WHAT IS THE REAL VALUE OF DIVERSITY IN ORGANIZATIONS? QUESTIONING OUR ASSUMPTIONS
KATHERINE W. PHILLIPS
It is hardly possible to overrate the value … of placing human beings in contact with persons dissimilar to themselves, and with modes of thought and action unlike those with which they are familiar.… Such communication has always been, and is peculiarly in the present age, one of the primary sources of progress.
—JOHN STUART MILL, Principles of Political Economy
When philosopher and political economist John Stuart Mill claimed diversity as a “primary source of progress,” it was 1848 and his initial concern was trade among nations. He claimed that being exposed to the goods and products of a foreign country served as a mechanism for economic growth. But more importantly he claimed that exposure to other countries served the intellectual and moral growth of the people. Mill’s concerns with difference did not stop there. Some have claimed Mill was the first feminist, as he advocated for the equality of women in society as well. Now, in 2017, we are having the same conversation about the “value … of placing human beings in contact with persons dissimilar to themselves” (that is, diversity) for organizational and team performance, educational development, technological advancement, and other important outcomes in society. The so-called business case for diversity has consumed the attention of scholars and practitioners alike as they seek to justify efforts to diversify organizations and open opportunities for a broader swath of society—women, racial minorities, persons with disabilities, the aging population, and other “nontraditional” and often undervalued society members. The fact is, this question of “What is the value of diversity?” is not a new one. Mill made proclamations about it in 1848, and now Scott Page’s treatise provides a compelling answer to this question through logic and empirical research on the promise of cognitive and identity diversity.
Page’s contribution to Our Compelling Interest has offered us an in-depth explanation of how diversity can provide a bonus under the right circumstances—when problems are sufficiently difficult. He has used simple and compelling logic, and provided numerous examples from the boardroom to the classroom to Hollywood studios, to explain how bringing a set of people with cognitive and identity diversity together can lead to better performance on complex tasks or difficult problems. Important to note is that Page does not use the word will or the word should in his statements. This is not an exercise of precision or guarantees, nor is it an exercise of morality. Instead, it is an exercise of functionality and utility—when and where is it possible for diversity to provide unique benefits over homogeneity? By understanding the possibility of diversity’s benefits, we might then understand what society is striving toward, so we can make some assessment of whether it is worth it to try to get there. I believe the logic and the evidence support the potential for a “diversity bonus,” and there is no doubt that it is worth it for organizations and for societies to try to capture that bonus.
I would like to add to the discussion of Our Compelling Interest and Page’s diversity-bonus premise by providing three contributions. First, Page mentioned that identity diversity’s bonus might emerge not from the direct connection between identity diversity and cognitive diversity but instead from indirect effects of the mere presence of diversity. Empirical research in this area reveals that identity diversity has an effect that is independent of its connection to cognitive diversity.1 I believe this is worth exploring further here, as it relates to the assumptions we tend to make about the value of diversity. Page argues that organizations must open opportunities for those who have traditionally been underrepresented in organizations so that cognitive diversity can enter, as in the example Page used about disabled employees being integrated into the federal government. I expand on one argument that goes a step further to say that identity diversity also benefits organizations because of the influence it has on every individual in the organization. Diversity makes people work harder, and that benefit can happen even if the identity diversity is not directly connected to new cognitive perspectives. My comments here lead to the fundamental question, what is the resistance to diversity in organizations? If diversity bonuses exist and can be captured, as postulated by Page and supported by empirical literature, why is it so hard for organizations to achieve them consistently?
Second, I believe it is important to recognize that diversity is hard. Diversity bonuses do not automatically emerge simply by putting diverse groups together. Groups are composed of people, and people are far from perfect. The people have to engage with one another in some meaningful and productive ways to garner the benefits of that diversity. Moreover, these groups of people are embedded in systems, in organizations, in historical times, that influence the potential for that diversity bonus to be reached. Even under the best of circumstances, when cognitive and identity diversity is in abundance, there are numerous factors that may facilitate or undermine the emergence of a diversity bonus. I use the example of functionally diverse groups—filled with cognitive and identity diversity to work on difficult problems, the basic prerequisites of Page’s model—to explore a few situational factors that might be important for facilitating diversity bonuses in organizations. In many of the successful examples used throughout Page’s book, I would postulate that an examination of those situations would find many of the features that are fundamentally built into the best practices of cross-functional teams. Without the right set of situational circumstances, it is difficult for any team to succeed. Teamwork is hard, and many teams fail to reach their potential. This is especially true when the teams are dealing with complex, difficult problems that require interdependence, coordination, and collaboration between a diverse set of individuals. Although organizations strive to facilitate these circumstances, they often fall short and might erroneously conclude that diversity does not work. It is not diversity that has caused the failure; it is a lack of organizational leadership, implicit and explicit biases, improper structures, and poorly designed processes that have caused the team to fail. The potential of the diversity bonus is still there.
Finally, I will challenge the reader with my own, somewhat philosophical, set of questions and assumptions: Why is it necessary to prove the benefit of diversity? Is there an equal push to prove the benefits of homogeneity to organizations and society? Why do some people (for example, women, minorities, the disabled, the old, the young, transgendered people, homosexuals, and so on) have to prove that their presence in a given environment is “beneficial” for the bottom line and others (that is, white people, males, cisgendered people, and heterosexuals) do not have that same burden of proof? There is power in framing and in the questions we ask. This raises explicit issues of representation, power, and status that are often left untouched by diversity researchers, including myself. The reality is that there is a power in the status quo that motivates the desire for “proof” of the business case for diversity—we all see homogeneity as the norm, which perhaps gives it more power than it deserves.2
DEFINING DIVERSITY
Before jumping into my contributions, I think it is important to link Page’s conceptualization of diversity with past research. Page divided the world of diversity into two main categories: cognitive diversity and identity diversity. This distinction aligns with distinctions often used in the organizational diversity literature. There has been a proliferation of labels used to distinguish different sources of diversity from one another—demographic diversity, value diversity, task diversity, surface-level diversity, deep-level diversity, and so on—but they all make the same basic distinction. In my own empirical research I have gravitated toward the use of the terms informational diversity and social category diversity to distinguish between diversity that results from what people know (that is, cognitive diversity) and diversity that results from how people identify themselves and others (that is, identity diversity). Importantly, I highlight in my work that informational and social category diversity coexist in groups. It is not possible to characterize a group as having only a level of informational diversity or only a level of social category diversity—a group constantly has some level of each of these types of diversity. Hence, you cannot study one without considering the other.
“Informational diversity captures the differences in information, opinions, perspectives, and modes of thought and action that are relevant for the task at hand being completed by a group.”3 This definition is consistent with the distinctions Page makes regarding cognitive diversity. The careful consideration of information, knowledge, heuristics, representations, mental models, and frameworks discussed explicitly by Page in the latter half of the book provides a logic and precision to our thinking about how informational or cognitive diversity influences groups. When individuals are brought together to solve problems in groups, they often possess different information or knowledge structures that can be utilized to inform the group. In fact, for many problem-solving and decision-making groups, the very reason they are brought together is to capture the diverse knowledge and perspectives that are uniquely held by different group members.4 However, in many cases there is no way to effectively measure cognitive diversity—what people actually know about the current problem. One does not know what another person knows until he or she shares that information. Page compels us to consider the fact that each person who comes to a group has a set of abilities and tools, knowledge, and mental models. These are shaped by a myriad set of experiences and identities possessed by the group member. Although not explicitly stated, Page and I share in the fundamental assumption that every group has some level, even if very small, of informational or cognitive diversity, as no two individuals are exactly the same. As groups work together, the goal should be to increase the presence of that cognitive diversity and the willingness to express those differences to capture the diversity bonus in groups that are facing complex, difficult, and uncertain problems.
Social category diversity exists alongside cognitive diversity in task groups. In my work I have used the term “social category diversity to refer to those distinctions that serve as a salient basis of categorization into in-group (people who are like me) and out-group (people who are not like me).”5 Salient demographic characteristics such as race, gender, nationality, or age are considered social category diversity, but this diversity may also stem from any characteristic that may not be immediately visible yet can be rendered salient in the context and thus be used to categorize group members. For instance, minimal distinctions such as an ostensible preference for a type of painting or for wearing a red shirt versus a blue shirt can be used to examine the effects of social category diversity, allowing for a connection of this research to the long tradition of social categorization and social identity research dating back to the 1970s.6 The critical feature here is that people use these social characteristics to tell themselves that some subset of the group of people is “like me” at a deeper level, in terms of what we know and how we feel about problems we are facing, and that some of them are not.7 For the rest of this chapter I will use the terms coined by Page, cognitive diversity and identity diversity, for consistency.
THE BENEFITS OF IDENTITY DIVERSITY
Page spent much of his treatise explaining how groups benefit from cognitive diversity. This is compelling and without much controversy. We can believe that if two people who aren’t necessarily the best at a task come together to share their specialized knowledge, they might benefit from the full complement of knowledge necessary to be successful on that difficult task. Diversity bonuses that stem from cognitive diversity can occur. However, the role of identity diversity in groups is often what people are thinking about when they ask the question, what is the business case for diversity? This question is a nice way of asking, why do we need people who look different from the majority of us here to be a part of our organization? Why do we need identity diversity to be successful? Although I find the question itself to be plagued by implicit or unconscious negative intentions (I discuss this further later), Page’s logic, along with empirical research on the impact of diversity in teams, provides at least two compelling answers to this question.
The first benefit of identity diversity is that it is a source of cognitive diversity. The premise here is that people from different identity groups will bring different knowledge, experiences, and mental models to the table for consideration, allowing for increased cognitive diversity and therefore better outcomes (predictions, creativity, decision making, problem solving, and so on). Just like one’s functional training in engineering, psychology, or cultural anthropology shapes one’s cognitive identity, so too does one’s gender, race, cultural background, (dis)ability, and so forth. If the training one has received for five to ten years in school is important for shaping a cognitive profile, one argument that follows is that the life experiences that are shaped by other identities should also be considered important for shaping that cognitive profile. People who have different identities walk through the world having different experiences, being exposed to different opportunities, different knowledge, and different mental models. Cultural differences, gender role expectations, marital status, parenthood, and a myriad set of other identities all influence the cognitive models an individual develops. Moreover, it is not unreasonable to believe that identity makes it possible that, for example, by virtue of being a woman, there may be some common experiences that shape one’s cognition in predictably different ways from the cognition of men. For example, women might have different insights about the design of a car because of predictably different ways that women might use the car in comparison to men. Here it is critical to recognize that representation may provide different benefits from diversity itself. When creating a product for women, it is critical to have a woman’s perspective represented, as she might have insights that a man could not. Likewise, if an American company is moving into the Chinese market, it is in their interest to work with Chinese people who have local knowledge to ensure development of a product that fits customers’ needs.
There is a caution here we must consider. No identity, even one’s functional identity, is 100 percent predictive of one’s cognitive profile. This is the beauty and the flaw of this logic about the congruence between identity and cognitive diversity. We should not assume that people from the same identity group have exactly the same cognitive perspective to bring to a group for a given problem, just as we should not assume that two people who are different necessarily have different cognitive perspectives.8 We use identity diversity as a proxy for cognitive diversity. It is not a perfect proxy. Page is essentially suggesting that if some people are kept out on the basis of identity, then organizations are potentially leaving out very valuable cognitive perspectives that could benefit groups, organizations, and society. It is better to err on the side of inclusion, given the complex and difficult problems often being faced in organizations.
It is easier to swallow the logic that identity diversity is important when we are talking about functional differences, but harder for people to accept that a female engineer and a male engineer, both trained at the same schools in the same “right” ways of doing things, might have different approaches to a problem by virtue of their gender. Functional background matters, but so do all of the other myriad identities that shape experiences in life—gender matters, race matters, nationality matters, and so on. The compelling example Page used about a person in a wheelchair having a different vantage point and seeing the world from a different perspective is apropos here. There are times when representation from different types of people is the only way to ensure that cognitive diversity is not only present in the group but also communicated and sought after by others in the group.9
A second benefit of identity diversity is simply that seeing differences on the surface makes people assume that there are more cognitive differences there in the group, prompting them to seek out this information. Even if the person who looks “different” does not bring any new cognitive differences to the table, his or her mere presence has been shown to change the behavior of the group’s members.10 People work harder in identity-diverse environments than they do in homogeneous environments when it comes to benefiting from cognitive diversity.
Let me review some of the psychological processes that support these claims. First, individuals expect individuals with the same identity to agree with them more on both task-relevant and task-irrelevant issues than people who are different.11 In particular, Katherine W. Phillips and D. L. Loyd’s findings support this psychological process in two experimental studies in which they examined two different sources of identity diversity (functional background and geographic affiliation) and argued that expectations can cause diverse groups to benefit from the mere presence of people who are different, regardless of whether those different individuals have unique perspectives to share.
In their first study, they told MBA students that they would be working in a three-person team that included the participant and either two other MBA students or one fellow MBA student and a medical student. This led to two conditions: the homogenous condition, in which all members of the team were MBA students (same functional identity), and the diverse condition, in which one of the members was a medical student (different functional identity). Given the composition of the groups, there was always a majority of MBA students, so the participant was always a member of the majority identity group. The participants were then told that their opinions about which market to target for a medical device were in disagreement with the opinions of the other two members of the group, who were in agreement with one another. Thus, the participant was always bringing some cognitive diversity to the group in the form of a unique opinion. The findings of this first study revealed that participants expected greater task perspective similarity with identity-similar individuals (that is, MBA students) than with identity-dissimilar individuals (that is, medical students). In addition, when their expectations for similarity were violated, participants were more surprised and irritated by disagreement from the identity-similar individuals in homogeneous settings than with the identity-similar individual in diverse settings—that is, they were more tolerant of cognitive diversity and disagreement in identity-diverse settings. Furthermore, participants in diverse groups expected a more positive and accepting group experience than those in homogeneous groups.
One might argue that the results of the first study using functional identity are not that surprising. In their second study, Phillips and Loyd examined the consequences of this assumption of similarity and dissimilarity even when the identity was irrelevant to the task at hand.12 In this study, the task-irrelevant distinction was the side of campus on which the participants lived (north or south campus); in the diverse condition, there was one group member from one side of campus and two from the opposite side of campus, whereas in the homogeneous condition, all group members were from the same side of campus. This was a meaningful source of identity for the participants, and the experimental methods made that particular identity salient for the participants. In this case, interacting groups were brought together to make a decision about the best company for another company to acquire. The task was an information-sharing task that allowed for the exchange of unique information and opinions about three possible options. In both the homogeneous and the diverse conditions, one of the group members from the majority identity group held an opinion that was different from that held by the rest of the group, just as in the first study. One member of the social majority had information about the companies that was different from that given to the other two members of the group.
The results of this second study suggest that even task-irrelevant characteristics (for example, geographic location) can trigger expectations of similarity. More specifically, Phillips and Loyd found that when a member of the identity majority voiced a different opinion, homogeneous group members had more negative feelings and engaged less in the task than diverse groups (that is, they finished their discussions earlier).13 Cognitive diversity was more acceptable in identity-diverse environments, even when the cognitive diversity was not aligned with the identity differences. When a member of the identity majority possessed a different opinion, diverse groups were perceived as more positive and accepting of alternative viewpoints, fostered more persistent and confident voicing of dissenting perspectives, and displayed greater task engagement in the same situation. Identity diversity, even when it is not the direct source of cognitive diversity, is beneficial for groups.
Further evidence for how the expression and consideration of cognitive diversity is facilitated by identity diversity in groups can be found in S. Sommers’s research on the effects of racial diversity on jury decision making.14 Sommers compared groups of six white jurors with groups that had four white and two black jurors. In comparing the behavior of the whites in the presence of racial identity diversity or not, Sommers found that whites changed their behavior. White jurors in diverse groups raised more novel case facts, had fewer factual inaccuracies in their discussion of the case, and identified more missing evidence during the deliberation than whites in homogeneous juries. The white participants in the homogeneous juries may have been comparatively less focused on information presented during the case because they expected that the homogeneous setting would be characterized by agreement and easy interaction. Conversely, white participants in diverse settings may have expected more disagreement and divergent opinions about the case, leading them to consider the facts of the case more thoroughly. Thus, Sommers’s study indicates that racial diversity in jury decision making can be beneficial, as it may allow white participants to express their opinions and consider information and alternatives that they would otherwise dismiss in homogeneous settings. These claims are also supported by the work of A. L. Antonio and colleagues.15 Their study directly supports the idea that social category diversity promotes critical thinking by individuals in groups. They found that individuals displayed more integrative complexity when they had been exposed to more racial diversity in their personal lives (as self-reported on a survey) and in their experimental study: following group discussion of a controversial social issue, whites demonstrated more complex thinking in writing an essay when assigned to a diverse group with a black minority-opinion holder than when assigned to a homogeneous group of all whites with a white minority-opinion holder. These results have been further supported by the work of Loyd and colleagues, who showed that even prior to entering a group discussion, people anticipating entering a homogeneous environment prepare less thoroughly for the meeting than those anticipating a diverse interaction.16 Even before entering into discussion, identity diversity promotes hard work.
A second psychological process supporting the conclusion that identity diversity has effects on groups that are independent of alignment with cognitive diversity comes from the fact that individuals prefer their opinions and beliefs to be more similar to those with the same identity than to those with a different identity.17 Ultimately, this may suggest that identity diversity among group members can be beneficial, not only because it serves to increase the breadth of knowledge available to its members but also because expectations about those identity differences may, under the right circumstances, facilitate expression and acceptance of differing mental models that will allow for the emergence of a diversity bonus. For example, Phillips argues that identity-diverse groups may be better able to garner the benefits of cognitive diversity (that is, opinion disagreement) when it comes from outsiders rather than members of the identity majority, because outsiders are more likely to state their dissenting views with confidence if they do speak up.18 For identity outsiders, there is a rationale for having a dissenting view that legitimates its expression. For identity insiders, it may be harder to justify why one is disagreeing with one’s fellow identity group members, leading to a dampening of confidence and of the expression of unique viewpoints. Cognitive diversity, even when it is present in an identity-homogeneous group, is less likely to be voiced with confidence and less likely to be accepted by others in the group as important and valid. There are caveats here that will be considered later. For instance, if the identity outsiders feel they are not welcome in the group, feel they are not respected enough by others, and expect a high cost for speaking up, then the benefits of identity diversity—the diversity bonus—are less likely to be captured.
The third psychological process underlying the value of identity diversity stems from individuals’ motivation to maintain balance in social relationships with identity-similar others. Specifically, Phillips, K. A. Liljenquist, and M. A. Neale argue that the members of the identity majority who agree with identity outsiders feel socially insecure because an alliance with outsiders threatens their social ties with other same-identity members in a group—imagine a man agreeing with a woman in a group with other men present.19 This threat motivates individuals to reconcile the divergent opinions in the group and contributes to better processing of the available information, better decisions, and better problem solving—a diversity bonus. The motivation to restore balance with similar others may ultimately improve group performance as members pay more attention to the task in an effort to reconcile the different opinions.20 Thus, even when identity outsiders do not bring divergent viewpoints, or cognitive diversity, to the table, their mere presence can fundamentally change the behavior of the identity majority and enhance group performance.
Taken together, rather than assuming that people who are different in terms of their identity are beneficial to a group only because they can introduce cognitive diversity to the group, the psychological processes just described provide additional explanations about why and how identity diversity can lead to a diversity bonus in groups. Identity diversity triggers expectations that cognitive diversity may be present in groups and legitimizes the expression of unique perspectives and knowledge from both identity insiders and outsiders. In addition, the presence of social category diversity can decrease conformity to socially similar others in a group, which ultimately leads everyone to voice unique perspectives more confidently. Finally, the desire to restore social ties with identity-similar others can benefit groups by increasing the discussion of differing information and knowledge.
Page acknowledges that cognitive diversity and identity diversity must both be considered and that identity diversity is important in groups—especially through the opportunity for representation. I have provided insight into the evidence that shows that the mere presence of identity diversity can alter how hard people work, how people think, and how teams perform. Identity diversity has an influence on people’s cognition and can paradoxically drive the processing and consideration of cognitive diversity toward more effective outcomes. It may be difficult to fully capture the benefits of cognitive diversity in the absence of identity diversity—this argument adds to Page’s claims regarding why we need identity diversity in teams and organizations. My premise is that cognitive differences might go unnoticed and underutilized if there are no triggers from identity differences to let us know that the cognitive differences are there, are valuable, and can benefit the group.
So, if there are so many clear and multifaceted benefits of identity and cognitive diversity in groups, why is there resistance or difficulty around increasing identity diversity in organizations? This question deserves an entire book of its own, but one conclusion I would like to highlight here that I have drawn from my research is that the hard work associated with diversity is not particularly welcomed by group members. Even if the leadership of an organization touts the benefits of diversity, the daily experience of diversity may lead people to feel as though the benefits do not outweigh the downsides of engaging in conflict and disagreement with others, being uncomfortable, working hard, and questioning one’s own perspectives and opinions. Diversity is difficult, and it is often hard to see the benefit in performance in organizations right away. How do we know if we have made the right decision, created the best product, or found the best solution to a problem when we are working on a daily basis in organizations? We often do not, especially in the places where Page highlights that diversity bonuses are likely to emerge. This can make it difficult to convince individuals that diversity is beneficial in comparison to the comfortable, more homogeneous environments that they have developed confidence in. In fact, in the Phillips, Liljenquist, and Neale paper, although the homogeneous groups were outperformed by the diverse ones in terms of finding the right answer, when asked how confident they were in their decision and how effective their groups had been, the homogeneous groups reported greater confidence and effectiveness than the diverse groups.21 There was a misalignment of actual performance and the feelings of the group members. This disconnect is a significant barrier for diversity, making this volume of Our Compelling Interests even more important. There are many other barriers to the acceptance of identity diversity, but it doesn’t make it any less true that diversity bonuses can emerge for complex, difficult problems. We should not stop chasing after the benefits of diversity because they are difficult to capture or because some are resistant. Diversity requires us all to work hard, and to embrace the change required in ourselves and in others around us.
SITUATIONAL FACTORS IMPORTANT FOR CAPTURING DIVERSITY BONUSES: THE EXAMPLE OF CROSS-FUNCTIONAL TEAMS
Let’s now turn to the example of functionally diverse teams to explore the situational factors that may play a role in the emergence of diversity bonuses. Nearly twenty years ago I wrote a seminal review of the organizational demography and diversity literature with collaborator Charles O’Reilly titled “Demography and Diversity in Organizations: A Review of 40 Years of Research.” We explored many, but not all, types of organizationally relevant sources of identity diversity in the paper, guided mostly by the available literature at the time. Of all types of identity diversity we reviewed—race, age, gender, tenure, and educational and functional background—functional background diversity was the most consistently positive for group and organizational performance.22 Let me be clear: introducing functional diversity was not easy, and it did not always work. Functionally diverse groups also face many of the same problems that other identity-diverse groups face. However, in many ways, the very impetus for creating cross-functional teams was a recognition of the same underlying parameters defined by Page. Organizations were facing difficult problems that could not be solved by one individual, or the best individuals in a given identity (that is, functional) group, alone. There was a recognition that if their organizations were going to be able to innovate the best products and find the best solutions, that all members of the organizations could buy into, they would need to work together across functional silos in the organization. Organizations rapidly expanded their use of functionally diverse teams to solve difficult problems, create new products, and change processes in the 1980s. The theory was that more innovative, efficient, and effective solutions might be found if barriers between functional groups were broken down from the beginning of a project, creating more interdependence and collaboration across the organization. Essentially, the use of cross-functional teams was a precursor to our current conversations about the value of cognitive and identity diversity. Diversity was being promoted in these organizations, not by bringing new types of people into the organization but by changing structures, breaking down boundaries, and creating teams of people from different functional groups to benefit the organization.
With the expansion of cross-functional teams, organizations began using the diversity they obviously already had within their walls. Organizations that did this were using the same basic logic of diversity bonuses introduced here by Page: difficult problems could better be solved by bringing together an identity-diverse group of individuals with diverse cognitive repertoires instead of working within homogeneous groups. People with different functional backgrounds (that is, identities) have been exposed to different information, knowledge, mental models, and representations. Thus, when exposed to new information in pursuit of a difficult problem, the diverse members of the cross-functional team might remember different things, encode the information differently, and highlight some information as important and relevant that others do not see as such. Cross-functional team members bring both cognitive diversity and identity diversity to the group in perhaps more predictable ways than most types of identity diversity. Functional identity differences serve as a trigger for inferring reasonable and acceptable cognitive differences that could benefit the organization.
The Congruence Effect
I believe there are a number of instructive reasons these functionally diverse teams were relatively successful when they were first introduced as a solution to the difficult problems organizations were facing at that time—factors that might be important for capturing diversity bonuses in organizations today. First, in functionally diverse teams, the cognitive diversity being brought to bear on the problem at hand is predictable. Group members know where they might be able to find that cognitive diversity within the group because the identity diversity (that is, functions) is meaningful and trusted. For instance, if a cross-functional team with a lawyer, marketing expert, finance person, and software engineer are put into a group together, there should be little to no concern with assuming that the finance person will be able to answer questions and bring relevant knowledge to the group that stems directly from her identity as the finance person on the team. In functionally diverse groups, identity diversity (that is, one’s functional identity) is safe. It is more likely to be aligned with cognitive diversity than in other cases. This alignment or congruence between identity and cognitive diversity is an important factor for helping teams effectively identify, share, and integrate disparate knowledge to solve problems.23
Page acknowledged that identity diversity is only correlated with cognitive diversity. This correlation should be particularly high for functionally diverse groups. In organizations, individuals who have different functions are supposed to have different information, knowledge, and mental models. The way they approach the problem, the concerns they have when creating a new product, and the perspective they bring to the process being examined all differ due to their functional identity. These differences are not only acceptable, they are desired, and this is the very reason the group has been put together. These differences in information, knowledge, and mental models develop deliberately for those with functional background differences. We want our finance colleagues to bring knowledge about finance to bear and our colleagues in marketing to bring marketing knowledge to bear. You would never hear someone lamenting, “I feel like I am representing all finance professionals when I speak up in this group,” though you might hear a similar sentiment stated by a woman or person of color when participating in a diverse group.
People like their expectations to be met, so this congruence can be beneficial for the group.24 As discussed previously, groups are more willing to discuss and accept novel information as useful and relevant when it comes from those they expect to have a different perspective.25 This use of information can then improve decision making and problem solving in the group. There is no controversy in believing that functional background diversity as an identity should be connected to more cognitive diversity. However, people are often wary when this same logic is applied to other identity categories in society, such as race and gender.
Page’s diversity-bonus statement does a great job of dispelling this concern for me, given the many examples he provides, but, for the sake of directly addressing the underlying problem here, let me say that it is one of status, not diversity. For instance, women and men might indeed have relevant cognitive repertoires that differ systematically for a given problem, but this does not have to be true for a diversity bonus to emerge. And if they do differ, this does not mean that men’s cognitive repertoires are better (more meritorious) than those of women or that diversity bonuses cannot emerge in gender-diverse groups. The assumption that identity and cognitive diversity will be congruent acts as a constraint on the value some people place on one subgroup’s cognitive diversity relative to another’s (for example, when Larry Summers was serving as president of Harvard University, he made a statement implying that women cannot succeed at science because their cognitive repertoires cannot be as good as the cognitive repertoires of men in this male domain because they are women). For functional background, it is easy to understand how the training, education, and deliberate development of particular mental models emerge. However, it is not so easy for other types of differences, such as race and gender. People may find it more difficult to believe that a female engineer will necessarily have a mental model about how to design a car that differs from that of a male engineer, and this difficulty will make it more difficult to understand the need for identity diversity in the field of engineering. Remember that the benefit of having identity diversity among the engineers arises not only because of the possible cognitive diversity the female engineer might introduce due to her life experiences but also because of the changes in behavior triggered by her mere presence. The bottom line here is that functional diversity is safe. People’s expectations about who knows what are aligned with who is from what identity group, and it is okay for people to make that assumption. For many other types of difference, this is more difficult.
What do we do to manage this then? The answer is simple in my mind: all identity groups have to be respected and valued in the same way that functional identities are seen as important. The cognitive diversity contributed to the group must be respected, and the diversity bonus that emerges from mere presence must be embraced for all types of identity differences, not just functional ones.
Common and Clear Goals
The second factor that makes cross-functional teams an interesting case study to consider is the fact that when they are put together, they are usually composed with the express purpose of achieving a particular goal. Having a common goal is one of the most important prerequisites for success in groups. Thus, one might find that as cross-functional teams become routine, and perhaps overutilized, in an organization, their success might decrease—not because diversity is no longer relevant, but instead because the reason for their existence, their common goal, is no longer clear. Tying the composition of a group to its goals may be an important precursor for the emergence of diversity bonuses. In the case of cross-functional teams, not only is there often a clear common goal, but accomplishment of that goal is directly served by the diversity that is present. This allows for the emergence of respect, acceptance, and a valuing of the identity and cognitive diversity in the group. When identity-diverse groups are composed, there is a possibility for cognitive diversity to emerge. The elements for success are there. However, the group members need to explicitly agree on what they are trying to accomplish together in order to harness the diversity bonus. Do not assume group members know what the goal of the group is; always make it explicit, repeat it often, and revise it as necessary to ensure it is understood by all group members.
Diminishing Status Differences
Third, even though functional differences can fuel the same negative concerns in groups that other identity differences can—less cohesion, increased interpersonal conflict, lack of trust, and difficulties with communication, to name a few—when cross-functional teams are put together, each different functional identity group may be more clearly needed for the success of the team. This helps reduce the status differences between different functional groups that might fuel negative effects in other identity-diverse groups. Even if the organization has a natural status hierarchy among functional identities—for example, the lawyers are seen as having a higher status than the marketing members—the team may experience more equality in status and see the value of all members for the team’s success. In Page’s statement, all cognitive repertoires, even if not all considered “the best,” are, within reason, considered important to the success of the team. When unconscious biases and stereotypes stop group members from integrating the cognitive repertoires of their teammates, diversity bonuses are more difficult to capture. When teams are purposefully composed, irrelevant status differences that might stem from identity diversity must be set aside for diversity bonuses to emerge.
Leadership Support
Finally, cross-functional teams are likely to be successful because they have the support of leadership in the organization. Identity diversity requires support by leadership to model appropriate behavior and attitudes about difference. Members of cross-functional teams, when they were initially being used, were often handpicked for participation by organizational leaders who not only believed in the concept of these functionally diverse teams but also invested time and resources in ensuring that the team met its potential. As Page noted, having a diversity bonus emerge is not a guarantee. Even under the best of circumstances, in which cognitive and identity diversity is abundant, there are no guarantees that diversity bonuses will emerge. Organizational leaders have to support diverse teams. Organizations must deliberately shape the environment, provide effective leadership, and support diversity efforts to ensure the success of diverse teams. There must be a regular commitment to the goal of diversity. Are all diverse teams supported by leadership? Are all diverse teams deliberately assembled for the cognitive and identity diversity of the group? Are all diverse teams constituted with a clear goal in mind? These are questions that Page did not pursue in his treatise, but they are important to consider in making diversity bonuses a reality.
CLOSING QUESTIONS AND ASSUMPTIONS
Despite the complexities of diversity, I agree with Page on the conclusion that cognitive and identity diversity can produce more ideas and possible solutions. As noted previously, this can happen because people who are different can bring different ideas to the table, but also and importantly because, as Mill proposed back in 1848, being exposed to difference changes people, helping them to generate more ideas and think more deeply about problems.26 The debate should not be over whether diversity can provide bonuses. It can, under the right circumstances. The question really becomes, will diversity provide bonuses? And I would say that really depends on us—it depends on how we manage our biases and stereotypes, how we lead in our teams, and how we manage our expectations about difference.
Important from my perspective is also the issue of why we need to answer this question of the business case for diversity in the first place. Why is it necessary to prove that diversity has benefits? Is it just as necessary to prove that homogeneity has benefits? The quest for a business case reveals important motivations and defaults in society. The default is homogeneity. People who look like the majority, the incumbent, do not have to justify their existence and presence in organizations. People who deviate from the prototypical majority do. The business case for diversity assumes that diversity is optional, that one could have diversity or one could not. In most organizations this really means that white, male, heterosexual, able-bodied individuals are welcomed as the default prototypical member. There is no surprise when they are members of the organization, when they are leaders.27 They are expected to be there in the organization. Anyone else who is added to the organization is a “choice,” a deviation from the norm, and they must prove that they deserve to be there—women, people of color, disabled individuals, and so on. The logic of the business case for diversity rests on the idea that those who deviate from the norm must demonstrate that their presence in the organization will result in the company making more money. Imagine walking into a company and having the burden resting on your shoulders to prove that you can make the company a million more dollars in profit. That is a heavy burden to bear. We should reflect on how seeking evidence of a business case for diversity reifies the status quo and legitimates the idea that some people belong and deserve to be included in organizations, while other people have to go above and beyond to prove their worth. The questions we ask about diversity have power. They have underlying assumptions that often go unspoken.
As I close this commentary, I would like to highlight fifteen assumptions one should make to help diversity bonuses emerge, though there are many others: (1) assume there is cognitive diversity in the room that just needs to come out; (2) assume that identity differences among group members can promote the expression of cognitive differences; (3) assume that one’s gender, race, sexuality, and so on, identities that she may have been walking around with her entire life, will have an influence on the way she sees a problem and what experiences she brings to the table, in the same way that you assume people’s functional training will affect them; (4) assume that there is not perfect overlap between the identity characteristics a person possesses and his opinions about a given problem—dig deeper; (5) assume that making difficult decisions, solving complex problems, creating processes, expanding into new markets, and developing innovative products will benefit from both representation of the constituents being served and diversity; (6) assume that people care about maintaining relationships with similar others and sometimes will sensor themselves to maintain harmony—this is just as big a force in diverse groups as the forces that are in operation between people who are different from each other; (7) assume that people will filter how they hear information depending on who is sharing that information; (8) assume people will filter what they are saying depending on whom they are talking to; (9) assume that diverse environments need to be managed to ensure all voices are heard and considered; (10) assume that great ideas may not be recognized as such when they are the product of a diverse group—take a closer look; (11) assume people will be more cautious in their support for diversity because the downsides are often easier to see and feel than the upsides (especially in the short term); (12) assume that status differences exist and that some people in the room are bound to be respected more than others, so group process is critically important for getting everyone’s ideas into the room and heard by others; (13) assume that everyone in the room is bringing implicit bias about others with them and that these implicit biases can have detrimental effects on interactions, judgments, and outcomes in the group, even when people have the best of intentions; (14) assume that structures need to be put into place to help combat these biases, as the problem does not solve itself; and, finally, (15) assume that you know something other people don’t know and that other people know something you don’t know.
Diversity bonuses are real, they can and do occur, and each of us has a role in making them happen.