative (Braun & Clarke, 2006). Integrating deductive and inductive strategies provided a comprehensive understanding of how predefined tenets were enacted and how novel, context-specific dynamics contributed to sustained STEM engagement.

Multiple strategies enhanced credibility, dependability, confirmability, and transferability. Triangulation of data sources (interviews and field notes) and stakeholder perspectives (students, teachers, administrators, university and industry partners) strengthened confidence in interpretations (Denzin, 1978; Creswell & Poth, 2018). Investigator triangulation involved multiple analysts independently coding a subset of transcripts and reconciling differences, reducing individual bias and increasing consistency in applying both deductive and inductive codes (Patton, 2015; Saldaña, 2003). Analytic decisions and reflexive memos documented how interpretations evolved and were evidence-based, consistent with qualitative standards (Lincoln & Guba, 1985). Member checking further ensured that findings reflected participants’ perspectives (Creswell & Poth, 2018). Together, these strategies ensured that findings were systematically derived, grounded in longitudinal evidence, and meaningfully connected to theoretical constructs and lived experiences within the STEM ecosystem.

5. Findings

The findings from this longitudinal qualitative study are organized around four interdependent, system-level mechanisms that shaped the development, functioning, and long-term sustainability of the STEM ecosystem: relational infrastructure, community recognition and collective pride, parental validation and family support, and youth identity development and leadership. These mechanisms emerged through a hybrid deductive and inductive analytic approach, in which STEM ecosystem tenets provided an initial coding structure while inductive analysis revealed the relational, social, and community-based processes most influential in this context. Importantly, these mechanisms were not produced by a single group of participants but emerged through the coordinated and role-differentiated contributions from multiple stakeholders, including students, teachers, university partners, parents, administrators, and industry partners. To clarify how these actors contributed to each sustaining mechanism,

Table 2. Stakeholder contribution to Sustaining Mechanisms in the STEM Ecosystem.

Stakeholder	Sustaining Mechanism	Function Within Ecosystem	System-Level Effect
Students	Relational Infrastructure	Built peer trust and collaboration	Strengthened belonging and sustained engagement
Teachers	Relational Infrastructure	Provided mentorship and accountability	Stabilized participation and expectations
University Partners & Mentors	Relational Infrastructure	Offered expertise and role modeling	Expanded capacity and program legitimacy
Students	Community Recognition & Pride	Participated in showcases and competitions	Reinforced identity and motivation
Teachers & Program Leaders	Community Recognition & Pride	Organized events and promoted student work	Increased visibility and community investment
Administrators	Community Recognition & Pride	Promoted and endorsed the program	Strengthened institutional stability
Parents & Families	Parental Validation & Support	Encouraged and supported participation	Sustained engagement beyond program settings
Parents & Families	Parental Validation & Support	Participated in events and provided advocacy	Reinforced community ownership and continuity
Students	Youth Identity & Leadership	Led teams and mentored peers	Built internal sustainability and leadership pathways
Teacher, Mentors & Partners	Youth Identity & Leadership	Structured leadership opportunities	Support identity development and role transition
Industry Partners	Community Recognition & Pride	Provided resources and external validation	Connected program to real-world pathways in STEM

Note. This table summarizes analytically derived stakeholder contributions to each sustaining mechanism within the STEM ecosystem. Roles reflect patterns identified through longitudinal qualitative analysis and are not exhaustive of all the possible interactions and system-level effects.

summarizes stakeholder roles across the ecosystem. Each theme is presented through analytic claims supported by participant accounts and interpreted to illuminate system-level processes shaping ecosystem sustainability. Together these findings address the research questions by illustrating how ecosystem tenets were enacted in practice (RQ1), how cross-sector interactions supported collaboration and sustained engagement (RQ2), and how program outcomes reinforced long-term commitment and shared responsibility beyond the initial grant period (RQ3).

5.1. Theme 1: Relational Infrastructure as the Core Mechanism of Ecosystem Sustainability

Relational infrastructure emerged as a significant mechanism through which STEM ecosystems translate structural arrangements into sustained participation and coordination across actors. Findings reflect that ecosystem effectiveness is contingent upon the configuration and quality of relational ties that organize interactions among stakeholders, rather than on the mere aggregation of programs of institutional actors. These relational ties structure the circulation of knowledge, resources, and opportunities across contexts, enabling STEM learning pathways and supporting continuity over time. The findings presented in this section focused on the relational contributions of students and university partners as central actors in the enactment of relational infrastructure. While not exhaustive of all stakeholder groups within the ecosystem, these actors were most prominently represented in the data and provided critical insight into how relational processes functioned to stabilize participation and sustain engagement.

Students, University Partner and Teacher

Students’ characterization of teachers as inspirational or familial figures reflects the affective dimensions of relational infrastructure, in which relationships extend beyond instructional support to encompass trust, care and personal investment. Although relational themes emerged across multiple years of the focus group interviews, the second and fourth year discussions are highlighted here to illustrate how students described their relationships with teachers. During a second-year focus group interview, Jarius reflected on his high school teachers, Mr. Morgan and Coach Randall, expressing admiration for how they engaged meaningfully with students while upholding high expectations and fostering accountability. He described feeling privileged to have been taught by educators whose character and commitment left a lasting impression on him.

Mr. Morgan and Coach Randall are so nice and so serious. They’ll make you laugh the whole day. Then they’re your problems of the day. They are your inspiration of the day. Anything you want to happen that day goes to Mr. Randall or Mr. Morgan. You’re going to get it done. They made sure you were down for business. They were making sure you’re having a good time doing it. They were going to make sure you’re doing it over a hundred percent. They were going to make sure that tasks that they told you to get done got done, plus more. Those guys, I can actually look up to them…They are not your average guys that you can just look at and say that they’re a good guy…I can’t come up with the words but they are some really good people. They’re really inspirational.

During a fourth year focus group interview, Ronald expressed appreciation for his teachers’ guidance and accountability, describing their support as resembling a fatherly presence in his life.

I found them easy to work with because my sophomore year Coach Morgan was my algebra teacher. Me and him bonded on a real close level. We carried that bond throughout my senior year… Throughout those years, these guys have been riding me, making sure even if I’m not in their class, I could hear their voice… It makes you smile every time you see them. They are like father figures.

These relational dynamics functioned to deepen students’ sense of belonging and accountability, suggesting that relational infrastructure operated not only as a structural connector within the ecosystem but also as an affective anchor that sustained engagement over time. While familial dynamics were articulated across multiple later years of the program, it is noteworthy that a university partner identified this theme at the beginning of the second year while reflecting on the previous year, even though that year was exceptionally challenging. While reflecting on the conclusion of the first year of the program, one of the university partners who taught electronics described how students made sense of persisting through the challenges of building the competition robot by drawing on their perception of the program as a family.

Yeah, we were listening to the kids, they talk about how they really thought that everybody and everything came together at the end. Even though there were a lot of ups and downs, and challenges, struggles, and frustrations, the kids talked a lot about some of those interesting dynamics and things last year. But then they said at the end, we went to Bellington [location of robotics competition], we were this team, this family, right? They kept using the word family.

While students demonstrated that affective domains and inspiration were core aspects of relational infrastructure, the teachers were also inspired by university partners, which also promoted persistence. During a focus group interview at the conclusion of the fourth year, Coach Randall, a teacher, stated,

What I brought to the program was, as far as my metalwork experience, was to make sure that the actual robot that it got built. Then you guys [referring to university partners] are doing good with the software and programming … It kind of encouraged me to kind of want to get into this and get into a little programming…It has challenged me, it has excited me. … You know, it’s kind of like when you give a person an inch, they’re going to take a mile, but I want more and more as far as that…I want to get into this computer world and into this thing about science and more technology, and of course the engineering…I’ve always been interested in engineering, but it’s just one of the things that’s kind of got me motivated.

Although these findings were drawn from different program years, they collectively indicate recurring patterns within the relational infrastructure that supported the initiative over time. Taken together the findings demonstrate that relational infrastructure in a STEM ecosystem is more than a coordination mechanism; it is an effective and motivational resource that shapes how students, educators and partners experience STEM while cultivating dispositions necessary for sustaining commitment over time.

5.2. Theme 2: Community Recognition and Collective Pride as Motivators for Persistence

Community recognition and collective pride emerged as a significant social force within this STEM ecosystem. When students’ STEM work became visible to peers, families, school district, and broader public through showcases, competitions, media coverage and community events, it signaled that their efforts have value beyond the classroom. The findings demonstrate that this visibility affirmed competence, strengthened STEM belonging and linked program achievement to community narratives of possibility and advancement. While not representative of every stakeholder group within the ecosystem, these actors were the most prominently reflected in the data and offered critical insight into how community recognition, along with collective pride and motivation tied to persistence, contributed to stabilizing participation and sustaining engagement.

Students, Teacher, Industry Partners, and Administrator

During a third-year focus group interview, a student, Jerry, reflected on the sources of his motivation, noting that it was closely linked to the positive recognition he received as a participant in his school’s robotics program. The team’s visibility through local news and newspaper coverage enhanced his sense of pride and affirmed his role within the broader STEM ecosystem. This public acknowledgment not only validated his individual contributions but also reinforced the value of the collective work undertaken by the robotics team. As a result, Jerry felt compelled to extend this sense of possibility to younger students in his community, encouraging them to join the program so they, too, could access similar opportunities for recognition, learning, and belonging within the STEM ecosystem.

Back to the TV part, I thought that smart kids get publicized for their creativity…Me being inside of the robotics class, it really opened a lot of doors, and not just within my school. I got my picture taken for the newspaper. My name is in the newspaper, on the news and all of that. When l go inside a store, people will say, “Hey! You were on the news.” It’s up to the point where I want to spread the word and get younger people to join and put them in my shoes. Where I’m from, the town doesn’t have a positive rep. Now it’s being known as, “oh that town.”

During the same focus group interview, a student, Jarius, emphasized that the public attention generated by the robotics team carried particular significance for him because it countered the negative portrayals of the town that he had grown accustomed to encountering. He expressed a sense of pride in contributing to an accomplishment that reframed the community in a positive light and offered residents a collective moment of celebration.

There is a lot of publicity about the robots that I did not know about until this week. This past week, we’ve had the demo day. Everybody that knew about the robot came in to see the robot, and take pictures…Even if you don’t even know about it. The publicity towards the robot in our town blew up quick. That also made me feel good. We don’t get too much attention. The attention that we do get is negative. To bring something positive to this town was a thought that kind of made you tear up a little bit.

Expressions of student pride resulting from community recognition were evident across the years. During the sixth-year teacher focus group, Mr. Morgan, a participating teacher, described how students outside the program expressed increasing interest after seeing participants showcase their robot on social media.

Everything is centered around the goal of the project in terms of building the robot, but it also goes back to the goal of the project of getting these kids interested in STEM jobs. The thing is that them just being able to be a part of this, has been so eye-opening for these kids. I mean, words can’t even describe what it has done for some of them already. I mean, it’s not just the talk amongst them. I’m in class, and I have kids telling me, I saw your robot last night on Snapchat, or I heard about it. I have kids talking about it, asking questions all day long now.

The heightened interest described in Mr. Morgan’s statement was also observed in year eight, as documented in field notes. During that year’s STEM night event, students who were not previously involved in the robotics program attended the demonstrations and subsequently enrolled in the program that same night after witnessing their peers showcase their robots. As a result of the program’s visibility through community-based showcases, including STEM Night, the robotics summer camp, and participation in regional robotics competitions, industry partners established collaborative relationships designed to support and sustain the ongoing work in the region. Various industry partners expressed enthusiasm and pride related to the program outcomes. During the eighth year, one of the local telecommunication companies sent a message to the robotics team stating,

I am reaching out to wish your team the best of luck in this season and to officially share that Ping will continue to support your team with sponsorship this year…We would love to arrange a visit with you this season to bring over some goodies and to hear how the team is doing.

During the same year, another industry partner of an e-commerce company stated,

Seeing these young people and their robots really brightens my day! I’ve been to multiple robotics events and know it is no easy task to get out there and compete! Thanks to the whole team for enabling their learning.

In addition to industry partners expressing pride, the school district began to provide more structural support in an effort to sustain the program. During a meeting in the eighth year, the administrator stated, Thank you all! The district is committed to doing whatever is required to ensure our robotics program remains a viable program.

Although these findings were derived from multiple program years, they collectively reveal consistent patterns in which community recognition and shared expressions of collective pride functioned as key motivational forces sustaining stakeholder engagement over time. Taken together, the findings suggest that public showcases of student work function as a catalytic mechanism within the broader STEM ecosystem, generating a reinforcing cycle of collective pride, heightened visibility, and renewed motivation among stakeholders. This positive feedback loop appears to contribute meaningfully to the ongoing commitment to sustain the program.

5.3. Theme 3: Parental Validation and Family Support as Reinforcing Commitment

Parents emerged as critical actors within the STEM ecosystem, providing forms of participation and validation that directly shape youth engagement and persistence. Their encouragement, logistical support and recognition of students’ growth and development functioned as relational resources that stabilized participation and signaled the program’s value within the community. When parents affirmed improvements in responsibility, confidence, or school performance, they offered an external validation that strengthened youths’ development. In this way, parental involvement operated not simply as background support but as sustaining mechanism that linked home, school, and community, contributing to the ecosystem’s resilience and its capacity to nurture long-term STEM pathways. Parents affirmed the program’s value in enriching their children’s high school experience. Their support was demonstrated through active participation in program events, assistance with fundraising efforts, and the provision of ongoing guidance that encouraged students to maintain their involvement in the program.

Students and Parents

During a third-year focus group interview, Darnell, a student in the program, emphasized that parents play an active role in strengthening the initiative. He asserted that parents carried a shared responsibility for ensuring the program’s continued sustainability. While he reflected on parental support, he stated, Parents are the ones that help boost up the program and we do fundraising and stuff like that. It is very helpful for the parents to do. Parental participation, which reinforced commitment, was also echoed by another student in the following years. During a sixth-year interview, Ronnie’s mother was actively involved in obtaining sponsorships for the program. He attributed his persistence in the program to her involvement in the program. Ronnie stated, Even if I thought about quitting my mom wouldn’t let me because she knows a lot of people and got them to help sponsor our program. From this perspective, parental validation functioned as an important sustaining mechanism within the STEM ecosystem, as their investment reinforced the social, cultural, and structural supports necessary for the program’s long-term continuity.

Within the STEM ecosystem, parents contributed not only to the cultivation of structural support for the program but also to its continued growth through informal recruitment practices. Their perceptions of the program’s educational and developmental value often led them to strongly encourage their children’s participation, thereby reinforcing a pipeline of sustained engagement and helping to stabilize the program’s long-term viability. During a fourth-year focus group, Paul, a student, attributed his decision to enroll in the program to his counselor and his mother, indicating that his mother encouraged him to spend time doing something other than playing video games in the evenings after school.

There were two people who influenced me, or helped me, and those two are my mother and my counselor. My counselor, she recommended me…. So, I said I would try it out, and then my mother, she influenced me because she wanted me to stop playing games in the house when I come home from school.

Parental influence operated not only as a motivational force shaping students’ decisions to participate in the program but also as a mechanism that reaffirmed and amplified the program’s perceived value within the broader STEM ecosystem. This perspective was observed during the seventh annual STEM Night. As parents entered the room, one of the students programmed their EV3 Lego robot to verbally greet guests at the beginning of the program. In response, one of the parents stated, Wow! We really needed this [referring to the pre-engineering robotics program] in our community so that our kids won’t be behind. Another parent, at the same STEM Night, after seeing her daughter’s robot prototype to solve a local issue, stated, I didn’t know you could do all of this. Why are you only considering majoring in business if you can do this? You should do STEM. You can do programming. Through their endorsements, expectations, and active engagement, parents contributed to the social and cultural conditions that legitimized the program and strengthened its role as a meaningful pathway for students’ STEM development.

5.4. Theme 4: Youth Identity Development and Leadership as Outcomes That Sustained the Ecosystems

The program’s outcomes illustrate the ways in which the interconnected components of the STEM ecosystem operated in concert to advance a shared educational mission. This coordinated interplay among students, families, educators, community organizations, and industry partners enabled the initiative to cultivate the knowledge, skills, and STEM identities that the program was designed to develop. Through this systemic alignment, the ecosystem functioned as an integrated support structure that both facilitated individual student growth and reinforced the program’s broader goals for sustained STEM engagement. Although these findings do not encompass every stakeholder group within the broader STEM ecosystem, the actors represented most prominently in the data offered critical insights into the ways youth identity development and youth leadership operated as sustaining mechanisms within the system. Their perspectives illuminate how young people’s evolving sense of competence, agency, and belonging contributed to the ecosystem’s ongoing vitality and its capacity to support long-term program sustainability.

Students

During the third-year focus group, Patrick reflected on the broader personal and social significance of his participation in the program. He explained that his engagement had earned the recognition of his parents, who viewed his commitment as evidence of his growing focus and discipline. Patrick further noted that his involvement positioned him as a positive role model within his community and contributed to a more favorable public perception of his high school. As he described:

My parents, they love it, because I’m not at home. I’m not out on the streets, and I’m doing something positive for my community. I am setting an example for those who are young and those that are on out the streets, and giving more credit to Williams High School, making Williams High sound like a better school.

Patrick’s reflection illustrates how his emerging sense of identity is closely tied to engaging in social contributions to his community, suggesting that his participation in the program functions not only as an educational experience but also as a meaningful context for constructing a purposeful and socially responsible self-concept.

Collective collaboration functioned as a key developmental process through which students constructed a more robust sense of self, signaling meaningful growth in their emerging self-concept. During the fourth year focus group, Isaac identified specific areas of personal development that he attributed to his participation in the program. He explained that he had cultivated a deeper appreciation for active listening, a skill that enhanced both his collaborative capacity and his overall effectiveness as a contributing member of the team.

I’ve become a better listener, because I usually don’t like to listen. I don’t follow the rules a lot. Well, I follow the rules, but I like to do my own thing. I’ve learned to listen that there’s more than my way to success.

During the same fourth-year focus group interview, another student, Courtney, reflected on the group’s collective achievements and articulated a strong sense of pride in what the team had accomplished. She acknowledged both the collaborative effort that contributed to the group’s success and the significance of her own individual project. The positive feedback she received from a university partner regarding her work further reinforced her confidence and her sense of accomplishment.

I’m proud. Today, when we were at the [national robotics] competition, and we got to compete in certain rounds, our team, we scored points and we won maybe two or three rounds, so that made me feel good because we worked hard, and we were still able to show off our skills. And I was also proud when I made the safety animation and Dr. Johnson said it was great. That made me feel good.

During the fifth-year focus group interview, Ronald articulated a deep sense of pride in deliberately pursuing a personal goal that he described as foundational to his emerging sense of identity. He recognized that the process of working toward this goal had expanded his problem-solving abilities, particularly as he succeeded in accomplishing something he had not initially believed himself capable of achieving.

For me, my most proud moment wasn’t anything about the robot building. My most important moment was deciding that I was going to do something important for myself. I got into it because I felt it was going to be something beneficial towards me. Not only that, it’s one of the things that I’m proud of. I mean I’m happy about it. I made a decision. I thought outside of the box. I was in the program. I had to be clever. I was doing things I didn’t expect to do, and as I said before, making a decision that was actually good for me. That’s what I’m most proud of.

Across the multiple years of the program, students’ experiences contributed to shifts in identity that unfolded both at the individual level and within the collective functioning of the group.

These emerging patterns also signaled the development of a strengthened sense of belonging, not only within the immediate STEM ecosystem that supported the program but also across broader STEM environments in which students increasingly perceived themselves as legitimate participants. As the students prepared to roll their robot onto the competition floor, one of the drivers, BJ, was approached and asked whether he felt nervous, given that this was his first time competing and operating the robot at a national event. His response captured a significant moment of identity affirmation within the broader STEM environments in which the team was now participating. He stated, No, I am not nervous at all. I belong here! This statement reflects more than confidence in a single competitive moment. It illustrates the culmination of multiple years of developmental experiences within the STEM ecosystem that enabled students to internalize a sense of legitimacy and membership in high-level STEM spaces. His assertion of belonging suggests that participation in authentic, community-validated STEM activities can strengthen students’ identity as capable contributors who see themselves as entitled to occupy, navigate, and succeed within competitive STEM arenas. Moments such as this provide evidence of how supportive ecosystem conditions can translate into durable shifts in self-perception, persistence, and long-term engagement in STEM pathways.

6. Discussion

This study traced how a STEM ecosystem took shape and persisted in a historically disinvested community by showing that durable capacity emerged first through relationships, community anchoring, and identity development, which then scaffolded formal governance and resource-sharing arrangements. Funding, teacher capacity, and multi-partner participation were important inputs, yet they proved insufficient as standalone drivers of stability. What converted these inputs into long-term outcomes was a robust relational infrastructure that reduced coordination costs, legitimized decision making, and generated a steady flow of effort, knowledge, and material support across partners. Relational mechanisms operated through repeated, trust-building interactions among students, educators, university partners, district leaders, families, and community sponsors. These ties supported boundary spanning, rapid problem solving, and shared norms about quality and inclusion. Community-based mechanisms grounded the STEM ecosystem in place: partners aligned activities to local priorities, parents and sponsors made the work visible and valued, and district leaders signaled institutional commitment. Identity-driven mechanisms translated participation into persistence. As students experienced leadership, mastery, and recognition, they developed a stronger STEM identity and a sense of ownership that increased effort, reduced attrition, and attracted new participants. These identity gains fed back into the STEM ecosystem by motivating volunteerism, near-peer mentoring, and positive reputation effects that drew in additional resources.

A central contribution of this study is the identification of relational infrastructure as the primary mechanism sustaining the STEM ecosystem. Although relational infrastructure is recognized as an important dimension of STEM ecosystems, prior research often treats it as one component within a broader constellation of structural and organizational features (Allen et al., 2020b; Vance et al., 2016). In contrast, the present findings show that relationships among teachers, parents, university partners and students formed the core of the STEM ecosystem’s functioning. These relationships provided emotional safety, academic support, and a sense of belonging that countered the instability and limited opportunities characteristic of the broader community context. This aligns with research demonstrating that relational trust and sustained adult–youth relationships are essential for participation and identity development in marginalized communities (Calabrese Barton & Tan, 2018). The prominence of relational infrastructure in this case suggests that in communities marked by chronic underinvestment, relational commitments may be more consequential for STEM ecosystem sustainability than formal governance structures typically emphasized in STEM ecosystem frameworks (Traphagen & Traill, 2014).

A second contribution concerns the role of community recognition and collective pride in sustaining engagement. Students’ experiences of being seen, valued, and celebrated by their community functioned as a powerful motivational force that reinforced persistence and strengthened their identification with STEM. Although existing STEM ecosystem literature highlights the importance of community engagement and place-based design (Bevan et al., 2020; Timko et al., 2023), it rarely examines how public recognition and positive community narratives operate as sustaining mechanisms. The present findings indicate that in historically marginalized communities, community pride and visibility can serve as symbolic resources that stabilize participation and reinforce the legitimacy of STEM pathways. This extends current understandings regarding STEM ecosystems research by demonstrating that sustainability is not only a function of organizational capacity or cross-sector coordination but also of the social meanings that communities attach to STEM participation.

The third contribution highlights the importance of parental validation and family support as sustaining forces within the STEM ecosystem. Parents reinforced students’ engagement by providing encouragement, logistical support, and advocacy, and by recognizing the program’s impact on their children’s confidence, behavior, and academic performance. These findings align with research showing that family engagement is a critical factor in STEM identity development and persistence (King et al., 2021; Means et al., 2017). However, the present study extends this work by showing how parental support contributes to the STEM ecosystem sustainability, particularly after external funding ends. In this context, families acted as stabilizing agents who helped maintain continuity and reinforced the STEM ecosystem’s value, thereby contributing to its longevity.

A fourth contribution concerns the role of youth identity development and leadership in sustaining the ecosystem. Students’ growth in confidence, communication, problem-solving, and STEM identity not only supported their individual trajectories but also reinforced their commitment to the STEM ecosystem. As students took on leadership roles and mentored younger participants, they contributed to a self-sustaining cycle in which program outcomes generated renewed investment from both students and the broader community. This finding aligns with research showing that identity development and leadership opportunities are central to long-term STEM engagement (Allen et al., 2019; Shah et al., 2018). It also demonstrates that youth outcomes are not only indicators of program success but also mechanisms that contribute to the STEM ecosystem’s resilience.

Finally, the study offers a critical perspective on the selective manifestation of STEM ecosystem tenets in historically disinvested communities. While relational infrastructure, equity and community engagement, and connected learning pathways were strongly evident, other tenets, such as distributed leadership, formal governance structures, and data-driven continuous improvement, were less visible in this data set. This selective enactment reflects the realities of communities with limited institutional capacity and long histories of structural inequity. It also suggests that STEM ecosystem frameworks may require adaptation when applied in such contexts. The findings indicate that sustainability in these settings may position relational and community-based mechanisms as highly significant to formal structures typically emphasized in STEM ecosystem models. This insight contributes to the field by suggesting that STEM ecosystem frameworks should be flexible and context-responsive, with greater attention to the relational and cultural dimensions of sustainability (Timko et al., 2023; Bevan et al., 2020). Taken together, these findings advance understanding of how STEM ecosystems develop and persist in historically marginalized communities. They demonstrate that sustainability is not solely a function of organizational design or resource alignment but is deeply rooted in relationships, community narratives, family engagement, and youth identity development. By documenting how these mechanisms operated over ten years, this study provides empirical evidence that expands current STEM ecosystem frameworks and offers practical guidance for designing and sustaining STEM ecosystems in contexts where structural inequities shape the landscape of opportunity.

7. Limitations and Conclusions

As with all single-case qualitative studies, the findings are shaped by the specific social, organizational, and historical conditions of the region, which constrain the transferability of the mechanisms identified to STEM ecosystems operating under different configurations. Additionally, because the study relies primarily on annual focus groups and field notes, it provides rich insight into stakeholder experiences but limits opportunities for triangulation with broader institutional documents, non-participating families, or longitudinal quantitative indicators of STEM pathways. These methodological boundaries narrow the scope for generalization, yet the study nonetheless offers a meaningful contribution to understanding how sustainability emerges through relational, organizational, and identity-related processes within STEM ecosystems.

Advancing the field will require research that interrogates how STEM ecosystems function in varied community and institutional environments, and how key relational and identity-driven mechanisms adapt under different conditions. Future work should use multi-site comparative designs across communities that vary in funding stability, institutional capacity, and industry presence to test where relational, community-based, and identity-driven mechanisms travel and where they require adaptation. Mixed-methods studies that pair case studies with social network analysis, longitudinal identity measures, and participation trace data would allow researchers to model how relational infrastructure predicts persistence, learning, and role uptake over time. Researchers should examine mechanisms formally by modeling parental validation as a mediator between participation and persistence, and community recognition as a moderator that strengthens identity formation and retention under resource stress. Additionally, the STEM ecosystem research would benefit from developing clear evaluation metrics of STEM ecosystem health, such as the strength of cross-role relationships, balanced partner contributions, student uptake of leadership roles, and diversified resource streams, so that ecosystem stability and growth can be monitored over time.

In policy environments where formal DEI offices, titles, or line items are restricted or politically contested, the study points to practical pathways for sustaining inclusive STEM ecosystems through core instructional and community practices rather than program labels. Relational infrastructure can be cultivated through universal student success strategies that are mission-aligned and compliance-safe, such as near-peer mentoring, structured recognition of mastery, parent liaisons, and regular partner huddles that solve problems and share assets. Equity work can be embedded in the everyday design of learning and governance by emphasizing access, belonging, clear role pathways, and transparent criteria for recognition, which improves outcomes without reliance on contested terminology. Communities can buffer fiscal and political shocks by diversifying supports through alumni networks, micro-sponsorships, in-kind contributions, and industry partnerships framed around workforce readiness and talent development. Measurement should focus on transparent outcomes that matter to broad audiences, such as persistence, credential attainment, problem-solving gains, and partner retention, while continuing to monitor distributional effects across student groups. By grounding inclusion in relationships, visible achievement, and community legitimacy, STEM ecosystems can remain durable and principled even when formal DEI infrastructures are curtailed.