Discipline-Based Diversity Research in Chemistry

Conspectus

We introduce the term discipline-based diversity research (DBDR) to capture the emerging field of research advancing diversity, equity, inclusion, and belonging with specificity to a given discipline. Contextualizing a human dynamic through a disciplinary lens has already given rise to discipline-based education research (DBER). The modalities through which students and practitioners think and process information are a reflection of a given discipline, and it tends to give rise to its professional practices. Through DBER, such specification is necessary in addressing evidence-based practices that are effective for teaching a particular subject. Likewise, the inequities and opportunities within a given field (and its professional culture) must be addressed within a disciplinary lens. Thus, the findings from social science in diversity in arbitrary contexts must be analyzed, interpreted, applied, and researched within a given discipline.

One specific challenge to academic chemistry is the lack of inclusion in the sense that the faculties in research-active chemistry departments are far from diverse. We recapitulate the percentage of women and under-represented person of color (URPOC) professors over the past 20 years reported by us and other sources. The data admits to linear fits with high confidence. Assuming this linearity holds, the gender gap in representation would be bridged only in 2062, and the threshold of 20% of the faculty as URPOC would be reached only in 2113. While the community has actively engaged in modifying practices and procedures to redress this grim projection, it should be clear that more needs to be done.

Toward this objective, we have been driven by the top-down hypothesis that solutions must be led intentionally through the top—that is, by department heads and chairs—because they are the stewards of the infrastructure. Department chairs and the chemistry community have engaged in DBDR through biennial workshops—that is, through the Open Chemistry Collaborative in Diversity Equity (OXIDE)’s National Diversity Equity Workshops (NDEWs)—to survey and evaluate existing policies and practices aimed at advancing inclusive excellence. This has led to research-based recommendations for the implementation of solutions in chemistry departments. This includes (i) engaging in community, (ii) conducting authentic and open searches, and (iii) recognizing and rewarding inclusive excellence. What makes them DBDR in chemistry is that we have to articulate and contextualize these solutions in terms of practices and procedures that we conduct in chemistry, assess their efficacy, and promote them across our discipline. Furthermore, we must offer theories of change for reforming them while offering frameworks that fit within how chemists think and practice.

In this Account, we demonstrate how DBDR has taken root in chemistry, forecast where this emerging field may go, and provide a blueprint for how it might be replicated in other disciplines.

Key References

• Hernandez R.; Stallings D.; Iyer S.; Cheng H. N.; Nelson D.. The Gender and URM Faculty Demographics Data Collected by OXIDE. Diversity in the Scientific Community Vol. 1: Quantifying Diversity and Formulating Success; ACS Symposium Series; American Chemical Society: Washington, D.C.,2017; Vol. 1255, pp 101–112 10.1021/bk-2017-1256.ch004(1)Yearly and congruent longitudinal demographics faculty data was summarized and analyzed to provide key findings about the progress of diversity solutions over the preceding decade.

• National Diversity Equity Workshops in Chemical Sciences (2011–2017); Hernandez R., Stallings D., Iyer S. K., Eds.; ACS Symposium Series; American Chemical Society: Washington, D.C., 2018; Vol. 1277 10.1021/bk-2018-1277.(2)The evolution of the practice of the top-down hypothesis across under-represented groups (URGs) in the United States was reported and analyzed in the context of four National Diversity Equity Workshops (NDEWs) held in 2011, 2013, 2015, and 2017.

• Stallings D.; Iyer S. K.; Hernandez R.. Removing Barriers. In Addressing Gender Bias in Science & Technology; Azad S., Ed.; ACS Symposium Series; American Chemical Society: Washington, D.C., 2020; Vol. 1354; pp 91–108 10.1021/bk-2020-1354.ch006.(3)A consensus analysis of diversity solutions based on the expert evaluation by department chairs and diversity subject matter experts.

• Hernandez R.A Cuban Campesino in Chemistry’s Academic Court. J. Phys. Chem. B 2021, 125, 8261–8267 10.1021/acs.jpcb.1c06073(4)To our knowledge, this article introduced the term DBDR.

1. Introduction

Inclusive excellence is founded on the fact that diverse teams lead to better outcomes.⁵⁻⁹ Unfortunately, the demographics of under-represented groups (URGs)—such as women, and under-represented persons of color (URPOCs)¹⁰ or persons excluded because of their ethnicity or race (PEERs)¹¹ —within chemistry have been far from commensurate with those of the availability pools, such as the general population or chemistry doctoral graduates.¹ This leads to critical questions over what the inequitable barriers are that lead to this incommensurability and what can be done systematically to remove such barriers and move the scientific work force toward commensurate participation.

We need research to answer these questions. Many of us have turned to social science research to help us solve them, and it has indeed been essential as cited throughout this Account. However, this alone is not enough because what we learn from the social sciences has to be reframed and often reinterpreted in the context of our professional culture of chemistry, and effective solutions for changing our practices, procedures, and policies must be implemented and assessed accordingly. The familiarity of images—about the work that we do as well as the ways in which we interact with each other—like the ones shown in Figure 1 provide a framework for the metaphors we use in conveying our science to each other, and it permeates how we organize ourselves within our professional culture of chemistry. This is why we have introduced the eponymous term in this Account, DBDR.⁴ It runs parallel to the now well-established field of DBER.^12,13 Therein, we have recognized that humans engage and learn concepts in discipline-specific ways. Likewise, humans interact with each other within our professional culture of chemistry in discipline-specific ways. Thus, chemistry education research (CER) and DBDR in chemistry are parallel subdisciplines.

Figure 1.

Image is meant to convey the breath of interactions in which chemists interact, viz., in small group meetings, in the laboratory, in one-on-one meetings, and in seminars. All of these are enhanced by diversity and inclusion, and DBDR is needed to learn about the sociology of these spaces and how to make them safe. The blue, green, orange, and red globes are meant to convey the broad span of the many subdisciplines of chemical research captured within the “OneChemistry” concept¹⁸ so as to be inclusive of all of chemistry. Art credits to Rigoberto Hernandez and Douglas Behr.

In this Account, we survey the work that has emerged from DBDR in chemistry without thus far having been framed formally within a specific subdiscipline of chemistry. This has, perhaps, led to additional challenges for such research because the absence of a specific journal supporting DBER has made it easy for papers and their findings to be lost. Here, we provide an account of several ways in which the chemistry community has researched and engaged inclusive excellence and, more broadly, diversity, equity, inclusion, and belonging (DEIB). In so doing, we advocate for the recognition of DBDR as a subdiscipline of chemistry. While DBDR efforts in other science, technology, engineering, and mathematics (STEM) fields are beyond the scope of this article, such efforts focused through a disciplinary lens have also been pursued for well over a decade as can be seen in, for example, Refs¹⁴⁻¹⁷. Thus, STEM fields broadly are in a position to formalize subdisciplines in DBDR.

2. Tracking the Numbers

As chemists, we like data. In principle, one can measure the demographics of any of the subgroups of, for example, staff, doctoral students, postdoctoral students, non-tenure-line faculty, and tenure-line faculty across all academic chemistry departments. Doing so is research which has to be well-defined and commissioned. The nature of the demographic groups—e.g., with respect to gender identity, and racial and ethnic status—has to be well-defined. The categories and nature of roles vary across institutions, and clearly defining them categorically is critical. The set of institutions that are surveyed could potentially affect the results. The mechanism of data collection can also give rise to varying concerns regarding the use of sensitive personal data and can require layered degrees of monitoring from Institutional Review Boards (IRBs). Presumably, the point of the data collection is not just to identify where the demographic distributions are incommensurate with the control group—e.g., the reference country’s population or the availability pool of possible hires—at a given instance but rather how they change as a function of time. Thus, on-going research projects are required to provide longitudinal data.

In the United States, for example, the National Science Foundation collects data from department and universities and publishes them biennially.¹⁹ It does not provide detail on how the demographics are distributed within individual departments by rank or within the aggregated disciplines with respect to URGs other than gender. This led to the need for the collection and dissemination of transparent data disaggregated across departments. In chemistry, there are at least two such surveys, and both are constructed so as to ensure that they require little IRB oversight.

The Nelson studies provide such data, with respect to gender and URPOC membership, for chemistry (and many other STEM departments), but they were not entirely longitudinal and the data collection involved the assignment of demographics by the data collectors rather than self-reporting by individuals.²⁰ As none of the data is provided by the individuals themselves, their database contains no publicly unavailable data that would constitute protected personal identifiable information (PII).

The C&EN scorecard was launched to address the need for yearly longitudinal data about tenure-track faculty disaggregated at the level of individual departments.^10,22,23 Their initial data reported only gender. They avoided IRB concerns because they collected the aggregated department-level data from department heads and chairs without ever seeing or retaining individual data. The potential challenge in this approach is that there may be differences in how individual departments report their data. Out of concern that under-representation might be under-reported if members of such groups are counted in multiple departments, the survey also instituted a rule that only faculty who held more than a 50% appointment should be included in the reported data. This and the choice of interpretation of the rule that only tenure-track faculty, restricted to exclude teaching-only faculty, can lead to some variations of the reported data in given departments. Another concern is that the scorecard reported only the top 50 chemistry departments, ordered according to total federal research expenditures in chemistry, that happened to be the top 50 the year that they were reported. That is, the list changed from year to year and so did the departments that were reported. We refer to this type of data collection as yearly congruent as opposed to the yearly longitudinal tracking that reports the demographics of the same set of identified schools. For example, through an expansion of the schools whose data were collected (and reported separately on the Open Chemistry Collaborative in Diversity Equity (OXIDE) web site²⁴), a yearly longitudinal data set can be reported for a given set of schools reconstructed retroactively as long as we track enough departments to encompass such different possible data sets. Through partnership between ACS C&EN and OXIDE starting in 2010, some of these concerns were addressed. The longitudinal data from 2010 to 2017 in the representation of women professors across the faculty ranks in the top 50 schools listed in the 2015 NSF ranking are reproduced in Table 1 and compared also to the congruent data of Nelson from 2007. It is notable that since 2016 women have held more than 20% of the faculty positions, but that remains far below their representation in the broad population, viz. 50%, or even among doctoral recipients in chemistry, viz. from 38% in 2011 to 42% in 2021 (according to the NSF Survey of Earned Doctorates.) The longitudinal data from 2012 to 2018 in the representation of URPOC professors across the faculty ranks in the top 50 schools listed in the 2016 NSF ranking is reproduced in Table 2. Approximately 5% of the chemistry faculty are URPOC professors, and this trails the representation in the broad population, viz over 40% in 2020, or even among doctoral recipients in chemistry, viz. 13 to 16% in 2021 (according to the NSF Survey of Earned Doctorates.) These trends in the demographics were found to be qualitatively similar for the top 50 schools whether one obtains them yearly congruently or longitudinally relative to different sets of the top 50 departments.¹ Moreover, the trends were also similar across the top 10, top 50, or top 75, indicating that they are not unique to a particular cohort.

Table 1. Longitudinal Relative Percentage of Women and Men Professors in the (Same) Top 50 Chemistry Departments Listed in the 2015 NSF Ranking by Federally Funded Expenditures in Chemistry as Reported in Reference (21) for Academic Years 2009 and 2010 through 2016 and 2017 and Compared to the Data Reported by Nelson in Reference (20) for the 2006–2007 Academic Year.

Rank	Gender	06–07	09–10	10–11	11–12	12–13	13–14	14–15	15–16	16–17
Assistant	All		19.4%	20.3%	18.3%	17.4%	17.7%	18.2%	19.7%	20.4%
	Women	21.7%	22.6%	25.1%	24.7%	28.1%	25.1%	26.1%	27.9%	27.1%
	Men	78.3%	77.4%	74.9%	75.3%	71.9%	74.9%	73.9%	72.1%	72.9%
Associate	All		15.5%	14.8%	14.4%	15.5%	15.7%	16.4%	17.8%	18.4%
	Women	21.3%	20.5%	19.1%	23.6%	23.6%	25.5%	29.7%	26.4%	26.7%
	Men	78.7%	79.5%	80.9%	67.3%	76.4%	74.5%	70.3%	73.6%	73.3%
Full	All		65.1%	64.9%	66.4%	67.1%	66.6%	65.3%	62.6%	61.2%
	Women	9.7%	11.4%	11.9%	12.5%	13.3%	13.3%	14.6%	14.9%	15.5%
	Men	90.3%	88.6%	88.1%	87.5%	86.7%	86.7%	85.4%	85.1%	84.5%
All	All		100.0%	100.0%	100.0%	100.0%	100.0%	100.0%	100.0%	100.0%
	Women	13.7%	15.0%	15.7%	16.4%	17.5%	17.3%	19.1%	19.5%	20.0%
	Men	86.3%	85.0%	84.3%	83.6%	82.5%	82.7%	80.9%	80.5%	80.0%

Table 2. Longitudinal Relative Percentage of URPOC Professors in the (Same) Top 50 Chemistry Departments Listed in the 2016 NSF Ranking by Federally Funded Expenditures in Chemistry^a.

Rank	11–12	12–13	13–14	14–15	15–16	16–17	17–18
Assistant	7.0%	7.9%	7.2%	6.9%	8.8%	8.9%	9.6%
Associate	8.4%	8.3%	7.5%	8.1%	6.6%	6.5%	7.3%
Full	2.9%	2.9%	3.4%	3.7%	3.5%	3.6%	3.8%
All	4.4%	4.6%	4.7%	5.0%	5.0%	5.1%	5.4%

^aThese numbers are updates from the longitudinal data reported in ref (21) based on the 2015 NSF rankings and extended to include the reported values above using the raw data from which ref (10) reported slightly different congruent relative percentages (which use the rankings from a given year as a basis for determining that set’s demographics).

A visual representation of the demographic data from Tables 1 and 2 is shown in Figure 2. There are certainly fluctuations among the representation within the ranks, but the representation of women and URPOC professors in all ranks combined has been very nearly linear. The slopes for the representation of women and URPOC professors are 0.670 and 0.154 percentage points/year, respectively. Assuming this linearity holds, the gender gap in the representation would be bridged only in 2062, and the threshold of 20% of the faculty as URPOC would be reached only in 2113. Generally, such relaxations slow from linearity well before they settle on an equilibrium value, and thus these grim numbers are an underestimate unless we actively engage in changing our policies and procedures as described below.

Figure 2.

Percentage representation of women and URPOC faculty in total and disaggregated ranks across several academic years. Academic years are labeled according to the final year of the pair; e.g., AY 16-17 is noted as 2017. For women faculty, the data corresponds to that listed in Table 1 and is longitudinal for the same top 50 departments listed in the 2015 NSF data set, except for the congruent data in 2007 from the Nelson study of ref (20). For URPOC faculty, the data corresponds to that listed in Table 2 and is longitudinal for the same top 50 departments listed in the 2016 NSF data set. Fits to the representation of women and URPOC professors across all ranks have coefficients of determination, R², equal to 0.97 and 0.96, respectively, and slopes as noted in the legend.

Although not a sustained longitudinal study, a hugely impactful survey in 2003 also surveyed the top federally funded chemistry departments.²⁵ They found that roughly 50% of the faculty received their degrees or postdoctoral training from one of the “top” 5–10 departments. This lack of diversity with respect to the training of students could have a root cause in pedigree bias²⁶ and can conflate arguments of mismatches between the demographics of the faculty and the availability pool if in fact the availability pool is skewed as found by Nolan et al.²⁵ Both of these concerns have provided drivers for the emerging solutions needed to remedy diversity inequities.

Finally, a related concern is the relative participation of members of URGs participating as editors, authors, and reviewers in chemistry journals. Several publishers have now committed to gather and share the demographics of these groups. That includes ACS Publications²⁷ and the Royal Society of Chemistry.²⁸ Similar to the demographics seen in the chemistry faculties in the United States reported above, the demographics of the publishing cohorts are also not commensurate with the availability pools.

3. Persistence of Personal Bias

In the early days of the discussions about under-representation, the lack of commensurate representation was difficult to deny, though documenting it was and remains necessary, as detailed in the previous section. However, some attempted to rationalize the incommensurability through a color-blind lens, claiming that it is the result of rigorous criteria for excellence. Disproving this baseless claim thus drove a significant effort within social science broadly and DBDR in chemistry in particular.

The recognition that implicit bias²⁹⁻³² plays a significant role in all kinds of evaluations is critical both to understanding some of the drivers for the incommensurability in URG participation in chemistry faculties and to partially removing the inequities by helping evaluators better manage their implicit biases towards equity in their evaluations. In representing the work of Nosek et al.,³³ Banaji provided a useful framing in saying that “implicit biases come from the culture. I think of them as the thumbprint of the culture on our minds. Human beings have the ability to learn to associate two things together very quickly—that is innate.”³⁴ Indeed, most if not all of us have biases associated with the schemas³⁵ that we have associated with the categories demarcated by URGs, as we can readily self-assess using implicit bias tests.^36,37

The challenge in discussing bias within STEM is that scientists have been trained to think of themselves as unbiased observers. Consequently, evidence about the bias that was observed in the selection of musicians for orchestras, for example, and the steps that were needed to remedy them³⁸ may not seem relevant to chemists. Similarly remote to some chemists are the results from studies that indicate the relative inequities in the success of applicants with female names, Karen vs Kevin, or black-sounding names, Jamal vs Greg, and how the disadvantage gap grows with increasingly elevated professional titles.³⁹ To overcome the reticence in the STEM community, significant effort has been needed to raise awareness of bias, its implications, and the tools to mitigate it within academic decisions and evaluation.⁴⁰⁻⁴²

For example, a seminal finding by Schmader et al.⁴³ found that recommendations for job applicants tended to differ dramatically along gender lines between the relative use of grindstone words, viz. descriptive of effort, and stand-out words, viz. descriptive of ability. Even when this is corrected, Moss-Racusin et al.⁴⁴ found that applicants for undergraduate research science positions identified as female in applications (identical to those of male applicants) were preferentially rated as less competent and consequently less likely to be offered positions by science faculty. Interestingly, the evaluations were not statistically different when disaggregated by gender or other demographics of the science faculty, providing an example that indeed all scientists have some bias and it is often directed towards those like themselves. Furthermore, such bias is inequitably affecting their observations. Yet another application of bias stems from the effects of the genius (or brilliance) myth, which correlates genius with success. Leslie et al.⁴⁵ found that the extent that a discipline holds brilliance as a stand-out descriptor necessary for success correlates with reduced diverse participation within their faculty. This is perhaps not surprising, given the finding that women and URPOC tend to be inequitably described by stand-out words such as brilliance and genius.

Thus, bias takes many forms. Uncovering how it manifests itself within the practice of chemistry specifically and how to teach ourselves to mitigate it are necessary components of DBDR in chemistry. Meanwhile, we cannot stop at characterizing and unraveling bias if we wish to resolve the incommensurability in the demographics between the availability pool and the faculties of chemistry departments. There are far too many other barriers that must also be addressed.^35,46 Indeed, Stockard and coworkers⁴⁷⁻⁴⁹ recently reported their findings about the inequitable impacts on the career progression of women and URPOC chemists due to several barriers, e.g., inequitable financial support, peer-to-peer support, advisor support, access to official and hidden rules, and access to resources (via limited socialization), which are perhaps conflated by bias but not entirely resolved if bias is eliminated. Yet another barrier lies in the use of student teaching evaluations in rating faculty performance as it has recently been reported to lead to significant inequities for instructors across the lines of gender, ethnicity, accent, sexual orientation, and disability status.⁵⁰

4. Top-Down Hypothesis

In order to make systemic change in policies and procedures, management or governing structures must somehow be engaged. Indeed, Dobbin and Kalev⁵¹⁻⁵⁴ found that in industry, holding middle-managers accountable, whether through reporting requirements or during periodic reviews of all of the reporting units convened by their supervisor, is an effective solution at increasing diverse representation. This lesson in the context of industrial chemistry was also reported over 20 years ago.⁵⁵ The extension of this concept to academia requires the identification of middle managers and their reporting structure. In organizing the first workshop on gender equity, the committee led by Friend and Houk⁵⁶ identified the chairs and heads of research-active chemistry departments as the cohort of middle managers that could be collectively held accountable.⁵⁷ As they are all in separate universities, however, they do not report to a common supervisor. Instead, they were effectively convened by the chemistry directors of several federal agencies. In a sense, they were held accountable by a budgetary means. An additional benefit of this conflation of equity and financial drive is that it clearly emphasizes the vision as inclusive excellence. The success of the gender equity workshop was documented in C&EN²³ and led to related one-time workshops on URPOC⁵⁸ and disability⁵⁹ in chemistry.

We have thus been driven by the top-down hypothesis that solutions must be led intentionally through the top—that is, by department heads and chairs—because they are the stewards of the infrastructure.^2,46 In current language, this is our change theory.⁶⁰ Its roots lie in the observation that holding middle managers accountable in industry is particularly effective at advancing equitable representation.⁵¹⁻⁵⁴ Note that the top here is unlikely to refer to presidents and deans of academic universities, though they can be the drivers of holding department heads accountable because they tend to be too far removed from discipline-based academic hiring, mentoring, and promotion. While in principle the top-down hypothesis can be applied within a single university across its many departments, in practice, those departments have very distinct professional cultures⁶¹⁻⁶³ and financial mechanisms. Effective practices do not always translate well, and the competitive drivers between them can sometimes become hazy because of the incompatible differences in their metrics of success. Instead, a cohort of chairs within the same discipline, e.g., chemistry, can form a community of practice⁶⁴ that can be held accountable to itself. Examples of such chemistry communities of practice include department chairs in the Big 10, Pac 10, AAU, and the southeastern regional chairs, all of whom meet regularly to analyze and share the state of their departments and emerging effective practices.

Through the convening of chairs from the top 50 federally-funded departments to discuss gender equity, URPOC equity, and disability equity, the workshop committees^56,58,59 that preceded OXIDE effectively created this community of practice. However, because these were all one-time events, these communities dissolved almost as quickly as they had been formed, and so did their communities of practice. The formation of OXIDE was thus driven by the change theory that the application of the top-down hypothesis required the existence of a standing community of practice among chemistry department chairs and a mechanism to keep them engaged through chair transitions. Such a community obviously need not be limited to the chairs from the top 50 because they are not the only producers of Ph.D.s or future Ph.D.s, so OXIDE’s efforts have been inclusive of department chairs across institutional types. Finally, as the objective should be to remove inequities along all under-represented groups, establishing equity across gender identity and orientation, which had not been a focus of the three NSF workshops, also had to be part of such an effort.

5. Engaging Evidence-Based Solutions to Systemic Bias, Not Training

The National Science Foundation, for example, has incorporated training focused on bias in the orientation of all of their reviewers.^65,66 It might be similarly tempting to create a program which provides diversity training resting on the change theory that such chairs simply are unaware of what to do or how to do it. Unfortunately, ample data suggests that mandatory diversity training tends to be ineffective and even counterproductive.⁵¹⁻⁵⁴

Thus, the convening of chairs in the initial workshops^56,58,59 were framed not just by the top-down hypothesis but also by the realization that they could not be described as training. Instead, they were framed as a scientific workshop in which experts from the social and related sciences would share their findings, and the participants, mostly chemistry chairs and their representatives, would study the findings, providing recommendations for systemic change that would lead towards equity for the targeted URG. Such a reframing is not a charade to better engage a broader (and perhaps not fully supportive) set of participants. It can be authentic in engaging scientists, who are accustomed to serving on advisory panels, to construct recommendations for systemic change based on the findings reported by the subject matter experts and interpreted to the professional culture of chemistry by the workshop participants.

The OXIDE project^46,67 built on the top-down hypothesis and the workshop framework in staging the first National Diversity Equity Workshop (NDEW) in 2011⁶⁸ and biennially since then.^21,69,70 Instead of focusing on only one URG, the workshops included discussions of barriers and solutions to equity for primarily four URGs: gender binary, URPOC, disability, and gender identity and orientation. The demographics of the first of these two groups, as discussed above in Section 2, is well below availability pools, and the designation under-represented is clear. However, the demographics of the latter two groups have not been as well documented, and this is clearly an area where more research is needed. Nevertheless, the challenges that have made such tracking difficult is sufficient evidence that these groups are not sufficiently visible and that their demographics among the faculty are also incommensurate with their availability pools. Hence, all four areas are under-represented in our faculties, and redressing this incompatibility is a driver for reducing diversity inequities intentionally through the work of OXIDE and many others. Meanwhile, each department has only one chair or head, and it would be unsustainable to expect them to attend multiple workshops addressing each of the URGs as well as the combinatoric areas due to intersectionality every year. Thus, NDEWs span all four areas of under-representation, including intersectionality⁷⁰, as they provide a periodic convening of department chairs and representatives to interpret advances in diversity research and practice and contextualize them within our discipline of chemistry.

With the recognition that not all barriers are affected inequitably across the four URGs, NDEWs typically include focus sessions allowing for deeper reflection on the practices affecting a given group. Prior to OXIDE, none of the workshops had focused on gender identity and orientation. This URG was included in the first NDEW⁶⁸ on an equal footing with the other three URGs, setting the stage for all barriers, specific and shared, across the URGs to be part of the discussion and subsequent recommendations. The second NDEW⁶⁹ included a focus session on gender identity and orientation that allowed the participants to engage in a deeper discussion about eliminating inequitable barriers to members and allies of that community. Thus, OXIDE’s NDEWs are constructed around exploring solutions to remove inequities through the work of a community of practice, leading to recommendations that invariably call on systemic changes to policies, procedures, and practices.⁷¹

6. Establishing and Disseminating Solutions

Just as the barriers to equity need to be contextualized within a discipline-based lens, so too do the approaches needed to remove them and to ensure that they do not recur. After staging a given NDEW, OXIDE’s recommendations are distributed through a website and in print. An early example of such a recommendation was the recognition that the term “intervention” to describe the implementation of a new practice or policy aimed at improving diversity equity did not resonate with the chemistry community. Instead, we adopted the term “solutions” not because of the chemical pun with solutes and solvents but rather because it appealed to chemists, who reframed this as a problem that needed to be “solved.” This reframes challenges to diversity equity as ones that can be solved through controlled experiments in which we redefine hidden and unhidden policies and practices. The need for an identification of evidenced-based solutions that are effective at improving diversity equity and inclusion has driven DBDR in the past few years. Examples of OXIDE’s initial recommended solutions are listed in the abscissa of Figure 3. The change in which solutions are targeted by the participants of NDEWs due to their partcipation is visible in the distribution of their intended implementation before and after attending a particular workshop, which happens to be NDEW 2015 in the data shown in the figure but is representative of the surveys from all of the NDEWs. In addition to an overall increase in the number of solutions to be implemented generally after attending an NDEW, there was an increased recognition of the amount of activity to remove inequities related to sexual orientation and disability status.

Figure 3.

Survey responses of possible actions, viz. interventions or solutions, related to diversity equity that 23 and 24 participants reported before and after NDEW 2015, respectively, that their department would consider within the next 12 months with respect to targeted URGs: gender (red), race and ethnicity (blue), sexual orientation (white), individuals with disabilities (green), and non-specific (purple). Participants could select actions for more than one group, and hence the total number of actions in a given category can exceed the number of participants. In the absence of color, shading and sequencing can be used to reveal which is which. Adapted with permission from ref (70). Copyright 2018 ACS.

But the problem is sufficiently complex that the chemistry community is also generating and sharing effective practices through multiple means.⁷²⁻⁷⁴ Notably, many of these solutions are rooted in recommendations from the broader social sciences^35,75 but are given specificity to chemistry within the frame of DBDR. I would be remiss if I did not also provide a short and targeted discussion of some of OXIDE’s recommended solutions, which are themselves outcomes from NDEWs, so as to challenge the DBDR community to further research their efficacy.

• (i)Engage in community. From the beginning of OXIDE, we recognized the need for having a community of practice that engages periodically and systematically in identifying effective solutions. One of the recommendations that emerged from the first NDEW was that department chairs should go back and engage in discussions with their departments. These are peer-to-peer conversations which can lead to more systemic changes in conduct and ultimately in climate. But that is not enough. The conversations must continue frequently and often across the entire department. To that end, we have also advocated that safety moments, now held increasingly commonly before weekly departmental seminars and other activities, should be extended to include diversity moments. Doing so is a part of creating an atmosphere of inclusive excellence by weaving our values through all that we practice, teach, and share with our community.
• (ii)Conduct authentic and open searches. We also advocated early on that departments should open faculty searches. Doing so helps to identify outstanding candidates who you might not have seen otherwise. The search must be authentic, and it should not be a surrogate for delaying divisional rivalries once candidates have been identified. Doing the latter would not be an authentic search. Indeed, several chairs reported failures in their search processes due to such inauthentic practices. Thus, a key to the success of searchers is the appropriate management of bias throughout every stage of the process. This does not apply only to the evaluation of the candidate applications. It is equally important to manage and reduce bias in the structure of the virtual and physical interviews. For example, stereotype threats should be removed wherever possible, and remaining differential threat conditions must be considered in discussions of candidates.
• (iii)Recognize and reward inclusive excellence. In the analogy that we have made between DBER and DBER, there is an additional parallel connection. Namely, education and diversity work have tended to be considered as additional service work and not as integral to the research mission of a department. As we increasingly recognize that both are critical to the success of a department (and the institution as a whole), then we must also recognize and reward the faculty that advance and promote inclusive excellence within the university and across the discipline broadly.

These recommendations may seem on face value to be simply practices that could be adopted in diversity research, broadly, and perhaps they could be. What makes them DBDR in chemistry is that we have to articulate and contextualize them in terms of practices and procedures that we conduct in chemistry. For example, the ways in which our community needs and uses equipment, both within individual laboratories and shared facilities, require us to engage in community practices that are very specific to our discipline of chemistry. The common practice of the so-called tenure tour, which appears to be quite specific to the chemistry community, is yet another example of how we engage in specific community practices. Ensuring that this practice is open and equitable to all of our assistant professors in chemistry is consequently a very specific need for our community. Thus, DBDR in chemistry is needed to identify the practices in chemistry that need reform, offer theories of change for reforming them that offer frameworks that fit within how chemists think, assess their efficacy, and promote them across our discipline.

The three overarching recommendations for advancing DEIB are based on the findings of OXIDE and NDEWs and the subsequent effects that we have seen in their implementation by the members of our community of practice, the chairs. A longer list is available in past publications^21,71,76 and on the OXIDE website.²⁴ However, holistic assessment is necessary to refine these practices and to ensure that effective practices are adopted by our academic leadership. To this end, faculties are encouraged not only to adopt quantitative target outcomes but also process metrics.⁷⁷ In turn, department strategic plans should be structured to include new directions, targeted outcomes, and metrics as part of the aim to achieve inclusive excellence. In the short term, diversity committees should be charged and judged accordingly.

7. Symbolizing Inclusion

If the mandate issued by the federal agencies in chemistry^56,58,59 almost 20 years ago was not enough, in the last few years journal editors and their editorial committees have issued statements making it clear that diversity and excellence come together. Examples include editorials from Analytical Chemistry,^78,79Organic Chemistry,⁸⁰ the Journal of the American Chemical Society,⁸¹ACS Catalysis,⁸²ACS Central Science,⁸³ and the Journal of Chemical Education,⁸⁴ to name only a few within the ACS Publications portfolio.

Fortunately, the community of journal editors has not only named the problem but many have also issued calls for actions with explicit recommendations for moving forward.^72,80 One such action is the cover art series introduced in 2021 by the editors of Analytical Chemistry to promote and amplify voices of members of URGs that had previously been under-amplified.⁸⁵ Its success led to the adoption by ACS Publications in 2022 of a cover art series across all of its publications,⁸⁶ and the writing of this Account is one response to that call. Another example includes the expansion of Viewpoint articles by the Journal of Physical Chemistry to explicitly promote URG physical chemists who had been under-amplified. Four such articles have already appeared in print^4,87−89 and provide inspirational accounts of the success through adversity that members of our community have experienced.

Is this type of work DBDR? Is it politicizing science?⁹⁰⁻⁹² Erin Cech argues that the introduction of depoliticization into a policy framework is a harbinger for reducing diversity within a profession.⁹³ Such depoliticization works against efforts to establish a strong and inclusive chemical workforce. It also, therefore, works against excellence. Thus, DBDR and the effort to broaden DEIB must necessarily be politicized because science is advanced by people. The need for chemists to intentionally drive inclusive excellence within our organizational structures is why the leadership of the American Chemical Society has issued many calls for establishing policies that promote and preserve diversity, inclusion, equity, and respect (vis-à-vis belonging) within all of its activities and operations.⁹⁴⁻⁹⁷ So DBDR must necessarily also include the social aspects of our profession. Research is needed to establish and promote effective ways to do so, not despite but because of our drive to advance chemistry.

8. Concluding Remarks

On the one hand, the students and faculty of academic chemistry departments form a cohort that could be considered a special case of the social or professional groups that are the objects of consideration of general principles related to diversity research. On the other hand, the practices within the professional culture of chemistry are baked not just with the language of chemistry but also with processes that echo the way that we do research and advance our field. This can lead to practices that are so specific that the general rules of the former at worst do not apply and at best need to be reinterpreted. As noted in the Introduction, the need for research in such discipline-based practices has already been accepted within the broad discipline of DBER and has led to, for example, the vibrant discipline of CER. So, too, we need to recognize the need for DBER in general and in chemistry, in particular.

Thus, the chemistry academic community is more than just a Petri dish or a reaction flask for the social sciences. We can change our professional culture and promote inclusive excellence by intention. To achieve this, we need to research ourselves through the lens of our professional culture in the context of the emerging literature from the social sciences. We can run the experiments, varying implementations of polices and practices, assess their efficacy, and share our learnings across our discipline through peer-reviewed publications. That is research. That is what DBDR can look like.

This Account is thus not meant to be exhaustive in reporting all of the work in DBDR that has already been done without a formal association to a journal or specific community. While I regret that I may have undoubtedly omitted important references, those that have been reviewed here show the breadth of questions that have been answered and those that remain to be researched. It should also be clear that DBDR is an emerging area of research in chemistry, and I hope to have inspired you to join this effort.

Glossary

Abbreviations

CER

chemistry education research

DBDR

discipline-based diversity research

DBER

discipline-based education research

DEIB

diversity, equity, inclusion, and belonging

IRB

Institutional Review Board

NDEW

National Diversity Equity Workshop

OXIDE

Open Chemistry Collaborative in Diversity Equity

PEER

person excluded because of their ethnicity or race

PII

personal identifiable information

STEM

science, technology, engineering, and mathematics

URG

under-represented group

URPOC

under-represented person of color

Biography

Rigoberto Hernandez received his B.S.E. in chemical engineering and mathematics from Princeton University in 1989 and his Ph.D. in chemistry from the University of California, Berkeley, with William H. Miller in 2003. After postdoctoral appointments with Eli Pollak at the Weizmann Institute of Science in 1994 and Gregory A. Voth at the University of Pennsylvania in 1995 and 1996, he took an assistant professorship at the Georgia Institute of Technology in 1996. After rising through the ranks to full professor there, he moved to Johns Hopkins University in 2016 as the Gompf Family Professor. He has also served as the Director of OXIDE since 2011.

Author Contributions

CRediT: Rigoberto Hernandez conceptualization (equal), formal analysis (equal), funding acquisition (equal), investigation (equal), methodology (equal), project administration (equal), supervision (equal), validation (equal), visualization (equal), writing-original draft (equal), writing-review & editing (equal).

The author declares no competing financial interest.