Download PDF Fall Bridge on Engineering a Diverse Future September 25, 2024 Volume 54 Issue 3 Guest edited by Wanda Sigur and Percy Pierre, this issue of The Bridge addresses the issues around sustaining a U.S. engineering workforce that builds on and integrates the talents and ideas of our diverse nation. The NASEM Diversity Science Report: Going Beyond Mere Participation Numbers Thursday, September 26, 2024 Author: Gilda A. Barabino and Susan T. Fiske Co-chairs of the NASEM diversity science report present a candid conversation in which they discuss a systems approach to tackling systemic racism, lived experience, implicit bias, and their perspectives on the wide-ranging recommendations of the report. We co-chaired the National Academies of Sciences, Engineering, and Medicine (NASEM) consensus study Advancing Antiracism, Diversity, Equity, and Inclusion in STEMM Organizations (NASEM 2023). The report makes evidence-based recommendations grounded in diversity science (Plaut 2010) and practitioners’ experience. The list below paraphrases the report’s recommendations: 2-1. Predominantly White institutions (PWIs) can gain from understanding the academic success of MSIs (minority-serving institutions), such as historically Black colleges and universities (HBCUs). Mutually beneficial partnerships might result. 3-1. To advance anti-racism and diversity, equity, and inclusion (DEI) requires new data on education paths for individuals in STEMM. 5-1. STEMM leaders should improve numbers of underrepresented minority (URM) trainees and improve their experience of belonging; (5-2) facilitate contact among URM ingroup peers; (5-3) facilitate mentoring; and (5-4) set norms of inclusion. 6-1. Organizational gatekeepers should be accountable for the gatekeepers’ own overall patterns of diversity or bias in the organization units they manage. 7-1. In both academia and industry, management should create explicitly interdependent teams, with incentives for performing well together. 8-1. Organizations should review criteria for entry, recognition, and promotion; (8-2) change agents should expect resistance to any proposed changes. 9-1. Organizations should evaluate their culture and do research advancing anti-racism. This article features our conversation about the report’s basis for generating its recommendations and conclusions. We begin by noting the history of systemic racism and then describe its continuing legacy in STEMM organizations. We examine the report’s relevance for engineers: analyzing racism as systemic, the evidence from diversity science, the approach of lived experience, awareness of bias, and working at all levels (societal, organizational, leadership, gatekeepers, teams, and individuals). Finally, we discuss next steps. Our discussion harkens back to our approach to co-chairing the study, as we shared in the study’s preface. Undaunted by the challenges associated with using evidence-based action to remedy systemic inequities, “we tackled our charge to identify racist and biased conditions that create systemic barriers and impede the full talent pool of our nation from pursuing and advancing in STEMM careers.” It is our hope that despite the inherent discomfort in tackling racism, readers engage the evidence-based, actionable strategies and recommendations in the report in order to “make STEMM settings more diverse, inclusive, and equitable—and more anti-racist.” Systemic Racism Susan Fiske: As a non-engineer, I’ve heard a lot about systems analysis. Could you explain in broad-brush ways how this report would fit an engineering approach, a systems analysis of the racism problem. Gilda Barabino: I like that idea precisely because people do not spontaneously take a systematic approach to these issues. People are looking for a magic wand or just one way to solve a complex problem. In something as complicated as anti-racism and DEI we need a systems approach. We need multi-pronged efforts. We need to understand the interrelationships and interconnectedness of issues, environments, people, backgrounds, identities, systemic policies, and habitual practices—and how they work together in a system. We need systems thinking to seek and to implement solutions. Fiske: That’s a really rich answer. One figure in the report (see figure 1) illustrates a system, breaking it down by level of analysis. It starts with individuals, then teams, gatekeepers, organizations, and, finally, society. Different causal factors would be relevant in each of those cases. It would be a mistake for people to think that implicit bias training for individuals is the answer to systemic racism. They are different levels of analysis, different levels of aggregation. It’s like putting a cork in a leak thinking that that’s going to change everything. People need to figure out the root causes of leakage from the system. Diversity Science, a Hard Science, Benefits from the Advice of Those Studied Barabino: One of the things that comes to my mind for you, Susan, is that I think that we don’t do a good enough job of engaging social scientists with the other scientists and engineers to study our systems. We tend to remain siloed rather than taking an interdisciplinary, collaborative approach to solving the problem. I’ll share one of the reasons why I began collaborating with social scientists: They often study other scientists and engineers without being those kinds of scientists and engineers themselves. Sometimes, groups of diversity scientists don’t include any URMs. So, they’re doing their best to pose and frame the questions to study these underrepresented groups. But maybe the questions would be posed and framed better if the people that they’re studying had some input into how they’re being studied. For example, interviewing highly successful STEMM professionals about their lived experiences reinforces the diversity science takeaways but can also provide additional insights. Let me further illustrate. While on my second sabbatical leave, I encountered a learning science professor and a psychology professor who were conducting a study on women’s experiences in research laboratories. Intrigued, I asked if their study would differentiate between the experiences of majority and non-majority women, and the answer was no. Knowing that I brought a different perspective from my own lived experience as a Black woman in engineering, I suggested another study focused on women of color. The study I conducted in collaboration with my colleague and psychology professor Kareen Malone resulted in a publication in Science Education, “Narrations of Race in STEM Research Settings: Identity Formation and its Discontents.” Fiske: I am reminded of the disability activists’ saying, “No about-us, without us.” Diversity scientists count themselves as scientists, but we don’t have all the answers. Diversity science involves both basic science and practice. But the larger point you raise is a perennial basic-science question: getting advice from people who are living the situation. What kinds of things are coming up that are not covered by the basic science of diversity? For example, the lived experience of exclusion—say, being ignored for study groups or not knowing about test banks could provide mechanisms to link belonging and performance (NASEM 2023). Do you think understanding that diversity science is a science would be surprising to most engineers and most members of the National Academy of Engineering? Barabino: Science and engineering get defined too narrowly, and who is seen as a scientist or engineer gets defined too narrowly as well. On the science and engineering side, although plenty of people care about these issues, there’s not a real understanding of diversity science at all. There’s just not. Fiske: I know engineering has infrastructure (grants, fellowships, programs) to broaden participation, often based on experience-based hunches about what should work. In diversity science, we test model-driven hypotheses. We have theories, we have frameworks, we do experiments, and we do surveys. And the biggest problem is measurement. So, we refer to it as a hard science. People are looking for a magic wand or just one way to solve a complex problem. In something as complicated as anti-racism and DEI we need a systems approach. Individual Awareness of Individual Bias Fiske: For one example of the measurement challenge at the individual level of analysis: It’s difficult to get people to talk about their feelings and reactions to other people without them stage-managing themselves. This is why diversity science has invented several indicators of individual bias, more or less explicit, with or without compunction, emotional or not (Banaji and Greenwald 2016; Eberhardt 2019, Fiske 1998; Richeson et al. 2024; Shelton and Turetsky 2024). Multiple indicators help triangulate on the self-report problem. But even before that, they have to know they are biased. So, the first question is, are they aware of their prejudice? Barabino: That’s right. Fiske: And many people are not aware. Then, even if they’re aware of their prejudiced attitudes on some level, will they tell us? After all, it makes them look bad. (That’s one kind of reason that we call diversity science a hard science.) Barabino: That makes sense. But as we’re talking, the power of how you and I can come together to point out some of these challenges—we could be filling a void. We could say, “Hey, all you folks mean well, but here’s something that all of you miss.” And I do think the existence and impact of prejudiced attitudes and behaviors (intended or not) are continually overlooked, not just by well-meaning STEMM professionals, but also by well-meaning diversity scientists. Fiske: I think it’s shortsighted to give people training just in implicit bias (Banaji and Greenwald 2016). Barabino: Very shortsighted. Fiske: I mean, if you have the experience of taking an implicit bias test (Banaji and Greenwald 2016), and you find yourself making bias a pattern, or exhibiting overall bias, most people have a moment of shock, like, “whoa.” But you have to be open to taking the test and believing that bias is a real thing and not just some made-up party game. Barabino: Once, when I had a group of engineers take the implicit bias test, they made comments like, “This is bogus.” “This is all fake, not measurable.” So, as you said, they have to believe bias is a real thing. In their minds, they are telling themselves, “It’s not a real thing, so I don’t have to pay attention to it.” Returning to the Big Picture: Working at All Levels Fiske: One way to tell if implicit bias is real is this massive online database of millions, literally millions, of people responding to implicit bias tests, and you can trace the county that they live in. With large numbers, the county level of implicit bias predicts discrimination against minoritized people in that county. The sort of aggregated individual persons’ bias predicting a general pattern of biased decision-making is typical in that ecology. So implicit bias is real if you look at aggregates like countywide indicators of bias. Barabino: That’s right. Fiske: But that means that in STEMM organizations, if you looked at the average level of implicit bias, you would be able to predict their overall patterns. At the individual level, it’s not so obvious because it’s automatic. Importantly, there are other ways that racism is not likely to be on an individual’s radar screen besides it being automatic. That’s what they call implicit bias, but it’s also ambiguous. People favor their own tribe, as much as, or more than, disfavoring others. They say, “Oh, well, I just don’t know that she’s our kind of person. She doesn’t feel like people like us.” And so, it’s that feeling of, “Who’s familiar? Who’s like me?” People like themselves. So, the ambiguity is: Favoring your own has this collateral damage of leaving out the others (Brewer 1999). After automaticity and ambiguity, the last part of it is the ambivalence; two forms of prejudice are disrespecting but liking (paternalism) or disliking but respecting (envy) (Fiske et al. 2002). All this means that individuals are unlikely to be able to track their own bias, so someone else has to notice the person’s overall patterns of decision-making. The organizational level, the context, may be a better predictor than the individual. Lived Experience of Individuals in the World Barabino: The context of everything matters, and lived experience matters. There’s something valuable about hearing from the people who authentically can speak to some of the toll that racism takes. One need not literally walk in someone else’s shoes in order to empathize and speak to their experiences, but there must be the capacity to hear and to learn and to enhance one’s understanding through scholarship, data, and other ways of knowing. Fiske: For example, in the field of diversity science broadly defined, about three decades ago, maybe four, all the White people in the field believed that Black people must have low self-esteem because how could you be subjected to the kind of garbage and not have low self-esteem (Allport 1954)? Of all the different ethnic and racial groups, Black people have some of the higher self-esteem, because when people take a piece of them, they say, “That person’s just a bigot.” So, you can have high self-esteem and just not care about what the public regard for your group is (Crocker and Major 1989). And the White people said, “Oh, I didn’t know.” So, there’s a lot of stuff like that that’s been found out the hard way. Different ethnic groups respond differently. Some groups tend to care a lot about what other people think of them some don’t. Barabino: Yes, you’re right about that. Fiske: So, if their group is seen in low regard, they feel it personally. But Black people have learned; it’s like the conversation that you have with your kid before they go to elementary school. Barabino: Exactly. Fiske: I mean, that’s what I gather. Co-Chairs Uniting Science, Practice, and Lived Experience: How to Create Metrics Together Barabino: Our chairing that study together was phenomenal because we meshed our strengths. We also brought our own perspectives in a way that made the overall product richer. Fiske: How very pragmatic. Barabino: Absolutely. And it’s hard for engineers and scientists to accept and be guided by what is perceived as being based on feelings rather than data and metrics. But feelings and attitudes are important and need to be acknowledged even if it’s hard to quantify them. Fiske: Diversity scientists quantify them. Barabino: To some extent, I guess. I think the readers need to understand, what does that mean? Fiske: I’m thinking of an example of a feeling thermometer (Allport 1954). It’s a very simple form of measurement of feeling toward groups. On a scale of 0 to 100, how do you feel about Native Americans? Or how do you feel about White people from Europe? And so that’s a very simple way, and it’s sort of engineering. Barabino: Then all of a sudden, engineers are thinking, “Hmm. Yeah, there’s some metric to it.” They’re always looking for metrics for— Fiske: Everything. That’s my instinct. And we look for reliability and validity. Engagement at All Levels Barabino: I like that concept. But that’s not the only way to engage. Let’s think of different ways to engage that are likely to be effective. For example, in speaking with a representative from the National Museum of African American History and Culture about our study, it came up that they do STEM days, and we were invited to participate. Fiske: We’re not science-fair material. Barabino: True. But maybe we could be involved in a seminar series or some other event or other way of connecting. It would be wonderful to connect to individuals’ personal experiences. An example of one of many connections for me is the museum exhibit about the Polar Bears, the exercise group that gathers on Inkwell Beach on Martha’s Vineyard, and my own participation with the Polar Bears during my vacation. Fiske: How cool. I’ve missed that somehow. Barabino: And so, my point is that in bringing our report to life, we draw upon the ways people, places, and times are connected. Just like me seeing the exhibit about the Polar Bears. Because connections like that, Susan, to me, are also dissemination. That’s getting in different rooms, reaching different audiences, and elevating the discussion about the report. I think that’s a big deal. Fiske: The people we want to reach are such a varied group of people in terms of what they do, but I bet they’ve all been subjected to bias training. Informing people that bias is not what they think it is and to consider the larger context is important (Dobbin and Kalev 2022; Paluck et al. 2021). Barabino: The other thing that I think is helpful for us is our constant ability to show how we, a diversity scientist and a practitioner, plan together. And that’s how the committee built our recommendations. The context of everything matters, and lived experience matters. Many Paths to Antiracism, Across Levels Fiske: The committee had a consensus to use strong language. This report uses the term “antiracism” frequently throughout. We intentionally included the word “antiracism” in the report title, and it was also a key part of our statement of task. The committee defined antiracism as an intentional set of actions that dismantle and disrupt racism. These actions may incorporate a range of behaviors, from reworking policies, to developing new systems, or changing practices. No matter what that specific behavior is, the goal is to initiate positive change. In the report, the committee included more than 25 evidence-based conclusions and recommendations. When developing our recommendations, we wanted to be sure that they would be consistent with and true to the principles of being antiracist. We authored recommendations that people can pick up and put into action, all with the goal of making STEMM more equitable, inclusive, and diverse. Over the course of this report, we hope people come away with the message that there is not just one, but many potential paths to successful and sustainable change. The report offers such a rich set of ideas and strategies; we think a book group or a weekly seminar would be a useful way to absorb and discuss these ideas. Next Steps Barabino: Currently, we are switching gears from talking about what is published in the report to talking about the next steps. This report was never intended to sit on a shelf. We want this report to inspire a body of future transformational and groundbreaking work. We are hopeful that this will be the first of many subsequent reports related to advancing antiracism, diversity, equity, and inclusion in STEMM. When you look at our report, it is evident that much work remains to be done. In the last chapter, the committee authored a robust research agenda that outlines the many research questions that still need to be answered. As the committee recognized, the work that comes next should not only fill these gaps, but it should propel the field forward in ways that are innovative, challenging, and necessary. There is much cause to be optimistic, and I am particularly optimistic about the continued impact and potential legacy this report can have. We look forward to other wonderful opportunities for all like-minded participants, in-person and remotely, to act on the lived experiences and evidence-based recommendations. References Allport GW. 1954. The Nature of Prejudice. Boston: Addison-Wesley. Banaji MR, Greenwald AG. 2016. Blindspot: Hidden Biases of Good People. New York: Bantam. Brewer MB. 1999. Psychology of prejudice: Ingroup love and outgroup hate? Journal of Social Issues 55(3):429–44. Crocker J, Major B. 1989. Stigma and self-esteem: Self-protective properties of stigma. Psychological Review 96(4):608–30. Dobbin F, Kalev A. 2022. Getting to Diversity: What Works and What Doesn’t. Cambridge, Massachusetts: Harvard University Press. Eberhardt JL. 2019. Biased: Uncovering the Hidden Prejudice that Shapes What We See, Think, and Do. New York: Penguin Random House. Fiske ST. 1998. Stereotyping, prejudice, and discrimination. In: The Handbook of Social Psychology, 4th edition, 357–411. Gilbert DT, Fiske ST, Lindzey G, eds. New York: McGraw-Hill. Malone KR, Barabino GA. 2008. Narrations of race in STEM research settings: Identity formation and its discontents. Science Education 93(3):485–510. NASEM [National Academies of Sciences, Engineering, and Medicine]. 2023. Advancing Antiracism, Diversity, Equity, and Inclusion in STEMM Organizations: Beyond Broadening Participation. Barabino GA, Fiske ST, Scherer LA, Vargas EA, eds. Washington, DC: The National Academies Press. Paluck EL, Porat R, Clark CS, Green DP. 2021. Prejudice reduction: Progress and challenges. Annual Review of Psychology 72:533–60. Plaut VC. 2010. Diversity science: Why and how difference makes a difference. Psychological Inquiry 21(2):77–99. Richeson JA, Rucker JM, Brown X. 2024. Race and racism. Handbook of Social Psychology, 6th edition. Gilbert DT, Fiske ST, Finkel E, Mendes WB, eds. Princeton: Situational Press. Shelton JN, Turetsky KM. 2024. Diversity. Handbook of Social Psychology, 6th edition. Gilbert DT, Fiske ST, Finkel E, Mendes WB, eds. Princeton: Situational Press. About the Author:Gilda A. Barabino (NAE, NAM) is president and professor of biomedical and chemical engineering, Olin College of Engineering. Susan T. Fiske (NAS) is Emerita Eugene Higgins Professor, Department of Psychology, and School of Public and International Affairs, Princeton University.