The illusion of inclusion: structural and methodological gaps in biology education research

ABSTRACT

This integrative literature review analyzes the corpus of biology education research published in the main biology education journals of major professional societies. The goal of this analysis is to determine which approaches (including groups of focus, research methods, and settings/perspectives) from social science fields (i.e., psychology, sociology, and anthropology) are utilized in published peer-reviewed biology education research relating to diversity, equity, and inclusion (DEI). Scoping how social science approaches are used in this area is important to understanding whether biology education research could benefit from complementary approaches that might advance praxis. This analysis found that research informing the biology education community draws heavily from psychological perspectives that are overwhelmingly not disaggregated (78% of articles identifying a group lumped the participant together), are by far more quantitative (58% used survey, 26% grades, 20% school data) than qualitative (17% used interview, 10% observation), and did not adopt structural approaches (72%). The addition of missing contributions from social science is critical to advancing interventions to broaden STEM participation, given that merging paradigms can offer more robust, multi-level explanations for observed phenomena. This has important implications for education, biology education, biology education research, social science, and research in related STEM fields.

INTRODUCTION

Biology education research (BER) is a recently emerging field at the nexus of the natural and social sciences. In recent decades, BER has increasingly centered its work on investigating and framing student learning and engagement, with an increased focus on promoting inclusion and equity in higher educational environments (1). Much like in other STEM fields, retaining and engaging students in college biology classrooms poses challenges, and this topic has taken on renewed importance in BER.

STEM education journals have become more distinct from each other over time, showing a clearer pronunciation of different foci (2). Over the last two decades, STEM education has coalesced as a field with multiple offshoots including BER. BER at the college level is often distinct from research in the K-12 setting when it comes to the conceptualizations of inclusion, with BER focusing on marginalization, while K-12 research concentrates on disabilities and the implications of the Individuals with Disabilities Education Act (3–6). An incipient body of literature suggests that BER favors quantitative methods, often missing out on insights from qualitative and mixed-methods approaches (7, 8). While using quantitative methods provides the advantage of statistical generalizability, there are well-known limitations to their sole use, particularly in understanding the everyday experiences of individuals of marginalized groups (9, 10). To address sample size issues as well as due to misunderstandings about social theorizing, quantitative approaches can often aggregate all marginalized groups together (e.g., including Black, Latinx, and American Indian in one group) as well as fail to see within-group differences (e.g., African American and African) (10). At the same time, much educational research relies on individual-level approaches to student success, rather than those that consider structural or systemic issues underpinning outcomes. These systemic issues include the lack of critical discussions around race, racism, and racial equity (11). Engaging in these conversations can foster understanding for minority students, help them recognize their significance in the field, and pave the way for their future success.

Given the importance of retaining students and DEI-focused work on broadening participation in the field, this article contributes to the literature by assessing the scope of studies in BER. Specifically, we take a first step toward this goal by investigating and analyzing the current state of the field of BER that focuses on DEI. To do this, we analyze papers published in four top biology journals overseen by the major biology/biology education professional societies for higher education. The rationale for focusing on BER society-based journals lies in both the impact of professional identity and community on teaching practices and the opportunity to explore the topic of DEI research in context. Hanauer and colleagues found that STEM faculty undervalue teaching when it does not reflect the learning and development that occur in professional research and that professional identity fostered through professional communities helps underpin the success of both research and inclusive teaching (12). We focus on publications by professional societies (e.g., the American Society for Cell Biology, the American Society for Microbiology, and the National Association of Biology Teachers), as practitioners often turn to them for information and professional development when it comes to their own pedagogies and educational research efforts. Because STEM graduate programs are considered training grounds for society’s scientists, training in how to teach has been generally absent or minimal (13). Discipline-based professional societies have stepped into this void, offering curriculum, special interest teaching subgroups, and a place for faculty to find a teaching community.

Indeed, society journals directly inform the research and teaching of practitioners, including journals such as CBE—Life Sciences Education and SABER conference papers (8). Focusing on the field of biology education allows for the exploration of the key ideas developed within this context. Another important element of that context is the political and social considerations that make examining DEI in the United States distinct (14). The key ideas found in a focused study like this one can then be examined more broadly in other disciplines and locations. Furthermore, focusing on the biology education community can help ensure that findings are directly actionable.

More broadly, scientific societies impact disciplinary culture and norms both by defining the disciplinary field and by influencing a range of stakeholders with the agency to make individual changes (15–17). Scientific societies have demonstrated an interest in promoting diversity, equity, and inclusion within STEM environments, with increasing interest from biology-related disciplines in particular (17–24). We add that, when it comes to consideration about what exactly gets published in BER journals, Dolan (25) highlights the tension between the desire of researchers to publish academic research (perceived as being more rigorous and a well-known expectation for tenure) and the publishing of implementable classroom tools sought by practitioners reading the journal.

We focus here specifically on BER to address the discussion in the literature that DEI-focused BER work has a quantitative bias (8). Many practitioners of BER are biologically trained researchers, who have transitioned into education research by virtue of their teaching-focused roles, and are more often preferred to have a strong background of training in biological sciences rather than in educational research. This phenomenon, considered to be a norm in biology, may also be applicable to education research in other STEM disciplines. Limiting our scope to this one discipline, where an issue has been raised, may inform further study and comparison of research methodologies for DEI-focused work across the STEM disciplines.

Background

Recent scholarship in BER provides some insight into the current state of the field, suggesting a biology education research tradition mostly excluding qualitative, disaggregated, and structural approaches. However, this existing research is somewhat dated and therefore might not reflect the current landscape, which is more likely shaped by the increased emphasis on DEI in STEM more broadly. For example, in one of the few articles that have undertaken a similar inquiry as ours, Lo and colleagues (8) performed a content analysis of peer-reviewed research articles in biology education research (i.e., in CBE—Life Sciences Education). They focused on research articles between 2002 and 2015 and also included conference abstracts for the Society for the Advancement of Biology Education Research (SABER) between 2011 and 2015. The authors concluded that these articles and abstracts relied heavily on causal research questions and quantitative methods (8). In addition, the authors found an important distinction among SABER abstracts: research talks were more likely to use quantitative methods, and posters were more likely to report qualitative methods, with poster presentations being perceived to have lower prestige. A crucial point made by Lo and colleagues in interpreting these findings is that most of the research conducted on and in biology classrooms at that time was carried out by practitioners who had “crossed over” to biology education research from the biological sciences where they had received training in causal, quantitative research methods, potentially without benefitting from training on social science methods and theories (8).

Lo and colleagues also found that the setting of biology education research draws more on a classroom context (8). On one hand, studies of this nature may focus on using discrete metrics, such as grades or academic performance indicators, to understand student success. On the other hand, fields such as sociology and anthropology draw on qualitative methods to better understand the in-depth perspectives of individuals and capture important elements of university, cultural, and societal contexts in understanding social forces at play in STEM education. Using narratives made possible through methods such as interviews, ethnography, and focus groups reveals insights not capturable in grades and other data (26–30).

Qualitative approaches help researchers understand how and why people experience what they do. For instance, Bucholtz et al. analyzed ethnographic data from a study with undergraduate math and science majors that yielded 439 hours of audio-video content of interactions among 118 participants who were part of a larger pool (n = 496) (31). Interviews and a survey offered additional data for triangulation and to complement the strengths of the other methods. The authors took a close look at how one lab group interacted during their weekly meetings in an honors-level first-year chemistry class over 10 weeks. There was one man and two women chemistry majors in the group (the participants’ race was not included by the researchers). The authors described how the team’s dynamic shifted during this time, as the two women first acknowledged the man as a competent team member but then later imposed “a socially and academically marginalized persona through habitual mockery and criticism,” despite his grades being similar, though sometimes slightly lower than theirs (31). In the end, the authors argue that understanding how student identities unfold in labs and classrooms requires attention to how time unfolds in interactions and the underlying ideology that links meaningful resources across contexts. Such focus on ideology and time when considering identity is a key consideration in sociology and anthropology, rather than psychology (32).

Relatedly, in scientific research, White male participants have been used as the primary population studied, and results were then generalized to all, including marginalized groups (8). Furthermore, research is often carried out under the White gaze (33), lumping together and naming marginalized groups as deficient, reproducing settler logic. For instance, work can aggregate Black/African American, American Indian, Asian, Latinx, and other Persons Excluded due to Ethnicity or Race (PEERs) (34–37) into one group. To help address this, NIH has supported a requirement to include racial/ethnic data in studies since 1993 (38). The requirements state that data on race, ethnicity, and gender should be reported, and participant populations should be diverse. However, African American and Latinx people account for less than 10% of study participants (39). Additionally, gender and sexual marginalized groups (e.g., LGBTQIA+) are regularly excluded from research studies (40), but when included, they can be lumped together in one group—despite the vastly different impacts of having gender marginalized (transgender and non-binary) versus sexual marginalized (lesbian, gay, and bisexual) identities (41). For instance, Coburn et al. (42) found that 92% of the biology education studies published in CBE–Life Science Education from 2018 to 2022 conflated sex and gender by using binary classifications as gender options in their data collection process. However, this exclusion of transgender and genderqueer students is often overlooked as a limitation. The invisibility of whole communities of participants can lead to inaccuracies in theoretical assumptions about how research impacts communities. This is particularly a problem in attempting to parse out how Asian students (e.g., underrepresented Southeast Asian students, such as Cambodian, compared to East Asian students, such as Chinese) experience STEM (43, 44). Better inclusion of diverse groups is required in order to fostering a BER paradigm that can better inform approaches to success (8, 39, 40, 45, 46).

At the same time as issues of methodological emphasis and group operationalization are germane, the level of analysis used in BER is also a paramount concern. Level of analysis relates to how students’ struggles are framed and whether individual versus structural frameworks are used to understand student experiences. A large body of STEM education research mobilizes individual-level approaches rooted in psychology to diagnose problems, such as poor outcomes, grades, and student attrition, without appropriately overlaying the structural conditions underpinning these issues. Such an approach can back classroom interventions (8) assuming small tweaks to student experiences can singularly drive success, without understanding how classroom contexts and pedagogies intersect and are affected by larger systemic issues.

Sociological and anthropological approaches often center the group and structural level. Theories from sociology and anthropology, among other social sciences, provide essential frameworks for understanding diversity, equity, and inclusion (DEI) within STEM environments. Symbolic interactionism emphasizes the importance of micro-level interactions, where individuals create meanings and identities through social engagement, which can reveal how inclusion or exclusion occurs in everyday STEM interactions (47). Critical race theory (CRT) focuses on the role of race and power dynamics in systemic inequality, offering an analytical lens for examining how institutional structures in STEM perpetuate or challenge racial marginalization (48). Structural functionalism considers how social institutions function to maintain order, a theory that can be applied to STEM environments to explore whether existing structures support or hinder diversity and equity (49). Finally, cultural relativism, a concept from anthropology, calls for the understanding of cultural practices in their context, which helps analyze the diverse cultural values STEM professionals bring to their workplaces and their impact on inclusion (50).

Qualitative methods from sociology and anthropology are well-suited for exploring DEI in STEM environments. Ethnography, rooted in anthropology, involves immersive participation and observation, providing a detailed understanding of the everyday experiences and organizational cultures that shape inclusion and belonging for underrepresented groups in STEM (51). In-depth interviews and focus groups, commonly used in both sociology and anthropology, allow researchers to collect detailed personal and collective narratives, offering insights into individual and shared experiences of discrimination, inclusion, and identity in STEM (52). Case studies, a method used across both fields, enable in-depth analysis of specific institutions or departments, highlighting the successes and challenges of implementing DEI policies (53). Finally, narrative analysis, a qualitative method that examines how individuals construct their personal stories, is crucial for understanding how STEM professionals perceive their roles, struggles, and contributions to DEI efforts (54). These methods, by capturing both individual experiences and broader institutional dynamics, offer a comprehensive approach to studying DEI in STEM environments.

For example, Campbell-Montalvo et al. draw from sociology as they mobilize social network theory (55) to understand how people in a student’s social network impact their persistence in their engineering major (56, 57). First, drawing on interviews with 55 women and underrepresented students, the researchers found that students experienced different engineering department climates based on how others reacted to their identities (20). For example, Black students experienced more stereotype threat (a concept born in social psychology) (58), a point the wider literature has already identified as anti-Blackness of STEM (59–63). This resulted in such students having to not only be aware of how others likely perceived them but to also make sophisticated calculations about how they should behave in order to manage others’ impression of them (a concept known as stereotype management) (64, 65). Yet, having people in an undergraduate engineering student’s social network helped them persist in their degree program—a pattern suggested in this qualitative research and confirmed through inferential statistics on surveys with over 2,000 undergraduates (66).

In Campbell-Montalvo et al., the mechanisms at play wherein social networks impacted persistence came from individuals who saw students as total people—individuals who were close to students like their parents, relatives, or teachers who were often engineers or scientists (56). These individuals advised women, Black, and Latinx students to expect to be stereotyped. When that happened, the advice served as an armor to help students make sense of what happened in a way that did not harm their sense of fit in engineering, allowing them to persist (20). Drawing from anthropology, the authors centered cultural model theory as they theorized about students’ cultural models of engineering success (67). Students’ cultural models of engineering success are the ways that students understand and think about how to be successful in engineering. One component of this type of cultural model is the notion of sense of belonging or fit, wherein fit is a structural outcome based upon the social and academic environment the student experiences within the engineering department, shaped largely on departmental, university, and societal policies, histories, and forces (56, 57).

Similarly, the researchers argued that students participating in professional engineering societies were more likely to persist than those who did not because of the common social identities of participants and those in the societies (57). Black students participating in the National Society of Black Engineers reported that being surrounded by other engineers like themselves, seeing them succeed, and receiving advice from them helped them maintain their fit in engineering. Thus, the mechanism at play is a range of advice received from people in a students’ social network that had much to do with those people sharing similar characteristics with the students. In these studies (56, 57, 66), the power in the advice comes from people with social characteristics similar to the students, a concept known as alter homophily—a term the authors borrowed from sociology (68). Ultimately, given the anti-blackness of STEM and the differential treatment of Black students in engineering compared to Latinx and other counterparts, the authors argue that social networks and their resultant effects on fit are important factors to consider when designing interventions to address STEM inequities. However, they caution that approaches aimed at fixing such students must also exist with approaches that seek to improve the STEM climate in which marginalized students are learning (20). Additionally, these findings explained here—relating to nuanced identity implications in which Black students had different experiences from Latinx students—were made possible by the team’s disaggregated approach. Yet more work needs to be done to further investigate intergroup differences, including those among Black, Asian, Latinx, and other students. In any event, this handful of exemplar articles illustrates the power of work on STEM inclusion research when multiple research methods, disaggregated grouping, and numerous levels of analyses are deployed to robustly measure and deconstruct phenomena, identify processes, and inform practice.

Reflecting on the state of BER when it comes to methods, group aggregation, and analytical level, we note that the characteristics of BER found in work reviewed almost about 10–20 years ago by Lo and colleagues (8) have led to ongoing calls for expanded approaches in BER. These calls have advocated for the inclusion of Discipline-Based Educational Research (DBER), the incorporation of social science fields and approaches, and the linkage between researchers and practitioners. For example, efforts to integrate with other DBER communities include expanding relationships with the long-established field of physics education to address STEM teaching and learning as well as inclusion and institutional change (69). Another example of calls for expansion is found in Dolan’s (25) perspective piece, which reviewed the literature from BER-focused journals and national reports echoing calls for greater collaboration and communication between BER researchers, social science researchers, and biology classroom practitioners.

In the same vein of collaboration, Dolan’s later (70) editorial in CSE—Life Sciences Education adds to the case to explore the utility of other fields, particularly in the social sciences, to incorporate new theories and methods into BER, which would allow researchers to explore important consequential research questions and outcomes (e.g., how biology students develop science identity and a sense of belonging in STEM) (70). Moreover, Peffer and Renken (71) suggest incorporating natural science traditions into DBER to provide a better and more robust understanding for science collaborative research to improve DBER. And when it specifically comes to BER, Akman and colleagues (72) highlighted the efforts of four education reform organizations—BioQUEST Curriculum Consortium, MathBench Biology Modules, QUBES (Quantitative Undergraduate Biology Education and Synthesis), and the Intercollegiate Biomathematics Alliance—to effect systemic change in the mathematics-biology education community (72).

Ultimately, it remains to be seen as to how these interdisciplinary calls have actually impacted BER, particularly BER related to DEI. Thus, understanding the scope of more current DEI-focused BER would illuminate the gaps and offer important insight to augment BER with additive approaches. Therefore, the following research questions guide this study.

1. What types of methodologies are used in biology education research?

2. How are marginalized groups defined and conceptualized in biology education research?

3. What types of frameworks do biology education researchers use to draw conclusions (with a focus on individual versus structural frameworks)?

Knowing the answers to these questions—and therefore, how far BER has come—will help inform more expansive approaches in the field, approaches seen in publications outside of biology societies, where researchers have utilized a range of methods to better understand and address STEM undergraduate inequity (64, 65, 73–83).

METHODS

In order to explore themes addressed above, we use an integrative approach. We collected a key sample of research articles on the topic of diversity, equity, and inclusion in biology education research and undertook data analysis of these articles to identify trends in this body of scholarship. We selected articles from journals associated with national biology professional societies, including such journals based upon the immense role of national professional societies in fostering community within STEM disciplines. These education articles also have a wide readership among biology instructors who are not education researchers. These journals are discipline-based, with articles written by higher education biology practitioners for their peers, illustrating a community of thought within the biology education field.

To address our research questions, we collected all abstracts from research articles from five professional society-run biology education research journals, including all articles spanning January 2017 through July 2020 (Table 1). The societies include the Association of College and University Biology Educators, American Society for Biochemistry and Molecular Biology, American Society for Cell Biology, American Society for Microbiology, and National Association of Biology Teachers. Collectively, their membership includes students, researchers, educators, and bioscience professionals across various biological science disciplines. Their journals include Bioscene: Journal of College Biology Teaching, Biochemistry and Molecular Biology Education, CBE Life Sciences Education, Journal of Microbiology and Biology Education, and The American Biology Teacher. These journals’ foci and scopes are specifically geared towards biology education (Table 2).

TABLE 1.

Journals identified for the study

Biology professional society	Membership defined by society	Acronym	Professional society journal	Journal acronym/abbreviation
Association of College and University Biology Educators	“ACUBE’s members are a group of post-secondary educators dedicated to excellence in teaching biology.” https://www.acube.org/Bio.phphttp://www.acube.org/index.php	ACUBE	Bioscene: Journal of College Biology Teaching	Bioscene
American Society for Biochemistry and Molecular Biology	“Biochemists and molecular biologists” https://www.asbmb.org/membershiphttps://www.asbmb.org/membership	ASBMB	Biochemistry and Molecular Biology Education	BMBE
American Society for Cell Biology	“Research scientists in academia and industry; Postdoctoral fellows; Educators (community college instructors); Graduate students; Undergraduate students...scientists who study the biology of the cell” https://www.ascb.org/membership/	ASCB	CBE Life Sciences Education	CBE-LSE
American Society for Microbiology	“Researchers, educators, and health professionals” https://asm.org/about-asm	ASM	Journal of Microbiology and Biology Education	JMBE
National Association of Biology Teachers	“Teachers from all settings, from K-12 to postgraduate to informal” https://nabt.org/Members-Area	NABT	The American Biology Teacher	ABT

TABLE 2.

Aims and scopes of journals

Journal	Aims/Scopes text	Source(s)
Bioscene	Not listed on their website.	N/Aa
BMBE	“Biochemistry and Molecular Biology Education (BAMBEd) is a peer-reviewed international journal sponsored by the International Union of Biochemistry and Molecular Biology that is devoted to improving the teaching and learning of biochemistry and molecular biology and promoting the scholarship of molecular bioscience education. The journal is committed to the publication of new ideas, literature reviews and original research on topics relative to these goals. The journal aims to be a central source of scholarly materials for researchers, teachers and students involved in biochemistry and molecular biology education.”	International Union of Biochemistry and Molecular Biology
CBE-LSE	“The journal publishes original, previously unpublished, peer-reviewed articles on research and evaluation related to life sciences education, as well as articles about evidence-based biology instruction at all levels. The ASCB believes that biology learning encompasses diverse fields, including math, chemistry, physics, engineering, and computer science, as well as the interdisciplinary intersections of biology with these fields. One goal of the journal is to encourage teachers and instructors to view teaching and learning the way scientists view their research, as an intellectual undertaking that is informed by systematic collection, analysis, and interpretation of data related to student learning. Target audiences include those involved in education in K–12 schools, two-year colleges, four-year colleges, science centers and museums, universities, and professional schools, including graduate students and postdoctoral researchers.”	American Society for Cell Biology
JMBE	“The scope of the Journal of Microbiology & Biology Education (JMBE) is rooted in the biological sciences and branches to other disciplines. JMBE publishes articles addressing such topics as good pedagogy and design, student interest and motivation, recruitment and retention, citizen science, and institutional transformation. JMBE may also choose to accept manuscripts for publication in special themed issues, which intersect with several scientific disciplines. Recent themed topics include ethics in science, scientific citizenship, and science communication.”	American Society for Microbiology
ABT	“The American Biology Teacher (ABT) is an award-winning, peer-refereed professional journal for K-16 biology teachers. Topics covered in the journal include modern biology content, teaching strategies for the classroom and laboratory, field activities, applications, professional development, social and ethical implications of biology and ways to incorporate such concerns into instructional programs, as well as reviews of books and classroom technology products.”	University of California Press

^a N/A indicates not applicable.

We note that the sample does include some special issues with calls for papers on topics of diversity, equity, and inclusion. The 3.5-year window (January 2017 to July 2020) of inclusion spans the most recent articles prior to our team’s embarkation on analysis, provides a large but feasible amount of work to analyze, and complements existing analyses on periods prior to this point.

As part of the initial review, we sought to estimate the number of research articles that might comprise the sample for close analysis. To do this, we first began with defining the search terms for articles of interest within the journal sample set. Two co-authors, both subject matter experts with experience in communities of biology education practitioners and an interest in inclusive science practices, met several times to discuss how inclusion is conceptualized at meetings of biologists and biology education practitioners. From these discussions, the co-authors met with the research team (which included additional subject matter experts from complementary fields such as social science) to discuss the search terms. Based upon this, the two co-authors conducted a search of articles in the journals identified using the keywords “diversity,” “equity,” and “inclusion.” Articles appearing with these three search terms were reviewed, and the keywords associated with the articles were discussed as a team to create the final list of search terms: “broadening participation,” “underrepresented,” “under-represented,” “disadvantaged,” “diversity,” “equity,” “inclusion,” “anti-racist,” “minority,” “disadvantaged,” “pell-eligible,” and “PELL eligible.” We counted the approximate number of all articles associated with these terms and estimated the total number of articles available for analysis to be 1,293 (Table 3).

TABLE 3.

Estimation of articles published in five national professional society biology education journals from 2017 to 2020 on diversity, equity, inclusion (2017 to July 2020)

Type of content	Journal
	BMBE	JMBE	CBE-LSE	Bioscene	NABT
Articles (Research/Empirical)	155	267	200	18	158
Perspectives (Editorials/Reviews/Essays/Letters/Notices)	31	16	20	2	161
“Innovations” (Case Studies/Teaching Approaches/Lab Exercises)	54	0	0	14	146
Miscellaneous (Technical Notes/Classroom Materials)	51	0	0	0	0

One team member then parsed these 1,293 articles out by type (i.e., research, perspective, and innovations) as shown in Table 3. This parsing was done based upon review of the journal’s Table of Contents, showing that the sample consisted of 798 research articles, 230 perspectives, and 214 innovations. There were also 51 miscellaneous articles in BMBE, which consisted of technical notes or specific classroom materials that could be used to teach biology. We note that JMBE and Bioscene include conference abstracts, which were not analyzed as part of this work.

First round of coding: Confirmation of research study and underrepresented group of focus

In the first round of coding, we sought to confirm that the set of 798 articles listed as research articles in the journals’ Table of Contents were indeed empirically-based studies. We also wished to establish the categories of underrepresented groups on which they focused. This was done through systematic coding by four coders using a codebook with two codes that included: (1) Abstract is a research study, and (2) Study is about any underrepresented groups. In the codebook, underrepresented groups were defined according to the National Science Foundation operationalization (84), including underrepresented racial/ethnic marginalized groups (URM), women, persons with disabilities, and people with sexual and gender marginalized identities, but we also counted any other groups underrepresented groups mentioned as such.

During this first round of coding, our team included four coders: one biomedicine PhD-holding interdisciplinary researcher, two HBCU undergraduate students (one majoring in sociology and one majoring in biology), and one graduate psychology student at a PWI. The team met bi-weekly to discuss any coding that lacked agreement of two or more coders. Two additional subject matter experts attended the meetings to provide further insight and assist with coding agreement: one was a sociology PhD-holding assistant professor from HBCU, and the other was either a biology PhD-holding former HBCU assistant professor who transitioned to an administrative role at a PWI, or a psychology PhD professor at a PWI. The specific codebook used during this time is shown below.

Codebook 1

Is this a research study?

• 1Yes
• 2No

Is the article about an underrepresented group?

• 3Yes
• 4No

The inclusion criteria we used in the initial review and first round of coding were as follows: the study was published in an education journal listed in Table 1; the study was a peer-reviewed article or report with empirical findings (i.e., conclusions from quantitative or qualitative data were included); the study was published between January 2017 and July 2020; and the study explored diversity, equity, and inclusion in the context of biology education. Articles were excluded if they related solely to the study of a biological process. Using these methods, we coded 798 abstracts listed in journals’ Table of Contents as research articles and determined that 645 were research studies based on a full review of the abstracts. Of the 645 research studies that appeared with our initial search terms, 93 articles focused on an underrepresented group.

To determine coders’ reliability with our approach, we calculated intercoder reliability across the four coders for the first two codes on 5% of the initial articles (n = 5). For the first code, regarding whether the article was a research study, the four coders applied the same code 87% of the time. For the second code, regarding whether the article was about an underrepresented group, the four coders applied the same code 97% of the time.

In the second round of coding, and in concert with our research questions, we identified the specific demographics discussed within each article, identified the research methods used, and considered the level of analysis. This second round of coding was conducted on the 93 articles from round one that were identified as research articles on diversity, equity, and inclusion involving an underrepresented group. This round involved three of the four coders, with the undergraduate biology student not participating in this round. These three coders coded each set of abstracts for the second round of coding. We divided the abstracts into three groups of 31 and utilized Codebook 2 on one group of abstracts at a time (see Table 4). After each individual coded a group of 31, the coders and wider research team met to discuss any coding discrepancies in that group and arrive at consensus in cases where less than two coders were in agreement. The specific codebook used during this time is shown below.

TABLE 4.

Codebook 2

Race/ethnicity (select all that apply)
Code	Example
African American	Survey data came from a national sample of 688 undergraduate students in science majors who were primarily Black/African American and Hispanic/Latino/a with a 2:1 ratio of females to males. Analyses yielded acceptable validity statistics, and race × gender group differences were observed in mean responses to several measures
American Indian	We piloted an easy-to-analyze qualitative survey of scientist stereotypes in a biology class at a diverse, 2 years, Asian American and Native American Pacific Islander–Serving Institution
Asian	We piloted an easy-to-analyze qualitative survey of scientist stereotypes in a biology class at a diverse, 2 years, Asian American and Native American Pacific Islander–Serving Institution
Latinx	We conducted a series of interviews with 17 undergraduate students from a Hispanic majority institution regarding their participation in STEM-focused summer research programs at nine universities.
White	To determine whether reading guide completion before class is associated with increased performance, we surveyed students about their use of reading guides in two sections of a large enrollment (400+ students) introductory biology course and used multiple linear regression models to identify significant correlations or In 28 interviews with Judeo-Christian students taking undergraduate biology classes, students reported a religious identity that can conflict with the secular culture and content of biology.
URM as one group	In this paper, we longitudinally examine the integration of URMs into the STEM community by using growth-curve analyses to measure the development of TIMIS’s key variables (science efficacy, identity, and values) from junior year through the postbaccalaureate year.
Other racial/ethnic grouping	A hierarchical Bayesian approach was used to estimate posterior standardized completion rates for 3-year undergraduate and 2-year postgraduate degrees undertaken by non–Māori-Pacific and Māori-Pacific students.

Prior to deploying the first codebook, the entire main research team of eight met over the course of several months to construct the full codebooks (one and two) based upon the project’s research questions, previous literature, and initial review. These meetings usually included the 4 coders and four researchers involved in this project. Right before using the first codebook, the team used multiple meetings to code several samples as a group. During this process, the team discussed the meaning of each code, when to use them, and also refined the codebook to ensure its usability. The four coders were then given a sample set of 40 abstracts to code, and the team met to review inconsistencies and learn how to code. During the initial review and the two rounds of coding, coders were trained on the codebooks and agreed-upon norms of deploying them, such as erring on the side of caution. If they were not sure whether an article should be counted as fulfilling a certain criterion, they should code it as a “yes,” and discrepancies would be further discussed during team meetings. The training included discussion of how to capture data within a code. For instance, the code of Latinx in the second codebook was to be deployed in instances where terms like Latino, Latinx, Hispanic, and others were used—although we acknowledge the terms are not synonymous. To be clear, the 93 articles were not read in full, and this study included only analysis of the abstracts.

RESULTS

The following is a description of the 93 abstracts about research articles relating to diversity, equity, and inclusion that were coded by the team. Although the initial sample drew more broadly from the range of journals, due to how the abstracts fit the criteria, we ended up closely reviewing a larger proportion of articles from some journals—specifically, CBE Life Sciences Education (n = 70) and the Journal of Microbiology and Biology Education (n = 15)—compared to others, such as the Biochemistry and Molecular Biology Education (n = 4) and the Journal of College Biology Training (n = 4), given that our review showed the former pair of journals had more abstracts than the latter that described research articles according to our operationalization described earlier.

In looking at the articles we analyzed, 60 of the 93 articles specified a focus on any race/ethnicity group in the abstract. Of the 60 articles that mentioned race/ethnicity, most (n = 53) of these articles named one specific racial/ethnic group: most common being a grouped URM or similar designation (n = 47), followed by African American/Black (n = 3), Latinx (n = 2), and then Asian (n = 1). The seven remaining abstracts emphasized more than one specific racial/ethnic group, with three focusing on Black/African American and Latinx, two on Black/African American, Latinx, and American Indian, and the final two had two additional group combinations. Arguably, papers that use a URM denotation and those that group together Black/African American and Latinx and Black/African American, Latinx, and American Indian populations have similar study foci.

Looking at all of the 60 abstracts together, with seven focusing on more than one group (i.e., both Latinx and Black), the URM category was emphasized in 47 (78%) of the abstracts that focused on a group. This was followed by abstracts centering on African American/Black participants (n = 9; 15%) and Latinx participants (n = 8; 13%). American Indian, other racial/ethnic grouping, and Asian were specifically mentioned in two to three abstracts each. Figure 1 depicts the list of the racial/ethnic categories used in the articles, along with the number of abstracts in which each category was found.

Fig 1.

Percentage of racial/ethnic categories in abstracts.

Two of the 93 articles specified a focus on sexual and gender identities, with each article addressing both aspects together. The 91 remaining articles did not mention a focus on any sexual or gender identities.

Of the 93 articles, our coding determined that 92 of them articulated the use of at least one clearly specified research method. Most articles (n = 68) described using one research method, while 19 identified two, and five identified three. Looking at the 92 articles together, noting that 24 used more than one method, Fig. 2 shows the distribution of methods across the articles. The most common methodological choices were quantitative: more than half of the abstracts described using survey methods, a quarter employed grades, and a fifth used school data. Qualitative methods were less often used and included interviews (n = 16) and observation (n = 9). Two articles employed reviews of the literature as their methods. Additionally, methods outside of those in our codebook were found in nine abstracts.

Fig 2.

Percentage of research methods in abstracts.

All of the 93 articles were able to be coded using one of the three codes developed by the team to reflect the level of intervention. However, one article was coded with two codes as it drew equally from both levels. A classroom level of analysis was the most common (n = 37, 40%, see Table 5), followed by individual focus (n = 30; 32%), and then structural approaches (n = 27; 29%).

TABLE 5.

Number of abstracts adopting particular levels of intervention

Code	Count	Percentage using method
Classroom	37	39%
Individual	30	32%
Structural	27	29%

DISCUSSION

To summarize our findings, research informing the biology education community draws heavily from psychological perspectives that are overwhelmingly not disaggregated (78% of articles identifying a group used a lumped together one), are by far more quantitative (58% used survey, 26% grades, and 20% school data) than qualitative (17% used interview and 10% observation), and did not (72%) adopt structural approaches. Because the articles reviewed come from journals of national biology professional societies, they generally come from authors who have primary disciplinary training in the biological and life sciences. We speculate that training primarily in a biological discipline drives a bias for quantitative methods (over qualitative or mixed-method approaches), given their ubiquity and familiarity in these fields. This would also be true for administrators who move into positions of leadership from their academic units.

As the recognition of social factors in learning continues to meet with the priorities of institutions to use data to promote student retention, it is important to consider the range of perspectives available to inform decision-making. BER would benefit from an expanded analytical and methodological repertoire. Including qualitative approaches (e.g., those from sociology and anthropology), ideally through interdisciplinary collaborations, could increase the ability to explain phenomena and support novel applications. This can further help scholars take into fuller account the diversity of marginalized groups and look beyond deficit approaches.

A consideration of additional levels of analysis and an understanding of the various sociological and anthropological perspectives may therefore require additional professional development to build awareness and collaboration. While there is a great need for the inclusion of qualitative and mixed-methods approaches in this work, we would recommend that, rather than embark on independent training in a new way of thinking and attempting to essentially take on the knowledge of a new field alone, there should be increased collaboration with those trained in qualitative methods, such as sociologists, anthropologists, and psychologists, who likely teach and work with the same students on the same campus as biology educators. Collaboration between those with a background in quantitative methods and biology, in combination with those trained in qualitative methods, will not only help biology education researchers to learn about research in this field but likely strengthen the work of researchers from these varied disciplines.

In addition, our review of these articles showed that we did not find a large number of empirical articles on diversity, equity, and inclusion. In examining the aims and scopes of the national biology education society journals we investigated, we observe that theoretical lenses and methodological practices are not emphasized. Instead, the focus is on the application of the research to education in the biological sciences. Thus, we suggest that biology education research journals that want to impact diversity, equity, and inclusion may want to consider expanding their aims, scope, or other criteria in order to draw on a wider and more diverse body of scholarship.

As shown, the biology education research in these journals draws heavily from psychological perspectives that are not disaggregated, are quantitative, and do not generally use structural approaches. Informed by broader changes in other STEM education research fields, it is increasingly clear that understanding the social aspects of higher education environments will be critical to broadening participation, particularly when it comes to PEERs.

There are some limitations to this work, many of which are balanced based on the goals we sought to achieve in this study. For example, analyzing only abstracts (rather than full articles) is a weakness, given that we are not able to make any inferences on the composition of samples in papers that did not specifically mention these concepts outside of their abstract. Papers that did not emphasize disaggregating by race in their abstract, for example, might have actually done so in the body of the paper. Yet, identifying research methods and explanatory frame/level are more common attributes/expectations of research abstracts. We balanced the feasibility of sample size with the robust coding schema and methodologies we sought to use, establishing a high degree of intercoder reliability and benefiting from the expertise of numerous coders and a larger team in reaching consensus on the coding.

In addition, our work is limited by the search terms used. We note that the initial search terms used to gather articles for analysis are not exhaustive and represent an informed and robust point of departure to make the claims we have made. Additional work using terms like “marginalized,” which are often favored by social scientists, especially more recently, would likely expand the pool. However, given that our search term creation was led by two biology education experts with decades of experience, we have confidence that they speak to the terms used by the community.

Our choice to include only the main biology education research journals created by societies eliminated BER work in other journals from being considered. For instance, we note that Dirks (1) found a larger quantity of BER articles in DBER journals not included in this study, such as the Journal of Research in Science Teaching and the Journal of College Science Teaching. Yet, we wished to focus on journals particularly studying BER, and journals such as the Journal of Research in Science Teaching and the Journal of College Science Teaching will, by virtue of their broader STEM remit, attract a different audience than BER-focused journals. Similar to the STEM education-focused systematic analysis by Irwanto et al. (7), we found that quantitative approaches were more common. These studies can provide added nuance to how BER research exists more fully. This speaks refers back to the opening of the present manuscript, which emphasizes the journal articles we chose due to their influence on biology educators and researchers specifically.

While this review provides valuable insights into the current state of DEI-focused biology education research within selected national professional society journals, it also highlights several important limitations. Future work can address these limitations in several ways. First, researchers could expand the analysis to include the full text of articles—not just abstracts—to capture deeper insights into sample compositions, research methods, and nuanced theoretical frameworks that may not be apparent in abstracts alone. Second, future studies could broaden the range of search terms beyond those used in the current study (e.g., “marginalized,” “inclusion,” “belonging”) to ensure the inclusion of articles that may have employed different terminologies to describe DEI efforts. Third, expanding the scope to include additional journals—particularly those that are interdisciplinary or focused on STEM education more broadly—would capture a more comprehensive picture of the field’s methodological and analytical diversity. Lastly, complementing this type of systematic review with qualitative interviews of authors or content experts could provide important context for understanding the motivations behind methodological choices and the practical constraints of implementing different research approaches.

Although this review focused on biology education research journals published by national biology societies in the United States, the patterns identified—such as a dominant quantitative orientation, aggregation of marginalized groups, and emphasis on individual-level analysis—mirror critiques in other STEM education fields. For instance, Ding (85) argued that quantitative physics education research (PER) often overlooks its theoretical foundations, privileging measurement and controlled experiments over understanding ontological and epistemological assumptions. Subsequent PER scholarship has urged greater inclusion of qualitative and mixed methods to capture rich, context-sensitive insights (86, 87). Similarly, engineering education research has been critiqued for favoring large-N, positivist paradigms that marginalize small but critical populations. Scholars have advocated for small-N qualitative and intersectional approaches to better represent underrepresented students (88, 89). Moreover, international studies in STEM education show how integrating qualitative data with quantitative findings can provide more nuanced understandings of equity, especially in diverse cultural settings (see, e.g., Aikenhead and Ogawa [90]). These critiques indicate that the methodological and analytical trends identified in this manuscript are not unique to U.S. biology education research but reflect broader challenges and opportunities across STEM education internationally. Our findings may thus inform cross-disciplinary and cross-national efforts to embrace methodological diversity, disaggregate participant data, and incorporate structural-level analyses to effectively address systemic inequities in STEM education.

In closing, we suggest that future research into this matter peel back additional layers to uncover more details about the specific theoretical lenses (or lack thereof) employed in biology education research. By understanding more about the extent to which such lenses, as well as the types of lenses that are utilized, those engaging in biology education research (often biologists) can engage in targeted training to develop these skills to advance inclusion. To address the identified gaps in methodology and theory, future biology education research (BER) should prioritize mixed-methods approaches that bridge the strengths of both quantitative and qualitative methods, leveraging their combined capacity for statistical generalizability and providing nuanced insights into lived experiences. Integrating sociological and anthropological theories can further enrich DEI-focused research by highlighting systemic inequities and offering actionable pathways for change. While outside of the timeframe of this research, the inclusion of work conducted in the aftermath of George Floyd’s murder offers a more visible lens into systemic inequalities. However, we note that there were a number of publicized murders—Michael Brown and Eric Garner—prior to our work that did not seem to have a large influence on research in the field, as indicated by our project’s findings. We do note the highly visible nature of George Floyd during the time of COVID when people that may not have otherwise been tuned in had a front row seat to these systemic issues. We propose that this heightened visibility may influence future research directions and bolster efforts toward meaningful structural change. To foster interdisciplinary collaboration, BER practitioners should consider partnering with researchers trained in qualitative and structural approaches, thereby ensuring methodological rigor and expanding the field’s capacity to address systemic challenges. This could help transform the field of biology education and usher in research and consequential changes to advance inclusion in new and important ways.

AFTER EPUB

[This article was published on 1 October 2025 with the incorrect affiliation for Yiwen Zha. The affiliation was updated in the current version, posted on 15 October 2025.]

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author upon request.