Parent’s Just Don’t Understand: Parental Support, Religion and Depressive Symptoms among Same-Race and Interracial Relationships

Abstract

Research finds that individuals in interracial relationships have poorer mental health than those in same-race relationships. Family support, or lack thereof, may play an important role in explaining the psychological risks for such individuals. Growing attention has focused on the complex interplay between religion, health, and family life, particularly the stress-buffering role of religious involvement. However, little attention has been given to the possible mitigating effects of religion in the face of limited family support among same-race and interracial couples. Using data from the National Longitudinal Study of Adolescent to Adult Health (Add Health), this study addresses two important questions: (1) Is weak family support associated with depressive symptoms among individuals in same-race and interracial relationships?; and (2) Does religious involvement buffer the association between weak family support and depressive symptoms for individuals engaged in these romantic ties? Results suggest that weak parental support is associated with depressive symptoms for individuals in both same-race and interracial relationships, however we find limited support of religion protecting against weak parental support for individuals in interracial unions. The results highlight the complex interplay between religion, health, and family in contemporary American life.

1. Introduction

Recent increases in the rate of interracial relationships, both in dating and marriage, suggest it has garnered greater social acceptance (Carroll 2007; Herman and Campbell 2012). As of 2015, nearly 17% of new marriages were between partners of a different race compared to less than 1% in 1970 (Pew Research Center 2017). A growing literature, however, also describes the persistent challenges interracial couples encounter, including negative reactions from strangers and diminished support from family and friends (Childs 2005; Dalmage 2000). Familial support continues to be an important and vital part of young adult adjustment (Arnett and Schwab 2013; Fingerman et al. 2012), and in its absence many young adults are vulnerable to psychological distress (B. Miller 2017). Findings from several recent studies suggest that when compared to their same-race counterparts, individuals in interracial relationships report higher rates of distress and anxiety (B. Miller 2017; Lykke 2017; Bratter and Eschbach 2006; Miller and Kail 2016; Kroeger and Williams 2011). There is mixed evidence that parental support helps to explain the differences in the association between depressive symptoms and interracial and same-race relationships (Tillman and Miller 2017). Surprisingly, limited work has identified social and cultural factors that may protect the mental well-being of individuals in interracial relationships from the loss of social support.

One such factor may be religion. First, several decades of work have documented the health-promoting effects of religiosity (Ellison and Levin 1998). Religious involvement is positively associated with better psychological well-being, including lower depression and anxiety and higher levels of self-esteem (McCullough and Larson 1999; Krause 1995; Shreve-Neiger and Edelstein 2004). Moreover, religion often plays a salient role in facilitating effective coping, and more favorable outcomes in the face of problematic life events and chronic stress, including financial hardship, role strain, discrimination, and other challenging events (Bradshaw and Ellison 2010; Henderson 2016; Bierman 2006; Sherkat and Reed 1992). Second, there is evidence surrounding a religion-family connection, such that religion is often identified as a salient force in the formation, quality, and preservation of romantic relationships (for reviews see Marks 2006; Mahoney 2010). Yet much of what we know about the role of religion in romantic relationships is centered on same-race, largely non-Hispanic white, unions (Mahoney 2010; Edgell 2013; Myers 2006; Vaaler et al. 2009). Little is known about how religion may work in the context of other, more diverse relationships. Religion, however, may play an important role across a variety of relationship contexts, including homosexual couples (Oswald et al. 2008; Rostosky et al. 2008), dating and cohabiting couples (Henderson et al. 2018; Freitas 2008), and “fragile” families (Wilcox and Wolfinger 2008; Sullivan 2008). Lastly, while the relationship between religion and attitudes and behaviors toward interracial relationships is complex (Perry 2013a, 2013b), there is some recent evidence to suggest that certain dimensions of religious involvement are positively related to interracial unions. Recent work by Perry (2013a), finds that more proximal dimensions of religious involvement—i.e., participation in private devotional practices and integrated churches—may incline individuals towards a more favorable attitude in interracial marriage. However, examining if and how religion works in the context of such romantic relationships has largely been ignored. More work on the interplay between religion, stress, and mental health in the context of romantic relationships, especially interracial couples, remains an unexplored avenue of research.

The aim of the present study is to help fill the gap in this area. Using Wave IV data from the National Longitudinal Study of Adolescent to Adult Health (Add Health), this study addresses two important questions: (1) Is weak parental support associated with depressive symptoms among individuals in same-race and interracial relationships?; and (2) Does religious involvement moderate (i.e., buffer) the association between weak parental support and depressive symptoms among individuals engaged in these relationships? We generate two conceptual models, including the stress-buffering model, to examine these research questions. Findings are discussed in terms of the research on religion and health, as well as the broader research on the relationship between religion, health, and family life. Study limitations are identified and several fruitful directions for further investigation are proposed.

2. Theoretical and Empirical Background

2.1. Depression in Same-Race and Interracial Relationships

Young adults involved in romantic relationships are more likely to experience symptoms of depression than their peers who remain romantically unattached (Connolly and McIsaac 2011; Davila 2008; Joyner and Udry 2000; Welsh et al. 2003). A variety of causes and consequences for the association has been posited. One such explanation includes the stress and coping model (Davila 2008), which suggests young adults in romantic relationships face unnavigated challenges and stress, including negotiation of partner’s needs, sexual feelings and desires, relationship conflict, and issues surrounding breakups and rejection (Larson et al. 1999; Davila 2008). Such challenges increase one’s risk for distress and poor psychological adjustment (Davila 2008; Connolly and McIsaac 2011). However, recent evidence suggests that individuals in interracial, versus same-race relationships, face additional burdens that may place them at greater risk for psychological distress (Childs 2005; Dalmage 2000; Tillman and Miller 2017; Wong and Penner 2018). Although interracial ties have become more accepted, a substantial portion of Americans continue to disapprove of interracial romance (Carroll 2007; Herman and Campbell 2012; Skinner and Hudac 2017). Consequently, individuals in interracial relationships are likely to face microaggressions, discrimination, and stigma from people who disapprove of their union (Solsberry 1994; Bonilla-Silva and Forman 2000; Skinner and Hudac 2017). For example, Dalmage (2000) finds that many interracially dating White youth report experiencing racism in ways that undermines their ideas of fairness and equality, and many young adults engaged in interracial unions report hiding their relationship from friends and family (Wang et al. 2006). Such maltreatment, whether real or perceived, may adversely affect psychological well-being (Pascoe and Richman 2009). Consequently, a substantial body of work finds that couple’s racial composition—same-race versus interracial—is a robust predictor of psychological well-being in young adulthood (B. Miller 2017; Lykke 2017; Bratter and Eschbach 2006; Miller and Kail 2016; Kroeger and Williams 2011).

2.2. Parents, Depression and Same-Race and Interracial Relationships

Family of origin, particularly parental support, is recognized as a salient factor in young adult psychological adjustment (Davila et al. 2009; Lorenzo-Blanco et al. 2013; B. C. Miller 2002; Whitbeck et al. 1993). Supportive parental relationships may protect against the negative effects of stress, including stress associated with romantic ties, on mental health (Hartnett et al. 2013; House et al. 1988). Therefore, in the absence of parental support young adults may experience elevated risk of distress for several reasons, including increased loneliness and a reduced sense of belonging, a withdrawal of instrumental aid, as well as a loss of emotional support in navigating the challenges of romantic ties. Indeed, research suggests that young adults who view their parents as warm and supportive (Jessor and Jessor 1975), and who feel there is open communication with their parents report better mental health and are less likely to engage in behaviors that risk their well-being (Fox 1980; Diiorio et al. 2003). A poor parent-child relationship, characterized by low instrumental and emotional support, has been found to be associated with risky sexual behavior and depression (Jessor and Jessor 1977). For example, Steinberg and Davila (2008) found that the link between dating and depressive symptoms is stronger for individuals who perceive their parents as emotionally unavailable.

In addition to health, parents play a significant role in the formation and quality of the romantic relationships their children experience as adults (Collins and Read 1990; Del Toro 2012; Xia et al. 2018). Parents offer children a foundation, via parental practices and family climate, on which to build future relationships that influence the quality, function, and health of young adult romantic ties (Raby et al. 2015; Xia et al. 2018). The loss of parental support, perhaps as a consequence of a child’s relationship, may result in unfavorable health consequences. Surprisingly, little is known about how partner characteristics, including partner’s race, influence parental support (for exception see Yahirun 2019). As previously suggested, however, there is some evidence that family opposition is a salient factor in interracial relationships. Individuals in interracial unions often report feeling ostracized from their family (Gaines 2001). Some have even argued that parental objection is the most prominent obstacle in pursuing and maintaining an interracial relationship (Mok 1999; Childs 2005; Dalmage 2000). A brief review of the historical and cultural norms surrounding interracial relationships in the United States may shed light on why parents may choose to withdraw their support from such unions, thereby placing their children at greater risk of distress.

For much of American history, the idea of an interracial union, particularly a Black/White partnership, was repugnant (de Guzman and Nishina 2017). Stemming from our nation’s history of slavery, which reinforced unequal race relations as natural, marriages between individuals of different races was often illegal and morally objectionable (de Guzman and Nishina 2017; Foeman and Nance 1999). The 1967 Supreme Court ruling, i.e., Loving v. Virginia, 388 U.S. 1967, overturning miscegenation laws, which barred sex or marriage between Blacks and Whites, substantially changed the cultural environment surrounding interracial relationships (Lombardo 1987). Today, interracial dating and marriage are often used to gauge the level of integration or assimilation for minority groups into the larger culture. Frequently more young adults are engaging in interracial romantic ties. Approximately 12% to 19% of young adults report they have been in at least one interracial romantic relationship (Kreager 2008; Wang et al. 2006) and marriage among different race partners continues to rise (Pew Research Center 2017). Yet despite these changes, individuals in interracial unions continue to face a series of obstacles that unfairly challenge their relationships and health (Childs 2005; Dalmage 2000; B. Miller 2017; Bratter and Eschbach 2006; Miller and Kail 2016; Kroeger and Williams 201). Identifying and understanding the tools and resources individuals in both interracial and same-race relationships use in the face of such challenges is needed.

2.3. Religion, Weak Parental Support and Depression

The stress and coping model used to explain the risk for elevated distress among individuals engaged in romantic experiences rarely investigates the coping-related mechanisms that may protect against the harmful effects of stress (Davila 2008). A growing body of evidence, however, suggests religion that may be a salient resource and strategy in protecting mental health (Ellison and Levin 1998; Koenig and Larson 2001; Koenig 2009). Various dimensions of religious involvement have been found to be positively associated with better psychological well-being, including lower depression and anxiety, and better self-concept (Levin et al. 1995; Ellison 1993; Koenig 2009). Moreover, religious involvement has been found to reduce the noxious effects of stress, including discrimination, family role strain, and loss of social ties, on psychological well-being (Bierman 2006; Bradshaw and Ellison 2010; Henderson 2016; Sherkat and Reed 1992).

Several decades of work have embraced the approach of defining religion as a complex multi-dimensional phenomenon, including organizational, non-organizational and subjective religiosity, which may exert direct and indirect effects on the mental health of individuals in same-race and interracial relationships (Levin et al. 1995; Mahoney 2010). First, organizational religious involvement, often measured via the frequency of religious services attendance (Ellison and Levin 1998), may cultivate well-being by establishing and reinforcing social networks of support and guidance. Religious congregations offer frequent opportunities to develop friendships and relationships that aid individuals in times of trouble (Ellison and Levin 1998; Krause 2006). Via both formal, e.g., sermons and official religious directives, and informal means, e.g., emotional support and practical guidance from coreligionists, religious service attendance brings together like-minded individuals who share faith commitments and values on a regular basis (Ellison and Levin 1998). Such opportunities often build solidarity among religious communities that contributes to a shared sense of meaning and purpose, promotes positive cognitions and emotions, and diverts attention from personal problems or challenges (Krause 2006; Strawbridge et al. 2001) Individuals in same-race and interracial relationships may receive messages from religious officials and co-religionists that reinforce their relationship, but also provide opportunities for personal development and spiritual growth that encourages well-being (Wilcox and Wolfinger 2008; Strawbridge et al. 2001; Rasic et al. 2011). The social support offered by organizational religious involvement may be particularly salient for individuals struggling with the loss of family support. However, for individuals in interracial relationships, who may be experiencing interpersonal sanctioning from other social sources, religious communities may offer a unique opportunity to provide support that encourages and promotes psychological well-being.

Although organizational religious involvement may positively influence the mental health of individuals dealing with the absence of parental support in the ways outlined above, it may be reasonable to expect that formal religious involvement may also negatively influence the health of individuals engaged in interracial unions specifically. Organized religion has a complex history with race and intimacy in which it has often played an unfortunate role in the stigmatization of interracial marriage and relationships in the U.S. (Coates 2015). American Christianity, via formal and unofficial church doctrine, historically prohibited interracial relationship formation, often vilifying such unions as being “unequally yoked” (Botham 2009). Although few churches today would openly admit to such beliefs and practices, most churches continue to be tightly constructed along racial lines (Emerson and Kim 2003). A small minority of churches and congregations are truly multiethnic (Yancey and Emerson 2003; Edwards et al. 2013). Therefore, individuals in interracial relationships, who remain involved in organized religion, may be the target of either real or perceived sanctioning as a result of their relationship that may negatively influence their mental well-being (Perry 2014). Therefore, religious attendance may have less (or even a negative) influence on the health of individuals in interracial relationships compared to their peers in monoracial relationships.

Second, religious coping, best conceptualized as a multidimensional construct involving the use of religious cognitions and behaviors, including private prayer, religious support and guidance (Taylor et al. 2003; Pargament et al. 1998), may also influence health and well-being (Siegel et al. 2001; Pargament and Brant 1998; Fabricatore et al. 2004). For example, through prayer, individuals may develop a close, personal relationship with God (or a divine other), who is thought to offer comfort and solace during difficult times (Pollner 1989; Kirkpatrick 2005). Such private activities may cultivate a belief of being a “child of God” that results in feelings of dignity and worth that alters the perception, experience, and reaction to negative events, including the loss of social support (Cooper-Lewter and Mitchell 1986). Such relationships often result in feelings of closeness and personal attachment, in which people are particularly likely to turn to God (or a divine other) during events perceived as stressful (Kirkpatrick 2005). Indeed, research suggests people often turn to prayer during difficult emotional states, such as loss, anger or fear (Ai et al. 2007). In this context, prayer may help to alleviate strong negative emotions and facilitates open communication and forgiveness in interpersonal relationships that results in better psychological well-being (Lambert et al. 2010). In this context, religious coping provides protection and security, and is therefore likely to reduce distress (Krause 2006).

Lastly, subjective religiosity is a dimension of religious involvement distinct from both public and private forms of religious belief and behavior. Subjective religiosity, generally conceived of as the personal importance or self-assessed strength of one’s religious identity, emphasizes an internalized (i.e., intrinsic) religious commitment (Allport and Ross 1967; Chatters et al. 2008). Although individual intentions for engaging in religious behavior may vary, those who are intrinsically motivated may find greater psychological benefits from their religious identity (Ryan et al. 1993). Additionally, subjective religiosity may influence health by offering a comprehensive framework for assigning attributions or explanations to mundane affairs, and chronic challenges that reduces psychological distress (Pargament et al. 1998). For example, intrinsic religiosity has been found to be associated with lower levels of anxiety and depression in response to negative life events, including events described as uncontrollable (Park et al. 1990). Among individuals in interracial relationships, subjective religiosity may be particularly helpful by creating opportunities to establish an identity based on internal qualities, such as spirituality, kindness, and generosity that may reduce the salience of external, racial differences. Work on multiracial congregations suggests participation in multiethnic religious communities may lead to ethnic transcendence (Marti 2008), or an alternative identity framed around religious interests instead of a race-ethnic identity. Perhaps it is the case that intrinsically religious individuals—i.e., committed to applying and living out their faith in all areas of their lives—in interracial unions are able to build a (shared) religious identity that transcends race and protects them in the face of stress and challenges related to their romantic union. For these reasons, subjective religiosity, or the salience of religion in one’s life, may results in higher levels of psychological well-being, even in the face of stressful situations like lack of family support.

Investigating similarities and differences in the role of religion in same-race and interracial relationships is largely an unexplored avenue of research. Surprisingly, little is known about how religion may influence the health and family life of individuals engaged in interracial relationships. Nevertheless, there may be some reasons to expect that such individuals may be less inclined toward religion when compared to their peers in same-race relationships. Research on the social factors that predict engagement in interracial romantic ties generally find these individuals are younger, politically liberal, live in urban areas, come from racially diverse backgrounds, and have higher incomes and education (Herman and Campbell 2012; Johnson and Jacobson 2005; Perry 2013a; Fujino 1997). Such factors are also associated with lower levels of religious involvement (Levin and Chatters 1998; Krause 2004). For these reasons it may be that individuals in interracial unions may be less likely to use and engage in religion in established ways, thereby distinguishing them from their peers in same-race relationships. However, theories surrounding resource substitution suggest social and interpersonal resources may be exchanged for one another in the face of loss or threat of loss (Ross and Mirowsky 2006). In the face of weak parental support, individuals in same-race and interracial relationships, who engage religion in the ways described above may indeed find themselves better off than their less religiously involved peers.

2.4. Conceptual Models

Based on the theory and research reviewed to this point, two conceptual models guide the way(s) in which weak parental support and religious involvement may be linked to the depressive symptoms of individuals in same-race and interracial relationships. In the first model, weak parental support is posited to have a positive association with depressive symptoms, while multiple dimensions of religious involvement are expected to have an inverse association with depressive symptoms. The effects of weak parental support and religious involvement, however, are thought to be largely or completely independent of one another. In the second model, or the stress-buffering model, we expect that religion will help reduce the impact of weak parental support on depressive symptoms. That is, religious involvement is expected to moderate, i.e., buffer or mitigate, the deleterious effects of weak parental support on depressive symptoms among individuals in same-race and interracial relationships. The buffering (or moderating) model is formulated as one involving an interaction—or cross-product term—between weak parental support and religion (i.e., weak parental support × religion; Ellison and Henderson 2011).

3. Data and Methods

Data come from the National Longitudinal Study of Adolescent to Adult Health (Add Health), a school-based study of a nationally representative sample of adolescents in grades 7–12 in 1994–1995 in the US. The Add Health used a multistage, stratified, school-based cluster sampling design and involves four waves of data collection. Add Health focuses on the social, economic, psychological and physical well-being of adolescents to young adulthood, and provides unique opportunities to study how social environments and behaviors, including families, romantic relationships, and religion, are linked to health. In 1994–95, Wave I data yielded a sample of 20,745 respondents with a response rate of 79% from all participating schools. Wave IV was a follow-up study of the individuals from Wave I conducted in 2008, in which respondents were 24–32 years old, and yielded approximately 75% of the original Wave 1 respondents (n = 15,701). The sampling methods of the Add Health have been described in detail elsewhere (Harris 2011).

3.1. Dependent Variable: Depressive Symptoms

Depressive symptoms was measured using nine items derived from the Center for Epidemiologic Studies Depression Scale (CES-D) (Radloff 1977). Respondents were asked how often in the past week (7 days) they: (1) were bothered by things, (2) could not shake off the blues, (3) felt as good as others, (4) had trouble concentrating, (5) felt depressed, (6) felt too tired, (7) enjoyed life, (8) felt sad, and (9) felt disliked. Responses ranged from 0 = “never or rarely (less than one day)” to 3 = “most or all of the time (5–7 days).” Per convention, responses were reverse-coded where necessary and the items were summed with a range of 0 to 27. The Cronbach’s alpha is 0.83.

3.2. Key Independent Variables

Weak Parental Support. Parental support was constructed by first averaging across four items regarding the respondent’s maternal and paternal figure. Specifically, the respondent was asked the following questions: (1) Are you satisfied with the way you and your mother [and father] figure communicate?; and (2) How close to do you feel to your mother [and father] figure? Responses to the questions were 1 = “strongly disagree,” to 5 = “strongly agree,” and 1 = “not at all” to 5 = “very much,” respectively. Responses were reverse coded so that higher scores reflect weak parental support; the original measure ranged of 2–20. However, due to the skewed distribution of the original index, we then dichotomized the index at the mean (x = 6.36), such that: 1 = “weak parental support (i.e., x ≥ 6.36)” and 0 = “high parental support (x < 6.36).”1 Respondents missing on both the mother and father support questions were dropped from the analysis (i.e., no parental figure present). The current measure of parental support is consistent with prior studies using Add Health data (LeCloux et al. 2017).

Religious Involvement. Three distinct aspects of religious involvement were included in the analysis. Organizational religious involvement was assessed by asking about the frequency of worship service attendance in the past 12 months (i.e., religious attendance). Responses ranged from 0 = “never” to 5 = “more than once a week.” Religious coping was assessed by standardizing and averaging the responses of two questions: (1) How often do you pray privately, that is, when you are alone in places other than a religious assembly?’’ and (2) “How often do you turn to your religious or spiritual beliefs for help when you have personal problems, or problems at school or work?” Responses to the questions ranged from 0 = “never” to 7 = “more than once a day;” and 0 = “never” to 4 = “very often,” respectively. The Pearson correlation coefficient for the two items is 0.76, p < 0.001. Lastly, one item assessed subjective religiosity, in which respondents were asked: ‘‘How important (if at all) is your spiritual life to you?’’ Responses ranged from 1 = “not important” to 4 = “more important than anything else.” Higher scores on all three religious involvement items reflect higher levels of the religious attendance, religious coping, and subjective religiosity.

Covariates.2 In addition, the models control for: race (measured in a series of dummy variables, including 1 = Black, 1 = Hispanic, 1 = Asian vs. 0 = White, non-Hispanic), gender (1 = female vs. 0 = male), age (measured in years), respondent’s marital status (measured in a series of dummy variables, 1 = dating, 1 = cohabitation, vs. 0 = married serving as the reference category), parental status (1=current living child(ren) vs. 0 = all other) and socioeconomic status (SES). Respondent SES is a composite index constructed from three standardized measures of the respondent’s education, occupation, and poverty threshold (i.e., income in relation to the federal poverty level for a given household size) measured in Wave IV. We also control for depressive symptoms at Wave I to account for previous mental health that may confound the relationship between weak parental support, religious involvement, and young adult depressive symptoms.

Our analytical sample was limited to those who participated in Wave IV in-home interview, reported being in a romantic relationship at the time of the interview, reported their partner’s race, and self-identified their race/ethnicity as either non-Hispanic white, non-Hispanic Black, Hispanic, or Asians. Respondents who reported their race as “other” were excluded from the analysis, resulting in an n = 13,044. After listwise deletion of missing cases (n = 873, <7%), including reports of no mother or father figure, the analytical sample is n = 12,171. The sample was stratified across individuals in same-race (n = 9789) and interracial (n = 2382) relationships.

3.3. Analytical Strategy

The data analysis progressed in several steps, and all analyses were stratified by same-race and interracial relationships. First, descriptive statistics are presented in

Table 1. Sample Characteristics by Same-race and Interracial Relationship, Add Health ^a.

		Same-Race Relationship (n = 9789)		Interracial Relationship (n = 2382)
Variable	Range	Mean (%)	St. Error	Mean (%)	St. Error
Dependent Variable
Adult Depressive Symptoms	0–27	5.26	0.08	5.90 a	0.15
Key Independent Variables
Weak Parental Support	0–1	(40)		(45) a
Religious Attendance	0–5	1.57	0.04	1.29 a	0.04
Religious Salience	0–3	2.48	0.03	2.35 a	0.03
Religious Coping	−1.53–1.23	−0.06	0.03	−0.20 a	0.03
Covariates
NH White	0–1	(72)		(59)
NH Black	0–1	(15)		(15)
Hispanic	0–1	(11)		(19)
Asian/Pacific Islander	0–1	(2)		(7)
Age (in years)	24–33	28.4	0.12	28.36	0.13
Female	0–1	(49)		(49)
Parent	0–1	(51)		(47) a
Respondent SES	−2.37–1.12	−0.07	0.03	−0.08	0.04
Married	0–1	(46)		(34) a
Cohabiting	0–1	(20)		(23) a
Dating	0–1	(34)		(43) a
Depressive symptoms, WI	0–27	5.63	0.08	5.97 a	0.14

Note: All data are weighted. ^a Mean differences between same-race and interracial relationship are significant at p < 0.05.

. To examine the proposed frameworks, we ran a series of five linear regression models. Model 1 reports the estimated net effects of weak parental support on depressive symptoms. Model 2 adds the key independent variables of religious involvement—attendance, religious coping and subjective religiosity—to Model 1. Models 3 through 5 introduce separate interaction terms between weak parental support and our three religious involvement variables (e.g., weak parental support × religion) across both same-race and interracial samples. Continuous variables were mean-centered before estimating interaction terms to reduce collinearity (Aiken and West 1991).

4. Results

Summary statistics for all variables are reported in Table 1 stratified by same-race and interracial relationships. Young adults in interracial relationships report more depressive symptoms relative to individuals in same-race relationships (values of 5.9 vs. 5.3, p < 0.05, respectively). Additionally, a higher percentage of individuals in interracial relationships report weak parental support (45% vs. 40%, p < 0.05). Of particular interest are the observed differences in religious involvement by relationship type. Specifically, individuals in same-race relationships appear to report higher levels of religious involvement across all three measures used in the analyses compared with their peers in interracial relationships. There appears to be few significant differences in our sociodemographic characteristics by relationship type, except for relationship status, where a higher percentage of individuals in interracial relationships appear to be dating (43% vs. 34%, p < 0.05) and cohabiting (23% vs. 20%, p < 0.05) and are less likely to be married (34% vs. 46, p < 0.05) compared to those in same-race relationships.

Table 2. Estimated for Weighted Linear Regression Models Predicting Depressive Symptoms, Same-Race (n = 9789), Add Health.

	Model 1	Model 2	Model 3	Model 4	Model 5
	b (SE)	b (SE)	b (SE)	b (SE)	b (SE)
Weak parental support	0.87 (0.12) ***	0.88 (0.12) ***	0.88 (0.12) ***	0.88 (0.12) ***	0.88 (0.12) ***
Religious attendance		−0.26 (0.05) ***	−0.24 (0.06) ***	−0.26 (0.05) ***	−0.26 (0.05) ***
Religious coping		0.59 (0.13) ***	0.59 (0.13) ***	0.58 (0.13) ***	0.59 (0.13) ***
Subjective religiosity		−0.20 (0.13)	−0.20 (0.13)	−0.20 (0.13)	−0.17 (0.13)
Black, Non-Hispanic a	0.67 (0.20) ***	0.58 (0.21) **	0.58 (0.21) **	0.58 (0.21) **	0.58 (0.21) **
Hispanic a	0.03 (0.20)	−0.01 (0.21)	−0.01 (0.21)	−0.01 (0.21)	−0.01 (0.21)
Asian a	0.73 (0.29) *	0.79 (0.29) **	0.80 (0.30) **	0.79 (0.29) **	0.80 (0.29) **
Age	0.06 (0.03)	−0.01 (0.03)	−0.01 (0.03)	−0.01 (0.03)	−0.01 (0.03)
Female	1.00 (0.11) ***	0.89 (0.11) ***	0.88 (0.11) ***	0.89 (0.11) ***	0.89 (0.08) ***
Parent	−0.22 (0.12)	−0.22 (0.12)	−0.22 (0.12)	−0.22 (0.12)	−0.22 (0.12)
Cohabiting b	0.33 (0.14) *	0.25 (0.15)	0.25 (0.15)	0.25 (0.15)	0.25 (0.15)
Dating b	0.53 (0.13) ***	0.49 (0.14) ***	0.49 (0.14) ***	0.49 (0.14) ***	0.49 (0.14) ***
SES	−0.92 (0.08) ***	−0.89 (0.08) ***	−0.89 (0.08) ***	−0.89 (0.08) ***	−0.89 (0.08) ***
Depressive symptoms WI	0.27 (0.01) ***	0.27 (0.01) ***	0.27 (0.01) ***	0.27 (0.01) ***	0.27(0.01) ***
Interactions c
Weak parental support × Religious attendance			−0.04 (0.07)
Weak parental support × Religious coping				−0.03 (0.13)
Weak parental support × Religious salience					−0.07 (0.11)
Intercept	2.83 (0.89) **	3.92 (0.88) ***	3.88 (0.88) ***	3.92 (0.88) ***	3.84 (0.87) ***
Adj. R2	0.16	0.17	0.17	0.17	0.17

Notes: * p < 0.05; ** p < 0.01; *** p < 0.001. ^a Reference category is Non-Hispanic White; ^b Reference category is married; and ^c Components of interaction terms are zero-centered as recommended by Aiken and West (1991) and are entered independently.

shows the results of the multivariate regression models predicting depressive symptoms by weak parental support and religious involvement for same-race couples. As predicted, weak parental support is positively associated with depressive symptoms (Model 1: b = 0.87, p < 0.001) net of covariates. That is, compared to individuals with high levels of parental support, those who reported weak parental support also report significantly higher levels of depressive symptoms. In model 2, we find mixed support for our hypotheses regarding religious involvement. After the inclusion of our religion measures, weak parental support remains positively associated with depressive symptoms net of covariates (Model 2: b = 0.88, p < 0.001). Religious attendance is inversely related to depressive symptoms (Model 2: b = −0.26, p < 0.001). However, surprisingly, religious coping is positively associated with depressive symptoms (Model 2: b = 0.59, p < 0.001) among individuals in same-race relationships. We found no significant interactions between weak parental support and our measures of religious involvement, suggesting religion does little to protect against the deleterious effects of weak parental support on depression among individuals in same-race relationships.

Table 3. Estimated for Weighted Linear Regression Models Predicting Depressive Symptoms, Interracial Relationships (n = 2382), Add Health.

	Model 1	Model 2	Model 3	Model 4	Model 5
	b (SE)	b (SE)	b (SE)	b (SE)	b (SE)
Weak parental support	0.34 (0.22)	0.32 (0.22) *	0.32 (0.22)	0.31 (0.22) *	0.31 (0.22)
Religious attendance		−0.22 (0.11)	−0.16 (0.14)	−0.23 (0.11) *	−0.22 (0.11) *
Religious coping		0.35 (0.20)	0.34 (0.20)	0.56 (0.21) **	0.35 (0.20)
Subjective religiosity		−0.15 (0.22)	−0.14 (0.22)	−0.12 (0.22)	−0.46 (0.29)
Black, Non-Hispanic a	1.11 (0.39) **	1.08 (0.39) **	1.07 (0.39) **	1.09 (0.39) **	1.07 (0.38) **
Hispanic a	0.49 (0.29)	0.52 (0.28)	0.51 (0.28)	0.52 (0.29)	0.52 (0.29)
Asian a	0.46 (0.38)	0.49 (0.37)	0.49 (0.38)	0.48 (0.38)	0.53 (0.29)
Age	−0.05 (0.06)	−0.05 (0.06)	−0.05 (0.06)	−0.05 (0.06)	−0.05 (0.06)
Female	1.19 (0.24) ***	1.13 (0.25) ***	1.14 (0.25) ***	1.16 (0.24) ***	1.15 (0.25) ***
Parent	−0.31 (0.28)	−0.31 (0.28)	−0.30 (0.28)	−0.29 (0.28)	−0.31 (0.28)
Cohabiting b	0.20 (0.32)	0.10 (0.32)	0.10 (0.32)	0.08 (0.32)	0.08 (0.32)
Dating b	0.71 (0.28) **	0.67 (0.28) *	0.68 (0.27) *	0.67 (0.27) *	0.66 (0.27) *
SES	−1.15 (0.18) ***	−1.13 (0.18) ***	−1.12 (0.18) ***	−1.12 (0.18) ***	−1.12 (0.17) ***
Depressive symptoms WI	0.27 (0.03) ***	0.27 (0.03) ***	0.27 (0.03) ***	0.27 (0.03) ***	0.27 (0.03) ***
Interactions c
Weak parental support x Religious attendance			−0.13 (0.14)
Weak parental support x Religious coping				−0.51 (0.24) *
Weak parental support x Religious salience					−0.46 (0.29)
Intercept	3.99 (1.79) *	4.61 (1.81) *	4.55 (1.81) *	4.63 (1.82) *	4.06 (1.87) *
Adj. R2	0.16	0.16	0.16	0.17	0.16

Notes: * p < 0.05; ** p < 0.01; *** p < 0.001. ^a Reference category is Non-Hispanic White; ^b Reference category is married; and ^c Components of interaction terms are zero-centered as recommended by Aiken and West (1991) and are entered independently.

shows the multivariable regression estimates for individuals in interracial relationships. In Model 1, weak parental support is not significantly associated with depressive symptoms among individuals in interracial relationships net of covariates. However, in Model 2, once the religion variables and covariates are held constant, a curious suppressor pattern emerges, in which weak parental support becomes significantly associated with depressive symptoms among individuals in interracial relationships (Model 2: 32; p < 0.05). Counter to our hypotheses, none of our religious involvement variables are significantly related to depressive symptoms among individuals in interracial relationships. Turning to our models examining the stress-buffering effects of religious involvement, it appears we find limited support for our hypotheses. Religious coping appears to protect against the harmful effects of weak parental support on depressive symptoms among individuals in interracial relationships based on the negative regression coefficient (Model 4: −0.51, p < 0.05). We present the interaction graphically in Figure 1. The vertical axis indicates depressive symptoms and presents the predicted probabilities. On the horizontal axis is religious coping, ranging from low to high. The lines in the figures show how the effect of religious coping depends on parental support. According to the figure, it appears that at low levels of religious coping, individuals in interracial relationships with weak parental support also report significantly higher levels of depressive symptoms compared with individuals reporting high parental support. The slope of the line for weak parental support, however, remains flat as religious coping increases. At the highest levels of religious coping there is a crossover effect, where individuals reporting weak parental support now report slightly lower levels of depressive symptoms than individuals reporting high levels of parental support. Surprisingly, however, our results suggest individuals in interracial unions with strong parental support, experience elevated symptoms of distress as religious coping increases. We found no significant interactions between weak parental support and religious attendance and subjective religiosity.

Ancillary Analysis

In ancillary analysis (not shown, but available upon request), we examined our results for differences by relationship status (i.e., dating, cohabiting and married). There is some evidence to suggest that relationship status is related to depressive symptoms, particularly among interracial couples (Connolly and McIsaac 2011; Bratter and Eschbach 2006), and parental support may also vary according to the level of commitment of the child’s romantic relationship. We found no significant differences in the relationship between weak parental support and religion by relationship status across both our same-race and interracial couples. Additionally, we examined our data for differences by racial composition² among individuals in our interracial relationship sample. Research suggests that white, minority couples may experience greater disapproval from peers and family than other combinations (e.g., minority, minority relationships; B. Miller 2017; Bratter and Eschbach 2006). However, we found no significant differences in our relationships of interest by couples’ racial composition.

5. Discussion

This research contributes to a growing literature on the mental health of individuals in same-race and interracial relationships. We examine weak parental support as a distinct psychosocial stressor and religious involvement as a coping-related mechanism in these romantic relationships. Our results suggest that indeed weak parental support acts as psychosocial stressor for individuals in both same-race and interracial relationships that increases their risk of depressive symptoms. These results are consistent with previous research on the importance of parental support in the psychological adjustment of young adults (Davila et al. 2009; Lorenzo-Blanco et al. 2013; B. C. Miller 2002; Whitbeck et al. 1993). Although there is little research surrounding the specific mechanisms, the absences of parental support may increase the risk of distress in young adults by increasing loneliness and isolation.

We find mixed support for our hypotheses on the main effects of religious involvement on depressive symptoms. Specifically, our results suggest that religious involvement, specifically religious attendance, is inversely related to distress for individuals in same-race relationships, but not individuals in interracial unions. A growing body of research indeed finds that social relationships within churches tend to be especially close and highly supportive, especially during times of trouble (Krause 2006; Nooney and Woodrum 2002). Organizational religious involvement offers opportunities for formal and informal social support (Krause 2006). Regular religious service participation may encourage beliefs and behaviors that promote and enhance the help-giving process (i.e., role models and shared beliefs on forgiveness and acceptance, etc.), which may be particularly helpful for those suffering from the absence of parental guidance. Contrary to our hypotheses, religious coping is positively associated with depressive symptoms among individuals in same-race relationships. What might explain this unexpected finding? Research finds that religion influences the formation of romantic relationships among young adults (Regnerus 2007). Religion has been identified as an important source in opposition to non-marital cohabitation and sex (Lehrer and Chiswick 1993; Barkan 2006; Regnerus 2007). Perhaps religious young adults engaged in romantic ties—especially if the union is not sanctioned by parents—experience elevated feelings of distress. Surprisingly, we find no significant main effects of religion on depression among individuals in interracial relationships. However, our results suggest such individuals are less likely to be religious. Recent work by Perry (2013b) finds that in some contexts, some dimensions of religious involvement are inversely related to interracial dating in adulthood. Such findings suggest that the relationship between religion, health, and family may be more complex among individuals engaged in interracial ties.

We find limited support for our hypotheses concerning the stress-buffering role of religious involvement on weak parental support and depressive symptoms. Specifically, while it appears that religious coping moderates the association between poor parental support and depressive symptoms for individuals in interracial unions, on closer examination it is among individuals with strong parental support that mental well-being declines as they seek comfort via private prayer and religious guidance. This finding is somewhat surprising, and an explanation may be beyond the scope of this paper. However, one possible explanation may be the issue of religious homogamy (Seshadri and Knudson-Martin 2013; Heaton and Pratt 1990; Kalmijn 1998). Religious homogamy, i.e., similarities between partners in religiosity, has been found to influence well-being and relationship quality among married and nonmarried couples (Heaton and Pratt 1990; Lehrer and Chiswick 1993). Perhaps highly religious individuals in interracial unions, with close family ties, experience distress if their partneris less religiously involved. Although such individuals may be able to turn to their parents for support, they may also feel guilt over the differences in their partner’s religious commitment, which leads to distress. Unfortunately, no measures of partners’ religiosity are available in Wave IV of Add Health. Future work should investigate these relationships at the couple-level.

Although this study addresses an important gap in the literature, it is also characterized by several notable limitations. First, while consistent with previous measures of parental support using Add Health (LeCloux et al. 2017), the items used to construct parental support do not directly reflect attitudes or feelings regarding the child’s romantic relationship. Future work should seek to incorporate a distinctive measure of parental support that reflects thoughts and attitudes about of the romantic relationship in which their child is engaged, particularly in the context of an interracial relationship. Second, due to the small sample sizes of some combinations, this study was also unable to investigate differences by specific racial dyads (e.g., Asian–Black)3. Third, while we control for prior mental health, the use of longitudinal, panel data to explore these relationships may also prove insightful for understanding the role of stress and coping across the life course among individuals in both same-race and interracial unions. Lastly, this study relies on the responses of a single partner. As previously discussed, future research should attempt to replicate these findings by examining relational spirituality (i.e., shared religious beliefs, shared affiliation and joint attendance; Mahoney 2010) via data from both partners on their shared religious involvement.

Despite these limitations this study has made a significant contribution to the research literature on the relationship between health and intimate relationships. Considerable evidence reveals romantic ties in young adulthood leads to worse mental health (Connolly and McIsaac 2011; Davila 2008; Joyner and Udry 2000; Welsh et al. 2003), particularly for those involved in interracial unions (B. Miller 2017; Lykke 2017; Bratter and Eschbach 2006; Miller and Kail 2016; Kroeger and Williams 2011). It is also clear parents play a critical role in the psychological adjustment of young adults (Davila et al. 2009; Lorenzo-Blanco et al. 2013; B. C. Miller 2002; Whitbeck et al. 1993). However, the potential influence of religious involvement as a protective factor in the absence of parental support has been widely neglected by researchers interested in understanding the health of individuals in same-race and interracial unions. This work adds to the research on the stress-buffering role of religious involvement (Ellison and Henderson 2011). Clearly, the potential influence of religious factors in such relationships is more complex and potentially important than previously recognized. Further investigation along the lines suggested above may clarify the connections between religion, health, and family among a diversity of relationships in the contemporary U.S.