Table 1
One of the first lessons teachers impart on young children is the virtue and value of cooperation. Think of kindergarteners tidying up their classroom together or helping each other with a shared task. These teachers are reinforcing the basic instinct that working together makes everyone better off. And the value of cooperation doesn’t end in the classroom.
Cooperation underpins the development of 21st century skills that are essential for advanced economies and civic life. It’s implicit in the Constitution’s democratic goals of establishing justice, ensuring domestic tranquility, providing for the common defense, promoting the general welfare, and securing the benefits of liberty. Equal protection under the law, codified by the 14th Amendment, requires cooperation between groups and government jurisdictions to interpret, enforce, and expand these protections.
The challenge is that the United States—with its history of racial stratification and subjugation—is one of the most diverse “classrooms” in the world, comprised of a multitude of people from different backgrounds. And we are still far from realizing the full, shared benefits of this diversity, evident in the significant racial gaps across wealth, real estate ownership, educational access, and more.
Today, divisive rhetoric and policymaking threaten to erode our basic understanding of why cooperation is essential to a functioning democracy—altering not only public discourse, but also distorting how we interact with each other on a daily basis. We regularly witness politicians pitting groups against each other for electoral gain; social media amplifying discord and racism; and political strategists leveraging gerrymandering and opaque rules to fracture civic cohesion.
In an environment in which racial competition is frequently emphasized over common civic ideals, it is essential to ask: How do Americans of different backgrounds actually engage and cooperate with each other in daily life?
The Brookings Institution’s Center for Community Uplift—a research center focused on increasing economic security and well-being across race and place—collaborated with Gallup to answer that question by documenting the extent of interracial contact among Americans and analyzing its links to well-being and satisfaction. Researchers surveyed a nationally representative sample of about 5,000 U.S. adults, asking detailed questions about work relationships, friendships, romantic ties, and family connections to assess how frequently these relationships involve people from other races. The survey also included a hypothetical hiring scenario to gauge racial preferences in selecting a candidate for a job. In addition, respondents reported attitudes toward discrimination and the relative importance of race versus other characteristics when choosing friends and business partners.
This first-of-its-kind survey assesses how Americans perceive their interactions with members of different racial groups as well as how they rate the quality of those interactions in work and social settings. It is not a survey that assesses the extent of structural racism or makes an argument for or against diversity initiatives; claims of that sort cannot be derived from the data presented.
There are two papers associated with this survey. The technical report provides detailed methods and major findings with minimal commentary. The analysis here presents the survey’s major findings, outlines the broader project’s motivation, contextualizes the results, and offers interpretive policy insights alongside necessary cautions.
To understand the state of interracial interactions in America, the Brookings Institution’s Center for Community Uplift partnered with Gallup to conduct a large, nationally representative survey.
Gallup mailed invitations to 100,000 U.S. addresses, using a sampling strategy that intentionally oversampled Black, Latino or Hispanic, younger, and less educated households to ensure their views were adequately captured. The final sample included 4,883 U.S. adults. Results were statistically weighted to accurately represent the entire U.S. adult population.
The survey asked Americans about their real-world relationships and attitudes on race in three key areas:
For more detailed descriptions of the survey’s methodologies (including recruitment, sample, survey questions, and the hypothetical hiring scenario), please refer to the technical report.
Workplace diversity in the United States is increasingly reflected in daily professional interactions, with a strong majority of Americans now regularly engaging with co-workers and clients from different racial backgrounds. Data from our survey show that 81% of U.S. adults have at least one co-worker or client of a different race.
However, exposure to supervisors of a different race remains less frequent. In our survey, 48% of workers report having a boss or supervisor from a different racial group. White workers report significantly less exposure to racially diverse managers, reflecting persistent racial sorting in workplace hierarchies. While workplace diversity has increased horizontally, there is still work to be done to increase minority representation in leadership and management positions.
While many Americans experience diversity at work, actual rates of interracial contact often fall short of what demographics would predict. This becomes clear when comparing the racial composition of workers’ professional networks to overall U.S. population figures.
A detailed comparison reveals that most racial groups work with same-race colleagues more frequently than national statistics would imply. For instance, although 88% of the U.S. population is non-Black, survey data show that only 60% of Black professionals’ clients are non-Black—indicating that their client networks remain disproportionately same-race. Roughly half of this divergence appears attributable to geographic segregation: The average Black respondent in our survey lives in a county that is 74% non-Black. A similar pattern emerges among Latino or Hispanic respondents.
In contrast, Asian Americans’ client networks more closely mirror national demographics: 88% of their clients are non-Asian American, which is close to the 93% non-Asian American share of the U.S. adult population. Finally, white Americans report having more nonwhite clients than their share of the population would suggest.
These findings highlight meaningful variation in exposure to racial diversity across groups and point to structural factors, such as residential segregation patterns, that may partly explain the gaps.
When compared to national demographics, all major racial groups report working with same-race co-workers more often than expected. This gap is smallest for white workers (just 1 percentage point) and largely explained by local population patterns. Black and Latino or Hispanic workers report much stronger same-race clustering: They are far less likely to have co-workers or managers of a different race than their share of the U.S. population would suggest. For example, while 91% of all U.S. managers are non-Black, only 64% of Black workers report having a non-Black manager.
In contrast, white workers report being managed by nonwhite supervisors at a rate relatively close to their population presence: Nonwhite managers make up 28% of all managers surveyed, while 19% of white workers reported having nonwhite managers. This also reflects occupational segregation; minority groups are more likely to be in roles or companies that limit their exposure to other racial groups, particularly in a supervisory relationship.
Looking at age, interracial contact in co-worker and manager relationships remains similar across age groups. Client relationships show the strongest generational divide: 84% of respondents under age 30 have a client from a different race or ethnicity, compared to 70% of respondents age 70 and older.
Notably, political affiliation has no meaningful bearing on workplace diversity. Democrats, Republicans, and independents report nearly identical rates of contact with co-workers, clients, or managers of different racial backgrounds.
Overall, interracial exposure in the workplace shows no broad association with job satisfaction among Americans, yet these effects vary meaningfully across racial and ethnic groups. While most groups report modest declines in job satisfaction when working with people of different races, Black workers stand out as an exception. They express greater job satisfaction when working with more diverse co-workers, managers, and clients.
Latino or Hispanic and white workers, in contrast, report lower job satisfaction when their supervisors are from a different race. Similarly, Asian American and white workers show slight reductions in job satisfaction when serving clients from other racial backgrounds.
When it comes to co-workers, having racially diverse teammates is linked to small but significant decreases in satisfaction for Asian American, Latino or Hispanic, white, and multiracial Latino or Hispanic-white workers. Once again, Black workers diverge, showing a significant positive effect.
These patterns suggest that the impact of interracial exposure in the workplace is not uniform. It reflects differing experiences, expectations, and workplace climates across groups. While these correlations are revealing, it’s important to note that exposure is not randomly assigned; underlying factors such as regional demographics, workplace power dynamics, organizational culture, and differences in how various groups experience interracial interactions at work may partly drive these outcomes.
A crucial caveat: Although these relationships are statistically significant, the effect sizes are minimal (the standardized co-efficients shown here rarely exceed 0.35, and most hover near zero). Put simply, if you want to understand why workers are satisfied or dissatisfied with their jobs, the racial makeup of their co-workers, supervisors, or clients is not a major driver. This suggests that concerns about diversity harming workplace satisfaction are largely unfounded, but it also means diversity alone—without supportive structures—won’t dramatically improve satisfaction either.
To assess the role of race in hiring preferences, the survey incorporated a randomized experimental design. Respondents were asked to evaluate two hypothetical job candidates for two types of jobs: a software developer and a home renovator. The job candidates were characterized by a varying set of attributes, including race (Black or white), professional experience, educational background, references, age, gender, nationality, interview performance, perceived intelligence, and social proximity. Race was systematically randomized among these attributes to isolate its influence on selection. A full description of the experimental design and analytical approach can be found in the technical report.
Overall, the experiment’s findings indicate that race plays a small but statistically significant role in shaping hiring preferences. Interestingly, respondents displayed a slight preference for Black candidates over white candidates for both job types. However, other professional and qualitative factors substantially outweighed this racial effect.
Candidate race accounted for just 4% of the variation in preferences for the home renovation role and 5% for the software developer position. Instead, interview performance was the dominant factor, explaining 37% and 44% of preference variation in the two roles, respectively.
For the software developer role, qualifications such as relevant experience, perceived intelligence, a computer science degree, and strong managerial recommendations proved far more influential than race. Similarly, for the home renovation job, trade-specific credentials, years of experience, neighbor recommendations, and demonstrated competence were stronger predictors of hiring choice than race. Demographic characteristics such as sex, age, and country of birth had less influence than race for both job types.
Consistent with intergroup contact theory—which proposes that “through meaningful, institutionally supported collaborative interactions between members from different social identity groups, prejudice will be decreased”—greater exposure to Black individuals among non-Black respondents was associated with an increased preference for Black candidates. This relationship was tested using a model that estimated the strength of pro-Black hiring preferences as a function of several measures of cross-racial exposure: share of Black co-workers, share of Black friends, and share of recent dates with Black individuals. In each case, higher levels of exposure predicted a significantly stronger preference for Black candidates, with the strongest effect observed for dating share, followed by Black friend and co-worker share.
Additionally, Black respondents who perceived anti-Black discrimination as particularly harmful to their lives were more likely to exhibit a pro-Black hiring preference. Political affiliation also emerged as a significant predictor: Democratic respondents expressed stronger support for Black candidates than Republicans.
Interracial friendship is a common feature of American social life, with 54% of survey respondents reporting they have at least one close friend from a different racial background. These close friendships—defined as relationships in which individuals can turn to one another for meaningful support—represent an important dimension of social cohesion in a diversifying nation.
Interracial friendship patterns vary meaningfully across racial groups in our survey. White adults are the least likely to have a close interracial friend, though nearly half (48%) do. By contrast, multiracial adults report substantially higher rates of interracial close friendships, especially those who are white and Latino or Hispanic (71%) and white and American Indian/Alaska Native (74%). Around 64% of Asian Americans, 60% of Black Americans, and 59% of Latino or Hispanic Americans report they have a close friend of a different race.
Interracial connections also extend into broader networks of casual friendships. Sixty-four percent of adults have at least one casual friend from another race, and 72% have some form of interracial friendship, whether close or casual. The average American has three close friends and seven casual friends, suggesting that friendship networks, while still shaped by racial homophily, are far from monolithic.
Notably, nearly all groups—with the important exception of Black adults—report greater racial diversity among their casual friends than their close friends.
Friendship patterns in the United States continue to reflect underlying social and geographic structures. To better understand the role of local context, survey researchers compared the racial composition of respondents’ friendship networks to the demographic profiles of the counties in which they reside.
The data reveal a consistent trend: Asian American, Black, and Latino or Hispanic adults form friendships that are less racially diverse than the populations of their surrounding areas would suggest. For instance, Asian Americans live in counties where, on average, 84% of residents are non-Asian American—yet only 46% of the friends in their social circles are non-Asian American. An even more pronounced discrepancy exists for Black Americans, who reside in counties that are 73% non-Black on average, but report that just 29% of their friends come from non-Black backgrounds.
In contrast, white, American Indian-white, and Latino or Hispanic-white Americans maintain friendship networks that are somewhat more diverse than their county’s racial composition would predict. These patterns highlight how social segregation often persists even within demographically mixed areas, suggesting that mere proximity is not always sufficient to foster cross-racial connections.
Although structural and social factors often lead Americans to form friendships within their own racial groups, explicit racial bias appears to play a minimal role in how respondents in our survey say they choose their friends. When asked how important race is in deciding whether to befriend someone, just 2% describe it as “very” or “extremely” important. The overwhelming majority (83%) say race is “not at all important.” Far greater importance is placed on shared interests, values, and personality.
This pattern holds across racial and ethnic groups. Majorities in every group report that race is “not at all important” in friend selection, though small differences emerge at the margins. Single-race white and Latino or Hispanic adults—as well as multiracial American Indian-white adults—are the most likely to dismiss the importance of race. Black and Asian American adults are slightly more likely to say race is important, but still, only 8% of Black adults and even fewer in other groups describe race as “very” or “extremely” important.
These results suggest that while actual friendship networks may reflect demographic and geographic segregation, very few Americans consciously prioritize race when considering whom to befriend. Interpersonal qualities such as shared values and common interests appear to be the dominant drivers of friendship formation in respondents’ stated preferences.
Interracial families represent one of the most profound forms of racial cooperation in American society, built on daily acts of shared life, mutual understanding, and often, the nurturing of multiracial children.
The prevalence of interracial marriage has risen dramatically over recent decades. In 1980, just 3% of married couples were interracial; by 2015, that share had grown to 10%. The most significant increases occurred among white and Black Americans: interracial marriage rates rose from 4% to 11% for white adults and from 5% to 18% for Black adults. Asian and Latino or Hispanic Americans continue to lead in intermarriage rates, at 29% and 27%, respectively.
This growth reflects a sweeping shift in social attitudes. In 1958, the first year Gallup surveyed attitudes about marriage between white and Black Americans, only 4% of U.S. adults approved of such marriages; by 2021, approval had reached 94%. Behavior has followed belief: Today, roughly 1 in 4 Americans are in an interracial romantic partnership, with Asian Americans most likely and white Americans least likely to have a partner of a different race according to survey results.
These relationships reshape not only couples’ lives, but entire family structures. According to the results of the survey, nearly a quarter (23%) of Americans now live with children of a different racial or ethnic background, which is more than twice the share who report having a parent of a different race (11%). This asymmetry tells us something important: We’re witnessing generational change in real time. Extended family networks are also diversifying: 19% of Americans have an aunt or uncle of a different race, and 25% have a cousin of a different race.
As interracial marriage becomes increasingly common, it creates organic, enduring bonds that reach across racial lines—transforming family life and broadening social networks in ways that reinforce intergroup cohesion.
Interracial family ties are far more common among younger generations, reflecting a decades-long rise in interracial relationships. While about one-third of Americans under 30 are in an interracial romantic partnership, only 1 in 7 Americans age 70 or older report the same. This generational divide is even more pronounced when it comes to parental relationships: Americans ages 18 to 29 are twice as likely as those 70 and older to have a parent of a different race.
Household composition reveals a more nuanced picture. Across all age groups, roughly 1 in 4 Americans live with a child of a different race. This includes a notable share of older adults (about 25% of those age 70 and older), many of whom likely serve as grandparents or kin caregivers to multiracial grandchildren or relatives. These patterns highlight both the deepening reach of interracial families and the evolving structure of American households across generations.
The presence of multiracial children in homes of all ages underscores how interracial relationships are reshaping not only couples’ lives, but also extended family networks—reflecting a broader societal shift toward more diverse and connected family experiences.
In a reflection of growing concern over the forces shaping public discourse, New York Times columnist Thomas Friedman said on an episode of “Real Time with Bill Maher” in 2025, “We aren’t divided; we are being divided by [social media] companies for profit…It is now a giant industry to make people stupid and angry.” Daily Texan Columnist Maria Vazquez expressed a similar sentiment in more industry-specific terms: “Engagement is the vitality of social media, and online outrage isn’t just a byproduct—it’s an essential component of its business model…Companies favor content that appeals to emotions over reason, which keeps consumers engaged but results in the rapid spread of misinformation.”
The findings from our survey on racial cooperation affirm what Friedman and Vazquez have sensed: The dominant discourse on racial division that digital and media feeds have propagated does not reflect our day-to-day feelings and behaviors. People are not as committed to division as their feeds may suggest; to the contrary, the evidence suggests that most people aspire to bridge divides.
Our survey and analysis establish a constructive baseline for tracking how Americans of different backgrounds live, work, and build relationships with one another. It’s a starting point, measuring the breadth of racial interaction rather than its depth, which involves trust, shared goals, and equitable power dynamics.
Still, the survey’s findings offer a largely optimistic snapshot. Today, a majority of Americans routinely engage with people of different races at work, in friendships, and within their own families. And exposure to other races predicts more positive interracial attitudes and perceptions, validating research showing that meaningful contact shatters stereotypes and builds empathy.
The data reveal that workplaces are often sites of organic diversity, with most respondents reporting interracial colleagues and clients. Perhaps even more encouraging, explicit racial bias appears minimal in personal choices: When selecting friends or partners, shared values and interests overwhelmingly outweigh race. Over 80% of Americans say race is unimportant in choosing friends or business partners—a powerful consensus that reveals a deeply held commitment to equality. The rising prevalence of interracial marriages and multiracial families further underscores a society increasingly building intimate bonds across racial lines.
There are three considerations that must be considered when absorbing these findings:
These self-reported attitudes, while meaningful, also reflect how we wish to see ourselves. They may represent our ideals more than our lived reality. Individual responses capture personal attitudes at a moment in time and should not be interpreted as explaining complex sociological phenomena, such as a supposed disconnect between stated values around equality and political behavior.
As the data show, evidence of discrimination in the workplace persists. As the technical report states, the absence of widespread anti-Black bias in our findings stands in contrast to results from audit studies in which researchers sent fictitious resumes—identical except for racially suggestive names—to real employers. Those field experiments consistently show a small but persistent anti-Black bias in callback rates. In addition, outcomes such as the homeownership gap between Black and white families have actually widened since 1968. Schools and neighborhoods remain deeply segregated, limiting opportunities for the very connections so many claim to value.
These contrasts are not contradictions—they are opportunities. According to the “contact hypothesis,” sustained cooperative interaction can improve intergroup attitudes, which numeric diversity alone does not guarantee. Progress requires more than circumstance; it demands intentionality. The workplace, often more diverse than other areas of life, offers a promising space to translate belief into behavior. To unlock the potential our diversity offers, business, education, and civic leaders must build environments where interracial cooperation is structured, sustained, and rewarded—where shared goals and genuine collaboration can bridge the gap between aspiration and outcome.
Renewing our shared capacity for connection will require not only individual choices, but also structural reforms that transform the systems shaping public attention, residential engagement, and the inequitable distribution of public resources.
Residential context plays a foundational role in shaping interracial exposure and social cohesion. Past housing policies such as redlining, racially restrictive covenants, single-family zoning, exclusionary school districting, and inequitable public school financing have systematically limited opportunities for meaningful interracial interaction. Because residential placement is not random, nondiscriminatory place-based policy is needed to foster diverse, connected, and inclusive communities.
To support meaningful racial cooperation, policymakers should prioritize:
Workplaces serve as critical sites for meaningful interracial cooperation, yet realizing their full potential requires moving beyond mere numeric diversity and toward intentionally inclusive environments. Research shows that the benefits of racial diversity are not automatic; they depend on reducing social categorization and fostering constructive interaction. While diversity can expand collective knowledge and innovation, it may initially strain cohesion if not supported by policies that promote equal status and shared goals among employees.
To maximize the benefits of diversity while mitigating its challenges, policymakers and organizational leaders should:
Future research based on these findings will take a deeper dive into the nature and quality of interracial interactions, examining issues surrounding productivity, economic security, and well-being at the firm, institutional, and community level.
Interactive feature created by Carie Muscatello and Alec Friedhoff