Moving Beyond Eurocentric Notions of Intellectual Safety: Insights from an Anti-Racist Mathematics Institute

Abstract

This paper reconceptualizes intellectual safety in mathematics spaces by centering the voices and lived experiences of BIPOC students. The marginalization of BIPOC students is compounded by structural racism, historical exclusion, and deficit narratives that continue to shape academic environments, especially in mathematics contexts. While definitions of intellectual safety reflect white, Eurocentric norms, we argue that for BIPOC students, intellectually safe environments must be anti-racist, culturally responsive, and rooted in belonging. We started with existing definitions of intellectual safety and incorporated a more critical approach to sense of belonging. Through ethnographic research design we gathered student interviews and daily journal entries from a 12-day anti-racist mathematics summer institute for secondary students. Analysis revealed that while existing attributes captured much of the scholar’s joy, cultural affirmation, and belonging, new themes, like pride/confidence, clarity/transparency, and being listened to, emerged directly from how students experienced intellectual safety in practice. This led to a refinement of our initial conceptualization. This study provides insight into how intellectual safety manifests in a space intentionally designed to support BIPOC youth in exploring mathematics in agentic and culturally sustaining ways.

1. The Case for Intellectual Safety

Intellectual safety aims to address how students feel comfortable challenging their own beliefs and assumptions in classrooms and feeling safe to ask questions without fear of repercussions from instructors or peers. Instructors ask and encourage questions from students in hopes to gauge students’ understanding of concepts and participation. A long-standing anecdote heard in academia is, “don’t be afraid to ask your question, there is most likely someone else in the room who has that same question as you.” So why is there hesitation or refusal in asking questions in classroom settings? What creates a safe place to learn? When or why do academic spaces feel unsafe for students and their learning? For some, the feeling of unsafety comes from fear of judgment, or fear of being deemed unintelligent by their peers or instructors. It could also come from feelings that your instructor does not believe that you are smart enough or capable (Lightfoot, 1986). Whatever the reasons may be, one thing is clear, we must create learning environments where students feel intellectually safe. It is important to think about the types of spaces we are intentionally creating for students to explore, question, and grow as individuals.

Intellectual safety has been defined by scholars in various ways. Guzzetti and Williams (1996) define intellectual safety as the ability of students to engage in science discourse with confidence and without fear of challenging dominant social norms. Their study also emphasized the need for educators to intentionally address gender inequities, as female students often experienced intellectually unsafe environments in male-dominated fields. Schrader (2004) defines intellectual safety through the lens of students’ epistemic perspectives and what she terms the “moral atmosphere” of the classroom. Her definition places emphasis on the tension between students’ epistemologies and the instructor’s role in creating psychological safety, describing intellectually safe environments as those that balance care and control, respect and fairness, and affirm students’ ways of knowing, belonging, and contributions. Building on Schrader’s (2004) work, Call (2007), in a study with 147 college students (74% female, 88% white), found that students defined intellectual safety in terms of two key categories: freedom of expression and comfort. In her definition, cultivating intellectual safety involved the ability to express oneself without fear of repercussions, as well as the classroom climate, including the professor’s care, openness, motivation, respect, and instructional methods. All these definitions emphasize that intellectually safe environments are built through intentional relationships, equitable classroom structures, and a commitment to affirming students’ diverse ways of knowing and participating.

We argue that a key limitation in existing definitions of intellectual safety was the lack of attention to students’ diverse social identities, particularly race, gender, and their intersections. Givens and Ison (2023) found that many current investigations of intellectual safety are rooted in the experiences of white male or white female students, failing to reflect the realities and needs of today’s racially diverse nation (Givens & Ison, 2023). Given these limitations, it becomes essential to reconsider what intellectual safety means in practice, especially in spaces like mathematics education where BIPOC students have been historically marginalized. Creating environments where students experience learning that is exciting, meaningful, academically challenging, and supportive is particularly crucial in mathematics spaces for BIPOC students. Research has explored the integration of antiracist pedagogy or culturally relevant practices to address the disproportionate lack of visibility and inclusion for racially and socially marginalized students in a field historically dominated by white students (Battey & Leyva, 2016; Dost, 2024; Gray et al., 2018; Gutierrez, 2012; Nasir & McKinney de Royston, 2013; Ortiz & Ruwe, 2021; Stinson, 2008). This is not to suggest that efforts are less important in other disciplines but rather that we will focus on mathematics contexts while recognizing the need for similar investigations across other fields. Having a climate that is equitable, where students feel valued, safe in learning, cared for, heard, etc., has been noted to be important in BIPOC students’ learning generally (Lightfoot, 1986) and in mathematics (Louie, 2018; Maloney & Matthews, 2020; Martin et al., 2019; Ortiz & Ruwe, 2021).

2. Learning Environments Beyond Inclusion

Although this study focuses on intellectual safety and its implications, particularly from a critical lens, it is deeply connected to broader movements in antiracist education and transformative approaches to mathematics teaching and learning. In the following text we briefly highlight this larger context below.

2.1. The Need for Culturally Responsive and Anti-Racist Education

Antiracist education intentionally challenges racism in schools by rethinking how and what we teach to how students are treated in the classroom. Antiracist education is grounded in Critical Theory and challenges deeply embedded racism in education by disrupting dominant, white-centered ideologies that normalize the white or Eurocentric epistemologies masquerading as merit in academic spaces. These ideologies, under the disguise of neutrality, are reflected in pedagogical practices that routinely rely on a color evasive discourse in the curriculum that centers white intellectual property at the front of academic spaces, limiting opportunities for BIPOC students (Chapman, 2013). Antiracist pedagogy pushes educators to create learning environments that value students’ cultural knowledge, encourage critical thinking about real-world issues, and disrupt systems that position white knowledge as the norm Blakeney (2005) in service to the liberation of all students (not just those with marginalized identities). Indeed, scholars have long argued that reimagining school curricula in ways that reflect cultural diversity expands the knowledge of all students (Baldwin, 2008; Oakes & Lipton, 2003).

Interventions that are centrally focused on cultivating critical consciousness, joint advocacy, and centering students in ways that intentionally address their experiences and sociopolitical realities are essential steps toward advancing liberatory learning environments (N. R. Davis et al., 2020; Hartlep & Xiong, 2018; Robbins et al., 2016).

2.2. Equity and Exclusion in Math Education

Traditional math classrooms often rely on rigid routines, like memorization and decontextualized problems, which privilege white, middle-class ways of learning while excluding the experiences and strengths of BIPOC students (Chapman, 2013; Oakes & Lipton, 2003). Westernized mathematical knowledge and competencies have also historically been a way to measure students’ intelligence and have primarily benefited white, male, and wealthy students (Gutierrez, 2013; Martin, 2019). BIPOC students often have limited access to advanced mathematics courses and are underrepresented in mathematics related careers (Maloney & Matthews, 2020). This is no surprise given the opportunity gaps imposed by education’s long history and antiblack roots. Although there has been tried improvement on a national scale (e.g., “mathematics for all”, common core practices, NCTM’s Principles to Action) many of these have fallen short in their tailoring towards white students’ ways of knowing in mathematics or for the economic advancement of the U.S. Instead, much of the equitable change has come from individual institutions and individuals relying on culturally sustaining pedagogical practices (Louie, 2018; Ortiz & Ruwe, 2021; Overby et al., 2022; Robbins et al., 2016; Wang et al., 2023). While feelings of connectedness are important to students’ mathematical identities, it is just as important for instructors to see them as mathematically capable (Louie, 2018; Maloney & Matthews, 2020).

2.3. Theoretical Framework

In Schrader’s (2004) definition of intellectual safety, she highlights the professor’s role in building meaningful connections with the content and instructor, emphasizing that feeling intellectually safe is built from this relationship. She mentions instructors’ disclosure, physical stance, classroom format, care and relationship with peers. An intellectually safe environment from Schrader’s (2004) point of view is described as “a caring environment in which the professor is open and caring, demonstrates respect, and embraces the uniqueness of students and their perspectives and does so in a classroom format where in which all are invited to participate actively, engage in personal self-disclosure while trusting the confidentiality of such openness, and where the professor maintains a sense of control and direction to facilitate learning.” (Schrader, 2004, p. 94). Schrader (2004) emphasizes including students feelings on themselves, the idea of care, connections to their peers and ties to the subject matter as important in cultivating an intellectually safe environment. In her study, the idea of care was thought of as identifiers that the instructor demonstrated in the classroom like compassion, interest in the subject matter and speaking on an equal level with students.

Call, on the other hand, describes an intellectually safe environment as “…an open and comfortable atmosphere in which students feel free to express their thoughts, feelings, or opinions without fear of retribution from instructors or peers.” (Call, 2007, p. 28). There is a greater emphasis on the importance of feeling safe from judgment, being deemed wrong, or perceived as incompetent. Call (2007) identified two primary components (freedom of expression and comfort) with the professor and classroom atmosphere operating as secondary contributors to an intellectually safe environment.

While both Schrader and Call’s frameworks provide a solid foundation, we argue that their approaches to intellectual safety are insufficient for addressing the sociopolitical realities of BIPOC students in mathematics academic spaces. Both Call (2007) and Schrader (2004) include the notion of care in their definitions, they conceptualize it in ways that reflect surface-level enactments, like general kindness or openness, without critically engaging with the power dynamics and cultural contexts that shape students’ experiences. As scholars have noted, a critical approach to care, one that explicitly attends to students’ social identities, cultural backgrounds, and experiences, has been shown to better support BIPOC students’ academic achievement and sense of belonging (Chhuon & LeBaron Wallace, 2012; DeNicolo et al., 2017). Deliberately attending to students’ political and social identity while challenging discriminatory practices is an aspect of care (or “cariño conscientizado”) that should be practiced in the classroom to move away from surface level caring (DeNicolo et al., 2017; Freire, 2000); especially for BIPOC students. With this in mind, we first turned to the literature on belonging to inform a more critical conceptualization of intellectual safety; one that attends to students’ sociopolitical contexts and affirms their cultural identities. Using these definitions and critical insights, we developed a reconceptualized framework for investigating how BIPOC students experience intellectual safety.

We recategorized Schrader’s (2004) major themes on connections to the self, peers, and professor and Call’s (2007) main components of freedom of expression and comfort, into our conceptual framework. This allowed us to hypothesize what we believe intellectual safety should entail to better reach a more diverse group of students. We began by adapting Schrader (2004) and Call’s (2007) framework into these attributes: exploration, joy, intellectual risk, and critical inquiry (Figure 1), belonging in mathematics classrooms (Figure 2).

We made connections with students engaging in exploration (authoritative but not authoritarian; flexibility in the classroom in ways that are authoritative but not authoritarian), experiencing joy within the classroom (connections to the self, a liking of the materials/subject matter), critical inquiry (classroom format; engagement in the classroom that speaks directly to students’ interest and discussion over lecture, encouraged questions, recognizing the intrinsic uniqueness of another respecting individuality), and intellectual risk (freedom to express themselves through various means without fear from and for repercussions) Figure 1.

A key adaptation of Schrader’s (2004) and Call’s (2007) definition of intellectual safety was the integration of a sociocultural lens that centers BIPOC students’ sense of belonging in mathematics spaces. We argue that recognizing the nuanced experiences shaped by diverse racial and cultural identities is essential to fostering true belonging in academic environments. Sense of belonging has been described to include feelings of connectedness with the community, peers, and instructor in a way that values students’ cultural and racial diversity and sees them as valued members of the learning community (DeNicolo et al., 2017; Duran et al., 2020; Maloney & Matthews, 2020). Historically, racially marginalized students have been framed as less likely to feel a sense of belonging in the classroom (Calabrese Barton & Tan, 2020). Student-teacher relationships are preeminently the strongest predictors of students’ sense of belonging. Sense of belonging has also been associated with future positive academic outcomes in short-term programs (Strayhorn, 2023; Williams et al., 2020). While Call (2007) does not directly attend to students’ sense of belonging in the classroom she does bring forth the importance of cultivating a relationship between the instructor and peers in ways that invoke trust and compassion through a positive classroom climate. Schrader (2004) on the other hand describes a sense of belonging built through the moral atmosphere, where students create shared norms and a supportive community, developing a sense of connectedness that creates a moral safety net for learning. We included Schrader’s (2004) themes that centered around the practices of the instructor: care (showed compassion and spoke to students at the same level), physical stance (approachable, acceptance of diverse perspectives and recognized the necessity of a caring connection to another person as an autonomous person), disclosure (shared personal stories, respected privacy and expressed honesty), and peers (opinions of other students influenced their participation, whether they would voice their reactions or questions). From Call’s (2007) study, we primarily used freedom to express without repercussions (expressing themselves through various means). These included the expression of their thoughts, ideas, feelings, concepts, theories, questions, beliefs, and views in the classroom without repercussions), care (trust in a teachers and their care for the students” and “an educator who cares about the students and encourages them individually in their education.”), openness (when a professor is open to his or her students, he or she might consistently use eye contact and “nodding to elicit more comment”) comfort (protection, attitude of the Professor, and nondiscrimination which reflect students’ views that intellectual safety involves feeling shielded from harm, supported by fair and non-imposing professors, and respected across diverse identities and perspectives.) and motivation (all of the positive feedback professors provide to their students, the enthusiasm they have for the subject, and their ability to stimulate interest.).

We align these attributes with Matthews et al. (2021) framework on belonging-centered instruction or BCI, highlighting how incorporating this framework offers a more critical lens on belonging, particularly for BIPOC students. Belonging-centered instruction attends to the ways in which teachers create classroom environments that respect BIPOC students’ autonomy, mathematical ideas, feelings of belonging in the classroom, and overall climate. This protocol was created by primarily centering BIPOC students. In belonging-centered instruction, there are two domains of teacher actions that promote a sense of belonging: instructional supports and interpersonal supports. Interpersonal supports is defined as, “teacher practices that actively facilitate social ties, social comfort, and interconnectedness between members of a classroom.” and instructional supports are defined as. “teacher practices that direct instruction toward making meaningful connections between content, the learning process, students’ identities, and broader society.” (Matthews et al., 2021, p. 6) where each in turn has subdimensions associated. To better articulate, or delineate, sense of belonging with an orientation towards criticality, we identify the following subdimensions: Social & Emotional Bridging, Communal Orientation, Empathetic Awareness & Support for the interpersonal domain and Safety to Be Wrong, De-centering Teacher Authority, Mathematics to Know Myself & My World, and High Standards & Rigorous Support for the instructional domain, see Figure 2. Moreover, we integrate two additional attributes: cultural distinctiveness and citizenship, borrowed from Gray et al.’s (2018) framework on belonging, where there is specific attention to cultivating experiences that validate students’ cultural identities through intentional activism and normalizing cultural uniqueness.

Finally, putting these aspects of intellectual safety together we proposed an initial conceptualization of intellectual safety that incorporates a critical approach to belonging that centers the experiences of BIPOC students in mathematics spaces. We note the bidirectional nature of intellectual safety with a sense of belonging as we expect that we can have an intellectually safe environment to some degree without a sense of belonging and for students to feel a sense of belonging without having an intellectually safe environment. Figure 3 presents our initial conceptualization of intellectual safety, with a particular focus on fostering students’ sense of belonging in mathematics spaces. This framework draws from Schrader’s (2004) and Call’s (2007) definitions of intellectual safety, as well as Matthews et al. (2021) model of belonging-centered instruction, or BCI, and Gray et al. (2018), to shape a critical approach that foregrounds students’ lived experiences and identities.

In this paper, we take a critical approach to investigating how BIPOC students experience intellectual safety. Using an antiracist mathematics summer research institute as illustration, we expand the current literature on academic belonging and centering students’ cultural identities. Within this context, this paper addresses the following questions:

• How can intellectual safety be reconceptualized to center the experiences and lived realities of BIPOC students?
• In what ways (if any) do secondary (8th–12th grade) students, from racially and socially marginalized backgrounds experience intellectual safety?

We further elaborate on this conceptualization and the development of our analytical codes in the Section 3.

3. Research Design

Below we examine the research context, methodology, and methods used in the study. The context centers on the Antiracist Scenario-Based Mathematics Summer Institute intentionally designed to center and support secondary BIPOC students. We include details on participant selection, data sources, and the development of the initial conceptualization. This study was guided by a research approach that emphasized researcher identity, the impact of systemic inequities, culturally grounded knowledge, and centering marginalized student voices. These guiding principles are further detailed below.

3.1. Research Context

The project serves BIPOC students and uses a co-design process to engage students as knowledge-creators and decision-makers fostering student autonomy and ownership. The storytelling approach for covering mathematical concepts deviates from traditional learning environments that often neglect explorative thinking, joy, storytelling, and experiential knowledge, as mentioned in STEM education literature which underscores the devaluation of Black, Brown, and Indigenous students’ knowledge and perspectives in mainstream mathematics academic spaces (Brayboy, 2005; Ladson-Billings, 1998). As mentioned earlier, the primary goal of the institute was to provide students with an opportunity to explore mathematics in a space where antiracist pedagogical practices were rooted in both instruction and the classroom climate. Coaches in this space normalized the concept of Black excellence and highlighted how systemic barriers prevent BIPOC students from receiving the same opportunities as white students. To emphasize our view of students as meaning makers, academic agents, and knowledge producers, we referred to them as “scholars” throughout the institute. Scholars worked in teams to develop storyboards for an antiracist scenario-based unit in mathematics in collaboration with math coaches. The research project focused on creating scenario-based mathematics assessments that addressed sociopolitical injustices relevant to the participants’ lived experiences or interests. The storylines and mathematics standards were chosen by the scholars. Prior to arriving at the institute, scholars were asked to complete a preliminary survey where they were asked if they had a particular interest in an injustice or issue that they would like to investigate. Their responses helped build a “bank” of potential topics scholars could choose or advocate for the storyboards. Examples of some of the potential sociopolitical injustices to be investigated included: reproductive rights, predatory pricing based on race, false incarceration, wage gap, health discrimination, racial discrimination, and hair discrimination. Math standards to investigate were primarily chosen from the Michigan K-12 Standards Mathematics (Michigan Department of Education, 2018). These assessments integrated mathematical concepts with real-world issues, centering students’ agency and amplifying their voices.

3.2. Participants

For recruitment, a flyer was created and sent out to students encouraging those who were interested in mathematics and making real-world change through the investigation of sociopolitical injustices. Students were encouraged to apply and were interviewed by the head of assessment for Center for Measurement Justice (CMJ). Scholars selected for the program were awarded a $500 stipend for their participation. There was intentionality to select scholars from racially diverse backgrounds. A total of 21 scholars were selected along with four math coaches and an on-site animator. Scholars were racially diverse with a majority of scholars self-identifying as African American or Black (13 students or 62%), 5 scholars self-identifying as South Asian (Indian), 2 scholars self-identifying as Asian, and one scholar self-identifying as white. Scholars self-identified their mathematical background and comfort level with mathematics. Answers ranged from Algebra 1, Geometry, and Algebra 2 for mathematics background and most students expressed positive feelings for mathematics (60% reported positive feelings, 30% reported negative feelings towards mathematics and 10% reported neutral feelings towards mathematics). Four coaches were hired to take part in the institute and three of them had prior experience in K-12 mathematics and the other coach had experience in graduate student teaching. Prior to the institute math coaches were invited to a professional development with Dr. Kari Kokka to discuss and practice teaching using the three dimensions of social justice mathematics (Kokka, 2022). See summer institute timeline (Table S17) for more information.

The storyboard topics and results were driven by scholars, with math coaches and other facilitators offering support as needed, in hopes to foster student autonomy and ownership. There were four person teams (4–6 scholars), and each team had a math coach as their lead support. To encourage and motivate scholars, “prizes” were awarded and those who fully completed their daily journals, attended every session, and showcased excellent participation (scholars were awarded an extra $100 bonus and scholars that halfway met these expectations were awarded an extra $50; this was decided amongst the math coaches and other helping staff). The final topics chosen by the scholars included: hair discrimination (HD), wage gap (WG), racial discrimination (RD), and health discrimination (HeD). Due to the large amount of data collected, we chose to focus on two of the four groups, selecting them based on which had the most complete data in terms of journals and interviews. With this criteria in mind we decided to fully investigate the hair discrimination (HD) group with scholars Brooklyn (Black/F), Nina (Black/F), Isaach (Black/Haitian/M), and LeBron (Black/M), Danai (African American/F), math coach Meryl (white) and the wage gap (WG) group with scholars Mark (Black/M), Sarmin (South Asian (Indian)/M), Alia (South Asian (Indian)/F), Kiara (South Asian (Indian)/F), Edward (South Asian (Indian)/F), Lidya (Black/Afro-Latinx/M), math coach Simon (Korean) [scholars were invited to choose their pseudonyms and that is the names we chose to include].

3.3. Methodology

We conducted an ethnographic study collecting interviews from young scholars and math coaches, daily journal entries from both scholars and math coaches, and observations between scholars and math coaches participating in the program. The study was conducted with attention to the following considerations, which shaped our research approach, data collection, and methods: the importance of researcher positionality, the role of race and racism in shaping educational experiences, and the significance of centering the voices and narratives of scholars and communities often marginalized in academic spaces. These guiding principles are explained more thoroughly below.

3.3.1. Researcher Positionality

Milner (2007) describes how messy and interconnected one’s social identities can be, especially when it comes to research. To uphold tenets of CRT we acknowledge our positionalities in ways that attempt to avoid performative tendencies and are instead self-reflexive and dynamic throughout the research process (Diemer et al., 2024; Milner, 2007). My, first author, positionality is deeply tied with my culture, ethnicity, race, and experiences in life. I am a cisgender, Guatemalan/Chicana, first generation college student, western trained researcher, who was born in LA county and was primarily raised in Bakersfield, California. I am a daughter of working-class immigrants who moved to the United States for a better life. Staying in my hometown for college, I noticed that there were racial and gender inequities within my major (mathematics/statistics). This gap widened coming graduate school for my masters in statistics because of its predominantly white population of students. Reflecting on my experiences as an undergraduate and graduate STEM student, I often felt uncomfortable in classrooms and struggled to feel secure in my learning journey or confident in seeking help. The classroom environments I encountered hindered my ability to develop a positive relationship with asking questions, embracing failure as part of the learning process, and trust in my own intellectual abilities. I found that was an important aspect of learning that I wanted to investigate both as a student and instructor. In my graduate classes I continued to refine and develop my understanding of intellectual safety through literature and class projects. These investigations, combined with my experiences as a graduate student instructor, motivated me to explore how students experience intellectual safety in learning environments. I, second author, also arrive at this work with multiple intersecting identities. I grew up poor, Black, and female in the deep American south. None of these identities (and their corresponding ways of knowing and understanding) were respected or valued in the public schools I attended. And, if I felt no intellectual safety in high school, I felt even less in college. Still, I believe that educational spaces can be a source of hope and liberation. At least that is the dream I carry with me into this work as I search for a better path forward—one that constantly reminds poor, Black girls in the deep American south just how amazing they have always been.

Using Milner (2007) nonlinear framework of critical reflection (seen, unseen, unforeseen), we planned to use the reflective questions to ground racialized and cultural understandings while centering the community. For example, with respect to conducting research within a short-term summer program we acknowledged the: (seen) attempting to avoid tokenization and model minority and seeing students as unique individuals, (unseen) reflecting on the power dynamics at play between the researcher and participants by being explicit on the research design and inviting scholars and coaches to take part in the research, (unforeseen) being intentional in the writing of this paper for whose experiences are told and validated. In order to combat white supremacist culture, integrating questions that help understand the insider and outsider perspective, uncover unconscious bias, and reflect understandings of the role of the researcher is necessary (Aldana & Richards-Schuster, 2021). We understand that our own realities and experiences bring a unique understanding of the phenomenon we investigate, and we want to explicitly consider that our evidential truth is not the same as the participants. Additionally, we take up Madison’s (2005) notion of having an activism stance when reflecting and positioning oneself as an ethnographic researcher. This can include being an advocate for our participants in the study, intervening in harmful situations that are hegemonic, racially, or politically motivated, and deliberate intention of transformative change through action oriented or organizational efforts (C. H. F. Davis et al., 2019).

3.3.2. Attention to Salience of Racism

According to Bonilla-Silva (2015), the era of “new racism” has established itself through various ways including the reproduction of racial privilege through continued separate and unequal opportunities in social areas of life (p. 1361). Paying attention to how “new racism” has permeated academic settings is one of the first steps to combat against discriminatory and oppressive practices. By acknowledging and reflecting that there is a contextual nature to race, and that systemic barriers and structures are in place, shifts blame from the individual to larger academia to allow for us researchers to mitigate and promote contributions to racial injustice (Milner, 2007). Additionally, not paying attention to how methodology and methods are racially considered perpetuates white dominant norms and white supremacy where blame is then placed on communities of color (Tachine et al., 2016).

One of the goals of the collaborative research project scholars worked on was that it focused on systemic injustices to counter white traditional curriculum in an environment that has supportive instructors. Scholars asked to reflect on their identity and cultural connectedness during the institute. As the researchers, we decided to not shy away from racial discourses and instead uplifted this whenever interacting with scholars and their projects. Chapman (2013) argues that in the absence of racially diverse content and a conversation of power and privilege, white privilege and traditional curriculum is perpetuated. The goal was to have math coaches reflect on their practices in ways where they thought of ways to uplift their scholars’ cultural, racial, and experiential knowledge in the curriculum (Chapman, 2013; DeCuir & Dixson, 2004).

3.3.3. Counter-Storytelling

Many scholars of color have gone back to the “field work” to explicitly attend to voices that are the same as them because they were “tired of bending their ears to the master’s book talk” (Ladson-Billings, 2000, p. 270). We know that it has been common practice to hear stories of white students and white ideologies/practices in academia. Using counter-storytelling in research is one way to position students who have long been seen and written as abnormal, disadvantaged, or in cultural deprivation (Ladson-Billings, 2000; Stinson, 2008). To take on a critical role in ethnographic research as deep and relational conversations are required (Madison, 2005). The ways in which the stories we hear collide, come together, bridge, and transgress are important observations to make in understanding the relationship between ourselves and the participants (Madison, 2005). During the summer institute the first author-built relationships with scholars by conversing about topics outside and inside academics, having lunch with scholars, mesmerizing student names, consciously reflecting on scholars’ interests to build future conversations and sharing personal connections with their interests.

We recognize that academic research has historically centered the experiences and ideologies of white students, often elevating their realities in both research design and methodology. Acknowledging how class, gender, sex, race, religion, sexuality and other social identities have complexly influenced the lives of the students we are collaborating with, challenged us to reflect on the goals and voices of those students. Scholars at the institute were made aware of the research goals and tasks and were invited to participate in the research design which included: which questions were asked in the scholar journals, who wanted to participate in interviews (and how), which groups were being observed, and how to help as collaborative researchers. For example, towards the end of the program we noticed that a particular scholar was less engaged with the storyboard creation and more interested in the work of actively researching the institute. This scholar became a researcher in training by placing the cameras in the room and recording interesting interactions. They also collected field notes during the last couple days of the institute. This scholar expressed that this was how they would have liked to participate in the institute. Another example included inviting the math coaches to create journal questions for both the scholars and themselves.

3.4. Data Collection and Analysis

We outline our methods and explain how our conceptualization of intellectual safety emphasizes the ways students’ sociocultural experiences shape their sense of belonging in academic spaces. This includes our data collection process and thematic analysis, which drew on a reconceptualized coding framework while also allowing for emergent codes to surface as we analyzed scholars’ responses.

In terms of the data collected from the institute, this included daily journals from scholars and coaches, field notes, coach-scholar interactions collected from videos and field notes, artifacts collected by the institute (scholar notes, google doc notes, written concept maps), reflective memos, and interviews with both coaches and scholars (see Student Journal Question Bank Figure S1 and Student Interview Protocol Figure S2 in the Supplementary Files). In regard to this paper, journal entries from scholars and math coaches and semi-structured interviews of scholars and math coaches were analyzed using the reconceptualized framework on intellectual safety.

An early version of the codebook was created following thematic deductive analysis using priori codes developed from previous literature and the research questions. The first author coded 50% of the scholar journals using priori codes developed, using the attributes of our approach to intellectual safety, and developed additional codes as new themes emerged from the data. New codes emerged as the first author began initial rounds of coding and responded to themes arising directly from scholars’ experiences. For instance, the idea of being listened to appeared frequently across scholar groups during the first round of coding. Since it did not align neatly with the existing codes, the first author determined it warranted its own distinct code. After this another 25% of the scholar journals were coded along with 50% of the scholar interviews to refine and develop an understanding of the codes. Throughout this process definitions of the codes were refined along with indicators to develop a coding frame for future coders to use. The coding process prioritized units of meaning over exact wording, acknowledging the nuanced and complex nature of scholars’ experiences with intellectual safety. During initial rounds of coding, there were instances where multiple codes could appropriately describe a single instance. Rather than forcing a choice between them, we allowed for code co-occurrence to better reflect the layered and intersectional dimensions of the data. This approach was intentional, recognizing that scholars’ experiences do not always fit neatly into distinct categories. After this the remaining data was coded by the first author. After completing all coding, the first author conducted a thematic analysis to identify trends and patterns in the data in relation to the research questions. We examined both within-group and across-group frequencies to detect recurring themes. As new codes emerged from the data, we revised our initial framework to better reflect how BIPOC students were experiencing intellectual safety. This revision was essential to the refinement of the framework, as it allowed it to be grounded in students’ lived experiences while remaining in conversation with existing literature and the original attributes to intellectual safety. We will discuss these emergent codes in the findings section.

Evaluating intercoder reliability (ICR) is widely recognized as a best practice in qualitative research. This iterative process enhances the study by improving the reliability and validity of the findings, minimizing individual biases, and fostering richer interpretations of the data. The inclusion of multiple coders also encourages reflexivity, allowing to critically assess and refine the coding framework for greater clarity, depth, and rigor. To ensure intercoder consistency and enhance the reliability, rigor, and transparency of the analysis (O’Connor & Joffe, 2020), three additional coders were hired on to conduct a second and third pass at the coding the data [which would include: scholar journals, math coach journals, and both scholar and math coach interviews]. Materials were shared via google drive with scholars’ and instructors’ names anonymized. Secondary coders were trained by the first coder on the coding frame by coming together and sample coding scholar data that would not be used in the analysis. Once there was an agreement and understanding on the codes, the three coders independently coded 25% of the scholar journals. After this first round of coding, primary and secondary coders met to discuss consistency of the codes and see if any clarifications or revisions need to be made (O’Connor & Joffe, 2020).

The primary coder generated a spreadsheet where all instances of the codes come up from all coders. After this, the primary coder flagged instances that did not appear across all coders for discussion to decide whether to keep the instance within the code or not. From this point, the secondary coders went through and commented on whether they agree or disagree with the codes they originally had not considered for that code. For every instance where at least one coder disagreed, there was a discussion and only codes where every coder agreed were kept. For example, in one instance there was one coder that agreed that a certain quote showcased social and emotional bridging. However, after reflecting on the quote, all coders disagreed with the quote remaining in that code and so it was removed. If coders did not agree due to code’s definition or language of indicators, refinement of the codes was performed. For example, with one of the emerging codes “clarity/transparency”, there was disagreement on the language being vague enough to include scholars’ personal goals. However, adapting to Schrader’s (2004) work, this should align more with the tasks scholars are expected to complete or do from the coach and so this code was changed for clarity. For instances where coders could not come to agreement on whether a specific excerpt met the criteria for a code, we made the intentional decision not to assign meaning to that instance. This decision was not taken lightly; it reflected our commitment to ensure that only data segments that clearly aligned with our coding indicators were included in building out our framework. These decisions were grounded in discussion where coders offered interpretations, justifications, and counterpoints. When an agreement between the coders could not be reached through refinement of the language of the code indicators or discussion, we interpreted that as exceeding the code’s boundaries and further change would compromise the code’s conceptual clarity. In this way, not assigning meaning to the instance was a safeguard against distorting the analytical categories we had defined and our attempt to minimize bias. At the same time, code co-occurrence was allowed and treated as methodologically meaningful rather than as a sign of redundancy. Co-occurrence often revealed the layered nature of scholars’ experiences especially when we want to consider a framework that seeks to reconceptualize intellectual safety through the lenses of race, power, and identity. Rather than indicating poorly defined codes, frequent co-occurrences often showed how themes like “being listened to” and “communal orientation” operated in tandem within the same interaction. These intersections were flagged and explored during analysis to represent the sociocultural complexity for reconceptualizing intellectual safety. Together, these methodological choices reflect our broader commitment to honoring the richness of BIPOC scholars’ voices.

Once all coders agreed on which instances belonged in each code, the three secondary coders coded the remainder of the data which included 50% of scholar journals, 100% of scholar interviews, and 50% of math coach interviews and journals. Because of the large dataset at least three coders worked on each set of data. Between independent coding, there were weekly check-ins for any questions or comments that may arise while coding. During one of the check-in meetings, language for definitions and indicators for instructor related data were refined to include coach’s perceptions. An additional code was added for scholars labeled clarity/transparency. The new code helped distinguish between instructor’s perceptions on providing clarity/transparency for their scholars and how the institute/program provided support/clarity for guiding scholars through their storyboards. After the independent coding was completed, the same process as above was repeated with only instances agreed by all parties being kept. Moving forward, once all the coding was completed, we analyzed any trends or themes in the data and its connections to the research questions. We used a thematic analysis to analyze the data and its trends. We looked across groups and within group frequencies to see if there were any potential patterns. Given that co-occurrence was permitted, we paid close attention to instances where multiple coders assigned overlapping codes. When two or more coders agreed on one code but diverged on the application of a second, we engaged in team discussions to determine whether the excerpt genuinely reflected both codes or required refinement. In some cases, this led to clearer definitions and indicators for each code; in others, excerpts were retained under only the agreed-upon code. All co-occurring codes were flagged for further analysis, and we examined occurrences to assess whether they pointed to meaningful intersections in the data or suggested a need for code consolidation. Ultimately, this process helped us ensure that our codes remained analytically useful while capturing the complexity of BIPOC scholars’ experiences.

4. What Emerged from the Work: Insights from Scholar’s Experiences

We examine how an anti-racist, scenario-based mathematics institute influenced secondary students’ sense of intellectual safety. By focusing on a space with key features aimed at fostering intellectual safety, we were able to explore how scholars engaged with those features and what emerged beyond our initial expectations. We argue that taking a more critical and sociopolitical approach to investigating intellectual safety for BIPOC students can create opportunities to challenge deficit narratives and provide actionable insights into fostering environments that affirm their identities, experiences, and ways of knowing.

As we started coding the data, we found an increasing necessity for these new emergent aspects: pride/confidence, clarity/transparency, and being listened to. We found that there were instances that we interpreted as important to establishing an intellectually safe environment that did not neatly fit into any of the current codes. For instance, pride/confidence was especially prominent across both groups’ daily journals with scholars recognizing their own strengths (like time keeping, leadership, research, and coding). Some scholars shared how peers helped refine their ideas without erasing their original contributions, highlighting intellectual collaboration. Edward reflected on this experience saying:

I think that that really ties into how my ideas right? I felt valued like in many, many different ways, really like whether it be like when we were like…even like, before we started working in groups, we were advocating for our topics. I felt like I had a voice, and I think that it felt like I was like, my ideas were actually important. And I was like able TO really, you know, add voice my opinion, and also like in groups who allows it play to our strengths, and also like after, like groups, when we’re like presenting, or even just looking at other groups, work like helping them in a sense, right? All of those aspects right? I felt empowered in unique ways… I think I was also able to like, suggest, like, Hey, do you think we should do this on the website. Should we do this? And like I was, I think that I was able to like contribute, because, like… like, I was like playing to my strengths, but also like I think that… you know, like, with the website, too, like, since we were kind of splitting it up into parts. And we were working on it at different times. I got a chance to do like couple of like the things, and I think that like helped. And it felt like because of the you know, roles being distributed right and kind of us like identifying those roles at the beginning of the project. We were all kind of able to have like somewhat of like an equal, contribute contribution to the project.

Edward’s comment in this instance showcased that they were proud of the opportunities and the work that they were doing. This was similar across other scholars’ comments where they had felt a sense of accomplishment, growth, or self-validation (which we argued was qualitatively different than their individual strengths being noticed by others. We argue similarly for the codes Clarity/Transparency and Listened To. Across both HD (hair discrimination) and WG (wage gap) groups, scholars repeatedly emphasized that they felt listened to, especially in group settings where their peers asked for input or when scholars felt “seen” as whole people, especially when their personal experiences were acknowledged. For instance, Brooklyn mentions, “when I’m working with my group, I felt like they were actually listening to me and that they like my ideas.” while Isaach stated that “The one time I felt cared about or heard was when we went over the class work and I spoke about one of my experiences which felt good.” There were other instances of scholars feeling like their voices were being heard and respected in this space, so much so that we developed a code for this. From our data analysis, of the 515 scholar journal entries, about 1/4 of the codes were labeled as one of the emergent codes not initially presented in our initial conceptualization. Based on these codes, we realized the need to reorganize them to better reflect the emerging themes in scholars’ experiences. This included clarifying the contextual relationship between scholars and teachers, distinguishing engagement related codes tied to scholars’ intrinsic motivation and the factors that best support it.

4.1. Reconceptualization: A Justice-Oriented Approach to Intellectual Safety

Using Schrader’s (2004) and Call’s (2007) conceptual framework as a basis along with our hypothesized components, analysis of scholar journals and interviews highlighted three key dimensions of intellectual safety: a sense of belonging, empowered engagement, and liberatory learning experiences. Building on the ways in which the three themes emerged across the data, we reconceptualized intellectual safety to account for the experiences of BIPOC scholars, primarily with respect to Black and South Asian (Indian) scholars. We build from the definitions brought on by Schrader (2004) and Call (2007), and argue that intellectual safety must also include the validation of students’ social identities, the disruption of power imbalances due to systemic structures, and the cultivation of agency and exploration in learning spaces. Prior research on intellectual safety has not fully accounted for the experience of BIPOC students. For example, when speaking towards the contextual relationship between students and teachers and the importance of nondiscrimination or attitudes of the professor, this did not consider the sociopolitical realities of students nor acknowledge the racist meritocracy that is centered in academic spaces. Prior definitions also did not consider the importance of sociocultural relationships in the classroom and how systems contribute to racial inequalities that shape students’ opportunities or experiences. There needs to be a validation of these students’ experiences, identities, and cultures in ways that are explicit, transparent, and built from the relationships of care and practice (Lightfoot, 1986; Louie, 2018). We justify this by borrowing Lightfoot’s (1986) themes of empowerment with respect to respectful and trusting relationships, commitment to diversity and inclusivity, encouragement of critical inquiry and dialogue, student autonomy, teacher involvement in decision making, and collaborative school culture (Lightfoot, 1986). For cultivating a sense of belonging, we build on Schrader (2004) and Call’s (2007) concept of belonging by examining it as a multidimensional space that includes a teacher’s pedagogical moves and supports, fostering an environment where all students, particularly BIPOC students, experience a sense of connectedness across various relationships (Matthews et al., 2021). We also argue for including mathematics for social justice where students can see themselves as being part of the solution to an injustice as a place for intellectual safety to counter the trivialization of students’ unique identities and ways of being (Gutstein, 2003). We reconstruct Schrader (2004) and Call’s (2007) concept of self to emphasize a sociocultural connection to identity, linking it to a deeper sociopolitical consciousness that attends to BIPOC students’ identity, purpose, and learning (Gray et al., 2018; Nasir & McKinney de Royston, 2013).

In revising the framework, we want to acknowledge the deeply interconnected nature of a justice-oriented approach to intellectual safety. Rather than presenting intellectual safety as the goal or foundation we are shifting toward a more bidirectional and interdependent framework. In this revised view, none of these elements stand entirely above or below the others; instead, they work together, each one essential to cultivating an intellectually safe environment. For example, you can certainly have a sense of belonging or moments of empowered engagement on their own. But when all three elements, belonging, empowered engagement, and liberatory learning experiences, are present and in conversation with each other, that is when we argue that intellectual safety becomes most fully realized. In this way, we are not treating intellectual safety as a prerequisite or an endpoint, but rather as something that exists through and because of these other conditions. The revised framework will reflect this shift visually and conceptually: it is not a hierarchy, but a system of mutual reinforcement where the presence or absence of one part meaningfully affects the whole, see Figure 4.

All the overarching themes work together within the sociopolitical context. Additionally, they also must work together to fully cultivate a justice-oriented approach to an intellectually safe environment. We argue to fully realize an intellectually safe environment, these aspects work together and build from each other. Liberatory learning experiences uplift the importance of connecting students’ interests and intrinsic motivations for students to explore content that resonates with their personhood that challenges power structures in ways that give ownership to the students. In turn, when students feel supported and motivated they share their perspectives and actively engage (Lightfoot, 1986; Moore et al., 2020). For example, critical inquiry and cultural distinctiveness are tied together in providing scholars with a sense of agency and meaningful participation. Oakes and Lipton (2003) agree that critically engaging with diverse and culturally relevant material in a proper way can enhance student’s view of themselves and ignite “a students’ passion for justice and demands for justification” (p. 117). Not only this, but having the ability to scaffold student’s thinking and build on what students already know has been proven to undo misconceptions teachers may have about students and build motivation in the classroom (Bricker et al., 2014; Oakes & Lipton, 2003).

To identify patterns in scholars’ experiences, we looked to see if any themes came across and within groups, paying attention to how frequently each code appeared in relation to the three central themes: Empowered Engagement, Belonging, and Liberatory Learning. Across both groups sense of belonging showed up the most in the scholar’s journal entries while the theme that showed up the least for the groups varied (for the HD (hair discrimination) group, empowered engagement showed up the least and for the WG (wage gap) group liberatory learning experiences showed up least in their journal entries).

When looking across the three themes, we saw clear patterns in how scholars described their experiences. The most prevalent codes within Empowered Engagement across both groups were citizenship (HD: 23%, WG: 45%) and listened to (HD: 30%, WG: 20%). For Liberatory Learning Experiences, pride/confidence was an emergent code that prominently surfaced for both the hair discrimination group (39% of responses within liberatory learning experiences were under this code) and wage gap group (38% of responses within liberatory learning experiences were under this code). The second most reported code was joy across both groups (31% for hair discrimination group, 37% for wage gap group). Across both groups, scholars expressed a sense of accomplishment and ownership over their projects. In Sense of Belonging, scholars across both groups talked most about feeling supported and emotionally connected, especially through empathy and shared experiences. Of the codes, most prominently came from interpersonal supports with the HD (hair discrimination) group having 47% of the responses within social & emotional bridging and Decentering Teacher Authority and WG (wage gap) group having 40% of their responses within empathetic awareness & support. Some codes, like “being listened to,” were especially important across both groups, showing just how central it was to feeling intellectually safe. Across scholars’ daily journals, approximately 13% of the belonging-related codes co-occurred with codes from LLE and EE, representing the highest rates of overlap. Additionally, co-occurrence between LLE and EE codes were less frequent, showing up in less than 2.5% of the coded instances. As noted earlier, allowing for code co-occurrence added complexity to our interpretations and prevented us from forcing instances into a singular code. While cooccurrences were relatively uncommon, their presence was still important to acknowledge.

4.2. Evidence of Belonging, Empowerment Engagement, and Liberatory Learning Experiences: Scholar’s Experiences of Intellectual Safety

The following sections unpack these findings in detail demonstrating how scholars experienced intellectual safety within the institute’s unique learning environment. Scholars from racially and socially marginalized backgrounds experienced intellectual safety through a strong sense of belonging, liberatory learning experiences, and empowered engagement. For belonging, scholars felt a strong connection to their peers and mentors, particularly because they shared racial or cultural backgrounds and interests in social justice. This connection fostered emotional safety, where scholars felt they could be authentic and contribute meaningfully without fear of judgment. For liberatory learning experiences, scholars described a sense of freedom and joy in exploring their identities through their research projects, especially when it came to real-life injustices that mattered to them. For empowered engagement, scholars were encouraged to take intellectual risks, engage in critical thinking and often leaned on collaborative problem-solving and community support. They felt seen and supported by peers and mentors with shared experiences, took pride in their identities and contributions, and engaged deeply with social justice topics. While some faced challenges with timelines or complex content, many described joy, motivation, and a sense of agency in doing their work at the antiracist summer institute.

4.2.1. Sense of Belonging

From our observations there was a strong emphasis on collaboration from the beginning from math coaches. This sets the stage for the future days at the institute. As time progressed scholars from day one, both HD (hair discrimination) and WG (wage gap) groups were showcasing many different aspects of belonging in this space with most instances reported in interpersonal supports. For social and emotional bridging, many scholars reported that they felt connected with their peers because of the environment they were in and were excited to collaborate with them. For instance, in interviews and journal entries, scholars expressed feeling “reassured” in the space because the people at the institute “looked like” them or shared an interest in discussing social justice issues that directly aligned with their racial backgrounds. They appreciated being able to talk with others who could “relate” to their experiences or “see” them as they truly are. For example, Lebron mentions, “I think having friends helps. So come up like, I’ll be working and they would come up like, “Wow, this is pretty good, Lebron.” People saying I did a good job, Giving feedback, and not just negative things…” showcasing that he felt connected to his peers (social & emotional bridging). Scholars shared similar sentiments with respect to the math coaches. They also expressed enjoying hearing their peers’ thoughts and contributions with respect to the social injustices investigated as well as making friends with everyone at the institute. This contributed to scholars feeling a positive classroom environment where social boundaries were deconstructed. For empathetic awareness & support, scholars expressed feelings of being able to share ideas during the institute because others valued their identities and ideas as well as feelings of having a meaningful role in the space. From our observations, one of the goals of the institute was to make sure that scholars had agency in their project and were guided in creating roles or ‘super’ prowess for themselves. Many scholars from across the groups reported feeling like they had a role or referencing their role or ‘superpower’ in positive and collaborative ways that showed a shared responsibility within groups. This added to scholars’ feelings of belonging in this space by showcasing that scholars’ individual strengths and resource needs were being valued. For example, scholars felt that “everyone had their own roles” or that “everyone got to “play an equal role”. This made scholars feel that everyone “worked collaboratively” or could “contribute without feeling like only some people were doing the work.” This shared responsibility also contributed to belonging through communal orientation. We also note that decentering teacher authority was a code that we saw often in both HD (hair discrimination) and WG (wage gap) groups. Many scholars felt that they were able to offer suggestions to instruction and that they had agency in the institute. For example, Isaach offered suggestions to the improvement of his learning experience,

“using physical and just like drawing it out. Especially with,… which can really help my generation, like my age group. Since we, we’ve gotten more where we need like that physical thing instead of like looking at it. Like we need to be more engaged like movement? Wise?”

For the HD (hair discrimination) group, scholars experienced many instances of safety to be wrong, particularly the younger scholars in the group. For instance in an interview with Brooklyn and Nina, both mention that their peers supported them in their learning process:

Nina: [About feeling comfortable making mistakes] Because like, I don’t think anyone here is like gonna be like mean or anything. Most of these people here are like older than me, and they’re more mature. So I don’t think they’ll be like that. They never been like that towards me or anything, so.

Brooklyn: Yeah, they’d like, I’m also like one of the younger ones, but they, since they’re older they’re nice. And then also because when we mess up we just like we just go back and fix it, and then we don’t make a big deal out of it, because other people would. But from what I’ve seen in a group, nobody really would do that.

For more student examples of sense of belonging instances, see the Supplementary Materials Tables S1–S7.

4.2.2. Liberatory Learning Experiences

We describe liberatory learning experiences as opportunities where scholars feel a deep connection, purpose, and agency in academic spaces to challenge systems of power. This focus on intellectual safety aligns with Schrader’s (2004) goals for cultivating learning spaces that attend to scholars’ self-awareness but also promotes learning that is relevant to their experiences, aspirations, and identities while taking on a critical consciousness lens. We situate scholars as active producers of knowledge where things like exploration, joy, and accomplishment are prioritized over passively taking in information. We argue that by emphasizing curiosity-driven learning and encouraging scholars to take pride in their identities and ways of knowing, learning moves from surface level motivations to liberation. They mentioned feeling excited to “show off” their projects or looking forward to presenting their work to each other. One scholar, LeBron, mentioned that seeing how his project looked similar in quality to one of last year’s projects made him feel proud of his project. Many of the instances centered around scholars’ individual contributions highlighting how their efforts enhanced the overall project with their peers. For example, scholars expressed a sense of joy and interest in the topics they were researching along with the other social aspects of the institute. Scholars were excited to get snacks, roam the university’s campus during lunch, and get paid for doing something that interests them.

When it came to exploration, scholars shared positive sentiments of experiencing “creative freedom”, feeling motivated to learn more about social injustices they were initially not as invested in, and taking part in different roles. For example, Nina mentions:

“My topic is hair discrimination at first I didn’t know much but the more and more I researched about it I really knew what I was doing and I was able to elaborate more on what I was thinking more to my peers.”

Additionally, Alia recounts:

“One thing I already knew a little bit about and learned more about today in terms of equity is like the differences in opinion on equality and equity coexisting. Equality is important when discussing rights, however, equity is important when we are talking about resources provided to different people. I plan to use this in my research when researching racial wage gaps. I plan on diving into what would be more just, equality or equity in pay. I think that maybe equity is not always better than equality.”

In interviews with scholars and from instances in their journals, all scholars reported that they had felt that their culture and identity were affirmed in this space. Scholars contributed these feelings to most of the institute being people of color, connecting their projects to their lived experiences, normalizing that injustices occur, and being understanding about learning about these topics. For instance, Brooklyn stated that she feels valued in this space because people did not look at her differently because of her skin. Isaach reflects on his experience at a predominantly white school and how being in this space made him feel valued because “there’s other people out there that have the same personality, looks, and everything.” These feelings contributed to feeling a sense of positive cultural distinctiveness. Edward mentions, “I was thinking of asking my parents about their experience with the wage gap. I wanted to implement their knowledge and ideas into my project” when reflecting on how to connect her personal story to her social justice topic. Altogether, scholars encountered opportunities to connect their experiences, interests, identities, and cultural backgrounds to learning in this environment contributing to a portion of intellectual safety. For more student examples of liberatory learning experience instances, see the Supplementary Materials Tables S8–S11.

4.2.3. Empowered Engagement

We describe empowered engagement as a learning environment where scholars genuinely feel supported in their educational endeavor. We argue that this is achieved through clear communication of expectations and goals, and the necessary resources for success. Empowered engagement encourages scholars to take intellectual risks, feel empowered in navigating educational content, and explore unconventional problem-solving. In inviting intellectual risk, instructors encourage scholars to challenge their epistemological stances in safe ways where diverse perspectives are necessary for a deeper understanding. With this framing, we ensure that scholars’ voices, experiences, and perspectives are put at the forefront and validated to promote not only a safe but inclusive environment.

For citizenship, scholars centered the idea of wanting to communicate their findings to the broader community. For example, Edward said “I want to inform more because the more people who know about this topic, the more likely we are to make a change.” With respect to being listened to, scholars expressed being able to share their ideas in ways where they felt genuinely respected and heard in group discussions. Scholars mentioned that their group members or math coaches used their ideas and that is how they felt heard. For instance, scholars reported how “[the coaches] take it [their ideas] to heart” and that “[the coaches] are at least going to process my opinion.” Scholars had similar sentiments towards their peers stating that their peers will take their ideas and “use it in the project” or have them “considered by my group”. In these instances, scholars used language that indicates that they are being heard, listened to, or included in the space in ways that felt meaningful to them. Scholars expressed tensions with the expectations and timeline of the project which contributed to their feelings and perceptions of clarity/transparency. Many wished for more time to work in groups for the project with two scholars from the WG (wage gap) group (and one from the HD (hair discrimination) group) stating that working over the weekend/from home on their project was necessary for its timely completion. For example, Sarmin notes “[the timeline is] a little bit tight, we just need more work time. We are also going to be working over the weekend and after class today” while Kiara mentions that further guidance and clarification would be useful:

“I would like to make steady progress on our project by ensuring that all the content is organized on the slideshow by the end of the week. I feel like having “tasks” for the day so we know what we need to get done to stay on track can help me not forget about what things to include.”

The HD (hair discrimination) additionally felt that the math component of the project was the most difficult to incorporate in their group, with one particular scholar requesting a math expert be added to the institute and two scholars expressing initial concern over understanding high school standards as younger scholars. For example, Brooklyn notes:

“Like with the math problem, like I didn’t understand it because it’s high school level, and I’m going into high school. So I’m like, I don’t understand it. So like now they’re [group members] helping me understand it and all that”

Additionally, many of our scholars reported the change in perspective of the severity of the social injustices they investigated. They had moments of determination and reflection while exploring various aspects of the project (researching, writing, conceptualizing math problems, coming up with solutions to social justice issues, etc.). Four out of five scholars in the HD (hair discrimination) group noted not realizing sooner the severity of the issue and how antiblack school/work policies were embedded in everyday life. Nina expressed her motivations stemming from her discussions with her math coach: “When one of my coaches helped us and told us about a guy who was forced to cut his dreads for the school match and that really is what motivated me to do this project.” Nina shared a desire to work and explore her chosen systemic injustice and ended up feeling a sense of accomplishment in her work. From the WG (wage gap) group, we had four out of six scholars reporting similar sentiments with the research process and institute. For example Kiara reflects on providing actionable tasks to the broader audience:

“For the wage gap project, we outlined some broad solutions such as encouraging students to research wage inequality further, have students find their own examples of wage inequality and present their findings, challenge students to think critically about the causes and effects of wage inequality, and inspire students to advocate for fair pay and equal opportunities for all. In addition to this, our group decided to make a website as we thought it would be accessible to everyone on the internet.”

Within empowered engagement, the codes that showed up the least were that of intellectual risk and critical inquiry. With intellectual risk we make the distinction that scholars need to actively engage in risk taking where they challenge their preconceived ideas, push beyond their current knowledge, and work through ambiguous tasks through trial-and-error learning. In one instance, Brooklyn showcased this when she talked about not understanding a math problem but continuing to work on it with the help of her peers, pushing past what she believed were math concepts that were more advanced than where she was currently. With critical inquiry, scholars expressed this through referencing diverse perspectives with contrasting ideas and synthesizing what they have learned. For example, in the WG (wage gap) group, Mark and Alia mentioned reaching out to people they know (parents or other coaches) to gain a better understanding of how wage gaps impact the world through people they are comfortable around. Within this theme, scholars challenged the power structures in the environment while validating their knowledge in tangible ways. For more student examples of empowered engagement, see the Supplementary Materials Tables S12–S16.

4.3. Scholar’s Experiences on the Interdependence of Belonging, Empowered Engagement, and Liberatory Learning Expereinces

All the themes (belonging, empowered engagement, liberatory learning experiences) work together within sociopolitical contexts to cultivate a justice oriented approach to an intellectually safe environment, and none should be treated as “completed” on their own. We argue that intellectual safety for BIPOC students emerges through the interdependence of these components: where student’s lived experiences shape how they may encounter each dimension simultaneously. In this section, we illustrate how these components intersect in students’ experiences and how their interplay shapes a justice-oriented approach to intellectual safety.

In an interview with Isaach, he spoke of his learning experience: how he felt connected to the issues addressed, and how activism could be enacted which showcased the gap in the world’s understanding while being reflective on social injustices; connecting empowered engagement and liberatory learning experiences, see Figure 5. In other instances, Isaach expressed enjoyment towards exploring the different stereotypes within the “hair community and how those can play a role towards someone” and finding “new ideas to the table and finding new and fun interesting ways to work and make the group project more interesting” further emphasizing feeling motivated to share their perspectives in agentic ways. We also wanted to showcase that there was a negative instance that showed up within our data including a scholar not experiencing feeling listened to. In Edward’s experience, she recounted experiencing the feeling that her group listening to her suggestions would have benefited the project.

In connecting both sense of belonging and liberatory learning experiences, we argue that both themes aim to promote an inclusive environment that affirms scholars’ unique lives and identities (see Figure 6). Gutierrez (2013) describes the importance of both students and teachers feeling a sense of belonging in the classroom in order for identity to serve as a learning tool. Sense of belonging has been shown to have a positive influence on students’ identities (Nasir & McKinney de Royston, 2013) and motivations in students’ learning in STEM spaces (Dost, 2024). Nasir and McKinney de Royston (2013) emphasize the importance of considering students’ race, identity, and power relations within academic settings and that mathematical reasoning is deeply embedded in cultural practices. Gray et al. (2018), notes that experiencing cultural distinctiveness along with positive achievement emotions like joy and confidence contribute to students’ sense of belonging. Scholars Alia and Edward both shared an excitement for learning in what ways mathematics plays a role in informing the general public about wage gap inequality see Figure 6. Whereas other scholars expressed uncertainties on how exactly this could be done but still felt like they belonged in this space as many felt they were able to safely be wrong in this space. Another example of a cooccurrence between themes can be portrayed with an excerpt from Isaach who showcased experiencing joy and feeling that they have a role in this space: “I felt happy and felt included in group work which is nice because sometimes I am quiet and tend to work on my own”. This further illustrates that the two themes can be intertwined in ways to explain more of the complexity of the lived experiences of students.

When students feel a sense of belonging in academic spaces, they are more likely to engage with purpose and when students feel that they can engage meaningfully and critically this reinforces a sense of belonging as they are more invested in the academic space, see Figure 7 (Maloney & Matthews, 2020; Matthews et al., 2021). Additionally, Conway Iv et al. (2023) emphasize the importance of integrating challenging but authentic math problems to allow for student critique, questions and reflection to promote a socially just mathematics learning environment. It is important to note that because the institute required students to investigate a social injustice, this environment fostered a sense of empowerment by supporting students in feeling like they could be part of the solution to the problem (Gutstein, 2003). The goals of sense of belonging and empowered engagement in fostering intellectual safety are to support collective efficacy, a sense of agency, sociopolitical consciousness, and critique in ways that allow students to participate safely individually and together. For instance, with Kiara and Lebron, they had felt that their ideas were not only valued in this space by their peers but also praised the reciprocative nature of sharing ideas with one another to create something of value together.

5. Discussion

This study offers insight into how intellectual safety can manifest in a learning environment intentionally designed to support BIPOC youth in exploring mathematics through anti-racist and agentic approaches. In the previous sections, we outlined our reconceptualization of a justice-oriented approach to intellectual safety and highlighted how scholars in a purposefully designed summer institute experienced this environment. We now turn to key challenges, implications, and future directions, recognizing that while this work offers important insights, its application may vary across different educational contexts.

5.1. Implications and Recommendations

While the context, an anti-racist summer mathematics institute, was unique in its structure and purpose, the findings offer valuable takeaways for classroom educators: (1) recognizing/understanding what intellectual safety actually looks like in practice; (2) the importance of affirming students’ identities; (3) valuing and planning for diverse entry points into learning; (4) criticality of incorporating their lived experiences. In the following text, we describe the implications in further detail.

First, intellectual safety is not a fixed condition but a dynamic, culturally embedded experience shaped by students’ racial, social, and cultural identities. For the students in this study, intellectual safety was deeply connected to: joy, pride, exploration, and being listened to. Which are often elements that are left out in traditional mathematics spaces. Educators can begin by asking, “What does intellectual safety look like for my students? How can empowered engagement, belonging, and liberatory learning experiences become central components to creating an intellectually safe space?”

Second, this study reinforces the importance of affirming students’ cultural identities and lived experiences. Rather than treating students as passive recipients of knowledge, we argue that students should be positioned as knowledge producers. When students engage with material that is relevant to their lives and communities, and are given the opportunity to lead, they are more likely to take academic risks and experience deeper intellectual engagement. This affirms research suggesting that culturally responsive content can lead to positive academic outcomes when students see the content as meaningful to their lives.

Third, not all scholars came in with the same education and at least one scholar in the program found the materials difficult to read. While some scholars reported having positive feelings toward mathematics, this was not universally the case across this group. In interviews, a few scholars shared that they found aspects of mathematics challenging. We identified this as an important area to consider for future iterations of the program. Peer support played a key role in helping scholars who were struggling. What mattered most was not whether scholars initially felt confident in mathematics, but that they had a space where they could acknowledge their struggles and still feel safe enough to engage and explore mathematical ideas. That said, we recognize that cultivating such spaces presents real challenges, especially in traditional K–12 classrooms. This program set out to treat students as researchers, where they had to read complex materials and design scenario-based math assessments. If this challenge emerged in a well-supported/small-group setting, it raises important questions for the challenges faced by public school educators who work with larger and more diverse populations. Educators aiming to build intellectually safe environments should anticipate and plan for students coming in with uneven entry points. Scaffolding, academic and emotional, must be embedded from the start. For instance, coaches in the program had to adjust their facilitation when they noticed scholars struggling. That being said, having flexible strategies and supports prepared ahead of time (rather than relying on in-the-moment improvisation) is critical and was noted by the coaches at the instate to be valuable. One practical suggestion is to identify team leads within student groups whose role is to ensure every team member’s strengths are leveraged and that support is offered in ways that affirm students’ capabilities. Educators can also consider asset-based approaches to grouping and collaboration that help build confidence and community.

Finally, this study underscores that when students are invited to engage with content that reflects their lives, while being treated as capable thinkers, they respond with higher forms of motivation and engagement. For classroom educators, this means intentionally leveraging sociopolitical issues relevant to students’ communities as part of instruction. Embedding real-world, justice-oriented questions into the curriculum can help foster a justice-oriented approach to intellectual safety and students’ sense of agency simultaneously.

5.2. Limitations and Future Work

Some challenges that emerged during the institute included moments that disrupted scholars’ sense of intellectual safety, like confusion around expectations, unclear timelines, and some difficulties collaborating on group tasks. Later in the instate coaches emphasized that projects could be works in progress, yet many scholars still felt pressured to finish by the institute’s end. A few scholars in both the HD (hair discrimination) and WG (wage gap) groups expressed frustration with uneven group contributions. However, one of the emphasis of the institute was to leverage individual strengths, recognizing that participation may look different across students. Additionally, positioning students as researchers and decision-makers initially proved challenging as scholars wanted more guided direction. Yet, once they recognized the freedom to explore, many embraced this agency and expressed it creatively in their projects.

While this study offers promising insights, there are important limitations to acknowledge. The analysis focused on a small group of scholars, predominantly Black and South Asian secondary scholars who self-selected into the program. Their participation suggests a pre-existing interest in mathematics and justice-oriented learning, which may not be representative of all scholars. Additionally, the summer institute served as an illustration of what is possible in an intentionally designed space. It is atypical compared to most school-year settings. Scholars were compensated for their participation, which we believe was ethically important; it affirmed their labor and contributions as researchers, rather than exploiting their experiences for academic purposes. However, we recognize that this level of support, financial and otherwise, is not always feasible in public school settings. We also note that the literature supports the idea that intellectually safe environments can be created in more traditional academic settings when educators are intentional about incorporating aspects of belonging, empowerment, and liberation into the classroom. As noted before, creating an intellectually safe environment must work in tandem with these three themes to encompass a fully intellectually safe space for students. While our goal was to highlight the positive aspects of cultivating an intellectually safe space, we also recognize the need for further research on the factors that hinder it. Some scholars experienced challenges in achieving this kind of space, including tensions around uncertainty, lack of clarity, and uneven distribution of labor during communal tasks. These complexities point to the importance of a more comprehensive study that explores both the supportive and limiting conditions of intellectual safety. Gender was not a central focus of this analysis, and future research should explore how race and gender intersect to shape experiences of intellectual safety.