Abstract
Opening up and feeling heard are central to close relationships; in fact, understanding and disclosure with close others are beneficial for individuals’ well-being and quality of life. However, for people of color, understanding and disclosure may unfold differently depending on whether their close others share their racial/ethnic background. We examine this question with young Black, Latine, and Asian people in a cross-sectional national U.S. sample (N = 1285) and a weekly diary study (N = 101). In Study 1, young people of color felt more understood in same-race than in cross-race close relationships. Moreover, feeling understood in both types of relationships distinctly predicted depressive symptomatology one year and two years later. In Study 2, same-race understanding was uniquely associated with depressive affect and flourishing, but cross-race understanding was not. In both studies, same-race and cross-race disclosure did not differ or predict outcomes. Together, these findings suggest that young people of color disclose similarly in their close same-race and cross-race relationships but feeling understood is more directly associated with their psychological well-being.
Subject terms: Human behaviour, Psychology
Introduction
Whether with friends, family members, or romantic partners, close relationships matter to people of color. In North America, for instance, Black, Latine, and Asian adolescents and adults value and invest in their ties with others, including their families and communities1–7. The quality of their close relationships matters too, as gratifying relationships are associated with better mental health and greater well-being8. For example, an intensive longitudinal study found that when people of color report having higher-quality close relationships, they also feel more content and grateful daily9. Another intensive longitudinal study found that, for people of color, higher-quality close relationships was associated with more weekly flourishing10—people’s sense that their lives are purposeful, rewarding, and engaging11.
Understanding and disclosure have garnered attention as two factors influencing relationship quality and outcomes12,13. Felt understanding arises when people believe their partner knows how they think, feel, and view the world14. In contrast, disclosure describes the process of revealing intimate details about the self to a partner15. Both are beneficial: Relationships characterized by greater felt understanding and disclosure are associated with greater trust, intimacy, relationship satisfaction, and closeness16–18. Also, both are essential to deepening intimacy: Understanding is a key mechanism that unlocks the closeness needed to reap many relationship benefits, yet felt understanding does not easily emerge without people self-disclosing in their relationships19.
Furthermore, feeling understood is associated with a wide range of personal well-being outcomes, including daily physical symptoms, life satisfaction, depression, anxiety, stress, loneliness, self-esteem, and meaning in life20,21. Although feeling deeply understood by close others and confiding in them are critical for relationships and personal well-being22,23, past research suggests differences in understanding and disclosure across racial lines. For example, in the early stages of friendship formation, people of color report disclosing more and feeling more understood in same-race than in cross-race friendships24. People of color also report feeling more misunderstood in cross-race than same-race interactions in daily life22,23.
Known cross-race challenges with understanding and disclosure
Due to racial differences, unique challenges to feeling understood can arise in cross-race dyads. For example, people may have different motives when engaging in cross-race interactions, depending on their racial/ethnic background. Members of less powerful groups may approach interactions seeking respect, while members of more powerful groups may approach interactions wanting to be seen as likable and unbiased25,26. These differences in goals can pose challenges in accurately assessing how one’s interaction partner thinks and feels27. For example, White people with a higher desire to affiliate also less accurately assessed the extent to which a person of color felt understood because their goals to affiliate led them to focus on themselves28. Moreover, people may focus on different aspects of cross-race interactions, depending on their racial/ethnic background. White people focus on their explicit behaviors towards Black people, whereas Black people focus on White people’s nonverbal behaviors towards them29, resulting in different interpretations of the same interaction. Thus, unsurprisingly, short cross-race interactions tend to be associated with lower understanding than short same-race interactions, which in turn explains why cross-race interactions lead to less positive outcomes23.
Furthermore, people of color are cautious when they disclose in cross-race dyads. They may be wary about sharing racial experiences: Interaction partners from different backgrounds may deny their experiences and dislike them for sharing30–32. These negative expectations may even hinder their desire to disclose in cross-race relationships. For example, one study examining new relationships found that first-term Black students disclosed less to their White than to their Black roommates24.
Possible cross-race challenges in close relationships
Thus far, differences in understanding and disclosure have been found between same-race and cross-race dyads engaging in single interactions or in new relationships. However, the close relationships of adults of color have received less attention8,33,34. It is unclear whether differences found in single interactions or new relationships persist in long-term close relationships35. One possibility is that the self-selection and positive relationship processes that characterize close relationships eliminate these initial differences. Consequently, the challenges posed by short-term interactions may diminish over time or be less significant in long-term relationships36. For example, repeated cross-race interactions are associated with less pronounced negative effects than one-time interactions37. With repeated interactions, people may learn more about each other, leading to greater understanding and disclosure. They may also become more adept at empathizing and signaling that they understand each other14. Moreover, because long-term close relationships persist due to mutual appreciation and respect, cross-race partners may pursue similar goals rather than the differing goals of being liked and respected that characterize isolated cross-race interactions. Thus, close relationship partners may focus more on understanding each other rather than on their self-focused goals.
Despite factors that may reduce gaps in understanding and disclosure in same-race and cross-race relationships, some challenges persist in close cross-race relationships38. In terms of understanding, people of color do not always accurately perceive how their cross-race friend or romantic partner views their racial/ethnic group, and these misunderstandings may damage the quality of their relationships39. Feelings of understanding may be more challenging to cultivate in cross-race dyads, as it can be difficult for individuals with differing racial or ethnic backgrounds to relate fully to each other’s life experiences. In terms of disclosure, people of color can be ready to share racial experiences with same-race and cross-race friends, but nonetheless expect that sharing would unfold better with same-race friends and actually disclose less to cross-race friends40. Indeed, Marshburn and Campos41 found that Black students seek out other Black friends when they want to share racially loaded experiences. Thus, even in close relationships marked by more trust and history than short-term interactions and where high understanding and disclosure can be expected, differences may nonetheless exist between same-race and cross-race dyads. In turn, these differences may be consequential for the well-being benefits provided by close relationships, particularly for people of color.
Close relationships and well-being for people of color
Differences between same-race and cross-race dyads and the resulting relational dynamics tend to be reflected in the well-being of people of color. In the U.S. and Canada, for example, cross-race interactions tend to be associated with lower satisfaction, positive mood, well-being, and higher negative mood than same-race interactions for people of color22,42,43. These outcomes may also extend to close relationships. Black and Asian adolescents who only have cross-race best friends report lower well-being than those who only have same-race best friends44. Moreover, although young Black, Latine, and Asian adults report feeling highly supported and accepted in both their close cross-race and same-race relationships, same-race close relationships offer more support and acceptance, and same-race support and acceptance are more strongly associated with their well-being10. Yet, the links between the quality of people of color’s same-race and cross-race close relationships and their well-being are not well established.
Understanding and disclosure are necessary for relationship development, maintenance, and benefits. Although understanding and disclosure pose challenges in short cross-race interactions, how they play out in established long-term close relationships is less clear. Moreover, these differences may influence the well-being of people of color. In two studies, we take a first step in addressing this gap by examining the closest relationships of young adults of color. Using a large, national, longitudinal dataset and weekly diary design, we examined whether levels of felt understanding and disclosure differed between same-race and cross-race relationships. We also tested whether they predicted depressive symptoms one and two years later (Study 1) and whether weekly changes in understanding and disclosure in cross-race and same-race relationships predicted depressive affect (Study 2). Because well-being captures more than the absence of illness, we also examined whether weekly changes in understanding and disclosure were linked to changes in flourishing (Study 2). Taken together, the present studies provide a deeper picture of understanding and disclosure in people of color’s same-race and cross-race close relationships.
Study 1
The data comes from the National Longitudinal Survey of Freshmen—a study about race, school achievement, and cross-race relationships with comparable numbers of young Black, Latine, and Asian students. The project was designed by Douglas S. Massey and Camille Z. Charles, approved by University of Pennsylvania in accordance to U.S. guidelines and regulations, and subsequently by all participating colleges and universities (access data at https://opr.princeton.edu/national-longitudinal-survey-freshmen-nlsf).
Method
Participants
The present analyses included a subset of participants who represent 54.2% of the original sample: Students of color who reported about both same-race and cross-race relationships with close loved ones during their sophomore year (N = 1285, see Table 1 for details). We coded relationships by examining whether participants’ self-identified race/ethnicity matched the race/ethnicity they listed for their close loved ones. The racial composition of participants’ close others was diverse (see Table 2).
Table 1.
Study 1 sample composition by year of data collection.
| Black participants | Latine participants | Asian participants | ||||
|---|---|---|---|---|---|---|
| Men | Women | Men | Women | Men | Women | |
| Sophomore year (N = 1285) | 148 | 255 | 156 | 262 | 184 | 280 |
| Junior year (n = 1106, 13.9% attrition) | 120 | 221 | 129 | 231 | 151 | 254 |
| Senior year (n = 859, 33.2% attrition) | 88 | 177 | 103 | 180 | 119 | 192 |
To not compromise participants privacy, deidentified Study 2 data is available upon request.
Table 2.
Distribution and composition of cross-race dyads in study 1.
| Number of cross-race dyads listed by participants | |||
|---|---|---|---|
| One dyad | Two dyads | Three dyads | |
| Four close others listed | 379 | 398 | 492 |
| Three close others listed | 5 | 9 | – |
| Two close others listed | 2 | – | – |
| Cross-race partners in dyads listed by Ps | |||
| Ps only listed a person or only listed people of color | 176 | 90 | 30 |
| Ps listed two people of color, one white person | – | – | 92 |
| Ps listed one person of color, one white person | – | 130 | – |
| Ps listed one person of color, two white people | – | – | 145 |
| Ps only listed a White person or only listed White people | 210 | 187 | 225 |
We conducted multiple sensitivity analyses to determine the smallest effect size we could detect with sufficient power (i.e., 80% power). For multilevel models, we used Monte Carlo simulations with 5,000 resamples to conduct sensitivity analyses. For correlations, we used G*Power 3.1. For multilevel models, we could detect a within-person effect of r = 0.04 and a between-person effect of r = 0.06. For analyses examining junior year and senior year outcomes, we could detect an effect of respectively r = 0.06 and r = 0.07.
Procedure
During their sophomore year, participants provided “the first names of the four people you consider to be closest to you.” They then answered questions about each of their four close relationships, including one understanding item (i.e., “How often does this person understand what you really are like?”) and one self-disclosure item (i.e., “How often do you share your inner feelings with this person?”), using a 5-point Likert scale that was reverse coded to range from 1 (never) to 5 (always). We computed four scores for each participant: how understood they felt in same-race relationships, how understood they felt in cross-race relationships, how much they disclosed in same-race relationships, and how much they disclosed in cross-race relationships. During their junior year, participants reported how often they experienced 19 symptoms linked with depression during the past week, using a commonly used scale designed to be provided to the population at large (CESD scale45,46, α = 0.89), on a scale from 0 (Never) to 4 (All of the time). During their senior year, participants completed a 13-item version of the same scale (α = 0.88).
Results
We first examined whether participants reported differences in understanding and disclosure in same-race and cross-race relationships. We then tested whether understanding and disclosure in same-race and cross-race relationships reported during participants’ sophomore year predicted depressive symptoms during their third and fourth years of college. For all results, we reported bootstrapped confidence intervals with 5000 resamples in square brackets to address the non-normal distributions of our variables (Table 3). We calculated the effect size (r) by converting the t-statistic from our output47.
Table 3.
Descriptive statistics for key variables in studies 1 and 2.
| Variables | Range of responses | M (SD) | Skewness | Kurtosis |
|---|---|---|---|---|
| Study 1 | ||||
| Disclosure | 1–5 | 3.81 (0.91) | -0.37 | -0.29 |
| Understanding | 1–5 | 4.20 (0.80) | -0.58 | -0.60 |
| Depression Symptomology (Junior Year) | 0–2.89 | 0.99 (0.51) | 0.66 | 0.47 |
| Depression Symptomology (Senior Year) | 0–3.38 | 1.04 (0.56) | 0.80 | 1.32 |
| Study 2 | ||||
| Disclosure | 1–5 | 3.37 (1.30) | -0.40 | -0.88 |
| Understanding | 1–5 | 3.78 (1.09) | -0.70 | -0.14 |
| Depressive Affect | 1–7 | 3.50 (1.64) | 0.80 | -0.84 |
| Flourishing | 1–7 | 5.02 (1.06) | -0.70 | 1.05 |
For depression symptomology, the full range of the scale is 0 to 4.
Differences in understanding and disclosure
To test whether understanding and disclosure differed between same-race and cross-race close relationships, we used two-level multilevel models with random intercepts and relationship type entered as an effects-coded variable (− 1 = cross-race; + 1 = same-race), participants’ ethnicity as two effects-coded variables (-1 = Asian, 0 = Latine, + 1 = Black; − 1 = Asian, + 1 = Latine, 0 = Black,), and their interaction. We tested for the omnibus effect of participants’ ethnicity and the ethnicity-by-relationship type interaction using likelihood ratio tests. For significant interactions, we used dummy-coded variables to examine the simple effects.
Understanding
On average, participants felt more understood in same-race (M = 4.24, SE = 0.02) than in cross-race relationships (M = 4.17, SE = 0.02), b = 0.04 [0.02, 0.06], SE = 0.01, t(4299.98) = 3.55, p < 0.001, r = 0.05. [Footnote 1: When we separated cross-race close others into White individuals or other people of color, close others’ race significantly predicted understanding, χ2(2) = 13.46, p = 0.001. Participants felt more understood by same-race close others than by cross-race close others of color, b = − 0.10 CI95% [− 0.15, − 0.04], SE = 0.03, t(4499.43) = − 3.45, p < 0.001, r = 0.05, and White close others, b = − 0.05 CI95% [− 0.10, − 0.00], SE = 0.02, t(4540.52) = − 2.14, p = 0.032, r = . 03. Understanding did not differ between White close others and close others of color, b = 0.05 CI95% [− 0.01, 0.11], SE = 0.03, t(4839.32) = 1.54, p = 0.124, r = 0.02.] [Footnote 2: Using t-tests, we examined whether the people with same-race and cross-race relationships (N = 1285) differed from people who reported having same-race close relationships only (N = 574) in terms of understanding and disclosure. These analyses revealed that people who reported having only same-race close relationships (Munderstanding = 4.24, SD = 0.59; Mdisclosure = 3.81, SD = 0.66) did not differ from those who reported about both types of relationships (Munderstanding = 4.25, SD = 0.68; Mdisclosure = 3.82, SD = 0.77) in terms of understanding, t(1263.79) = 0.07, ptwo-side = 0.941, or disclosure, t(1269.36) = 0.23, ptwo-sided = 0.817.] Participants’ ethnicity was also a significant predictor, χ2(2) = 11.35, p = 0.003. Asian participants felt less understood than Latine, b = 0.13 [0.05, 0.20], SE = 0.04, t(1300.47) = 3.37, p = 0.001, r = 0.09, but not Black participants, b = 0.06 [− 0.01, 0.14], SE = 0.04, t(1285.84) = 1.69, p = 0.092, r = 0.05. Black and Latine participants did not differ, b = − 0.06.
[-0.14, 0.01], SE = 0.04, t(1307.09) = -1.61, p = 0.108, r = 0.04. The interaction was not significant, χ2(2) = 3.20, p = 0.202.
Disclosure
Participants reported similar levels of disclosure in same-race (M = 3.82, SE = 0.02) and cross-race (M = 3.81, SE = 0.02) relationships, b = 0.01 [− 0.01, 0.03], SE = 0.01, t(4250.11) = 0.61, p = 0.543, r = 0.01. However, participants’ ethnicity was a significant predictor, χ2(2) = 7.36, p = 0.025. Asian participants disclosed less than Latine, b = 0.12 [0.03, 0.20], SE = 0.04, t(1301.27) = 2.71, p = 0.007, r = 0.07, but not Black participants, b = 0.05 [-0.03, 0.13], SE = 0.04, t(1287.86) = 1.15, p = 0.252, r = 0.03. Black and Latine participants did not differ significantly, b = -0.07 [− 0.16, 0.02], SE = 0.05, t(1306.82) = -1.50, p = 0.134, r = 0.04. [Footnote 3: We also tested whether network racial composition moderated these effects. For disclosure, there was a significant network racial composition by participants’ ethnicity interaction, χ2(2) = 7.24, p = 0.027. For Latine participants, participants with more same-race close others reported lower disclosure than those with fewer same-race close others, b = − 0.20 CI95% [− 0.37, -0.03], SE = 0.09, t(1275.08) = -2.30, p = 0.022, r = 0.06. This association was not significant for Black or Asian participants, ts < 1.35, ps > 0.176.] The interaction was not significant, χ2(2) = 0.11, p = 0.947.
Understanding and disclosure as predictors of depressive symptoms
To examine the links between understanding and disclosure during participants’ second year and their depressive symptoms during their third and fourth years, we calculated partial correlations. In each set of correlations, the same-race (or cross-race) predictor was partialed out to examine the unique contribution of the focal cross-race (or same-race) predictor. Full results are reported in Table 4 with bootstrapped confidence intervals.
Table 4.
Study 1 Bivariate and partial correlations between understanding and depressive symptoms, and disclosure and depressive symptoms.
| Bivariate correlations | Partial correlations | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| n | r | Bootstrapped 95% CI | p | df | r | Bootstrapped 95% CI | p | |||
| Lower | Upper | Lower | Upper | |||||||
| Same-Race Disclosure (Partial Correlations Control for Cross-Race Disclosure) | ||||||||||
| Junior Year CESD | 1106 | − .05 | − .11 | .01 | .089 | 1103 | − .04 | − .10 | .02 | .194 |
| Senior Year CESD | 859 | − .04 | − .10 | .03 | .281 | 856 | − .05 | − .11 | .02 | .148 |
| Cross-Race Disclosure (Partial Correlations Control for Same− Race Disclosure) | ||||||||||
| Junior Year CESD | 1106 | − .04 | − .10 | .03 | .230 | 1103 | − .02 | − .07 | .05 | .622 |
| Senior Year CESD | 859 | .02 | − .05 | .09 | .598 | 856 | .04 | − .03 | .10 | .271 |
| Same-Race Understanding (Partial Correlations Control for Cross− Race Understanding) | ||||||||||
| Junior Year CESD | 1106 | − .21 | − .26 | − .15 | < .001 | 1102 | − .15 | − .20 | − .09 | < .001 |
| Senior Year CESD | 859 | − .14 | − .21 | − .08 | < .001 | 856 | − .10 | − .17 | − .04 | .003 |
| Cross-Race Understanding (Partial Correlations Control for Same− Race Understanding) | ||||||||||
| Junior Year CESD | 1105 | − .19 | − .25 | − .13 | < .001 | 1102 | − .12 | − .18 | − .06 | < .001 |
| Senior Year CESD | 859 | − .13 | − .20 | − .06 | < .001 | 856 | − .08 | − .15 | − .01 | .021 |
Biased-corrected and accelerated bootstrapped 95% confidence intervals with 5000 resamples are reported.
Students who felt more understood by same-race close others during their sophomore year reported experiencing fewer depressive symptoms during their junior and senior years, even when partialing out cross-race understanding. We obtained similar effects when cross-race understanding was the focal predictor: Students who felt more understood by cross-race close others during their sophomore year reported fewer depressive symptoms during their junior and senior years even after partialing out same-race understanding. [Footnote 4: We also recoded cross-race disclosure and understanding predictors to differentiate between disclosing to and feeling understood by White individuals and disclosing to and feeling understood by other people of color, resulting in four continuous predictors. We tested one model with disclosure predictors, and one with understanding predictors. We used these disaggregated predictors to predict depressive symptoms in junior and senior years. To determine whether these effects were unique to cross-race relationships with White individuals or other people of color, we controlled for the corresponding same-race predictor (i.e., understanding or disclosure) in a subsequent model. The resulting analyses revealed that none of the four disaggregated cross-race predictors predicted senior year depressive symptoms. We also tested whether disclosure to White individuals and disclosure to other people of color predicted junior-year depressive symptoms. Disclosure to White individuals did not significant predict depressive symptoms, t(310) = − 0.16, p = 0.871; however, cross-race disclosure to another person of color did, b = − 0.07 [− 0.13, − 0.001], SE = 0.03, t(310) = − 2.11, p = 0.036, r = 0.12. Greater disclosure to other people of color predicted fewer depressive symptoms during students’ junior year. When controlling for the effect of same-race disclosure, the effect of disclosure to other people of color was no longer significant. Same-race disclosure was not a significant predictor of depressive symptoms either. Finally, we tested whether feeling understood by White individuals and feeling understood by other people of color predicted junior-year depressive symptoms. Feeling understood by White close others did not significantly predict depressive symptoms, t(310) = − 1.66, p = 0.098; however, feeling understood by another person of color did, b = − 0.11 [− 0.19, − 0.03], SE = 0.04, t(310) = − 2.68, p = 0.008, r = 0.15. Feeling more understood by another person of color was associated with fewer depressive symptoms. In addition, this association remained significant when controlling for same-race understanding, b = − 0.09 [− 0.17, − 0.005], SE = 0.04, t(309) = − 2.17, p = 0.031, r = 0.12.] In contrast, neither same-race nor cross-race disclosure was associated with depressive symptoms (although cross-race disclosure and understanding involving other people of color, not White people, was associated with depression; see Footnote 4).
Discussion
Study 1 revealed that participants felt more understood in same-race than in cross-race relationships, although high levels of understanding were observed in both. Participants frequently felt understood, with scores leaning towards “almost always” in both same-race and cross-race relationships. Additionally, feeling understood in close same-race and cross-race relationships independently predicted fewer depressive symptoms one and two years later—but disclosure did not. Therefore, Study 1 offers a snapshot of how understood young adults of color felt in their closest relationships during their sophomore year. In Study 2, we assessed understanding and disclosure on multiple occasions over 7 weeks to capture a richer and more nuanced portrait of understanding and self-disclosure over time.
Study 2
In Study 1, people of color felt more understood in same-race than cross-race relationships, although they reported similar levels of same-race and cross-race disclosure. Feeling understood in both types of relationships also predicted fewer depressive symptoms over time. In Study 2, we sought to replicate whether people of color perceived differences in understanding in same-race and cross-race close relationships. We also examined whether same-race and cross-race understanding and disclosure predicted less depressive affect and more flourishing in daily life over seven weeks by testing three models: same-race predictors only (first model), cross-race predictors only (second model), and same-race and cross-race predictors (third model).
Method
Participants
First-year students were recruited from a Midwestern U.S. university during the first three weeks of their first academic term. The sample (N = 101) included 62 women, 31 men, and 8 participants who did not disclose their gender (M = 18.14 years old, SD = 0.60). All participants identified as people of color (28 Black, 29 Latine, 43 Asian, 1 Multiracial with a Black and Asian background; we chose not to classify them into one group). Since participants did not provide consent for their data to be posted publicly, deidentified Study 2 data is available through a private OSF link that can be obtained upon request.
Procedure
In a background survey, participants provided details about their three closest relationships, including whether these relationships were same-race or not. Five participants named cross-race close others only, 22 named two cross-race and one same-race close others, 33 named one cross-race and two same-race close others, and 41 named same-race close others only.
Next, participants were invited to complete diary surveys every Sunday night about their experiences in these relationships and during the past week for the remaining seven weeks of the term using a custom smartphone app48. For each relationship, participants rated how much they disclosed to their close other (“This past week, I shared my inner feelings with [close other’s name]”) and how understood they felt (“This past week, I felt that [close other’s name] understood what I am really like”) during the past week using a 5-point scale (1 = None of the time; 5 = All of the time). Each close other’s name was piped into the item wording, allowing us to track relationship quality for each dyad. Participants also indicated their depressive affect with one item (e.g., “This past week, I had little interest or pleasure in doing things”49), chosen for its concision and wide use, on a 7-point scale (1 = Strongly disagree; 7 = Strongly agree). Using the same scale, participants also completed six items measuring flourishing (“This past week, I led a purposeful and meaningful life”11, ωwithin = 0.78); this scale was designed to be used with the population at large, with higher scores reflecting higher levels of functioning and psychological well-being. On average, participants completed 4.75 of the seven surveys (Medianresponse rate = 71.43%).
Because participants varied in the racial composition of their dyads, each model we tested had a different number of participants. The first model included only same-race predictors (n = 96) and was sufficiently powered (i.e., at least 80%) to detect a medium between-person effect, r = 0.28, and a small-to-medium within-person effect, r = 0.15. The second model included cross-race predictors only (n = 60) and was sufficiently powered to detect a medium between-person effect, r = 0.35, and a small-to-medium within-person effect, r = 0.18. The third model included both same-race and cross-race predictors (n = 55) and was sufficiently powered to detect a medium between-person effect, r = 0.34, and a small-to-medium within-person effect, r = 0.19.
Results
We first examined whether participants reported differences in understanding and disclosure in same-race and cross-race relationships. We then tested whether understanding and disclosure in same-race and cross-race relationships predicted weekly depressive affect and flourishing.
Differences in understanding and disclosure
We averaged weekly understanding and disclosure ratings for each relationship to test whether participants reported greater understanding or disclosure in same-race than cross-race relationships. In these analyses, relationships (Level 1) were nested within participants (Level 2). We tested our hypotheses using two-level multilevel models with random intercepts and relationship type entered as an effects-coded variable (− 1 = cross-race; + 1 = same-race).
Consistent with Study 1, participants disclosed a similar amount in same-race (M = 3.41, SE = 0.09) and cross-race relationships (M = 3.23, SE = 0.11), b = 0.09, CI95% [− 0.02, 0.20], SE = 0.06, t(270.93) = 1.53, p = 0.128, r = 0.09. [Footnote 5: When we separated network racial composition into its within- and between-person components, we found that participants with more same-race close others disclosed more relative to those with fewer same-race close others, b = 0.34 CI95% [0.07, 0.60], SE = 0.13, t(99.00) = 2.52, p = 0.014, r = 0.25. This effect was not significant for understanding, t(99.00) = 1.35, p = 0.181.] In contrast with Study 1, same-race (M = 3.82, SE = 0.07) and cross-race (M = 3.71, SE = 0.10) relationships did not differ in understanding, b = 0.05, CI95% [− 0.05, 0.15], SE = 0.05, t(281.14) = 1.01, p = 0.315, r = 0.06. [Footnote 6: This difference in understanding was moderated by participants’ race, χ2(2) = 6.87, p = 0.032. Asian participants felt more understood in same-race than cross-race close relationships, b = 0.19 CI95% [0.04, 0.34], SE = 0.08, t(252.44) = 2.49, p = 0.013, r = 0.15. This difference was not significant for Black or Latine participants, ts < 1.36, ps > 0.175.]
Understanding and disclosure as predictors of well-being
Analytic strategy
In our dataset, relationships (Level 1) were nested within weeks (Level 2) and weeks within participants (Level 3); however, in multilevel modeling, lower-level predictors (i.e., relationship-specific understanding and relationship-specific disclosure) cannot be used to predict higher-level outcomes (i.e., weekly depressive affect and flourishing). Because we were interested in whether experiences in same-race and cross-race relationships during a particular week, rather than experiences in specific relationships, predicted individuals’ well-being during the same week, we calculated four weekly averages to represent how understood participants felt and how much participants disclosed in their same-race and cross-race relationships (i.e., same-race understanding, same-race disclosure, cross-race understanding, cross-race disclosure). These averages were used as predictors of weekly well-being. The resulting data structure was weekly diaries (Level 1) nested within participants (Level 2). [Footnote 7: Alternatively, we could have modeled our outcomes using a cross-classified model with weekly surveys nested in relationships and participants. When we tested this model, the model was singular and the variance estimate for the relationships intercept was 0 for both weekly outcomes, suggesting that there was not enough variability to support this random structure. Moreover, because participants selected their three closest relationships, it would not make sense to treat relationship as a random variable because these findings may not generalize to other relationships that are less close.]
Because understanding and disclosure varied between and within participants, we separated these predictors into their between-person (each participant’s mean across all their weekly diaries) and within-person (centering each participant’s responses on their own mean) components of the predictors and entered both into our models. This allowed us to examine whether participants in higher-quality relationships differed from those with lower-quality relationships (between-person effect) and whether changes in each participant’s relationship experiences, from week to week, predicted changes in their psychological well-being (within-person effects). In all models, we included a random intercept for each participant, allowing the average level of psychological well-being to vary between participants. Despite no apparent time trends and specific time-related hypotheses, we included time as a grand-mean centered variable as a fixed effect to control for any variables co-varying with time.
For each outcome, we tested three different models: same-race predictors only, cross-race predictors only, and same-race and cross-race predictors. Full results are presented in Tables 5 and 6.
Table 5.
Full results from understanding models.
| Predictors | Depressive affect | Flourishing | ||||
|---|---|---|---|---|---|---|
| Model 1: Same-Race Predictors Only Nobservations = 447 Nparticipants = 96 |
Model 2: Cross-Race Predictors Only Nobservations = 291 Nparticipants = 60 |
Model 3: All Predictors Nobservations = 264 Nparticipants = 55 |
Model 1: Same-Race Predictors Only Nobservations = 447 Nparticipants = 96 |
Model 2: Cross-Race Predictors Only Nobservations = 291 Nparticipants = 60 |
Model 3: All Predictors Nobservations = 264 Nparticipants = 55 |
|
|
Same-Race Understanding (Within-Person) |
b = − 0.12 (0.12) [− 0.36, 0.11] p = .325, r = .05 |
b = − 0.14 (0.14) [− 0.42, 0.15] p = .325, r = .07 |
b = 0.05 (0.06) [− 0.06, 0.16] p = .371, r = .05 |
b = 0.02 (0.07) [− 0.12, 0.16] p = .825, r = .02 |
||
|
Same-Race Understanding (Between-Person) |
b = − 0.49 (0.13) [− 0.74, − 0.23] p < .001, r = .36 |
b = − 0.51 (0.15) [− 0.82, − 0.22] p = .001, r = .41 |
b = 0.40 (0.10) [0.21, 0.61] p < .001, r = .39 |
b = 0.30 (0.14) [0.03, 0.57] p = .036, r = .29 |
||
|
Cross-Race Understanding (Within-Person) |
b = 0.07 (0.13) [− 0.19, 0.33] p = .610, r = .03 |
b = 0.11 (0.13) [− 0.15, 0.37] p = .391, r = .06 |
b = − 0.10 (0.07) [− 0.24, 0.04] p = .121, r = .10 |
b = − 0.12 (0.07) [− 0.24, 0.01] p = .083, r = .12 |
||
|
Cross-Race Understanding (Between-Person) |
b = − 0.52 (0.17) [− 0.87, − 0.19] p = .003, r = .37 |
b = − 0.19 (0.18) [− 0.54, 0.15] p = .291, r = .14 |
b = 0.39 (0.15) [0.11, 0.70] p = .010, r = .33 |
b = 0.19 (0.16) [− 0.14, 0.50] p = .250, r = .16 |
||
| Time |
b = 0.07 (0.04) [0.00, 0.14] p = .055, r = .10 |
b = 0.04 (0.05) [− 0.05, 0.13] p = .388, r = .05 |
b = 0.08 (0.05) [− 0.01, 0.18] p = .075, r = .12 |
b = 0.00 (0.02) [− 0.04, 0.04] p = .990, r = .00 |
b = 0.00 (0.02) [− 0.05, 0.05] p = .988, r = .00 |
b = − 0.01 (0.02) [− 0.06, 0.04] p = .583, r = .04 |
Bootstrapped 95% confidence intervals with 5000 resamples are reported in square brackets.
Table 6.
Full results from disclosure models.
| Predictors | Depressive affect | Flourishing | ||||
|---|---|---|---|---|---|---|
| Model 1: Same-race predictors only Nobservations = 446 Nparticipants = 96 |
Model 2: Cross-race predictors only Nobservations = 291 Nparticipants = 60 |
Model 3: All predictors Nobservations = 263 Nparticipants = 55 |
Model 1: Same-race predictors only Nobservations = 446 Nparticipants = 96 |
Model 2: Cross-race predictors only Nobservations = 291 Nparticipants = 60 |
Model 3: All predictors Nobservations = 263 Nparticipants = 55 |
|
|
Same-Race Disclosure (Within-Person) |
b = 0.07 (0.11) [− 0.15, 0.28] p = .535, r = .03 |
b = 0.08 (0.13) [− 0.18, 0.35] p = .533, r = .04 |
b = − 0.02 (0.05) [− 0.12, 0.08] p = .706, r = .02 |
b = − 0.13 (0.07) [− 0.26, 0.01] p = .064, r = .13 |
||
|
Same-Race Disclosure (Between-Person) |
b = − 0.17 (0.12) [− 0.41, 0.07] p = .160, r = .14 |
b = − 0.21 (0.17) [− 0.54, 0.13] p = .216, r = .17 |
b = 0.28 (0.09) [0.10, 0.46] p = .002, r = .30 |
b = 0.17 (0.14) [− 0.10, 0.44] p = .237, r = .16 |
||
|
Cross-Race Disclosure (Within-Person) |
b = − 0.01 (0.10) [− 0.21, 0.19] p = .943, r = .00 |
b = − 0.07 (0.11) [− 0.28, 0.13] p = .500, r = .05 |
b = − 0.02 (0.05) [− 0.12, 0.08] p = .723, r = .02 |
b = 0.00 (0.05) [− 0.20, 0.40] p = .954, r = .00 |
||
|
Cross-Race Disclosure (Between-Person) |
b = − 0.06 (0.16) [− 0.37, 0.25] p = .699, r = .05 |
b = 0.11 (0.18) [− 0.24, 0.47] p = .546, r = .08 |
b = 0.22 (0.13) [− 0.05, 0.48] p = .107, r = .21 |
b = 0.10 (0.15) [− 0.10, 0.11] p = .520, r = .09 |
||
| Time |
b = 0.07 (0.04) [0.00, 0.15] p = .048, r = .10 |
b = 0.04 (0.05) [− 0.05, 0.13] p = .375, r = .06 |
b = 0.08 (0.05) [− 0.02, 0.18] p = .091, r = .11 |
b = 0.00 (0.02) [− 0.04, 0.03] p = .920, r = .01 |
b = 0.00 (0.02) [− 0.04, 0.05] p = .889, r = .01 |
b = 0.00 (0.02) [− 0.05, 0.04] p = .890, r = .01 |
Bootstrapped 95% confidence intervals with 5000 resamples are reported in square brackets.
Understanding and depressive affect
In the first model, there was a significant between-person effect of same-race understanding. People who felt more understood in same-race close relationships reported lower depressive affect each week than people who reported lower same-race understanding. The within-person effect was not significant.
In the second model, there was a significant between-person effect of cross-race understanding. People who felt more understood in cross-race close relationships reported lower depressive affect each week than people who reported lower cross-race understanding. The within-person effect was not significant.
In the final model, the between-person effect of same-race understanding remained significant after including cross-race understanding. The between-person effect of cross-race understanding was no longer significant. The within-person effects were also not significant.
Understanding and flourishing
In the first model, there was a significant between-person effect of same-race understanding. People who felt more understood in same-race close relationships reported more flourishing each week than people who reported lower same-race understanding. The overall within-person effect was not significant. [Footnote 8: We found that participants’ race moderated the within-person effect of same-race understanding, χ2(2) = 6.02, p = 0.049. For Black participants, during weeks when they felt more understood in their same-race relationships than usual, they reported higher levels of flourishing relative to weeks when they felt less understood in their same-race relationships than usual, b = 0.22 CI95% [0.03, 0.43], SE = 0.10, t(344.92) = 2.23, p = 0.026, r = 0.12. This effect was not significant for Latine or Asian participants, ts < 1.21, ps > 0.228. However, this interaction was no longer significant when we included the cross-race understanding, χ2(2) = 5.75, p = 0.056.]
In the second model, there was a significant between-person effect of cross-race understanding. People who felt more understood in cross-race close relationships reported more flourishing each week than people who reported lower cross-race understanding. The within-person effect was not significant.
In the final model, the between-person effect of same-race understanding remained significant after including cross-race understanding. The between-person effect of cross-race understanding was no longer significant. The within-person effects were also not significant.
Disclosure and depressive affect
Disclosure did not predict depressive affect in any model.
Disclosure and flourishing
In the first model, there was a significant between-person effect of same-race disclosure. People who disclosed more in same-race close relationships reported more flourishing each week than people who reported lower same-race disclosure. The within-person effect was not significant. [Footnote 9: The within-person effect of same-race disclosure was moderated by network racial composition, b = 0.17 CI95% [0.06, 0.28], SE = 0.06, t(348.00) = 3.05, p = 0.002, r = 0.16. For participants with same-race close others only, during weeks when they disclosed more to their close others, they experienced more flourishing, b = 0.21 CI95% [0.03, 0.39], SE = 0.09, t(347.92) = 2.32, p = 0.021, r = 0.12. In contrast, for participants who had same-race and cross-race close others, during weeks when they disclosed more to their same-race close others, they experienced less flourishing, b = -0.13 CI95% [− 0.25, − 0.003], SE = 0.06, t(348.16) = − 2.00, p = 0.046, r = 0.11.] No predictors were significant in our second and third models.
Discussion
Study 2 reveals that students of color who felt more understood in their same-race relationships also experienced less depressive affect and more flourishing. These effects were unique to same-race relationships; they held when accounting for experiences in cross-race relationships. In contrast, feeling understood in cross-race relationships did not uniquely predict depressive affect or flourishing. Moreover, the extent to which students of color disclosed in their same-race or cross-race relationships was not associated with depressive affect or flourishing (but see Footnote 9 for additional analyses). Although we expected these associations to also be present at the within-person level, we did not find any consistent within-person effects. These nonsignificant effects, however, may be due to limited statistical power because we obtained at most seven observations from each participant. It will be important for future research to test for within-person effects with a fully powered sample to more rigorously test which aspects of relationship quality are likely causes of well-being.
Taken together, these findings are consistent with Study 1: Same-race understanding is associated with depressive affect, whereas neither same-race nor cross-race disclosure is associated with it. Moreover, these findings extend and strengthen Study 1 by linking understanding with depressive affect and flourishing in a shorter, week-to-week period.
General discussion
Being offered the space to open up and feeling heard are among the kindnesses that people, including those of color, seek in close relationships. Yet, challenges may arise across racial lines even in people of color’s closest relationships. Moreover, these challenges may extend to understanding and disclosure, which are essential to unlocking relationship and personal benefits. Using a national and high-powered longitudinal study, we found that Black, Latine and Asian students who had both same-race and cross-race close relationships felt significantly more understood in their same-race than in their cross-race close relationships (Study 1). In turn, same-race and cross-race understanding were independently associated with less depressive symptoms one and two years later. In Study 2, same-race and cross-race dyads did not differ in levels of understanding measured over seven weeks; however, only same-race understanding was uniquely associated with depressive affect and flourishing during students’ first college term.
In both studies, disclosure generally did not differ between same-race and cross-race relationships, and neither was associated with psychological well-being. One notable exception: in Study 2, disclosing to same-race close others was associated with less flourishing. This pattern could have emerged because participants of color disclose to same-race close following racially charged moments, which could also be associated with lowered flourishing. This finding illustrates that the role of qualitative differences in disclosure could be central, where people “use” their relationships for different purposes and thus share about different topics. For example, people of color may disclose less about racial experiences in cross-race relationships but more about other experiences. They may also form close relationships with individuals they feel a unique connection to, including cross-race others. Therefore, examining what people of color disclose in which relationship is an important avenue for future research—and could allow examining whether such circumstances explain that on weeks when they disclosed more to same-race close others, they also experienced reduced flourishing. Overall, and despite their distinct experiences and histories, Black, Latine and Asian students presented similar patterns of same-race and cross-race understanding; however, future research could investigate the circumstances under which they differ. Moreover, future research could further examine the distinct experiences of people of color engaged in both close same-race and cross-race dyads (as in Study 1), as opposed to people of color who engage in same-race dyads only. It could also establish whether the present findings extend to well-being generally, or are specific to flourishing, depressive affect and depressive symptomatology.
The present findings highlight that understanding is central to people of color’s well-being outcomes and the associations with same-race understanding are more consistent. Thus, they reaffirm the importance of community ties for people from marginalized groups, suggesting that being understood may be especially meaningful in this context. They also reveal that cross-race understanding can be uniquely associated with their well-being. Indeed, we found this association in a well-powered national sample (Study 1) but not in a smaller sample (Study 2). Therefore, the present research does not suggest that people of color have strong relationship experiences solely in either same-race or cross-race relationships. Rather, this research highlights the critical contributions of same-race relationships to the psychological well-being of people of color and the possibility that cross-race close relationships are also rewarding when characterized by high levels of understanding. Determining when and why close relationships are most beneficial and most costly to people of color matters, as close ties and community represent strengths that can foster and protect their well-being.
Acknowledgements
The authors thank Mesmin Destin and Maya Yampolsky for their feedback on the studies, as well as Ivy Onyeador and Dorainne Green for their feedback on the manuscript.
Author contributions
Régine Debrosse and Sabrina Thai share first authorship for this paper. Régine Debrosse designed and conducted analyses for S1, Sabrina Thai designed the software used in S2 and conducted analyses for S2. Regine Debrosse and Sabrina Thai conceptualized and designed S2 together and both wrote large portions of the paper. Emilie Auger substantially contributed to interpreting data from both S1 and S2; she also substantially revised the paper with the first authors. Tess Brieva substantially contributed to data collection, and revised the paper. When this project started, Régine Debrosse was affiliated with the School of Education and Social Policy and with the Department of Psychology, Northwestern University, Evanston, Illinois, USA. She is now affiliated with the School of Social Work, McGill University, Montréal, Quebec, Canada. This work was supported by a postdoctoral fellowship from the William T. Grant Foundation (188,943), by a Small Research Grant from the Society of Personality and Social Psychology in Fall 2017, and by a William Dawson Award offered by McGill University.
Data availability
More information and access to the data used in Study 1 is publicly available at: https://opr.princeton.edu/national-longitudinal-survey-freshmen-nlsf.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Boykin, A. W., Jagers, R. J., Ellison, C. M. & Albury, A. Communalism: Conceptualization and measurement of an Afrocultural social orientation. J. Black Stud.27, 409–418. 10.1177/002193479702700308 (1997). [Google Scholar]
- 2.Cahill, K. M., Updegraff, K. A., Causadias, J. M. & Korous, K. M. Familism values and adjustment among Hispanic/Latino individuals: A systematic review and meta-analysis. Psychol. Bull.147, 947–985. 10.1037/bul0000336 (2021). [Google Scholar]
- 3.Christophe, N. K. & Stein, G. L. Facilitating the study of familism across racial/ethnic groups: Creation of the short attitudinal familism scale. J. Fam. Psychol.36, 534–544. 10.1037/fam0000954 (2022). [DOI] [PubMed] [Google Scholar]
- 4.Corona, K., Campos, B. & Chen, C. Familism is associated with psychological well-being and physical health: Main effects and stress-buffering effects. Hisp. J. Behav. Sci.39, 46–65. 10.1177/0739986316671297 (2017). [Google Scholar]
- 5.Johnson, V. E. & Carter, R. T. Black cultural strengths and psychosocial well-being: An empirical analysis with Black American adults. J. Black Psychol.46, 55–89. 10.1177/0095798419889752 (2020). [Google Scholar]
- 6.Schwartz, S. J. et al. Communalism, familism, and filial piety: Are they birds of a collectivist feather?. Cult. Divers. Ethnic Minor. Psychol.16, 548–560. 10.1037/a0021370 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Stein, G. L. et al. Familism through a developmental lens. J. Latina/o Psychol.2, 224–250. 10.1037/lat0000025 (2014). [Google Scholar]
- 8.Fehr, B. & Harasymchuk, C. Conceptions and the experience of friendship in underrepresented groups. Pers Relations29, 451–487. 10.1111/pere.12431 (2022). [Google Scholar]
- 9.Chung, J. M. et al. An examination of personality expression in everyday life in a Canadian sample of racialized undergraduate students. J. Pers92, 16–37. 10.1111/jopy.12797 (2024). [DOI] [PubMed] [Google Scholar]
- 10.Debrosse, R., Thai, S. & Brieva, T. When skinfolk are kinfolk: Higher perceived support and acceptance characterize close same-race (vs. interracial) relationships for people of color. J. Soc. Issues79, 21–49. 10.1111/josi.12534 (2023). [Google Scholar]
- 11.Diener, E. et al. New well-being measures: Short scales to assess flourishing and positive and negative feelings. Soc. Indicators Res.97, 143–156. 10.1007/s11205-009-9493-y (2010). [Google Scholar]
- 12.Reis, H. T. (2017). The interpersonal process model of intimacy: Maintaining intimacy through self-disclosure and responsiveness. In Foundations for Couples’ therapy (pp. 216–225). Routledge.
- 13.Reis, H. T. & Gable, S. L. Responsiveness. Curr. Opinion Psychol.1, 67–71. 10.1016/j.copsyc.2015.01.001 (2015). [Google Scholar]
- 14.Reis, H. T., Lemay, E. P. Jr. & Finkenauer, C. Toward understanding understanding: The importance of feeling understood in relationships. Soc. Personality Psychol. Compass11, e12308. 10.1111/spc3.12308 (2017). [Google Scholar]
- 15.Altman, I., & Taylor, D. A. (1973). Social Penetration: The Development of Interpersonal Relationships. Holt, Rinehart & Winston.
- 16.Cahn, D. D. Perceived understanding and interpersonal relationships. J. Soc. Pers. Relationsh.7, 231–244. 10.1177/0265407590072005 (1990). [Google Scholar]
- 17.Laurenceau, J.-P., Barrett, L. F. & Pietromonaco, P. R. Intimacy as an interpersonal process: The importance of self-disclosure, partner disclosure, and perceived partner responsiveness in interpersonal exchanges. J. Pers. Soc. Psychol.74, 1238–1251. 10.1037/0022-3514.74.5.1238 (1998). [DOI] [PubMed] [Google Scholar]
- 18.Rempel, J. K., Holmes, J. G. & Zanna, M. P. Trust in close relationships. J. Pers. Soc. Psychol.49, 95–112. 10.1037/0022-3514.49.1.95 (1985). [Google Scholar]
- 19.Reis, H. T. & Shaver, P. Intimacy as an interpersonal process. In Handbook of Personal Relationships: Theory, Research & Interventions (eds Duck, S. et al.) 367–389 (Wiley, 1988). [Google Scholar]
- 20.Du, X., Livingstone, A. G. & Adlam, A. L. R. Felt understanding as a bridge between social identity and wellbeing among international university students. J. Community Appl. Soc. Psychol.34, e2722. 10.1002/casp.2722 (2024). [Google Scholar]
- 21.Lun, J., Kesebir, S. & Oishi, S. On feeling understood and feeling well: The role of interdependence. J. Res. Pers.42, 1623–1628. 10.1016/j.jrp.2008.06.009 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Mallett, R. K., Akimoto, S. & Oishi, S. Affect and understanding during everyday cross-race experiences. Cult. Diversity Ethnic Minority Psychol.22, 237–246. 10.1037/cdp0000032 (2016). [DOI] [PubMed] [Google Scholar]
- 23.Shelton, N., Douglass, S., Garcia, R. L., Yip, T. & Trail, T. E. Feeling (mis)understood and intergroup friendships in interracial interactions. Pers. Soc. Psychol. Bull.40, 1193–1204. 10.1177/0146167214538459 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Shelton, J. N., Trail, T. E., West, T. V. & Bergsieker, H. B. From strangers to friends: The interpersonal process model of intimacy in developing interracial friendships. J. Soc. Pers. Relations.27, 71–90. 10.1177/0265407509346422 (2010). [Google Scholar]
- 25.Bergsieker, H. B., Shelton, J. N. & Richeson, J. A. To be liked versus respected: Divergent goals in interracial interactions. J. Pers. Soc. Psychol.99, 248–264. 10.1037/a0018474 (2010). [DOI] [PubMed] [Google Scholar]
- 26.Richeson, J. A. & Shelton, J. N. Negotiating interracial interactions: Costs, consequences, and possibilities. Curr. Direct. Psychol. Sci.16, 316–320. 10.1111/j.1467-8721.2007.00528.x (2007). [Google Scholar]
- 27.LaCosse, J. & Plant, E. A. Internal motivation to respond without prejudice fosters respectful responses in interracial interactions. J. Pers. Soc. Psychol.119, 1037–1056. 10.1037/pspi0000219 (2020). [DOI] [PubMed] [Google Scholar]
- 28.Holoien, D. S., Bergsieker, H. B., Shelton, J. N. & Alegre, J. M. Do you really understand? Achieving accuracy in interracial relationships. J. Pers. Soc. Psychol.108, 76–92. 10.1037/pspi0000003 (2015). [DOI] [PubMed] [Google Scholar]
- 29.Dovidio, J. F., Kawakami, K. & Gaertner, S. L. Implicit and explicit prejudice and interracial interaction. J. Pers. Soc. Psychol.82, 62–68. 10.1037/0022-3514.82.1.62 (2002). [DOI] [PubMed] [Google Scholar]
- 30.Hafer, C. L. & Bègue, L. Experimental research on just-world theory: Problems, developments, and future challenges. Psychol. Bull.131, 128–167. 10.1037/0033-2909.131.1.128 (2005). [DOI] [PubMed] [Google Scholar]
- 31.Kaiser, C. R. & Miller, C. T. Stop complaining! The social costs of making attributions to discrimination. Pers. Soc. Psychol. Bull.27, 254–263. 10.1177/0146167201272010 (2001). [Google Scholar]
- 32.Phillips, L. T. & Lowery, B. S. The hard-knock life? Whites claim hardships in response to racial inequity. J. Exp. Soc. Psychol.61, 12–18. 10.1016/j.jesp.2015.06.008 (2015). [Google Scholar]
- 33.Marinucci, M. et al. Intimate intergroup contact across the lifespan. J. Social Issues77, 64–85. 10.1111/josi.12399 (2021). [Google Scholar]
- 34.McGorray, E. L., Emery, L. F., Garr-Schultz, A. & Finkel, E. J. “Mostly White, heterosexual couples”: Examining demographic diversity and reporting practices in relationship science research samples. J. Pers. Soc. Psychol.125, 316–344. 10.1037/pspi0000417 (2023). [DOI] [PubMed] [Google Scholar]
- 35.Shelton, J. N., Turetsky, K. M., & Park, Y. (2023). Responsiveness in interracial interactions. Curr. Opin. Psychol. 101653. 10.1016/j.copsyc.2023.101653 [DOI] [PubMed]
- 36.Page-Gould, E., Harris, K., MacInnis, C. C., Danyluck, C. M., & Miller, I. D. (2022). The intergroup perspective on cross-group friendship. In B. Gawronski (Ed.), Advances in Experimental Social Psychology (Vol. 65, pp. 1–56). Academic Press. 10.1016/bs.aesp.2021.10.001
- 37.Toosi, N. R., Babbitt, L. G., Ambady, N. & Sommers, S. R. Dyadic interracial interactions: A meta-analysis. Psychol. Bull.138, 1–27. 10.1037/a0025767 (2012). [DOI] [PubMed] [Google Scholar]
- 38.Yampolsky, M. A., Rossini, A., Pagé, J., Leanza, Y. & Lalonde, R. N. Intimate racism from one’s partner in young intercultural couples. J. Soc. Issues79, 161–195. 10.1111/josi.12558 (2023). [Google Scholar]
- 39.Lemay, E. P. Jr. & Teneva, N. Accuracy and bias in perceptions of racial attitudes: Implications for interracial relationships. J. Pers. Soc. Psychol.119, 1380–1402. 10.1037/pspi0000236 (2020). [DOI] [PubMed] [Google Scholar]
- 40.Sanchez, K. L., Kalkstein, D. A. & Walton, G. M. A threatening opportunity: The prospect of conversations about race-related experiences between Black and White friends. J. Pers. Soc. Psychol.122, 853–872. 10.1037/pspi0000369 (2022). [DOI] [PubMed] [Google Scholar]
- 41.Marshburn, C. K. & Campos, B. Seeking just us: A mixed methods investigation of Racism-Specific Support among Black college students. J. Black Psychol.48, 67–99. 10.1177/00957984211034961 (2022). [Google Scholar]
- 42.Carey, R. M., Stephens, N. M., Townsend, S. S. M. & Hamedani, M. G. Is diversity enough? Cross-race and cross-class interactions in college occur less often than expected, but benefit members of lower status groups when they occur. J. Pers. Soc. Psychol.123, 889–908. 10.1037/pspa0000302 (2022). [DOI] [PubMed] [Google Scholar]
- 43.Doucerain, M. M., Morin, L. & Grégoire, S. Are daily intergroup interactions distressing? An end-of-day diary study of associations between daily intergroup interaction proportion and psychological distress/well-being. Int. J. Intercult. Relations95, 101827. 10.1016/j.ijintrel.2023.101827 (2023). [Google Scholar]
- 44.McGill, R. K., Way, N. & Hughes, D. Intra-and interracial best friendships during middle school: Links to social and emotional well-being. J. Res. Adolesc.22, 722–738. 10.1111/j.1532-7795.2012.00826.x (2012). [Google Scholar]
- 45.Eaton, W. W., Muntaner, C., Smith, C., Tien, A., & Ybarra, M. (2004). Center for epidemiologic studies depression scale: Review and revision. The Use of Psychological Testing for Treatment Planning and Outcomes Assessment, 363–377.
- 46.Radloff, L. S. The CES-D scale: A self-report depression scale for research in the general population. Appl. Psychol. Meas.1, 385–401. 10.1177/014662167700100306 (1977). [Google Scholar]
- 47.Page-Gould, E., Sharples, A. E., & Song, S. (2019). Effect sizes for models of longitudinal data. In P. Shrout (Chair), Modeling Mediation Processes in Longitudinal Data. Symposium conducted at the annual meeting of the Society of Experimental Social Psychology, Toronto, ON, Canada.
- 48.Thai, S. & Page-Gould, E. ExperienceSampler: An open-source scaffold for building smartphone apps for experience sampling. Psychol. Methods23, 729–739. 10.1037/met0000151 (2018). [DOI] [PubMed] [Google Scholar]
- 49.Bowling, A. Mode of questionnaire administration can have serious effects on data quality. J. Public Health27, 281–291. 10.1093/pubmed/fdi031 (2005). [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
More information and access to the data used in Study 1 is publicly available at: https://opr.princeton.edu/national-longitudinal-survey-freshmen-nlsf.