(Un)equal demands and opportunities: Conceptualizing student navigation in undergraduate engineering programs

It is well known that earning a bachelor's degree in engineering is a demanding task, but ripe with opportunity. For students from historically excluded demographic groups, this task is exacerbated by oppressive circumstances. Although considerable research has documented how student outcomes differ across demographic groups, much less is known about the dynamic processes that marginalize some students.

describe the process of students overcoming obstacles they encounter when responding to the demands and opportunities associated with their pursuit of an undergraduate engineering degree.We focus on navigation because engineering is not a land of equal opportunity: the engineering learning environment produces different demands, obstacles, and opportunities for students.We define equal opportunity (Nickel, 1987) as the presence of means (e.g., resources, student support) that provide students with a reasonable chance (emphasis on reasonable) of overcoming obstacles, leaving none insurmountable.Although the word "equal" might suggest uniformity (i.e., taking an approach of equality), that is not our intention.Instead, we emphasize the need for everyone to have a reasonable chance of overcoming obstacles, and note that the presence of means intended to provide this chance must account for differential circumstances (i.e., taking an approach of equity).In their current state, engineering degree programs lack equal opportunity for all students.
Our model establishes (a) undergraduate engineering learning environments as entities full of obstacles that must be navigated, and (b) marginalization as a phenomenon that can be disrupted by structural change to said learning environments.Rather than focusing on inputs and outputs, we focus on processes.We define marginalization as a context-specific phenomenon that occurs when a learning environment denies equal opportunity by imposing varying levels of demands on students.This definition builds on prior conceptualizations of marginalization (e.g., Hall et al., 1994;Messiou, 2012;Mowat, 2015;Young, 1990Young, /2011)).
We hope these insights from our conceptual model will prompt reflection, analysis, and action among faculty members, university administrators, and policymakers-actions that involve making structural adjustments that anticipate and interrupt structures placing extra demands and obstacles on some students.

| BACKGROUND
The existence of marginalization and a lack of equal opportunity in engineering education is not a novel concept, evidenced by the large foundation of literature dedicated to understanding how engineering environments operate inequitably and are experienced by students from historically underrepresented groups (e.g., Blackburn, 2017;Museus et al., 2011;Ong et al., 2020).Marginalization has previously been studied using three main approaches: (1) deficitbased approaches, (2) asset-based approaches, and (3) critical approaches.Deficit-and asset-based approaches are often presented in binary opposition, where deficit-based approaches anchor on the notion that marginalization occurs because individuals have cognitive, motivational, or cultural deficiencies (J.Mejia et al., 2018), and asset-based approaches (or anti-deficit approaches) emphasize unique strengths (e.g., grit, resilience) that individuals bring to an environment as the result of overcoming disadvantages and resisting oppression (Harper, 2010).The unit of analysis for these viewpoints tends to be individuals or communities.Expanding this focus, critical approaches (or critical theoryguided approaches) intend to challenge oppressive social practices, belief systems, and structures of domination (J.Mejia et al., 2018).The unit of analysis for this viewpoint is typically society or dominant cultures, often accompanied by historical considerations.
Although we present these viewpoints as three equally distinct approaches to studying marginalization or historically underrepresented groups, we would be remiss if we did not note that some asset-based approaches are also critical, depending on how they are used.For example, Community Cultural Wealth (Yosso, 2005) and the Anti Deficit Achievement Framework (Harper, 2010) are both asset-based critical frameworks that assert that individuals have agency and the ability to overcome oppressive conditions.We would also be remiss if we did not join other scholars in rejecting deficit-oriented explanations of marginalization or underrepresentation (e.g., Basile & Lopez, 2018;Harper, 2010;J. Mejia et al., 2018).Accordingly, we aim to build only on the traditions of asset-based and critical frameworks.Though not exhaustive, these traditions can be seen in the work of prior scholars who have begun summarizing the list of critical frameworks potentially used in engineering education.Two notable and recent examples are J.Mejia et al. (2018) and Dietz et al. (2022).
In J. Mejia et al. (2018), the authors investigated how 33 engineering education research articles used critical theoretical frameworks.The authors note the use of frameworks such as Critical Race Theory, Feminist Theory, Intersectionality, Community Cultural Wealth, and Funds of Knowledge.Summarizing one of their key findings, they state the following: "The variety of critical theoretical frameworks indicated the openness and effort from the engineering education research community to integrate theoretical lenses to challenge the status quo.However, in terms of praxis, several studies did not combine theory and action while engaging with minoritized groups to ask deeper questions related to power, oppression, and normative practices."(p. 6).
In Dietz et al. (2022), the authors describe nine theoretical frameworks that reject deficit paradigms, with a specific focus on theories focused on race: (1) racial and vocational identity development, (2) color-blind racism, (3) funds of knowledge, (4) anti-deficit achievement, (5) community cultural wealth (CCW), (6) Nepantla, (7) in/authenticity, (8) ruling relations, and (9) racialized organizations and the history of racist ideas.In addition to describing each theory, the authors also describe how they have been applied to engineering education research questions and contexts.At the conclusion of their paper, they offer the following piece of advice: "Use these theories as a means to provide explanatory power [emphasis added] to your study.A tendency, particularly in qualitative research, seems to be to code data and present findings that are purely descriptive rather than interpretive" (p.21).
Across both papers, it is evident that there is no shortage of theoretical frameworks that can be used to explain the underlying reasons why marginalization occurs.J. Mejia et al. (2018) and Dietz et al. (2022) both highlight the tradition of anti-deficit scholarship focusing on domination and oppression as opposed to perceived student deficits.As they describe it, traditional scholarship tends to focus on describing society, whereas critical scholarship diverges by focusing on critiquing and changing society as well as exploring the circumstances that lead to oppression.Although there are existing theories that help us to better understand the experiences of marginalized students, a more comprehensive conceptual model is needed to capture the conditions and dynamics between an environment and person that lead to their marginalization in a particular setting such as engineering or STEM (science, technology, engineering, and mathematics) education.
We build on this tradition by developing a conceptual model that depicts the process of navigating the obstacles of a learning environment (i.e., undergraduate engineering program).In doing so, we aim to advance scholarship related to understanding college students 0 success.Prior researchers have explored college students 0 success most often through constructs such as involvement, integration, and engagement (Astin, 1984;Kitchen et al., 2021;Milem & Berger, 1997;Tinto, 1975).Wolf-Wendel et al. (2009) concisely describe these constructs with involvement being student-driven, integration resulting from a relationship between a student and their campus, and engagement relating to creating opportunities for students through engaging campus environments.Our conceptualization of navigation considers each of these perspectives.However, although previous work has conceptualized various constructs associated with navigating the higher education environment, we have not found any conceptual models or frameworks that theorize how those constructs fit together to produce varying obstacles for students in engineering and to understand how students navigate those obstacles.
It is important to note that different types of "frameworks" can be used for different purposes in the engineering education research process (Magana, 2022).There has also long been inconsistent messaging around what constitutes a theory and how it should be used (e.g., Magana, 2022;Suppes, 1974;Sutton & Staw, 1995).To minimize confusion, we will clarify our use of the term "conceptual model" and its distinction from theoretical frameworks before proceeding.By definition, a conceptual model is an abstraction of meanings and characteristics associated with an event or process experienced in the real world (Meredith, 1993).Conceptual models are typically created by combining and adapting existing frameworks.They are theory-driven and aim to describe constructs salient to a phenomenon and map relationships among those constructs (Magana, 2022).The purpose of a conceptual model is not to explain the underlying reasons why a phenomenon occurs (e.g., racism, sexism)-this is where theoretical frameworks come into play.Instead, the goal of a conceptual model is to assist people in the process of classifying and categorizing constructs that are salient when the phenomenon occurs.These constructs can be associated with one another, function in tandem, mitigate the phenomenon, or play a conditional role with respect to the phenomenon (Meredith, 1993).
We posit that, although existing critical theories/theoretical frameworks are suitable for explaining why marginalization occurs more broadly, additional models are needed to help stakeholders to classify and categorize constructs most salient when marginalization occurs in the context of engineering education.Therefore, we aim to answer the following overarching question: How might the process of a student navigating an undergraduate engineering program be visually represented in a conceptual model?
We arrived at the answer to this overarching question by addressing the following sub-questions: 1. What propositions are key to understanding how engineering students, particularly those from historically marginalized groups, navigate learning environments?2. What relationships exist between these propositions?

| METHODOLOGY
To address our research questions with full transparency, we begin this section by presenting a description of ourselves and our roles in this project, highlighting the sociopolitical contexts that brought us to this article.We then describe the model development process.

| Positionality
We considered our positionality from two perspectives, integrating insights from both Secules et al. (2021) and Hampton et al. (2021).The first approach considers positionality from the position of its influence on one's research topic and questions, epistemology, ontology, methodology, instruments, and communication (Secules et al., 2021).This approach influenced which questions we asked ourselves in preparing this article.The second approach frames positionality from the perspective of disclosing one's identities, experiences, and journeys (Hampton et al., 2021).This approach influenced how we chose to present this information to the readers.

| Our identities and experiences
We (i.e., Lee, Hall, Josiam, and Pee) each have personal interests, identities, and experiences that drew us to this project.We all received bachelor's degrees in engineering from predominantly White institutions (PWIs), have professional experience working in student support for underrepresented groups, and have research experience in engineering education.Lee and Hall have PhDs in engineering education and Josiam and Pee are working toward the same.As members of groups that have been historically underrepresented in engineering based on race and/or gender, each member of our team has emotional proximity to how members of certain demographic groups can encounter obstacles that negatively impact the pursuit of an engineering degree.Our research efforts are largely driven by the desire to address and improve this issue.
Although we could have focused more narrowly on race or gender, we chose a broader scope for this project, exploring previous work about a wide range of identities: race, gender, disability, socioeconomic status, and so on.Although much of the literature, as well as the examples we include in the following sections, does focus on race and gender, our project is intentionally not about one specific marginalized group because we believe that oppression-and consequently liberation-is interconnected and collective; we aim to be inclusive so that our findings will be widely applicable.This decision meant that we had insider status for some topics and outsider status for others, which influenced what we noticed and highlighted from previous work.
The members of our research team have epistemologies ranging from constructivist to critical, which drove conversations and considerations about how to frame this work.To develop a model, we relied on documents, as opposed to human participants, interacting with first-and second-hand sources to gather the information we needed.We heavily relied on previous literature, trusting the scholarship that came before us.We value collective knowledge, which we view as what we know and what others know.In highlighting and honoring previous scholarship in this field, we assert the value of those works and take our positions as researchers contributing to an ongoing academic conversation.
Our methodologies were driven by an interdisciplinary approach, whereby we considered ideas from literature in many fields outside of engineering education and higher education.We also sought recommendations from people beyond our author team to gain new perspectives on how previous work aligned with ours.We used a discovery approach, allowing the research process to be driven by what we needed to know.
Lastly, our communication style was driven by our goal of improving transparency and documenting the messiness of our project.This led us to use a formal, process-oriented approach to communicate the information presented in this document, disclose our decision-making methods, and develop a visual model to clearly communicate our findings.

| Our journey to this work
We began this work in May 2020 in response to the COVID-19 pandemic disrupting the original data collection plans of a larger project led by Lee.The aim of the larger project was to advance the extent to which the engineering education community understands how marginalized students navigate undergraduate engineering programs and make decisions with respect to seeking help.Because the pandemic significantly altered the experiences of undergraduate engineering students, we redirected our focus to better understanding the underlying phenomenon (i.e., marginalization) before moving forward to more traditional modes of collecting data (e.g., interviews).More specifically, the disruption provided Lee and Hall the time and space to embark on this conceptual project in a systematic way.Josiam and Pee joined the project subsequently.
The beginning of this project was also situated in a time of national awakening to systemic racism and infrastructure failure.This sociopolitical context provided additional motivation to begin synthesizing insights from literature for a broader audience that appeared more willing to talk about and eager to understand how oppressive systems work to marginalize people of color.This article describes our efforts toward that goal.Our hope is to provide a foundation for future work focused on how students experience marginalization as they matriculate through undergraduate engineering programs.With refinement, our model will contribute toward illuminating how the learning environment creates different obstacles for some students (i.e., marginalization), ultimately guiding universities toward disrupting marginalization altogether.

| Model development
We developed our model using an iterative process (Figure 1) involving four key stages: (1) clarifying the purpose of the development process; (2) identifying concepts and insights from prior research; (3) synthesizing the concepts and insights into propositions; and (4) visualizing the suspected relationships between the salient constructs in the propositions.One aspect of our research not represented in Figure 1 is that we advanced research quality (Tracy, 2010) by embedding reviews and feedback throughout the model development process.We reviewed prior works related to the phenomena, checked the comprehensiveness of our review through a peerreviewed feedback process and multiple rounds of stakeholder feedback, and pursued transparency by means of providing detailed documentation of the methodology for added accountability.We describe the process in more detail in the following.

| Clarifying the purpose
The first phase of developing our model was clarifying its purpose.We began our conceptualization of student navigation by considering what is already documented about the experiences and perspectives of students who are presumed to be marginalized (e.g., students from underrepresented groups are often assumed to be marginalized, minoritized, or otherwise underserved).We conducted this work from the stance that people are not born marginalized but, rather, there are particular structural features and power dynamics that result in certain individuals being treated as peripheral based on group membership.In alignment with engineering education scholars before us (e.g., Holly, 2020;J. A. Mejia et al., 2020;Pawley, 2019;Secules et al., 2018), we chose to position marginalization as something that is done to people as opposed to focusing on the absence of resources that someone is born into.
Guided by Lee's initially proposed research plan, we began the clarifying process with the theoretical lens of the person-environment fit theory, or PE fit (Caplan, 1987;Fraser & Fisher, 1983;Jansen & Kristof-Brown, 1998;Muchinsky & Monahan, 1987;Porter & Umbach, 2006).Briefly, PE fit suggests that individual-level outcomes such as behavior and attitudes result from the relationship between a person and his or her environment (Ahmad, 2010;Caplan, 1987).Fit between a person and an environment has been shown to improve outcomes such as satisfaction, well-being, commitment, and performance.In the context of our work, environment refers to the learning environment, which we operationalize as the physical space, contexts, and cultures in which students learn (Great Schools Partnership, 2013).The PE fit framework helped us to refine the purpose of our conceptual work, begin unpacking the process of marginalization, and situate marginalization as an environment-dependent phenomenon.Muchinsky and Monahan (1987) defined two types of PE fit: supplementary fit and complementary fit.Supplementary fit exists in a learning environment when a student and university possess matching or similar characteristics (Muchinsky & Monahan, 1987).Supplementary fit is conceivably the type of fit implicated when discussing a sense of belonging (Muchinsky & Monahan, 1987), or misalignment (or incongruence) of cultural values between students and institutions (Cable & Edwards, 2004).On the other hand, a complementary fit exists when the demands of the university match the abilities of the student and/or when the needs of the student are met by the university (Cable & Edwards, 2004;Muchinsky & Monahan, 1987)-ideally both.Complementary fit is therefore represented by need fulfillment (Cable & Edwards, 2004;Muchinsky & Monahan, 1987).
While recognizing that a person (i.e., a student) can experience misalignment with his or her environment without being marginalized, herein we focus on misalignment and unmet needs that result from "othering" on the basis of personal identities.As Caplan (1987) points out, when it comes to making improvements to the fit between a person and his or her environment, we must decide whether to change the person and/or the environment.With respect to combating marginalization, we posit that the target for change should primarily be the environment, particularly the structural features and power dynamics of the learning environment.Otherwise, a disruption of othering would likely necessitate a complete overhaul of the learning environment.To that end, we chose to begin by focusing on complementary fit, as it more immediately relates to improving the working conditions (i.e., demands and supplies) of undergraduate engineering students, inspired by a question posted by Mejia and colleagues: "Why do we think that engineers must work in unpleasant work conditions in order to do rigorous calculations to ensure, as they say, the bridge doesn't fall down?"(J. A. Mejia et al., 2020, p. 24).Through the lens of complementary fit, a student's working conditions can be better understood by examining the process of the student continually encountering and responding to demands generated by the environment while simultaneously having his or her own needs met.
However, because the nuances of marginalization depend on person-specific features and are significantly impacted by social context, we needed a more holistic approach to studying the phenomena than PE fit alone could provide.Specifically, we needed a conceptual model that (1) represents concepts and insights already captured in the literature; (2) highlights the role of environment-specific features in terms of the structure, power dynamics, and resources, or lack thereof, that create a sense of insignificance or peripheral treatment for a person or group; and (3) considers students as active agents in their engineering learning environments.To this end, we considered additional theoretical frameworks that complemented PE fit to inform our conceptual model of student navigation.

| Identifying concepts and insights
Upon clarifying the purpose of the model, the second phase of our model development process was identifying sensitizing concepts (i.e., concepts that would sensitize us to possible lines of inquiry) (Blumer, 1954) and insights in prior literature.We leveraged methods from two different review types.We used search and review methods associated with state-of-the-art reviews to identify insights about the current state of knowledge and search and review methods associated with critical reviews to identify sensitizing concepts from a range of perspectives (Grant & Booth, 2009).We needed to simultaneously consider mechanisms associated with the phenomenon (navigation) and the contexts (e.g., undergraduate engineering).Utilizing only one process would not have facilitated us to consider both the phenomenon and the context.Lee and Hall were each responsible for different parts of this process.
Lee was responsible for identifying insights that focused on STEM learning environments and higher educational attainment for students from underrepresented groups.This portion of the process focused on conceptual induction, that is, identifying key insights from prior literature that explain, provide understanding, or suggest testable hypotheses (Meredith, 1993).As Meredith (1993) points out, the goal of conceptual induction is to go beyond an accurate description of a phenomenon and explain how it occurs.To capture the complexity that marginalized undergraduate engineering students face, Lee reviewed prior studies related to diversity, equity, and inclusion in both STEM education and general higher education settings.His selection of studies was prompted by the following rhetorical question: What is required for an undergraduate engineering student to be successful, and what does it mean for them to be marginalized?Lee conducted searches in journals that specialize in publishing literature pertinent to higher education, STEM learning environments, and/or marginalized STEM undergraduates.His search began with the most recent issues at the time of our search (2020) and progressed backward in time until he began reaching multiple journal issues that no longer provided new information.Using a combination of searching journals issue by issue and searching generally based on prior knowledge, he identified many studies describing the phenomena undergraduate students experience while navigating an engineering program, including marginalization and academic, social, and professional situations.The journals that Lee searched issue by issue are provided in Table 1 in alphabetical order, with the years/issues he searched noted in parenthesis.Other journals that he found articles in as he searched generally are provided in Table 2, in alphabetical order.
Hall was responsible for identifying sensitizing concepts related to our theoretical framing.This part of the process was approached as a form of philosophical conceptualization aiming to integrate several different works on the same topic (Meredith, 1993).Specifically, she worked to further define/operationalize "navigating" from a theoretical stance by identifying the different ways that scholars from various disciplines have studied and discussed (1) someone's navigation of a system, and ( 2) relationships between people and their environments.Hall's process was initiated by the following question: What does it mean for a person to navigate an environment?She searched for articles that considered multiple ways to conceptualize what it means to "navigate," agnostic of context.This part of the process was not bound by the context of higher education or engineering education.Lee and Hall also sought the suggestions of other researchers to identify any theoretical frameworks that immediately came to mind when the purpose of the model was discussed.
In total, Lee and Hall read over 150 articles in search of insights and sensitizing concepts.They synthesized these insights and sensitizing concepts through bi-weekly discussions about how their latest readings either supported what other literature stated or offered new understanding.For example, although the initial searches for literature by Hall supported the use of concepts from PE fit as useful for studying student navigation, it was determined that information processing, a key component required for a person to navigate their environment, was not sufficiently foregrounded in that framework.Accordingly, they chose to consider how a person copes with stress as an additional framework to understand how people may navigate an environment (Folkman et al., 1986;R. S. Newman, 2002;Peacock & Wong, 1990;Slavin et al., 1991).
Through this iterative process, Lee and Hall identified sensitizing concepts from four agentic perspectives of student navigation (Table 3): students as workers, students as people, students as patients, and students as consumers.Ultimately, each perspective offered a unique way to consider student navigation, although no one view on its own quite captures the complex and dynamic relationship between students, the learning environment, and the institutional agents.

| Synthesizing concepts and insights into propositions
Lee and Hall then synthesized the concepts and insights into a list of propositions.Briefly, propositions are normative statements that summarize links between concepts from an array of related literature (Meredith, 1993).Propositions are rooted in a collection of prior research, reasonable assumptions made by subject matter experts, and existing correlative evidence.Our propositions summarize descriptive evidence of marginalization in undergraduate engineering programs and its relationship with students navigating their learning environments.
Lee and Hall developed initial propositions by clustering insights from the literature using sensitizing concepts and then crafting statements to summarize the resulting clusters.For example, insights related to the role of office hours (Briody et al., 2019), co-curricular support (W. C. Lee & Matusovich, 2016), counterspaces (Ong et al., 2020), family (e.g., Arellano & Padilla, 1996;Carrigan et al., 2019), fictive kin (e.g., Carrigan et al., 2019;Simmons & Martin, 2014), peers (e.g., J. P. Martin et al., 2013;Ong et al., 2020), faculty (e.g., C. Newman, 2011), and other university personnel (e.g., J. P. Martin et al., 2013;Qaqish et al., 2020) are all related to the sensitizing concepts "Supplies" and "Characteristics of Delivery System."After discussing this comparison, we drafted a proposition that focused on the breadth of support sources a student may turn to in a time of need, including people both within and beyond the university.
After creating an initial draft of the propositions, Lee and Hall incorporated the perspectives of Pee as a researcher who had not been as involved in Phase 2 and could pose clarifying questions.This group then reviewed the list of propositions for clarity, logic, and compatibility, ensuring that the propositions did not contradict each other.(Reviewing the propositions included reviewing the visual representation of the propositions' constructs and relationships, which is discussed further in the subsequent subsection.)This group also sought feedback from members of the project's T A B L E 3 Overview of perspectives drawn from theoretical frameworks and their corresponding sensitizing concepts.

Perspective
Framework Sensitizing concepts Sources

Stress-coping
Coping strategies; adaptive process Folkman et al., 1986;R. S. Newman, 2002;Slavin et al., 1991 Students as patients Health care access Policy; characteristics of the delivery system; characteristics of the population at risk; utilization of services; consumer satisfaction Aday & Andersen, 1974;Aday & Andersen, 1981 Students as consumers

Service quality
Expected service; perceived service; perceived service quality; decision on support offering Parasuraman et al., 1985 Note: Each perspective offers a unique way to consider student navigation and, thus, each informs the development of our conceptual model.
advisory board.The advisory board consisted of a combination of five engineering education researchers, practitioners, and administrators carefully selected for the expertise they brought to the larger project.In total, this process served as a way to corroborate the comprehensiveness and logic of the insights we planned to incorporate into our conceptual model.

| Visualizing the proposition's constructs and relationships
Lastly, we used the list of propositions to develop a conceptual model to visually represent the salient constructs and the relationships between them.We checked the alignment of our proposed model with our list of propositions by gathering outside perspectives on clarity and our logical reasoning.Josiam was added to our team to help refine the conceptual model and assess how well it depicted what the list of propositions stated.Lee also presented our work in progress to multiple audiences, enabling them to check the validity of our development process and gather feedback on our interpretations and visualization of the relationships between constructs.This process included receiving feedback from STEM education researchers and graduate-level engineering students.
Based on feedback and questions from each of the two audiences, we refined our list of propositions in an iterative process, going back and forth between the list of propositions and the visualization of the model, to reduce redundancy and have the model reflect changes related to the categories of influence.

| Limitations
Our method for developing a conceptual model was shaped by intentional choices, including how we incorporated prior work into our model.Although we thoroughly searched the engineering/STEM education literature for insights about the student experience, our search identifying theories, concepts, and insights was limited to a subset of fields.So, although our model offers one explanation, we could have identified other explanations had we relied more heavily on other disciplines that might view phenomena such as navigation differently.
We pursued several strategies to overcome this selection bias.First, we intentionally created an author team to ensure the quality of our process.Josiam was added to our team to help refine the conceptual model and assess how well it depicted what the list of propositions stated.Pee was not involved in Phase 2 and could pose clarifying questions.
Next, we also sought recommendations from people beyond our author team to gain new perspectives on how previous work aligned with ours.We embedded reviews and feedback throughout the model development process.We reviewed prior works related to the phenomena, checked the comprehensiveness of our review through a peer-reviewed feedback process and multiple rounds of stakeholder feedback, and pursued transparency by means of providing detailed documentation of the methodology for added accountability.We also sought feedback from members of the project's advisory board.Additionally, we presented our preliminary results to members of engineering education via invited talks to get community feedback.In summary, we included a lot of check-in points during the development of our conceptual model for people to identify concepts we might have initially overlooked.

| CONCEPTUAL MODEL OF STUDENT NAVIGATION
In this section, we present a conceptual model of student navigation, which is underpinned by a list of propositions.Our final list of propositions is presented in Table 4 and addresses both research subquestions.The table arranges each proposition according to its corresponding fundamental relationship in the model.The fundamental relationships are the learning environment, personhood, embedded contexts, sensemaking, and responding.Then we present and discuss the model, which addresses the overarching research question and research subquestion 2.
In phrasing the propositions, we chose to center student responses to the marginalizing dynamics of the learning environments (i.e., student experiences, demands, and decision-making).We made this choice to foreground student agency (e.g., student decisions) and the reality of individual students being negatively impacted by their environments.Combined, these propositions convey the complexity of student navigation in undergraduate engineering programs.
In Figure 2, we propose a conceptual model of student navigation in the context of undergraduate engineering programs.The model illustrates how localized structural features unjustly shape the demands and opportunities encountered by students and influence how they respond.It focuses on the dynamic interactions between the abilities, attributes, and characteristics of a student and the embedded contexts and support infrastructure of the learning environment in which they are situated.Our model addresses our overall research question by visualizing the relationships between the propositions and their respective constructs.
We use visual elements to improve the readability of the model.First, enclosed boxes (i.e., boxes with solid lines) denote constructs.Second, dashed lines denote fundamental relationships between constructs.Third, arrows represent the direction of the process, with double-headed arrows symbolizing the dynamic interactions and intersections of the primary relationships of the model.And lastly, gray shading denotes cognitive processes that are not easily observed or captured in demographic questions.Essentially, the model shows the bidirectional process of a student entering a learning environment with their personhood, giving meaning to their collective experiences (i.e., sensemaking), and responding to their obstacles, demands, and opportunities in coordination with the support infrastructure in the learning environment.
The remainder of this section is organized around the fundamental relationships in the model that posit how the constructs fit together.We will use the propositions from Table 4 to explain the process in more detail.
T A B L E 4 List of propositions that convey the complexity of undergraduate student navigation in engineering programs.

Model relationship
No. Propositions

Learning environments 1a
Student decision-making is burdened by the need to coordinate their preferences given the availability of options in the context of their learning environment's climate and dominant culture.
1b Student experiences occur within the organizational processes (e.g., the application of grading scales) and organizational characteristics (i.e., structure and culture) governing and shaping the local learning environment.
1c Student experiences must be situated within the broader context of historical harms (e.g., social, economic, political) and ongoing systems of oppression.

Sensemaking 4a
The full scope of student demands is best understood in the context of simultaneous realities (e.g., being a student enrolled in college, pursuing an undergraduate degree in engineering, and facing marginalization in an undergraduate engineering degree program).Note: The propositions establish the relationship between marginalization in undergraduate engineering programs and student navigation of their learning environment.Each proposition is labeled with a number and letter that corresponds to the fundamental relationship in the model that the proposition pertains to.

| Learning environment
The first fundamental relationship relates to recognizing and understanding that the university is both an organization and a learning environment.Simultaneously, the university is situated in a societal context.As a result, students' navigation of the university is contextualized by the learning environment that the university fosters and is embedded within broader societal contexts.The learning environment includes the climate and dominant culture of the university, organizational processes and characteristics of the university, and the historical harms and ongoing systems of oppression of the university and society.To represent this dynamic, the learning environment surrounds the navigational constructs in the model, and the arrow from the learning environment to personhood denotes the power of the learning environment to dynamically shape the characteristics of the student while they are in the learning environment.Further implications of the university being the context that students must navigate are discussed in Propositions 1a-1c.
Proposition 1a.Student decision-making is burdened by the need to coordinate their preferences given the availability of options in the context of their learning environment's climate and dominant culture.
We begin by acknowledging the additional work required to navigate a learning environment for some students.We recognize that choosing to move from decision to action is not without cost or coordination (Skinner et al., 2003).Some actions require students to modify their internal dialogue or behavior (Ong et al., 2020;Rodriguez & Blaney, 2021), focus on the advantages as opposed to the disadvantages (Powell et al., 2009), adapt or assimilate to a new culture (Morales, 2012;Ong et al., 2020;Powell et al., 2009), accept discrimination, or subscribe to exclusionary practices (Powell et al., 2009), to name a few.Proposition 1b.Student experiences occur within the organizational processes (e.g., the application of grading scales) and organizational characteristics (i.e., structure and culture) governing and shaping the local learning environment.
It is also important to acknowledge that universities are constructed of hierarchical bodies (e.g., colleges, schools, departments, programs) that can create both formal and informal power dynamics.This additional organizational F I G U R E 2 Conceptual model of student navigation of the undergraduate engineering learning environment.The model consists of five fundamental relationships (learning environment, personhood, embedded contexts, sensemaking, and responding).It represents the bidirectional process of a student entering a learning environment with their personhood, giving meaning to their collective experiences (i.e., sensemaking), and responding to their obstacles, demands, and opportunities in coordination with the support infrastructure in the learning environment.This model aims to illustrate how the learning environment creates different obstacles for engineering students to understand how students navigate those obstacles.The white boxes represent observable constructs, and the gray boxes represent cognitive processes that are not easily observed.
The notion of inequality regimes is useful to understand why organizational equality efforts have either had only modest success or have failed outright (Acker, 2006).Briefly, inequality regimes are defined as "loosely interrelated practices, processes, actions, and meanings that result in and maintain class, gender, and racial inequalities within particular organizations" (Acker, 2006, p. 443).Several scholars have highlighted the local influences (e.g., practices, processes, actions, and meanings) maintained by structure and culture that perpetuate inequalities within learning environments and account for readily observed differential student outcomes (Acker, 2006;Aday & Andersen, 1974;Harris, 2013;Parasuraman et al., 1985).For example, in an organization where people feel pressure to appear superior, competent, or independent, people who need help may be reluctant to seek it (Lee, 2001).
Proposition 1c.Student experiences must be situated within the broader context of historical harms (e.g., social, economic, political) and ongoing systems of oppression.
Lastly, because universities are situated in societies, contexts beyond the institution must be considered when seeking to understand student experiences.First and foremost, broader societal contexts influence the local learning environments and contribute to the oppression of some students (F.Lee, 2001;L opez, 2003;Mwangi et al., 2018).For example, Mwangi et al. (2018) examined how Black students contextualize their campus racial climate in light of "broader race issues, tensions, and movements occurring across the nation" (p.456) and found strong evidence indicating that the campus climate mirrored the racial climate of society.The racial climate also influenced Black students' future planning, perceptions of self, and engagement in racial movements on campus.By acknowledging the influence that broader contexts and harms can have on college students, it becomes clear that some students pursue an engineering education while facing a number of challenges stemming from societal or political influences, which indirectly impact their educational experiences.
Given this evidence, our conceptualization of student navigation must consider the complexity of multiple contexts, which can impact learning environments as well as a student's decision-making with respect to navigating their environment.Works such as Hurtado et al. (2012) and Perna and Thomas (2008) offer useful examples of models that recognize the complexity of multiple contexts and place an emphasis on the dynamics at play within and around institutions.Essentially, we recognize that the broader context and ongoing systems of oppression contribute to the overarching climate, educational practices, and student outcomes in engineering (Hurtado et al., 2012).Therefore, it is important to develop a comprehensive understanding of student success (i.e., considering structural as well as personal contexts; Perna & Thomas, 2008) that is properly situated within the broader context of historical harms.
By acknowledging that contexts are overlapping and unbounded, we recognize that events happening in society at large have an impact on students.Students at a university are attuned to societal events and practices that directly impact one or more of their identities, and this influences their educational experience.

| Personhood
The second fundamental relationship relates to recognizing and understanding students as individual people who are pursuing engineering degrees while attending a university.As people, students enter the learning environment and are immediately and continuously situated within the embedded academic, professional, and social contexts (represented in Figure 2 by the arrow from personhood into the learning environment) with an assortment of features and qualities that are inherent to them and shaped by their identities and prior experiences.These features and qualities are represented by the constructs within the personhood box on the left side of the model outside of the learning environment.Personhood is intentionally placed outside of the learning environment to emphasize that (a) the student's personhood exists outside of the learning environment and (b) it is the learning environment that interacts with this personhood to produce marginalization.The implications of the personhood constructs are captured in Propositions 2a and 2b.Proposition 2a.Student decisions are mediated by characteristics they have upon entering the learning environment, such as (a) their demographic identities and the visibility of those identities, (b) their familial and social networks, (c) their psychological characteristics, (d) their student status classification (e.g., transfer student), (e) their past experiences, and (f ) their goals and desires.
In addition to environment features, our model also accounts for the attributes that each student brings with him or her, such as demographic identities (Aday & Andersen, 1974;Perna & Thomas, 2008).People make assumptions about students' capabilities, and these stereotypes about students' demographic identities impact their experiences (F.Lee, 2002;Link & Phelan, 2001).For example, Faulkner explores the paradox of female engineers being simultaneously visible and invisible in the workplace.They point out that female engineers are highly visible as women yet invisible as engineers.This results in additional work for their gender identity to fit in with male colleagues' assumptions and for their professional identities to be taken seriously (Faulkner, 2009b).E. O. McGee and Martin (2011) provide another example in their study exploring stereotype management among Black mathematics and engineering students.They point out how students begin implementing strategies to lessen the threat and effect of negative stereotypes that assume Black inferiority in mathematics and science.Although the students in this study used their exposure to stereotypes to motivate them academically, this exposure did not come without psychological costs, which they should not have been expected to pay.
Moreover, the extent to which demographic identities are either visible or concealable (e.g., sexual orientation, depression/anxiety disorders, socioeconomic status) also impacts the uniqueness of students' experiences as well as whether an engineering environment is considered or experienced as marginalizing (Cantrell, 2001;Link & Phelan, 2001;Morales, 2012;Pawlowska et al., 2014).For example, although the additional fees and overall higher tuition costs of pursuing an engineering degree may put a strain on the financial ability of students from lower socioeconomic backgrounds and lead to marginalization, these students are not necessarily immediately recognized as poor by other people (Major, 2019).Thus, their financial status might influence their interpersonal interactions to a lesser extent.
Familial and social networks must also be considered because student experiences are influenced by the personal relationships and resources accessible to them.The value or capital provided by parental support (Arellano & Padilla, 1996;Carrigan et al., 2019) and familial support more broadly (Ong et al., 2020) is not to be understated.In addition, peer networks or friends also aid students during their educational journey (Subethra & Nirmala, 2018).From peer-led study groups (Qaqish et al., 2020) to the development of fictive kin (Carrigan et al., 2019;Simmons & Martin, 2014), students build relationships and source additional forms of support that can aid their navigation through the learning environment.
Additionally, students' individual psychological characteristics such as their attitude with respect to help-seeking (Aleven et al., 2003;Bamberger, 2009;R. S. Newman, 2002;Skinner et al., 2003), self-confidence (Litzler & Young, 2012;McLoughlin, 2009;Meyer & Marx, 2014), satisfaction with the quality of their interpersonal interactions, and perceptions of engineering (Litzler & Young, 2012;Meyer & Marx, 2014) are all characteristics influencing their engineering degree-seeking experiences.Some of these psychological characteristics may be related to their position with respect to the institution, such as being a first-generation college student (Morales, 2012), or any other nontraditional student status as determined by age, residence, employment, or having children.
Lastly, prior educational experiences, personal goals, and desires that each student has inherently shape his or her education expectations and desired trajectory while pursuing an engineering education.With this view, students' needs, prior educational experiences, and word-of-mouth communication help formulate their expectations for their educational experience.Their future decision-making, in relation to their education, will depend on their perceptions of the service(s) received (Parasuraman et al., 1985).Proposition 2b.Student decisions are further mediated by their (a) self-awareness, (b) situational awareness in the context of their environments, (c) and critical consciousness.
In Proposition 2b, we acknowledge the different ways marginalization can be recognized.It can be recognized or denied by both the individual and others, in various combinations (Mowat, 2015).Consequently, in addition to in/visible demographic identities, student experiences are further influenced by their respective levels of awareness regarding both themselves (i.e., self-awareness) and their environment (e.g., situational awareness, critical consciousness, sociological imagination).For example, E. McGee and Bentley (2017) illustrate how students' understanding of the world can influence what type of engineering work they engage in.They found that students who expressed collectivist values also displayed altruistic tendencies in their own STEM pathways, which may result in their experiencing conflict between STEM environments and their personal desire to focus on social justice and integrate equity into scientific work (E. McGee & Bentley, 2017).
Because the constructs in Proposition 2b are not typically included alongside other student characteristics and thus may be new to some readers, we will further expand on each.By self-awareness, we are referring to how well a person understands his her own actions, thoughts, emotions, skills, values, personality, and so on: in essence, how well they understand themselves as a person.This idea is similar to the self-concept, which has been used in prior STEM education research (e.g., Sax et al., 2015;Starobin & Laanan, 2005), but differs in that it accounts for the possibility of inaccurate self-perceptions.By situational awareness, we are referring to how well a person understands the environmental elements within their surroundings.In our context, this might entail understanding the department, discipline, and university in which they are located.And by critical consciousness, we are referring to the extent to which a person is aware of and has reflected on social inequities within the world, and desires to change it for the better.The development of this concept is often associated with Paulo Freire (1970), and it is similar to C. Wright Mills's (1959Mills's ( /2000) ) notion of the sociological imagination, or the capacity to see how historical context shapes our lives.
We emphasize the constructs because it is important to acknowledge that students have different levels of awareness and consciousness, which influence their capacity to understand (and desire/ability to change) the relationship between themselves and society, their university, and even their college and department.It is not enough to assume that their demographic identities are sufficient indicators of each.Therefore, the extent to which a student is conscious and capable of perceiving himself or herself to be marginalized and understanding what that means must be considered when analyzing their experiences (Mowat, 2015).For example, some students have situational awareness and are equipped with skills and abilities to navigate the systems which marginalize them.These "skills and abilities to maneuver through social institutions" have previously been referred to as navigational capital (Yosso, 2005, p. 80).Yosso's description of navigational capital combines individual agency with community navigation, which ties into the familial and social capital identified and discussed in Proposition 2a.Because these skills can translate into decisions or strategies with respect to help-seeking in the learning environment, they are important to consider when analyzing a student experience to ensure that what is initially viewed as thriving is not actually the student effectively dealing with undue labor.Although navigating an oppressive system is possible, it does not come without a cost.

| Embedded contexts
The third fundamental relationship relates to recognizing and understanding the embedded contexts of a university as subsystems of the learning environment.In the model, the embedded contexts are grouped within a box on the leftmost side of the learning environment to represent that students contact these embedded contexts immediately upon entering the learning environment.These contexts are where obstacles, demands, and opportunities are generatedwhether explicit, implied, or inferred.The implications of these embedded contexts are described in Propositions 3a and 3b.Proposition 3a.Student experiences must be understood in relation to the different embedded contexts of a learning environment (e.g., academic, social, professional) within the university, which simultaneously generate student demands (e.g., financial, logistical, psychological, physiological, and cognitive).
There are general demands germane to being a student in pursuit of post-secondary education, such as interactions with the university and its repository of services, staff, and program offerings (W.C. Lee et al., 2018;W. C. Lee & Matusovich, 2016) as well as more specific demands across the academic, social, and professional contexts that students have to navigate.We assert that the demands of learning environment subsystems are not mutually exclusive.Having acknowledged the separate contexts in the environment, we also looked at how those systems act to simultaneously influence student experience (Parasuraman et al., 1985;Perna & Thomas, 2008;Subethra & Nirmala, 2018).In the academic context, students encounter demands related to expectations for coursework and demonstration of knowledge attainment in evaluations (W.C. Lee & Matusovich, 2016;W. C. Lee & Matusovich, 2018).The social contexts also place demands on students in terms of interpersonal relationships and interactions with peers, faculty, and staff in their respective programs (W.C. Lee & Matusovich, 2016;W. C. Lee & Matusovich, 2018).Although we have largely focused on stakeholders employed by a college or university, it is important to consider that harmful prejudices, dispositions, and behaviors can also be operationalized laterally by fellow students.Lastly, there are professional contexts that call on students to engage with engineers and practitioners (e.g., networking skills, co-ops, resume building) in preparation for the workforce (W.C. Lee et al., 2018;W. C. Lee & Matusovich, 2016;W. C. Lee & Matusovich, 2018).Collectively, these demands have a cumulative impact on a student's undergraduate experience.These demands can be collectively referred to as student demands.
Although all students will experience some assemblage of the demands from these contexts, students with different demographic identities will likely employ different strategies to meet the demands encountered.Moreover, students may also employ similar strategies for dissimilar reasons.For example, F. Lee (2002) found that social demands mediated help-seeking behavior in professional contexts for male-dominated occupations.Her work suggests that men in engineering may be more reluctant to seek help from female peers and superiors because of threats of appearing incompetent or inferior based on their gender narratives (F.Lee, 2002).Essentially, the gender of a person mediated navigational strategies (i.e., help-seeking patterns/behavior) in professional contexts with a significant gender imbalance.In another study, E. O. McGee and Martin (2011) highlight a similar reluctance from men-this time Black men-to request (or rely on) help from superiors.However, the source of the reluctance was now due to negative stigmatization based on race narratives.As one of their participants noted, in a concise summary of how anti-Blackness hinders the ability to educate Black students, "the teacher can't teach somebody she is afraid of."In this example, race and gender combined to mediate a person's navigational strategies in academic contexts.
Because the social, academic, and professional aspects of a learning environment each require their own demands in undergraduate programs, the complexity of performing as a student is best captured when we acknowledge the cumulative demands across all systems and subsystems.Proposition 3b.Student experiences are shaped by (a) the extent to which the embedded context(s) are designed to meet students' needs, and (b) the extent to which students are prepared to meet the demands and challenges of each embedded context.
Although the environment exerts significant demands on a student's undergraduate experience, simply identifying those demands and assessing the availability of support to help students meet those demands is not sufficient.The scope of support within an environment must be understood as it relates to availability, accessibility, and suitability (Aleven et al., 2003;R. S. Newman, 2002;Skinner et al., 2003).As an example, J. P. Martin et al. (2013) point out the role of social capital (or lack thereof) in relation to student demands.A major finding of their study was that "delayed recognition" of available resources slowed access to and activation of resources, leading to difficult university transitions for students lacking social capital (J.P. Martin et al., 2013).This finding was specifically related to participants having trouble understanding procedures related to the college application and enrollment processes-information that was not readily available to them through their personal networks and had to be accessed through broader social ties.Viewed more critically, this finding demonstrates a failure of the learning environment to proactively ensure that the demands placed on students do not place undue demands on their available social capital.
When analyzing the design of a learning environment, it is important to acknowledge that resource allocation decisions are made by those who designed and structured the educational systems (George et al., 2019).Unfortunately, the needs of students do not always drive educators' decision-making.Resource allocation is commonly approached in a manner that is either reactive to external pressure applied to an organization (i.e., coercive isomorphism) or replicative of other organizations thought to be exemplars (i.e., mimetic isomorphism) (George et al., 2019).Each of these approaches falls short of proactively meeting students where they are with respect to their support needs.Given the multiple layers of differences across institutions, both coercive and mimetic isomorphism are inadequate at providing suitable sources of support and, in some ways, both approaches ignore the needs or abilities of students for which an intervention is targeted.George et al. (2019) underscore the need for more information that can better inform how universities assess and respond to student needs: "As postsecondary institutions and the contexts that shape them change, the ability of administrators and institutional leaders to proactively respond to and stay abreast of [the people in their] environment [and their needs] remains important" (p.15).The ability of an institution to respond to changing student needs could be enhanced by continuously evaluating student perceptions of support offerings and shifting resource allocation based on feedback.
Lastly, we must also acknowledge the influence of prior educational experiences, including the quality of those encounters and students' future expectations, on how students meet demands (Parasuraman et al., 1985;Subethra & Nirmala, 2018).A practical example of a student's preparedness to meet academic demands would be sufficient early experience with STEM (Cheryan et al., 2017) and core math and science readiness (Meyer & Marx, 2014;Minichiello, 2018;Seymour & Hewitt, 1997).Another example would be a student's history of trauma within or beyond the learning environment.
In seeking to advance complementary fit as a means of addressing instances of marginalization in undergraduate engineering education, especially for underrepresented students, the importance of the environment cannot be overstated.Environments should be intentionally designed to accommodate the people who will be navigating them.

| Sensemaking
The fourth fundamental relationship relates to recognizing and understanding the role sensemaking has in the student experience process.Sensemaking captures the plausible stories that a student can use to narrate their interactions with the university and its subsystems.Students-particularly students from demographic groups that have been historically excluded from the field-are burdened with the cognitive task of determining why an event happened.It is not always easy to identify instances of unfair treatment or their source(s).Thus, it is not always easy to identify obstacles, demands, and opportunities in the learning environment.Any obstacles, demands, or opportunities identified by an individual's own judgment should be considered.However, instances in which a student's assessment is incorrect due to their sensemaking process should also be explored, as some students may lack the cultural knowledge needed to identify what an unfamiliar situation might require of them.In the model, sensemaking surrounds the embedded contexts box and the obstacles, demands, and opportunities box to represent the student's processing of the obstacles, demands, and opportunities generated by the embedded contexts.The complexity of this process is described in Propositions 4a and 4b.Proposition 4a.The full scope of student demands is best understood in the context of simultaneous realities (e.g., being a student enrolled in college, pursuing an undergraduate degree in engineering, and facing marginalization in an undergraduate engineering degree program).
The undergraduate experience is far from one-dimensional and is best represented as a conglomerate of multiple realities that overlap.We must account for these realities.For instance, there are the general demands of attending college (regardless of the chosen course of study), in addition to the unique demands at the program level that engineering majors encounter.All of these demands must be considered because misalignment at any level can negatively impact a student's experience.
We will first focus on the reality of pursuing an undergraduate degree in engineering.One attempt to examine the demands and requirements of undergraduate engineering students looked at the undergraduate experiences of students who chose to leave engineering (Meyer & Marx, 2014).Findings indicated that the alignment between the demands of the learning environment and student abilities is key to understanding students' decisions to persist or leave engineering.For example, common themes identified by non-persisting engineering undergraduates included individual factors (such as poor performance, feeling unprepared for the demands of the engineering program, difficulty fitting into engineering) and institutional factors (such as disappointment with engineering advising) (Meyer & Marx, 2014).
In a similar study, Rulifson and Bielefeldt (2017) look at students who chose to leave engineering, this time through the lens of social responsibility.Their findings indicated that the alignment between students' needs and the supply of the learning environment is also key to understanding students' decisions to persist or leave engineering.For example, a central theme from this study was the impact that decontextualized technical courses (i.e., what was supplied) had on students with prosocial motivations (i.e., what was needed or desired).
If we stop our analysis here, we will have failed to account for the more general reality of being a college student as well as the particulars of marginalization.Knowing that some students are marginalized on the basis of assumptions, beliefs, and preferences about inherent student characteristics in certain environments, we acknowledge that marginalization produces and potentially exacerbates the misalignment or lack of fit for some students in some departmental contexts.Instead of discussing the details here, we will use insights from our model development process to summarize what is already known about the typical demands of most engineering students in higher education and discuss what we considered to be atypical demands (or obstacles) more common to marginalized students in STEM higher education contexts (Table 5).
Summarizing the obstacles column in Table 5, it can be deduced that marginalization is associated with encountering additional, unnecessary demands that impede a student's ability to both develop relationships and meet the typical demands they must meet as undergraduate engineering students in general.These demands and obstacles are present across academic, social, and professional contexts and must be considered.Marginalization not only involves a sense of not belonging but is also accompanied by feeling unable to make valuable contributions within a community, or inaccessibility of services and/or opportunities available to others (Mowat, 2015); this conceptualization provides more insight into why relationship building, among other things, may be more difficult for students facing marginalization.In addition to being physically and relationally excluded, marginalized students must also grapple with the impacts of exclusion.
Lastly, to add to the complexity that marginalization places on student demands in a learning environment that is not prepared to adequately support all of its students, marginalization places stress on students' health and wellness, including psychological factors.Work by Cech and Rothwell (2018) demonstrates the extent to which LGBTQ students face greater marginalization, devaluation, and health and wellness issues relative to their straight and cisgender peers.Proposition 4b.Student experiences are influenced by the dominant ideologies, hierarchical structures, organizational beliefs, and normalized behaviors of departmental faculty and the overall engineering culture created by present and former engineers and engineering students.
Each local environment is subject to the influence of broader disciplinary cultures and practices upheld by the academic units within them (Perna & Thomas, 2008).Broadly speaking, the following historically rooted dominant ideologies shape students' educational experiences in engineering: colorblind and gender-neutral narratives of meritocracy (Cech, 2013;Conefrey, 2001;Museus et al., 2011;Rohde et al., 2020;Seron et al., 2018), masculinity (Secules, 2019), heteronormativity (Faulkner, 2009a), depoliticization (Cech, 2013), and competitiveness (Secules, 2019).Each of these ideologies presents its own unique demands and challenges.For example, given the cultural norms of masculinity, it is important to consider the extent to which certain behaviors are considered either masculine or feminine within an T A B L E 5 Some demands and obstacles in undergraduate engineering programs.
engineering context (Blosser, 2017;Cheryan et al., 2017;Conefrey, 2001;Faulkner, 2009a;Secules, 2019), because, in such a culture, feminine behaviors are often devalued.The cultural myths espoused by engineering faculty and the broader organization they represent also lead engineering educators to perpetuate and underestimate the impact of oppression.For example, although significant, sexual discrimination is underestimated at a national level, and is thus an under-addressed problem for women in science and engineering programs (Conefrey, 2001).Beyond the ideologies mentioned above, myths such as each scientist's classroom is his castle or pretty girls do something else (Conefrey, 2001) dissuade engineering educators from making substantial efforts to address sexual discrimination practices in the science and engineering fields by implying that structural changes are off-limits and "gendered issues" are the result of inherent genetic differences between men and women.
In addition to shaping the experiences themselves, the dominant ideologies in engineering also influence how people react to incidents brought to their attention.For example, the myth of meritocracy can result in people assuming failure is always a personal matter."If women are not succeeding, it is because they lack ability or effort or both-gender has nothing to do with it.This myth ignores the fact that scientists and scientists-in-training are social beings who do not work in isolation" (Conefrey, 2001, p. 175).
In addition to these dominant ideologies, the rigid hierarchical order between faculty and students within a learning environment acts as a key barrier to faculty-student interactions.For example, in a learning environment with a strong hierarchy, students may be reluctant to initiate contact with their professors and instead choose to rely on other sources of support, such as their friends or online resources (Briody et al., 2019).This power imbalance may also make students less likely to confront or report faculty who are covertly or overtly prejudiced and treat students badly because of their demographic identities.
Lastly, as indicated by Baber (2015), normalized behaviors in STEM education practices (e.g., compositional diversity, a cost-benefit approach to diversity, and benefits for faculty from majority populations) prioritize increases in enrollment numbers without reshaping policies and practices to benefit students of color.Our conceptualization of what it is to be, and make sense of being, marginalized in engineering builds on the knowledge of these systemic influences within disciplinary fields.

| Responding
The fifth fundamental relationship relates to student response, which involves coordinating their adaptive processes with their recognition of support in the learning environment.The responding part of the model refers specifically to the actions a student takes once he or she has made sense of the obstacles, demands, and opportunities.These actions involve coordinating their adaptive processes with the support infrastructure available within the learning environment.Therefore, in the model, the responding box includes the adaptive reasoning and support infrastructure boxes, where the bidirectional arrow between them represents a student's coordination between the two systems.Adaptive reasoning relates to recognizing and understanding the conceptual parts of the navigation process, where students must make decisions that are contingent on the availability, accessibility, and proximity of supplies within their environment.The decisions are represented as a student's adaptive processes, which are influenced by the obstacles, demands, and opportunities he or she faces (represented by the bidirectional arrow between obstacles, demands, and opportunities and adaptive processes).Adaptive processes include the need to coordinate preferences and available options and coordinate both in-class/curricular and co-curricular/extra-curricular actions and contingencies.These adaptive processes are described in more depth in Proposition 5a.
Support relates to students' recognition and understanding of the role of student support in the learning environment and the support infrastructure available to them.The arrow connecting the support infrastructure box with the embedded context box represents the fact that support infrastructure can be targeted toward or contained within specific embedded contexts.The implications of student decisions regarding support systems are discussed in Proposition 5b.Proposition 5a.Students must make decisions related to participation (e.g., whether to engage and to what extent) in (a) in-class and curricular activities, and (b) co-curricular and extra-curricular activities.
A key consideration pertaining to students' abilities to navigate a given environment relates to the need to coordinate their actions and contingencies (Skinner et al., 2003) within the learning environment.Students' decisions about participation in in-class, co-curricular, and extra-curricular activities have a direct influence on their experiences in an undergraduate engineering program.This decision includes coordinating multiple behaviors and actions such as study habits (Qaqish et al., 2020;Treisman, 1992) and organizational skills (McLoughlin, 2009).For example, the process of a student deciding how to complete simple academic tasks may be more complicated than it appears on the surface.This decision can result in a student either interacting with other members of the engineering community or working alone.Although we could assume that this is a matter of personal preference, Hora and Oleson (2017) demonstrate that studying alone is a multifaceted process that can be initiated by either instructor-or self-generated cues.
The work of Karabenick and Knapp (1991) raises additional considerations for evaluating how marginalized students in engineering enact help-seeking behaviors and how this impacts their navigational strategies.Student experiences must be understood in conjunction with the work strategies they selected (i.e., coordinating their actions) when working to meet the demands of engineering.For example, when a student is struggling academically, it is important to know whether that student was working alone, had the option to work collaboratively with peers on assignments, or engaged in forming study groups for engineering assignments (Briody et al., 2019).
Students also coordinate their actions regarding whether and to what extent they will participate in co-curricular and extra-curricular activities.These decisions are made based on a host of contingencies, such as the students' familiarity (Shehab et al., 2012), connections, comfort, interests, and time restrictions (W.C. Lee et al., 2018).A commitment to participating could mean subjecting oneself to uncomfortable situations for some students.For example, "Women in Engineering" programs have been suggested to contribute to spotlighting.McLoughlin (2005) defines spotlighting as singling out female students by gender in ways that make them uncomfortable.McLoughlin noted three types of spotlighting: Type I is singling out with the intention to harm (overt sexism); Type II is doing so with neutral intentions (tacit sexism); and Type III, a new type of gender bias, is singling out women with the intention of helping them (McLoughlin, 2005).Though this example is focused on women, it is worth noting the different ways engineering educators may target students and considering how that targeting can be perceived by students (e.g., patronizing, performative, supportive).Even if the aim is to help marginalized students, misguided attempts can negatively impact students' decisions to participate in certain activities.For some students, deciding to participate is not just about enriching their educational experience: engagement in co-curricular activities can represent a real cost, such as one's self-esteem or confidence (Karabenick & Knapp, 1991), due to the stereotypes and social scripts associated with who needs help.A final aspect of navigating one's learning environment as a student is making decisions about which of the available resources to rely on (Skinner et al., 2003).These potential resources include relationships with faculty (C.Newman, 2011), office hours (Briody et al., 2019), interactions with school personnel (J.P. Martin et al., 2013;Qaqish et al., 2020), and university services (Parasuraman et al., 1985).Students also have the option to utilize the programming available on campus as a source of support to navigate their learning environment.Some types of programs exist as co-curricular support (W. C. Lee & Matusovich, 2016), and there are also extra-curricular support programs (Ong et al., 2020;Simmons & Martin, 2014).
In addition to programming, there are often designated physical spaces for supporting specific groups of students in a learning environment (Cabrera et al., 2016).For example, Hoffman et al. (2019) examined the discursive framing of spaces on campus that support students of color and found two predominant types of spaces, each with a different focus of support: assimilative spaces and subversive spaces.Assimilative spaces were found to be those that assisted students to navigate the predominantly White campus culture and institutional systems, while subversive spaces allowed students to connect individually with their affinity/community peers (Hoffman et al., 2019).In a perfect world, learning environments would be designed such that neither form of support would be too heavily relied upon, as it can be psychologically taxing to engage in assimilative spaces and isolating if subversive spaces are the only places you feel able to be yourself.
Nonetheless, given all of these sources of university support, students have other options available, including whether to rely on sources of support not directly provided by the university or to rely on themselves.Sources of support not directly provided by the university (i.e., personal relationships and resources) are featured within the learning environment in the model because if/when students make the decision to utilize non-university support, they are often making the decision from within the learning environment.We also recognize that despite having personal relationships and resources, some students rely on their own abilities, aptitudes, and competencies to meet the demands of their environment.Reviewing the support options and coordinating their behavior is not a straightforward task.
Student decisions about using support have a direct impact on outcomes and are influenced by their awareness of, proximity to, and activation of the resources available in their environments, as well as the accessibility of those resources.For example, students may resort to self-reliance because they deem themselves the best source of support, they are unaware of external support, or, worse, they perceive external support to be unavailable to them or unreliable based on prior negative experiences.Nuances in student decision-making are driven by the perceived equity in the environment in which the decision-making takes place.Given the number of reasons why students may or may not rely on external versus internal support, and the range of support options that exist, we view coordination of support from the perspective of coping mechanisms (Cech & Waidzunas, 2011;Powell et al., 2009), support access (Aday & Andersen, 1974;J. P. Martin et al., 2013), students' assessments of their own interactions (W.C. Lee & Matusovich, 2018), and perceptions of organizational justice (C.C. Martin et al., 2019).Taken together, these insights illustrate the complexity of student decision-making regarding which resources to use when navigating an environment (Aday & Andersen, 1974;Perna & Thomas, 2008).

| SCHOLARLY SIGNIFICANCE AND PRACTICAL IMPLICATIONS
If we designed learning experiences and assessments from a place anticipating such grinding chronic trauma, how might we rethink what is important to teach, or what kinds of behaviors we hold our students to? (J. A. Mejia et al., 2020, p. 25) In the spirit of the question quoted above, we offer the engineering education community a carefully developed way to analyze student experiences and identify obstacles that must be removed to disrupt marginalization.Our motivation for developing a conceptual model of student navigation was a desire to make undergraduate engineering programs more supportive and easier to navigate, particularly for students from underrepresented or underserved groups.Our model highlights salient constructs influencing the experience, learning, or persistence of students in undergraduate engineering programs, which is comparable to other models (e.g., Hurtado et al., 2012;Terenzini & Reason, 2005) with similar purposes.This lens should prove useful for examining elements of the embedded contexts and support infrastructure of the learning environment (e.g., when experiences are available, how much they cost; Secules et al., 2019) over which faculty members and administrators, staff, and policymakers have programmatic or policy control.
The primary utility of our model is in its ability to uncover elements of a learning environment (i.e., universityembedded contexts and support infrastructure) that negatively impact students and lead to unproductive struggleelements over which faculty members and administrators, staff, and policymakers have programmatic or policy control.Doing so could assist the engineering education community in identifying ways for institutions to become more responsive and work toward ensuring equal opportunity.To support this goal, educational researchers and evaluators should analyze learning environments in each of the embedded contexts to advance our understanding of how and when their designs produce student experiences that lead to obstacles such as social isolation, discrimination, harassment, and so on.For example, educational researchers could examine how students respond to the demands and opportunities associated with various high-impact practices (e.g., undergraduate research programs, co-ops or internships, study abroad programs), paying close attention to obstacles that arise.By examining the influence of structural features on how students navigate and thus participate in undergraduate engineering, the engineering education community can better understand what kinds of behaviors it holds its students to.Our conceptual model supports such an investigation by foregrounding important constructs to consider when collecting or analyzing data about such undergraduate experiences.
To help engineering education practitioners operationalize our model and reflect on how their learning environments may marginalize some students, we offer reflection questions in Table 6 that correspond to each part of the model.
The secondary utility of our model inheres in its ability to uncover students' decision-making or navigational reasoning.More specifically, our model may prove useful to educators, administrators, and student support practitioners who need to better understand the subjective behavior (e.g., decisions, whether adaptive or maladaptive) that students they wish to support make in undergraduate engineering programs.To support this goal, educational researchers and educators should explore the perspectives of students to advance our understanding of when and why certain obstacles, demands, and opportunities elicit certain navigation strategies and responses.For example, given what we know about spotlighting (McLoughlin, 2005), educational researchers could examine women's perspectives of which demands and opportunities are easier to respond to because of gendered sources of support, such as those offered by "Women in Engineering" programs or the Society of Women Engineers.By examining students' perspectives of the existing support infrastructure, the engineering education community can better understand how to create support systems that are more responsive.By foregrounding the importance of sensemaking and adaptive processes, our conceptual model raises the types of questions that should motivate such an investigation.
We encourage researchers and practitioners who use the model in either of these ways to also use asset-oriented and critical theoretical frameworks that encompass the interplay between structure (i.e., learning environments) and agency (i.e., the capacity of students to act independently).For those who are unsure where to start and are looking for anti-racist theories, Dietz et al. (2022) offers a helpful starting place as the authors conventionally direct you to foundation texts, summarize key tenets, and provide examples of use.Beddoes and Borrego (2011) similarly offer a helpful starting place for those looking for an introduction to feminist theory in engineering education, as the authors discuss the importance of theory in engineering education research and provide an overview of five different branches of feminist theory.Beyond engineering education publication venues, there is also no shortage of books and articles focused on these theories.Thus, we close by echoing the advice of Dietz et al. (2022): "Read, read, and then read some more.To use any framework, it is necessary to read extensively to understand its roots, its fundamental concepts, and the contexts in which we may appropriately use it."(p.21).

| CONCLUSION
We offer scholars and practitioners a more comprehensive, integrated understanding of student navigation and marginalization in the context of undergraduate engineering education.The resulting model illustrates the influence of structural features on how students respond to demands and opportunities and navigate differential obstacles created by the learning environment.Although the focus of our model is on marginalized students in engineering degree programs, it may also be applicable to STEM higher education more broadly.It is our hope that this model will be used in the traditions of asset-based and critical frameworks, by STEM educators, administrators, and researchers alike who wish to disrupt marginalization and advance equal opportunity.

ACKNOWLEDGMENTS
This material is based upon work supported by the National Science Foundation under Grant No. 1943811.Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
T A B L E 6 List of reflection questions for each relationship of the model to guide stakeholders on how the model can be used to change the learning environment to better support students.

Model relationship Reflection questions
Learning environments What structural features in our learning environment most significantly impact how students navigate and thus participate in undergraduate engineering?Personhood What elements of students' personhood does our learning environment privilege?What elements does it penalize?
Embedded contexts How are embedded contexts in our learning environment more or less easy to navigate for some students than others?
Sensemaking How are we supporting students with making meaning of the experiences they are having in our learning environments?

Responding
To what extent is our support system responsive to the ways students are actually responding to the demands and opportunities in our learning environment?

F
I G U R E 1 Overview of the conceptual model development process showing the four phases we undertook to develop the conceptual model.Phase 1 involved clarifying the purpose.Phase 2 involved identifying the concepts and insights.Phase 3 involved synthesizing these concepts and insights into propositions.Phase 4 involved visualizing the propositions' constructs and the relationships between the constructs.
List of journals searched issue by issue to identify studies describing the phenomena undergraduate students experience while navigating an engineering program.Journals (target years of publication) 1. Engineering Studies (2009-2020) 2. Equity and Excellence in Higher Education (2015-2020) 3. International Journal of STEM Education (2014-2020) 4. Journal of College Student Development (2017-2020) 5. Journal of Diversity in Higher Education (2015-2020) 6. Journal of Engineering Education (2010-2020) 7. Journal of Women and Minorities in Science and Engineering (2010-2020) Note: The list is arranged in alphabetical order.T A B L E 2 List of journals where we found subsequent articles to identify studies describing the phenomena undergraduate students experience while navigating an engineering program.Journals 1.American Educational Research Journal 2. Annual Review of Sociology 3. Educational Researcher 4. Journal of Higher Education 5. Journal of Hispanic Higher Education 6. Journal of Negro Education 7. Journal of Student Affairs Research and Practice 8. Review of Educational Research 9. Review of Higher Education 10.Sociology of Education Note: The data is arranged in alphabetical order.
Personhood 2a Student decisions are mediated by characteristics they have upon entering the learning environment, such as (a) their demographic identities and the visibility of those identities, (b) their familial and social networks, (c) their psychological characteristics, (d) their student status classification (e.g., transfer student), (e) their past experiences, and (f) their goals and desires.2b Student decisions are further mediated by their (a) self-awareness, (b) situational awareness in the context of their environments, and (c) critical consciousness.Embedded contexts 3a Student experiences must be understood in relation to the different embedded contexts of a learning environment (e.g., academic, social, professional) within the university, which simultaneously generate student demands (e.g., financial, logistical, psychological, physiological, and cognitive).3b Student experiences are shaped by (a) the extent to which the embedded context(s) are designed to meet students' needs, and (b) the extent to which students are prepared to meet the demands and challenges of each embedded context.

4b
Student experiences are shaped by the dominant ideologies, hierarchical structures, organizational beliefs, and normalized behaviors of departmental faculty and the overall engineering culture created by present and former engineers and engineering students.Responding 5a Students must make decisions related to participation (e.g., whether to engage and to what extent) in (a) inclass and curricular activities, and (b) co-curricular and extra-curricular activities.5b Students often rely on a variety of sources for support, such as (a) university representatives, (b) support programming and infrastructure, (c) physical spaces (and counter-spaces), (d) personal relationships, and (e) themselves.

Proposition 5b .
Students often rely on a variety of sources for support, such as (a) university representatives, (b) support programming and infrastructure, (c) physical spaces (and counterspaces), (d) personal relationships, and (e) themselves.