We Are What We Read: Assessing Bias in the Implicit Curriculum of a Social Work Program

: A school of social work devised a process to assess the implicit curriculum by auditing the required readings to identify the race and gender of the authors. As a profession, we espouse a strong commitment to social justice and diversity. Yet we know that there are limitations to our objectivity and that auditing is a valuable tool that can reveal biases. The concrete data provided by an audit can help reveal and disrupt entrenched patterns. The audit was conducted by reviewing the syllabi for required BSW and MSW courses. For each text, we collected the names, gender, and race for each author. Across all programs, authors were disproportionately White as compared to the general U.S. population, professional authors, professional social workers, and students in the programs. Similarly, men were over-represented as compared to all of the benchmarks, except for the authors in the BSW program, which was more feminized as compared to the U.S. population. This assessment process adds to the existing toolset by measuring current levels of representation—including over and underrepresentation. It is hoped that auditing will prove an effective tool for doing antiracist and anti-oppressive assessment, however an audit can only reveal where work is needed.

The best elements of social work's legacy are the values of social justice and human rights (Murdach, 2011). Today these values are still found in the profession's foundational documents. However, a history of strong activist leadership is not enough; higher education, and social work specifically, is not part of the vanguard of contemporary social justice movements (Healy, 2008). Social work, as a profession and in the academy, often fails to confront inequality and instead aligns in practice with neoliberalism, individualism, and corporate entities while providing lip service to the values of social justice and equity (Corley & Young, 2018). Higher education offers a lot of symbolic talk about commitment to diversity, but the literature suggests there is a gap between rhetoric and practice (Bonilla-Silva & Baiocchi, 2001;Harper, 2012;Shapiro et al., 2017). This gap creates a tension between social work values and actions, and may fail to produce professionals that are prepared to work in a diverse society or challenge inequalities (Mehrotra et al., 2017). This paper follows one program's assessment process, attempting to identify aspects of their implicit curriculum-specifically the race and gender of assigned textbook authors. The authors are cognizant of the legacy of inequality and, in particular, White supremacy and patriarchy that are endemic to American culture. Accordingly, we thus hypothesize that despite a commitment to social justice, social work educators will disproportionately select textbooks written by authors with dominant racial and gender identities.

Literature Review
Addressing the challenge of achieving equal opportunity and justice is difficult to imagine without also considering how inequalities are maintained within higher education. Most campus norms currently support elective opportunities for those seeking antioppressive and antiracist coursework. Yet, few places require anti-bias classes or cocurricular programs (Hart Research Associates, 2015). Underneath the commitment to diversity messaging, there seems to be an even greater commitment to guarding tradition and business as usual. Some professions, like social work, do mandate that students demonstrate competency in "diversity and difference" although they fall shy of requiring a discrete class. Nor is there currently a mandate calling specifically for antiracist, antioppressive curriculum. Instead accreditors continue to use the concepts "diversity and difference" which does not explicitly call for the dismantling of white supremacy and systems of oppression. The accrediting body, The Council on Social Work Education (CSWE; 2015) defines diversity as the following: The dimensions of diversity are understood as the intersectionality of multiple factors including but not limited to age, class, color, culture, disability and ability, ethnicity, gender, gender identity and expression, immigration status, marital status, political ideology, race, religion/spirituality, sex, sexual orientation, and tribal sovereign status. (p. 7) While social work may be more robust in their requirements than other disciplines, scholars, social workers, students, and critics are vocally demanding reform and a more significant commitment to anti-oppressive work (Abrams & Moio, 2009). The CSWE annual survey of accredited programs reports that the United States has more than 800 accredited social work programs (CSWE, 2017). Social work scholars have cited the need for stronger anti-bias standards and measurement tools appropriate for accreditation and program evaluation (Abrams & Moio, 2009;Brown, 2016;Carnevale et al., 2018). While these calls for change and awareness from scholars are encouraging, systemic inequality in higher education outcomes remains (Espinosa et al., 2019;Jack, 2019). The lack of exposure to antiracist and anti-bias content is, in part, due to the historic inequalities built into the system of higher education (Jack, 2019;Yee, 2016). The landscape is full of aging institutions with generations of experience with the status quo. The history of disparity, discrimination, marginalization, and oppression within higher education is substantial (Byrd, 2017;Jack, 2019;Warikoo, 2016).
The profession of social work and social work education have historically held a certain discomfort towards the academy, remaining wary of the elitism and inequality built into the system and a nervousness about being co-opted, yet also struggling for professional acceptance (Healy, 2008;Olson, 2008). These tensions are located around the values of social justice and a more radical vision of a just society versus those who have sought to build a credentialed profession with similar structures and hierarchies found within the larger adjacent system of health care (Olson, 2008). Some scholars have written that this trend towards the clinical has led the profession away from its social justice roots (Specht & Courtney, 1994). Others suggest that debate continues to exist within social work about whether the profession serves the more social justice-oriented human rights agenda or the micro practice-oriented, human needs agenda (Murdach, 2011).
Institutions are tasked with providing the next workforce, and colleges and universities are eager to demonstrate their commitment to diversity in response to changes at the grassroots level. Social work education, like higher education in general, is moving quickly to address the call for anti-racism initiatives-at least in rhetorical terms. However, the toolset remains undefined and limited (Bogo & Wayne, 2013;Grady et al., 2011;Peterson et al., 2014). Universities and social work educators need new and better ways of measuring equity work, including finding common definitions and reliable ways to quantify this work (Abrams & Moio, 2009;Bell & Hartmann, 2007). At the same time, accreditors are increasingly demanding detailed reports. Auditing readings offers a new and useful tool for assessment writers to respond to these demands.
A primary stakeholder in this discussion, CSWE, has a robust interest in understanding inequality, including a center for diversity and social justice. As an accrediting body, they also include diversity within accreditation standards and competencies but fall short of clearly articulating these goals in ways that align with anti-oppressive scholarship and best practices. Currently, social work educators can use any approach to teach this content, including outdated modes, in part because the competencies are so ill-defined. Programs are also asked to address the implicit curriculum, but diversity is just one of many means of completing this requirement, and this too is a liminal area without clear standards or measurements (Bogo & Wayne, 2013). Unlike the practice content, each school is largely left to its own devices concerning the qualifications of the instructors charged with teaching diversity and inclusion material, and there is too little research about these courses (Mehrotra et al., 2017). Still, other programs offer electives, but students can easily miss this so-called "elective" content. Students themselves recognize the value of diversity skills and content (Larson & Bradshaw, 2017). Recently Columbia University's MSW program experienced student protests demanding stronger anti-oppressive curriculum (Williams & Reidel, 2018). This highlights the important role that CSWE plays as a critical leader in setting standards and preparing the next generation of social workers.

A Case for Auditing
Within the higher education landscape, the diversity lens used by CSWE is commonplace. The current landscape for diversity work on campuses includes a variety of tools and vendors. There are consultants, climate surveys, trainings, and workshops, among other options for schools to choose from. Auditing does exist in this context. One mainstream example, from the Annenberg Inclusion Initiative's audit of recent topgrossing films, revealed substantial underrepresentation of women and received national press attention (Smith et al., 2018). Still, audits are not regularly included as mainstream tools, and accreditors have yet to embrace the practice as part of the diversity assessment in cyclical reports. The existing tools are helpful and should continue to be used. Still, they do not provide the baseline information of an auditing process. Many current measurements look at specific diversity offerings and talk about campus climate, but more normative aspects of the implicit curriculum are less scrutinized. This greater campus climate-which is considered a combination of environmental factors and social relations-is directly related to the implicit curriculum. Relying solely on climate surveys to assess the implicit curriculum can inadvertently produce a misreading of the campus climate, leading administrators to believe the climate is more positive and inclusive than it is (Hurtado et al., 2008). While there are several definitions of the implicit curriculum, long-time education scholar, Eric Eisner, provides a succinct description, "The implicit curriculum of the school is what it teaches because of the kind of place it is" (Bogo & Wayne, 2013, p. 5). The hidden or implicit curriculum is a challenging area of campus assessment, yet accreditors, including CSWE, are beginning to direct programs to address it (Bogo & Wayne, 2013). However, this direction has not arrived with clear standards or measurement tools despite calls from the literature (Grady et al., 2011;Peterson et al., 2014). The auditing process can help schools and programs examine the implicit curriculum and measure their learning environments from various vantage points. This paper proposes building on the concept of implicit bias auditing to gather information specifically about the organization (Milkman et al., 2012). Implicit bias auditing is the process of examining the social and media choices at the personal level to reveal patterns of self-segregation in a variety of categories. The auditing process relocates and expands this idea for use at the organizational level. However, the basic premise of auditing remains the same-to make the invisible and hidden clear and discernable. In this study, we consider textbook author identity and representation as a significant element of what has been called the hidden or implicit curriculum (Bogo & Wayne, 2013). We did not analyze the content of the texts-the explicit academic content-but instead the race, gender, and discipline of the authors. Our analysis is not concerned with the topics, theories, or content of the required readings as we understand that this is part of the explicit curriculum. In the same way that the diversity of the faculty contributes to the implicit curriculum, the diversity of the authors contributes to the implicit curriculum.
Creating standards and measurement techniques that make it easier to identify disparities aligns with social work's earliest values (Murdach, 2011). For social work educators, scrutinizing how to best prepare students for the field is regular work. Professional schools with predictable accreditation cycles are uniquely well-positioned to engage in meaningful assessments that will help them see and address these often hidden and damaging patterns of White supremacy, patriarchy, and other systems of oppression. It is clear that believing in diversity and social justice are not enough to create equitable learning environments. Critical self-reflection and tangible measures are needed at the teacher, program, and college-level, to reduce inequalities in the implicit and explicit curriculum. Producing an accurate picture of the implicit curriculum, that can be used repeatedly to measure current program performance and change over time, is an important tool for program renewal and creating truly anti-oppressive learning environments.

Theoretical Research Framework
By focusing on outcomes of behavior, auditing provides a countermeasure to methods that seek to understand psychological or interpersonal expressions of racism and sexism that center on intentionality. A growing body of scholarship argues that we should eschew focusing on motivations or intentions but instead on the behaviors and artifacts themselves-policies, structures, and practices that, intentionally or not, reproduce inequalities by implicitly or explicitly centering on the best interests of those in power (Kendi, 2019). This disregard for intentions or the implicit vs. explicit nature of inequality replication is particularly useful in measuring the actual behaviors of neoliberals who claim a post-inequality frame and for espoused adherents to social justice, like social workers, who claim to have an anti-inequality agenda (Roberts, 2002).
Instead, we rely on QuantCrit or quantitative inquiry that is grounded in Critical Race Theory (CRT) and assumes that racism is pervasive and impacts structural and individual behaviors (Garcia et al., 2018). Originating in legal studies, CRT historically used qualitative methods, including narratives, storytelling, and content analysis, to identify the subtle ways that white supremacy is replicated in the post-civil rights era. CRT's rejection of objectivity within our racist society makes the use of seemingly positivist quantitative methods an unlikely tool of inquiry. Scientists, and of particular import to our work, social scientists, have long used the purported objectivity of science and quantitative research to demonstrate the racial unfitness of people of color and to naturalize the domination of Whites. Because of this history and its continuance by contemporary researchers, a postpositivist approach to quantitative research that focuses on the structural, historical, and personal liminality of objectivity is the only way to integrate CRT and quantitative research (Crawford, 2019).
Auditing the selection of textbooks aligns with CRT as it attempts to understand the subtle everyday behaviors that reinforce white supremacy and the emerging subfield of QuantCrit scholarship. The tenets of QuantCrit are laid out by Gillborn and colleagues (2018) who identify: 1. The centrality of racism as a complex and deeply rooted aspect of society that is not readily amenable to quantification; 2. The acknowledgment that numbers are not neutral and they should be interrogated for their role in promoting deficit analyses that serve white racial interests; 3. The reality that categories are neither "natural" nor given and so the units and forms of analysis must be critically evaluated; 4. The recognition that voice and insight are vital: data cannot "speak for itself" and critical analyses should be informed by the experiential knowledge of marginalized groups; 5. The understanding that statistical analyses have no inherent value but they can play a role in struggles for social justice. (p. 158) This evaluation of assigned readings in one social work program's curriculum is grounded in self-reflexivity and its liminalities, because we teach within the program we are auditing. We do not trust that our own antiracist and feminist ideologies are producing reading lists that center people of color and women. Using a measurement tool, rather than relying on our perceptions, accomplishes the goal of challenging the taken for granted ways that inequalities are replicated through daily choices within the academy. Focusing on who gets to contribute to the chorus of voices in the classroom by auditing and strategically making new textbook selections to include more diverse authors also aligns with the CRT goal of questioning dominant ideologies and utilizing counternarratives (Solórzano, 1997). The process of auditing ensures these notions are tested and makes space for counternarratives.

Method
During the 2017-2018 school year, an audit was conducted of all of the required readings in one social work department that offered both BSW and MSW programs. The BSW program is a residential program located in the rural western U.S. The MSW program consists of a two-year program made up of a generalist year and clinical year. The clinical year of the program is also taken as a one-year advanced standing clinical program for entrants who have strong grade point averages and graduated with BSW degrees from CSWE accredited programs. The MSW programs, both the one and two-year programs, are taught in three different locations (sites A, B, and C) in two states in the rural western U.S. To account for the two programs and three sites, we assessed each program option and each location separately to identify the particularities of each year of the program and on each campus.
Syllabi for every required course were included in the sample. The syllabi follow a singular format, with one section labeled "Required Texts." All books listed in this section of the syllabi were included in the sample. Arguably, this did not represent every book and article required for a class, but it did capture those readings deemed important enough to name in the "Required Texts" portion of the syllabus. The BSW program included 30 books written by 52 authors over four years of the curriculum. The first year of the MSW program at site A had 19 books by 33 authors, the second year at site A had 18 books by 29 authors. At site B the first year courses required 25 books by 39 authors, and the second year at site B had 31 books written by 53 authors. The first year at site C had 15 books authored by 22 authors, and the second year at site C had 27 books and 41 authors.
All of the titles and author names were recorded in an Excel spreadsheet. Next, a researcher identified each author and assigned each a gender and race. We intended to collect information about the sexuality, religion, ability, and other elements of identity for each writer, but found it too difficult to consistently locate this information for many of the authors. The invisible nature of some of these elements of identity made it impossible to infer their identity with any reliability. Assigning gender and race information to each author followed this multi-step process. Researchers first visited the publishers' and authors' personal webpages to determine gender and race. Gender was determined by first looking for clear statements in author biographies, the use of pronouns, and lastly, photographs of the authors. All of the authors identified as either male or female.
The same process was followed for race, where explicit self-identifications were sought out, followed by using photos to assign race to the authors. In a few cases, biographical information did not specifically name an author's race, and photos were unavailable. In these instances, we used contextual clues to assign their race. For these individuals, their race was often inferred as White because they attended traditionally white universities in the mid-twentieth century (or before), their biographies and often obituaries did not contain any information that would mark them as anything other than White. The unmarked nature of Whiteness means that it is often not explicitly named as part of one's biography or identity (McIntosh, 1988). The authors included in the study were assigned one of the following races: White, Black, Asian American, Latinx, Native American. A multi-racial or other category was not used, so authors like Malcolm Gladwell, who has a White parent and a Black parent, is identified as Black in our assessment. Our assessment revealed only one multi-racial individual, so we chose not to create a dedicated category for a single person.

Bachelor of Social Work Program
In the BSW program, most of the books were authored by women; 55.77% of authors were women, while 44.23% were men (see Table 1). The majority of the authors of the required readings were White. In our sample, 84.62% of all authors were White, 5.77% were Latinx, 5.77% were Black, 1.92% were Asian American, and 1.92% were Native American (see Table 2). Table 1 shows the genders of authors from the required readings across programs and program sites. At site A, in the generalist year, 51.52% of books are authored by women and 48.48% by men, and in the clinical year, 44.83% of authors were women, and 55.17% were men. At site B, in the generalist year, 33.33% of authors were women, and 67.67% were men. In the clinical year, 35.85% of authors were women, and 64.15% were men. Women authors were much more underrepresented in the authors of books at site B. At site C, 22.73% of authors in the generalist year were women, and 77.27% were male; in the clinical year, 43.90% were women, and 56.10% were men.

Master of Social Work Programs
The race of the authors varied across sites and program concentrations (see Table 2). At site A, in the generalist year, the authors' races were as follows: 87.88% White, 6.06% Latinx, 6.06% Black, 0% Asian American, 0% Native American. In the clinical year at site A, 92.86% were White, 7.14% Latinx. Zero authors were identified as Black, Asian American, or Native American. In the generalist year at site B, 97.44% of authors were White, 0% Latinx, 0% Black, 2.56% Asian American, and 0% Native American. At site B, in the clinical year, 90.57% of the authors were White, 1.89% Latinx, 1.89% Black, 5.66% Asian American, 0% Native American. At site C, in the generalist year, 95.45% were White, 4.55% Latinx, 0% Black, 0% Asian American, 0% Native American. At the same site in the clinical year, 95.12% of authors were White, 4.88% were Latinx, 0% Black, 0% Asian American, and 0% Native American.

Representation of Dominant Identities in Textbook Selection
The ongoing racial justice and #Metoo movements have made clear that Americans need to do better at understanding the insidious functions of White supremacy and patriarchy. Evaluations of institutionalized, structured power are necessary before justice can be sought out. We believe that this auditing assessment is a substantial aid in understanding patterns of white supremacy and patriarchy as it quantifies the sum of our choices and moves them from implicit to the realm of explicit-where they can fundamentally be addressed. This paper examined the program assessment of the implicit curriculum at a multicampus social work program. The hypothesis was supported as educators did disproportionately select books written by authors who were male and white. The proportionality was measured based on the demographics of students in the programs as self-reported at admission, the social work profession, the general U.S. population, and those who identify their profession as author or writer on the U.S. Census.

Race
The study supports the hypothesis regarding race, finding that across all programs and sites, there was an overrepresentation of White authors as compared to U.S. Census data on national and professional author demographics, professional social workers, and the racial demographics of the students enrolled in the BSW and MSW programs. In our sample, the racial diversity of the authors of the required readings are very White, at 84.62% of all authors in the BSW program and 93.06% in the combined MSW programs. Latinx authors were included in the BSW required readings at a much higher rate compared to the combined MSW programs at 5.77% and 3.70%, respectively. Black and Asian American authorship was low at the bachelor and master levels. Three Black authors and one Asian American author were assigned as required texts in the BSW program. Of the 216 authors assigned in the combined MSW programs, only four were Black, and three were Asian American. One of the MSW sites did not assign any required readings by Black or Asian authors. BSW students had merely one book authored by a Native American. Native American authors were not represented at all in the MSW program. Despite Native American enrollment in all the programs on each campus. Of note, Native Americans represented 12.5% of student enrollment on one of the campuses.
This overrepresentation of White authors is mirrored in the national data from the Census Bureau, which indicates that 86% of all authors are White, 5.7% are Black, 4.5% are Asian American, and .4% are Native American (Data USA, 2018). This census data on authors only includes race and not ethnicity, so Latinx individuals are included in this data based on their racial identities. This makes for a difficult comparison of data. The racial demographics of the U.S. in 2018 are as follows: 60.7% White, 18.1% Latinx, 13.4% Black, 5.8% Asian American and, .2% Native American (U.S. Census, 2019).
Among active social workers with bachelor's degrees at the national level, 67.5% are White, 25.7% are Black, 1.8% are Asian American, 1.2% are Native American, 1.9% are another race, and 2.1% are two or more races (Salsberg et al., 2017). Among this group of BSW-prepared social workers, 89.3% identified as not Spanish, Hispanic, or Latino, and 10.7% designated that they were Spanish, Hispanic, or Latino. The racial demographics for active social workers with a master's degree or greater are 72.6% White, 19.1% Black, 3.3% were Asian American, .5% were Native American, 2.0% were some other race, and 2.5% were two or more races. Among the social workers with at least a master's degree, 9.5% identified as Spanish, Hispanic, or Latino (Salsberg et al., 2017).
As Table 2 shows, in the combined BSW and MSW programs, 75.67% of graduates were White, 11.03% were Latinx, 1.65% were Black, 2.15% were Asian American, 6.34% were Native American, .25% identified as multiracial, and 2.91% of graduates did not specify a racial or ethnic group. Across all programs and sites, White authors were selected 91.42%, compared to 4.1% Latinx, 2.24% Black, 1.87% Asian, and .37% Native American authors. Races of authors are compared to the races of graduates on a given campus, revealing noteworthy gaps in representation. In the BSW program, 26.88% of graduates were Latinx; however, only 5.77% of authors were Latinx. At site A, 16.84% of graduates were Latinx, while only 6.06% of generalist text authors and 7.14% of clinical year authors were Latinx. We see a similar disparity at site B, where 12.5% of our graduates were Native American, and there were no Native American authors in the sample.
Using all four measures-students, professional writers, professional social workers, or the general US population-the textbooks in this sample all exceeded proportionality in the assignment of White authors except for one data point. The 84.62% White authors assigned in the BSW program is lower than the percentage of White professional authors at 86% (Data USA, 2018). The authors are more White than the student body, across all programs. This varies by site and year, with the authors in the first year of site B almost exclusively White, with only one Asian American author in the sample (see Table 2).
Examining the race of authors provides a singular but important piece of data about who has the authority to construct knowledge within a course. It is acknowledged that some of the authors included in the reading list are committed White antiracists and their books focus on issues of diversity and equity. Their voices are important too, particularly for our White students who need antiracist White role models. That said, we align ourselves with CRT and believe that people of color and other oppressed individuals should be able to tell their own stories (Solórzano, 1997). We need to address this disproportionality. Students should be able to see themselves reflected in the curriculum. This is happening for our White students, but not our students of color.

Gender
When examining the authors' gender identity, the study found that the majority of the authors for the combined MSW programs were men at 60.83%. The BSW program audit found that most of the authors were women at 55.77%. The percentage of women authors in the BSW program does exceed the national population estimates for 2019, which indicated that 50.8% of people in the U.S. were female (U.S. Census, 2019). Additionally, the Census Bureau also indicated that 61% of all authors in 2016 were women (Data USA, 2018). In total, across all programs and sites, males composed 57.62% of all authors. This is an overrepresentation of men as compared to the national demographics of the profession which indicate that 83.3% of professional social workers were women in 2017 (Salsberg et al., 2017). The gendered representation of our students aligns with the national data as 84.95% of BSW students are women, and 83.19% of all MSW students are also women. Using these benchmarks, only the BSW program exceeded proportionality in just one of the four measures-as compared to the U.S. population. Therefore, we concluded that the outcome supported the hypothesis that a preference for men existed in the book selection process for both the BSW and MSW programs.

Implications and Limitations
The reading audit harvested institutional information to give course decision-makers a richer understanding of the program and their individual choices as teachers. Often privileged voices are accepted without question or understood as neutral (Phillips & Lowery, 2018). The audit of assigned readings tells a story about whose voices matter to teachers and in constructing their students' understanding of content and the world around them. Students spend hours throughout their education reading the books assigned by their professors. The authors assigned telegraph to students, and the campus community, who is worthy of an audience and who gets to be a creator of knowledge. The practice of auditing for race and gender will help decision-makers confront choices and see who is getting too much time at the proverbial mic and who has been neglected. While many decisions about texts are made one at a time over the course of the year, the audit aggregates these decisions to make representational patterns clearer and easier to see-across courses and programs of study.
For assessment and accreditation, additional areas can be included in an audit of textbooks, such as author discipline. In an interdisciplinary field like social work, this can be very helpful in understanding what views the program may be intentionally or unintentionally emphasizing. Faculty members are often completely unaware of their peers' course material selection or the overall implicit messaging of their department until an audit is completed. Although the auditing tool was designed to measure social work programs, it is possible to imagine other useful applications. Colleges could elect to audit their general education programs as part of an internal review or accreditation. These audits could be completed by external researchers as schools use websites as an archival repository for syllabi, making auditing from a distance more feasible. Similarly, an audit of authors of the most commonly used textbook publishers could reveal needed information about the demographics of current authorship.
Audits can highlight patterns and trends that were there all along, but so normative, they went unnoticed (Smith et al., 2018). While this is useful to disrupt norms, there are a few potential negative effects to consider. First, using auditing to achieve better representation is a goal of the auditing process. Improved representation, measured in numbers, may not achieve deeper goals like adopting an antiracist perspective or an awareness of toxic masculinity. Some may view simply getting the numbers right as the whole picture, instead of a part of the picture. This kind of approach could lead to another negative consequence, manipulating the numbers to appear to be achieving race and gender representation while preserving the status quo. Social workers are ethically driven by the National Association of Social Workers Code of Ethics (NASW, 2017) to empower and advocate for oppressed groups, but not all in higher education have a code of ethics requiring this commitment.
Organizations may prefer not to know what the performance of the institution is in terms of representation. Some research suggests that the dominant group is often invested in ignorance or maintaining "racial innocence" as a form of plausible deniability and to maintain existing privileges (Phillips & Lowery, 2018). Norms are durable and resistant to change. As social phenomena, racism and sexism are responsive to changing social standards and adapt quickly (Berdahl et al., 2018;Bonilla-Silva, 2006). Secondly, accreditors may decide to select other mechanisms for measuring the implicit curriculum. Ideally, the auditing tool would become an accepted process of accreditation assessment.
Auditing is a straightforward process, but assigning gender and race to individuals who do not explicitly identify themselves is an instance where the researchers' decisions impact the results. Using a gender binary with only male or female as an option should be addressed in future audits. Similarly, not using a multi-racial category produced results that emphasized one element of racial identity over others. This was true in racializing Malcolm Gladwell as Black and not multi-racial.
Lastly, as researchers and program administrators, we elected to study our own four programs and to pilot the auditing process on our courses and faculty. This certainly is a limitation that impacts our ability to approach the work with neutrality. Yet, as discussed above, we take a postpositivist theoretical approach that begins without an expectation of objectivity. We do not know if these findings hold beyond our program. Virtually nothing is known about the diversity, or lack of, among textbook authors in the academy or specifically social work education. We believe that the process was useful for our critical self-reflection, particularly the ability to name our biases, and for targeted program renewal by consciously selecting textbooks written by authors with a broader array of social identities.

Conclusion
The world of higher education was already facing substantial pressures, including higher costs to provide education and smaller pools of potential applicants before COVID-19 forced institutions to send their students home and complete the school year entirely online. At the same time, the nation has finally woken to the truth of structural racism across the country. In particular, the murders of Breonna Taylor, Ahmaud Arbery, and George Floyd have inspired protests, rallies, and national soul-searching that has been compared to the Civil Rights era (Berry, 2020). The Black Lives Matter movement was launched in response to the death of Michael Brown in Ferguson, Missouri, in 2014, yet, early public opinion polls and reports indicated that most Americans did not align themselves with the movement until recently. In the run-up to the 2020 U.S. Presidential Election, polls suggested that opinion has changed dramatically, especially among political progressives and liberals (Yglesias, 2019). As the national conversation around racial justice may wax and wane, the current context within academia appears supportive and eager for new tools with the potential to answer the harder and more critical questions students, alumni, and accreditors are sure to be asking soon.
Much more work is needed within higher education to break down the deeply entrenched norms of patriarchy and White supremacy. Even forty years after the Civil Rights movement and the passage of critical pieces of legislation, there is evidence that access to higher education-especially elite higher education-may have decreased for students of color in recent decades (Espinosa et al., 2019). Moreover, students are aware that they need skillsets for working with diversity and difference in their future workplaces and are demanding more preparation (Larson & Bradshaw, 2017;Warikoo, 2016). Social work education has made a start by developing a stand-alone competency for diversity and requiring programs to specifically address diversity in their self-assessments. If schools want to close the divide between rhetoric and practice and embrace diversity, equity, and inclusion as an integral part of the implicit and explicit curriculum, additional tools and approaches are needed. The hope is that the audit will help to uncover the unconscious biases of faculty, stimulate further research, and facilitate meaningful course and program level change that is grounded in creating more just educational programs that reflect the intersectional diversity of our students.