Much of the racial disparities in math can be attributed to the structural and systemic barriers that disproportionately affect Black and Latina/o students. One key factor is the inequitable access to high-quality math learning opportunities, which persists throughout the K-12 pipeline in the U.S. (Davis et al., 2019). In math classrooms, students experience both covert and overt biases, including stereotypes, microaggressions, and being overlooked for advanced classes (Copur-Gencturk et al., 2020; Grossman and Porche, 2013). Such biases directly and indirectly shape students’ math engagement (Wang et al., 2024). For example, recent studies of middle and high school students from diverse racial and ethnic backgrounds found that perceptions of math/STEM teacher discrimination were associated with lower math/STEM engagement (Del Toro et al., 2024; Mulvey et al., 2022). As another example, some math curricula may lack cultural responsiveness or fail to reflect diverse perspectives (Collins, 2018; Mathews et al., 2022), which could hinder Black, Latina/o, and other marginalized students’ sense of belonging in the classroom, their motivation, and their ability to fully engage in math and reach their academic potential.

Despite the structural and systemic barriers that they disproportionately face, Black and Latina/o youths continue to demonstrate resilience in math. The community cultural wealth model (Yosso, 2005), highlights that, although often unrecognized by educational systems, Black, Latina/o, and other students from marginalized backgrounds bring tremendous assets into the classroom. These include educational aspirations and resilience despite obstacles (aspirational capital), knowledge and skills for navigating complex and often racialized systems (navigational capital), motivation to fight against unfair systems (resistant capital), diverse communication skills (linguistic capital), access to community networks and support (social capital), and cultural and experiential knowledge passed through families (familial capital) (González et al., 2013; Nasir et al., 2008).

Studies from a cultural-wealth perspective suggest that Black, Latina/o, and other students from marginalized backgrounds bring these strengths into math learning spaces. For example, a recent systematic review found positive relations between parents’ STEM-specific support and Black and Latina/o middle and high school students’ STEM motivational beliefs (Starr et al., 2022). McGee and Spencer’s (2015) study with high-achieving Black students in STEM showed that parents played a major role in fostering STEM success through instilling persistence, encouraging their self-efficacy, and serving as role models. Relatedly, studies have found that Black and Latina/o parents foster their children’s math and STEM learning by emphasizing the value of education as a pathway for better opportunities (Martin, 2006; Soto-Lara and Simpkins, 2022). These studies illustrate the rich familial capital and broader community cultural wealth that Black, Latina/o, and other marginalized students bring to math learning.

Over time, Black, Latina/o, and other students from marginalized backgrounds may internalize aspects of their community cultural wealth as personal strengths. For example, STEM can elicit assets such as critical consciousness that are particularly critical for the thriving of students from marginalized backgrounds (Mathews et al., 2025; Upadhyay et al., 2021). Relatedly, the barriers that these students navigate in math may strengthen their motivation to disprove stereotypes (Fries-Britt and Onuma, 2020; Wilson and Matthews, 2024), though not to downplay the substantial emotional tolls that such a process of resiliency takes. These internal assets contribute to the development of strong and healthy math/STEM identities (Berry et al., 2011; Collins and Jones Roberson, 2020). Despite the valuable assets that Black, Latina/o, and other marginalized students bring to math and STEM, the community cultural wealth framework remains underutilized in STEM research. A recent systematic review suggested that the community cultural wealth framework is still not widely used in STEM compared to studies that examine marginalized students from a deficit-based perspective (Denton et al., 2020). This gap underscores the need for more research that centers the strengths and lived experiences of marginalized students in STEM.

With the goal to understand the various and interrelated barriers and assets that Black and Latina/o students face and utilize for their math engagement, we use the cultural microsystem model developed by Vélez-Agosto et al. (2017). The cultural microsystem model adapted Bronfenbrenner’s bioecological theory (Bronfenbrenner and Morris, 2006) by centering cultural processes as fundamental to development. Specifically, the cultural microsystem model drew upon Vygotsky’s Sociocultural Theory, Weisner’s Ecocultural Theory, and Rogoff’s transformation of participation perspective to argue that culture is inseparable from human development and its layered contextual influences. As such, the cultural microsystem model recognizes that “mathematics, like all other forms of knowledge, is situated within a cultural context” (p. 262, Leonard et al., 2010).

We use the cultural microsystem model to guide this study for two main reasons. First, its bioecological foundation provides a framework for organizing math engagement barriers and assets across different ecological levels. Examining these factors at multiple levels allows us to understand how immediate environments, connections between those environments, and broader societal influences collectively shape math engagement. At the most immediate level, the microsystem includes students’ direct interactions within classrooms, families, and peer relationships. Beyond this, the exosystem consists of broader school structures and policies, such as access to resources, tracking, and school climate, which influence students indirectly. The macrosystem encompasses the larger societal and systemic forces that shape math engagement, including racialized perceptions of ability, structural inequities, and dominant cultural narratives about who is “good” at math. Lastly, the chronosystem captures the influences of time or historical context on students.

The other main rationale for using the cultural microsystem model to guide this study is that the model centers the role of cultural processes, which aligns with our focus to examine the inequities that Black and Latina/o students navigate in math learning. Importantly, centering the role of cultural processes also makes room to examine the assets and strengths that marginalized students draw upon to navigate the barriers, which ties back to the community cultural wealth perspective (Yosso, 2005) that we reviewed in the previous section.

This study examines two central research questions: (1) What are the barriers to math engagement from the perspective of Black and Latina/o students? (2) What strengths-based strategies and assets are Black and Latina/o students utilizing to navigate these barriers?

We intentionally focus on the developmental period of adolescence for two main reasons. Firstly, students’ math motivation and engagement typically decline during this developmental period (Fredricks and Eccles, 2002; Denner et al., 2019; Gottfried et al., 2007; Martin et al., 2015; Musu-Gillette et al., 2015); making it critical to understand the factors that hinder and support students’ math engagement during this developmental period. Secondly, a significant milestone of adolescence is the negotiation of identities, including one’s math identity (Berry et al., 2011; Collins and Jones Roberson, 2020). By examining both the barriers and the strategies that adolescents use to navigate them, this study offers important insights into how to better support Black and Latina/o students’ engagement and success in math.

This study is part of a larger, three-year project that aims to develop a measure of Black and Latina/o students’ math engagement. The project took place in a school district that is located in the Minneapolis-St. Paul (MN) metropolitan area in Minnesota, U.S. All three middle schools (one is predominantly attended by White students, one is racially/ethnically diverse, and one is predominantly attended by student of color; see the notes section of Table 1 for more detail) and two high schools (one is predominantly attended by student of color, the other is predominantly attended by White students) in the district participated in this project. In this school district, most students take math all throughout middle school, but not necessarily all throughout high school since the state only requires three credits of math for graduation.

It is important to disclose these contextual information for this study as we attend to issues of educational/math inequities and systemic barriers that students from marginalized backgrounds face. For example, according to recent (2019) data from the districts’ North Star system results (i.e., state-wide accountability indicators), Black students had a math achievement rate of 27 percent and Latina/o students had a rate of 31 percent, a stark contrast to the 54 percent state average (Minnesota Department of Education, 2019).

The data for this study came from focus groups conducted with 50 Black and Latina/o students conducted in Spring 2023, and with 57 Black and Latina/o students conducted in Spring 2024. See Table 1 for more detailed participant demographics. A qualitative approach was used to capture how Black and Latina/o students conceptualize math engagement, allowing for a deeper understanding of their experiences than surveys or standardized assessments might reveal. Focus groups were selected as the primary method because they provided a structured yet flexible setting where students could share their perspectives, reflect on their engagement in math, and build on one another’s insights (Creswell, 2012).

The focus group protocols were designed to explore how Black and Latina/o students experience engagement in math, with questions aimed at understanding both barriers and strength-based strategies (see Appendix 1 for full protocols). The Spring 2023 focus groups focused on understanding students’ experiences with math engagement and the factors that influence it. The discussions included questions about what makes students feel engaged or disengaged, the role of their learning environment, and the challenges they encounter in math classrooms. Students were also asked about barriers to engagement, including instructional practices, classroom interactions, and broader perceptions of who is seen as a “math person.” Additionally, the protocol included questions about strength-based strategies, prompting students to describe the resources and approaches they use to stay engaged, such as seeking support from teachers, peers, and family or using personal strategies to persist in math.

The Spring 2024 focus groups expanded on these discussions, incorporating questions to deepen understanding of how students experience support in math. Students were asked to recall specific moments when they felt supported, who provided that support, and how it influenced their engagement. The discussions also explored how students perceived others’ views of their math abilities and whether those perceptions affected their confidence and participation. These follow-up questions were intended to provide a more detailed understanding of the types of support that contribute to students’ engagement and persistence in math.

Each focus group consisted of 4 to 7 students, lasted approximately 90 min (Spring 2023) or 60 min (Spring 2024), and was conducted in person. The student participants were recruited by teacher co-researchers who are part of the larger project team. The teacher co-researchers were informed of the purpose and goal of the focus groups and were asked to recruit students who identify as Black and/or Latina/o. We intentionally asked the teacher co-researchers to recruit students from a range of engagement levels in math to capture a diverse perspective of student’s lived experiences, which helps strengthen the transferability and thus trustworthiness of our data. When participants signed in for their focus group, they filled out a paper slip that asked for their grade, gender, race/ethnicity, and their response to the question “do you like math?” (see Table 1). All study procedures were approved by the Child Trends Institutional Review Board (IRB#: FWA00005835). Each student received a $50 electronic gift card as compensation for their time and insights.

All focus group discussions were audio-recorded and transcribed verbatim using an online transcription service. We conducted a thematic analysis (Braun and Clarke, 2012), which involves identifying, analyzing, and interpreting patterns of meaning across qualitative data. Our analytic process was both theory-driven and flexible, guided by the cultural microsystem model (Vélez-Agosto et al., 2017), which helped us consider how students’ math engagement is shaped by factors at different levels—from classroom relationships and school environments to broader social and cultural dynamics.

To develop our initial codebook, the first author reviewed approximately half of the transcripts, creating preliminary codes informed by prior literature on math engagement and equity (Berry et al., 2011; Leonard et al., 2010; Denner et al., 2019). Our research team then collaboratively tested and refined this codebook by discussing examples and adjusting definitions based on additional student responses. For example, as students described personal strategies—such as independently managing distractions or maintaining focus during challenging lessons—it became clear these did not fit neatly within existing ecological categories (micro, exo, macro). Thus, we added an intrapersonal category to more accurately reflect these individual-level approaches. As analysis progressed, we noticed that some strategies students described were closely connected to specific challenges, while others offered more general support that cut across multiple types of barriers. We approached the analysis with this flexibility in mind, remaining grounded in what students said while also considering how their strengths related to the barriers they described.

All coding was conducted in Dedoose. Each transcript was initially coded by a team member and then reviewed by another to check for accuracy and consistency. The research team met regularly to review the transcripts together, clarify discrepancies, and ensure shared understanding of how codes were applied. After the coding phase, the first and second author met to cluster related codes into themes and finalize their definitions. We also used memoing throughout the process to document emerging insights and reflect on how our interpretations evolved over time. Overall, the collaborative and iterative approach to coding helps strengthen the dependability and hence trustworthiness of our data analysis.

This study was a team effort, and we are mindful of the various perspectives each of us brought to the team and how they might shape interpretation. Specifically, the study team, who are also the authors of this manuscript, consisted of five researchers (from four research organizations), as well as six students and five math teachers (from each of the five participating middle and high schools). With a team of this size, we drew on a wide range of personal, professional, and practice-based experiences related to equity and math education. We are also diverse in terms of our lived experiences with math. While all student co-researchers identified as Black and/or Latina/o, most of our teachers and researchers identified as White or Asian (American). Beyond these difference in our racialized experiences with math, the teachers and researchers in our team are also cognizant that our math upbringings differed from the students, partly due to generational differences. As part of the larger study, we met regularly and drew on each other’s expertise to inform our understanding and interpretation of the focus group data. For example, while the researchers were responsible for most of the data analysis in terms of coding, we relied on the student and teacher co-researchers for member checking, which helped strengthen the credibility and thus trustworthiness of our findings.

To answer research question 1, which focused on identifying barriers to math engagement, this section describes the challenges Black and Latina/o students reported across multiple ecological levels, organized using the cultural microsystem model. Table 2 summarizes the major themes, definitions and illustrative quotes related to these barriers.

To answer research question 2—what strength-based strategies and assets Black and Latina/o students draw upon to stay engaged in math—this section highlights how students navigated the barriers they encountered. These findings are organized using the same ecological framework as before, with an added intrapersonal level to reflect self-directed strategies and internal motivations. No macro- or chrono-level responses emerged, as students focused on actions and supports that were immediate and within their control—such as their own habits, classroom interactions, and relationships. Table 3 summarizes the major themes, definitions and illustrative quotes related to these strength-based strategies and assets.

Taken together, our findings align with the call to conceptualize student engagement more holistically, socioculturally, and critically (Lawson and Lawson, 2013; Wang and Hofkens, 2020). A holistic perspective encourages schools to see engagement as more than student behavior or attention—it’s also about whether students have the resources they need to participate. For example, something as simple as coming to school with a pencil or notebook—basic but essential—was tied to whether students could meaningfully engage in math (Martinez and Ellis, 2023).

A sociocultural lens emphasizes the importance of instructional approaches that reflect and build on students’ lives and experiences. Tying back to our community cultural wealth framing (Yosso, 2005), the first step in engaging students from marginalized backgrounds is to recognize and uplift the diverse assets they bring into the classroom, which too often get overlooked. Many students in this study responded more positively to interactive, hands-on lessons than traditional lectures—highlighting the need for culturally responsive and engaging math instruction that taps into students’ funds of knowledge (González et al., 2001; Thomas and Berry, 2019). Similarly, students’ references to family motivation and community pride underscore the importance of affirming familial capital and making space for students’ identities and cultural values in math learning (Williams et al., 2016).

Representation also matters. Students’ comments about who they see in advanced math classes—and who they do not—underscore the importance of creating inclusive environments where all students can see themselves as capable math learners. This includes recruiting and retaining diverse educators, showcasing relatable math role models, and shifting dominant narratives about who “belongs” in math (Gladstone and Cimpian, 2021; Lee et al., 2022; Martin and Fisher-Ari, 2021; Yonas et al., 2020).

Lastly, a critical perspective asks schools to reflect on their own structures and power dynamics. Students in our study did not always have consistent access to help, and sometimes had to work around instructional barriers. As Gutiérrez (2018) argues, rehumanizing math involves shifting power and authority toward students—designing classrooms where they can shape the learning, not just receive it.

One strength of the study is its attention to both contextual barriers and student agency, highlighting how engagement is shaped by systems yet also navigated by youth with intention and care. Our ecological approachsurfaced themes and patterns across layered contexts. However, because contextevolve over time, our study is limited in that it captures only snapshots of how students responded to challenges. Future research could take a more longitudinal or narrative approach to trace how students’ math engagement strategies evolve over time or across settings.

Another strength of this study is it being grounded in the voices of Black and Latina/o students. We intentionally did not compare between our Black and Latin/o participants, as our goal was to understand shared challenges and strategies. This scope, though, comes with limitations as we recognize that group-specific experiences exist and deserve further exploration. Future studies might take a more targeted approach to examine whether different student groups experience or respond to math engagement challenges in distinct ways. In fact, our data shows that there are tremendous diversity and nuances even within the Black and Latina/o student groups (e.g., see Table 1 for the various racial/ethnic and gender identities that students self-reported), this calls for studies that take a within-group approach to understand the unique, and often intersecting, lived experiences of students from marginalized backgrounds.