Cultural Dynamics and Marital Relationship Quality in Mexican-origin Families

Abstract

Prior research suggests that acculturation may influence relationship outcomes among Mexican-origin married couples, including marital adjustment and distress. Despite much theory and research on parent-child cultural differences and disruptions in the parent-child relationship, no previous research has investigated possible associations between husband-wife cultural differences and marital relationship quality. With a sample of Mexican-origin married couples (N = 398), the current study investigated the relations between husband-wife differences in acculturation (American orientation) and enculturation (Mexican orientation) with husband and wife reports of positive marital qualities (warmth and relationship satisfaction). To clarify and extend previous research, the current study also investigated within-person models of cultural orientation domains as related to positive marital quality. Results provide partial evidence showing that dyadic cultural differences are associated with lower positive marital quality while cultural similarity is associated with higher positive marital quality; however, the relations are complex and suggest that the associations between wife cultural orientation and positive marital quality may depend on husband cultural orientation (and vice versa). Findings also implicate the importance of assessing spouse bidimensional cultural orientation by showing that the relation between spouse acculturation level and relationship quality may depend on his or her enculturation level. Additional nuances in the findings illustrate the importance of assessing multiple domains of cultural orientation, including language use and cultural values. We highlight several future directions for research investigating nuances in spouse cultural dynamics and relationship processes.

Disrupted marital processes are associated with poorer functioning of family members, including poorer mental (Papp, Goeke-Morey, & Cummings, 2007) and physical health (Kiecolt-Glaser & Newton, 2001) among husbands and wives. On the other hand, positive marital relationship quality may help to buffer negative health outcomes among spouses (e.g., Kiecolt-Glaser & Newton, 2001). However, research on factors that influence marital relationship dynamics is limited by a lack of ethnically diverse groups (Fincham & Beach, 2010; McLoyd, Harper & Copeland, 2001; Zimet & Jacob, 2001), particularly with respect to Latinos (Gonzales, Pitts, Hill, & Roosa, 2000; Flores, Tschann, Marin, & Pantoja, 2004). Additional research investigating martial relationships within Latino families is needed given the size and projected growth of this group, which is predicted to make up 30% of the total U.S. population by 2050 (U.S. Census Bureau, 2008). A focus on Mexican-origin families, the largest Latino subgroup (63%) (Humes, Jones & Ramirez, 2011), is particularly relevant at this time.

Culturally grounded research is needed to investigate unique influences on marital relationship processes and to address the notable heterogeneity within the Latino population. We define culture as shared meanings, understandings, or referents held by groups of people (Shore, 2002; Triandis, 1995). In particular, we focus on acculturation and enculturation, which represent important cultural domains of variability among Latinos in the United States. Acculturation is defined as the adoption of American cultural practices, values, and identity. Enculturation is defined as the maintenance or retention of culture of origin, also evaluated in terms of cultural practices, values, and identity. Contemporary acculturation theory emphasizes dynamic changes on these multiple domains of culture across both acculturation and enculturation dimensions (Berry, 2006; Schwartz et al., 2010), also labeled acculturation and enculturation, respectively (e.g., Gonzales et al., 2002; Knight et al., 2010). Acculturation and enculturation are linked theoretically with changes related to the immigration process, which is important given that individuals born in Mexico make up approximately 40% of the U.S. Mexican-origin population (Grieco, 2010), and are also relevant for the substantial proportion of Mexican-Americans who have been in the U.S. for one or more generations, especially in the border region of Texas, New Mexico, Arizona, and California.

Several authors have proposed that changes and stress associated with acculturation and enculturation may disrupt family processes (e.g., Portes & Rumbaut, 1996), and a growing literature investigates the link between acculturation variables and family relationships (e.g., Smokowski, Rose, & Bacallo, 2008). Research has examined both the role of individual cultural orientation, and of dyadic similarity and discrepancy in cultural orientation (e.g., the acculturation gap) in relation to family relationship qualities, yet, the majority of this research has focused on parent-child relationships (e.g., Bornstein & Cote, 2006), and these questions have not been adequately addressed within the marital dyad. To address this gap in the literature we examined dimensions of acculturation and enculturation within the marital dyad and tested how (a) dyadic cultural orientation dynamics and (b) individual husband and wife cultural orientations were associated with positive marital qualities.

Sociocultural Dynamics and the Marital Relationship

Sociocultural factors exert a dynamic influence on Mexican-origin immigrant couples both in the culture of origin (i.e., Mexico) and in the receiving culture (i.e., the U.S.). In Mexico, awareness of the global modernization of women’s role and changing marital patterns have intersected, precipitating a transformation of gender roles and the marital relationship (Hirsch, 2003; Oropesa & Landale, 2004). Contrary to the stereotypical portrayal of dominant, authoritarian husbands and loyal, submissive wives (e.g., Cauce & Domenech-Rodríguez, 2002; McLoyd, Cauce, Takeuchi, & Wilson, 2000), recent qualitative studies of Latino couples in both Mexico and the U.S. have suggested that love, trust (confianza), and friendship are endorsed as key components of a healthy marriage (Harris, Skogrand, & Hatch, 2008; Hirsch, 2003). Furthermore, the risky and sometimes dangerous experience of immigrating to America may require more flexibility in traditional gender-typed roles and responsibilities within couples (Hirsch, 2003; Hondagneu-Sotelo, 1994; Oropesa & Landale, 2004). In addition, immigrant couples arrive in an American cultural landscape that is traditionally individualistic, increasingly egalitarian relative to gender (Phinney & Flores, 2002), and in which traditional marriage patterns have been shifting over time (Oropesa & Landale, 2004). In this regard, exposure to the United States may serve to further to transform traditional marriage patterns among Hispanic immigrants, with this transformation mutually influenced by socioeconomic and cultural characteristics (Landale & Oropesa, 2007; Mcloyd, Cauce, Takeuchi, & Wilson, 2000).

The heterogeneity of exposure to Mexican and American culture (Cruz et al., 2012) may be related to changes in individual cultural ties (e.g., Knight et al., 2010), which in turn may have negative effects on family relationships (Smokowski, Rose, & Bacallo, 2008). In addition to possible changes in traditional gender roles described above, for some individuals, exposure to the U.S. may reduce the traditional emphasis on familismo, a set of values about the importance of strong family bonds, mutual assistance, and considering one’s family when acting as an individual (Cauce & Domenech-Rodriguez, 2002; Knight et al., 2010), and personalismo, the value of preserving positive and harmonious interpersonal relationships over individual considerations (Cauce & Domenech-Rodriguez, 2002). For some individuals, mainstream American cultural values, which tend to assert the importance of individualism and competition (Knight et al., 2010), may become more prominent over time and shape relationship characteristics. Spanish or English language use and preference, which provide important connections both to Mexican and American culture, also vary based on contextual demands and individual preference, and husbands and wives may face unique contextual demands and have discrepant individual language preferences. Ultimately, variation in the values and behaviors reflective of acculturation and enculturation may be associated with different ideals, expectations, and behaviors that influence qualities of the marital relationship.

Although sparse, prior research provides some evidence that cultural factors such as nativity (i.e., country of birth), overall acculturation level, and specific dimensions of acculturation and enculturation are related to marital relationship dynamics for Mexican Americans. In an early exploratory study, Casas and Ortiz (1985) found that Mexican-born husbands and wives endorsed more positive marital relations compared to their U.S.-born counterparts on all subscales of the Dyadic Adjustment Scale (DAS; Spanier, 1976). In studies using the Acculturation Rating Scale for Mexican-Americans (ARSMA and ARSMA-II; Cuellar, Harris, & Jasso, 1980; Cuellar et al., 1995), an overall measure of acculturation largely based on English and Spanish language use, Negy and Snyder (1997) and Parke et al. (2004) showed that increased acculturation was related to higher marital distress and problems for wives, but unrelated to marital problems for husbands, while Wheeler, Updegraff and Thayer (2010) found no relation between either husband or wife cultural orientation and relationship quality. Although these previous findings are somewhat mixed, they suggest that greater acculturation may be related to increased marital distress for wives, but possibly not for husbands. It has been suggested that greater acculturation associated with living in the U.S. may reflect a renegotiation of traditional marital roles (Casas & Ortiz, 1985) resulting in a more egalitarian marital relationship (Negy & Snyder, 1997), or may also reflect a more individualistic orientation (Parke et al., 2004), which disrupts traditionally strong family bonds.

These studies examining broad relationship functioning are informed by additional research investigating specific dimensions of marital relationship functioning. Prior work has found that women of lower acculturation status were more likely to experience their spouse as not reciprocating in the “give and take” of the relationship (Vega, Kolody, & Valle, 1988), which may fit with traditional gender role expectations within the marital relationship. On the other hand, more acculturated husbands and wives experience more direct marital conflict and are more expressive of their feelings in an argument than less acculturated husbands and wives (Flores, Tschann, Marin, & Pantoja, 2004), and more acculturated wives are less likely to withdraw from conflicts (Wheeler, Updegraff & Thayer, 2010). This increased approach towards conflict may be reflected in prior research showing that acculturated husbands view their wives as more aggressive both physically and verbally (Flores et al., 2004). Ultimately, greater acculturation may be associated with greater relationship distress via greater engagement in relationship conflict and perceived aggression, unless these approach-oriented behaviors are balanced with other positive relationship qualities. When jointly examining American and Mexican cultural orientations, Wheeler, Updegraff and Thayer (2010) found that greater endorsement of both Anglo and Mexican cultural orientations were related to greater use of cooperation and compromise in conflict resolution for both wives and husbands, thus it may be that spouses’ biculturalism helps them to navigate relationship demands by effectively balancing the shifting couple power dynamic and competing values of individualism and familism.

A notable strength of several of the prior studies is that they have included both husband and wife reports of cultural orientation and perception of marital quality, which is important to account for differential relations among these variables between husbands and wives (Negy & Snyder, 1997; Parke et al., 2004; Wheeler, Updegraff & Thayer, 2010). However, these few prior studies have several significant shortcomings, including limited attention to (a) jointly assessing both acculturation and enculturation dimensions, in particular the role of cultural values, and most notable, a lack of attention to (b) dyadic similarities and differences between spouses in their cultural orientation, which would better inform our understanding of the interplay between cultural and marital dynamics. As illustrated by Wheeler, Updegraff, and Thayer (2010), jointly assessing both American and Mexican dimensions of cultural orientation may provide unique insights into the associations between spouse cultural orientation “profiles” and relationship outcomes. The current study examines unique and joint effects of individual acculturation and enculturation to understand whether bicultural individuals may report more positive marital qualities. Importantly, we examine multiple cultural domains including both language use and cultural values as possible influences on the marital relationship, which provides an opportunity to address convergence and divergence of findings across cultural practices and values.

Although researchers have previously dedicated substantial attention to the effects of intergenerational (i.e., parent-child) cultural differences on parent-child relationships (Telzer, 2010), no study to our knowledge has examined whether cultural differences or similarities influence relationship dynamics within immigrant Latino marital couples. This is a notable omission given that cultural differences may emerge not only inter-generationally between parents and children, but between any family members (Garcia-Coll & Magnuson, 2000; Keefe & Padilla, 1987; Szapocznik & Kurtines, 1980). Therefore, consideration of cultural similarity and discrepancy within Latino marital dyads is strongly warranted.

The acculturation gap

According to the traditional acculturation gap distress model (Birman, 2006; Szapocznik & Kurtines, 1993, Szcapocznik, Kurtines, & Fernandez, 1980), youth from immigrant families tend to acculturate to American culture at a faster rate than their parents due to greater engagement in mainstream contexts (e.g., American schools), producing increasing cultural dissonance (Phinney, Ong, & Madden, 2000), which results in disrupted family processes and, ultimately, youth maladjustment. Similarly, we propose that as Mexican-origin couples integrate into U.S. society, husbands and wives may interact in divergent social and work contexts, leading to varying adoption of receiving cultural values and practices, and varying retention of values and practices from the culture of origin, which may in turn disrupt positive marital dynamics. On the other hand, we also expect that cultural similarity may promote positive marital relationships. This proposition is informed by the literature on marital similarity (e.g., Deal, Wampler, & Halverson, 1992; Gaunt, 2006), which suggests that spouse similarity in personality and values may be important determinants of marital relationship qualities, including marital satisfaction. Thus, the current study examined the links between cultural orientation and marital quality, focused specifically on understanding within-person associations between acculturation and enculturation in relation to positive relationship quality, and importantly, is the first study to address the effects of cultural differences (i.e., the acculturation gap) and similarity between husbands and wives on positive relationship qualities.

Methods for testing effects of the acculturation gap

Despite strong theoretical support for the acculturation gap-distress hypothesis, empirical studies testing parent-child acculturation gaps have provided a mixed picture (Telzer, 2010). For this reason, considerable debate has emerged about methods used to assess acculturation differences (Birman, 2006). Following Birman (2006) and Telzer (2010), we examined the effects of cultural discrepancy and similarity by testing interactions between husbands and wives on dimensions of acculturation and enculturation, which addresses the hypothesis that a spouse’s cultural orientation will have differential associations with marital relationship dynamics depending on his or her partner’s cultural orientation. The interaction method avoids interpretational difficulties of other methods (e.g., match/mismatch and difference score methods; see Telzer, 2010) and also allows for a test of the main effects of each spouse’s cultural orientation. The acculturation gap distress hypothesis would be supported if a significant interaction were found that showed decreased relationship quality when one partner is higher while the other is lower on a particular dimension of cultural orientation. Positive effects of cultural similarity would be supported if a significant interaction demonstrated that positive relationship characteristics were promoted when both spouses were similar in their endorsement of cultural practices and values.

Failure to detect effects of acculturation gaps in prior studies also may stem from a mismatch between theory and the measures on which acculturation discrepancies are based. Theoretical discussions typically focus on value discrepancies between family members, but most studies use language-based measures to evaluate gaps. In the current study, we used both linguistic markers of acculturation (English language use) and enculturation (Spanish), as well as a values-based measure developed for Mexican Americans (Knight et al., 2010) that examines both traditional Mexican and mainstream American cultural values. This allowed us to hone in on the nature of discrepancies that best predict marital relationship dynamics, and test core theoretical propositions underlying the acculturation gap distress hypothesis.

The Current Study

Drawing on bidimensional models of acculturation, such as Schwartz et al. (2010), we examined the interactive effects of husband and wife acculturation dimensions (English Use, American Cultural Values) and enculturation dimensions (Spanish Use, Mexican Cultural Values) on marital satisfaction and warmth, among Mexican-origin marital dyads. The current study aimed to investigate the effects of acculturation and enculturation differences and similarity within the marital dyad. We predicted that there would be dynamic relations between husband and wife cultural orientation, such that increasingly discordant endorsement of acculturation or enculturation would predict lower levels of marital relationship quality, while higher similarity would be related to greater marital quality. In addition, we tested unique and joint relations between specific acculturation and enculturation dimensions with aspects of marital quality to understand how cultural dimensions may interact to influence relationship quality. We predicted that positive marital quality would be greatest when spouses were bicultural, that is, above the mean in both Mexican and American cultural domains. In all models, we accounted for possible structural or demographic effects by adding husband and wife average income, education, and age, as well as both spouses’ exposure to U.S. as covariates.

Method

Overview of Research Design

Data for the current study were drawn from the California Families Project (CFP), an ongoing longitudinal study of Mexican-origin families in a metropolitan area in Northern California. Participants were enrolled based on Mexican/Mexican-American heritage, as determined by their self-identification and ancestry, as well as having a child in the 5^th grade in a public or Catholic school. Children and their families were drawn at random from rosters of students from school districts in this metropolitan area. First, second, and third generation children of Mexican origin were eligible for the study. Either two-parent or single-parent families were eligible to participate. Participants were recruited by telephone or, in cases where they did not have a telephone, by a recruiter who went to their home. Trained research staff interviewed the participants in their homes using laptop computers equipped with audio computer-assisted self-interviewing (ACASI), and participants were paid for their participation. Interviews were conducted in Spanish or English based on the preference of the participant. Data were from the first wave of the CFP and collected between 2006–2008.

Participants

From the larger sample of 674 families participating in the CFP, we limited the current analyses to those families (n = 398) in which the child’s mother and father were married and both parents participated in the interview. On average, husbands were 40 years old (SD= 6.12; Range = 28 to 65) and had nine years of education (SD=3.75). Wives, on average, were 37 years old (SD= 5.80, range= 27 to 57) and had nine years of education (SD=3.75). The vast majority of both husbands (89%) and wives (87%) were born in Mexico, and on average had been living in the U.S. for less than half of their life (.49 and .44 respectively)(See Table 1).

Table 1.

Descriptive Statistics for Study Variables

	Husband Report	Wife Report
Variable (Range)	Mean (SD)	Mean (SD)
Partner Warmth (1 – 4)	3.38 (.57)	3.40 (.64)
Relationship Satisfaction (1 – 4)	3.63 (.49)	3.55 (.55)
American Cultural Values (1 – 4)	2.80 (.44)	2.68 (.47)
Familism (1 – 4)	3.55 (.30)	3.53 (.31)
Traditional Gender Roles (1 – 4)	2.57 (.68)	2.47 (.70)
English Use (1 – 4)	2.43 (.71)	2.24 (.86)
Spanish Use (1 – 4)	3.34 (.69)	3.47 (.70)
Proportion of Life in U.S. (0 – 1)	.49 (.22)	.44 (.27)
Nativity (0 = Mexico, 1= U.S.)	89% Mexico	88% Mexico
Family Income (<$5k to >$95k)	(Median) $40,000– $44,999 ($20,000)
Couple’s Mean Age (28 – 61)	38.42 (5.47)
Couple’s Mean Education (Years) (1 – 19)	9.40 (3.19)

Note: n = 360– 398.

Measures

Two dimensions of relationship quality were assessed for each member of the marital dyad: warmth, and relationship satisfaction. Partner warmth was measured using the Behavioral Affect Rating Scale (BARS; Kim et al., 2003). The BARS warmth scale (9 items; α_Wife = 0.93; α_Husband = .90) asked each member of the dyad to report how often in the past three months his/her spouse engaged in specific actions that communicate caring and mutual respect. Sample items include: “how often did he let you know he really cares about you?” and “act supportive or understanding toward you?” This scale was rated on a 4-point frequency scale (1= almost never or never, 4 = almost always or always). Relationship satisfaction was measured with 5-item self-report scale (α_Wife = .92; α_Husband = .91). Four items (e.g., “Your relationship is strong”) were rated on a scale ranging from 1 “not at all true” to 4 “very true”, whereas the fifth item (“Everything considered, how happy are you in your current relationship?”) was rated on a scale ranging from 1 “very unhappy” to 4 “very happy.”

We used the Acculturation Rating Scale for Mexican-Americans–II (ARSMA–II; Cuellar, Arnold, & Maldonado, 1995) to assess English use (5 items; α_Wife = .91; α_Husband = .87) and Spanish use (5 items; α_Wife = .89; α_Husband = .87). Items asked participants to report how frequently they spoke, wrote, thought, listened to music, and watched television in each language, using a four-point frequency scale (1= Never or almost never, 4= Always or almost always). Although the original ARSMA-II also measures ethnic identity and affiliation, we only used language use items for both subscales.

To assess culturally related values we used the Mexican American Cultural Values Scale (Knight et al., 2010). The items comprising this measure were developed through focus groups of immigrant and U.S. born Mexican-origin individuals that were asked to identify traditional values that they ascribed to both Mexican and American culture. Specifically, we assessed American Cultural Values (ACV) (14 items; α_Wife = 0.75; α_Husband = .82), which measured values related to self-reliance, material satisfaction, competition, and independence. The ACV scale was supported as a higher order factor in the original validation study. Using the same measure by Knight and colleagues we also assessed traditional Mexican cultural values (MCV) including gender-role attitudes, religion, respect, and three forms of familism: support, obligations, and family as referent. We chose to investigate familism and traditional gender role values as specific, theoretically relevant domains of traditional MCV that may particularly influence marital relationship quality. We computed a mean for the gender-roles subscale (5 items, α_Wife = 0.64; α_Husband = .64) and the familism subscale (16 items; α_Wife = 0.79; α_Husband = .79), which combined items related to family support, obligations and family as a referent. All items used a four-point rating scale indicating endorsement of cultural values (1= Not at all, 4= Very much).

Control variables entered in all regression models included family income, husband and wife mean age and mean years of education. To control for U.S. exposure, we tested separate models controlling for spouse proportion of life in U.S. (range = 0 – 1) or nativity (0 = Mexico, 1= U.S.). Models did not differ substantially across these different methods of controlling for U.S. exposure, thus we present models controlling for proportion of life in the U.S.

Data Analysis Plan

After generating descriptive statistics and bivariate correlations among the key study variables, we then examined the effects of husband and wife cultural dynamics on marital relationship quality by testing two sets of regression models using Mplus 7 (Muthén & Muthén, 1998–2012). The first set of models examined dyadic effects of cultural orientation, testing main effects and interaction effects of each husband and wife cultural dimension (e.g., husband ACV x wife ACV) on their report of positive marital qualities. The second set of models examined within-person associations and interactions between American and Mexican cultural orientation (e.g., husband Spanish use x husband English use) on report of positive marital qualities. Variables were centered prior to testing interactions. Regression analysis in Mplus allowed us to simultaneously model the effects of our predictors and covariates on the continuous outcomes. For all models we evaluated husband and wife outcome variables in separate regression equations. We used maximum likelihood with robust standard errors (MLR) estimation for model parameters, which is robust to non-normality (Muthén & Muthén, 2012, p.603). We probed interactions using the framework provided by Bauer and Curran (2005), performing simple slopes analyses at low (−1SD below mean), medium (mean), and high (+1SD above the mean) values of the moderator. Significance tests of simple slopes parameters were obtained from entering re-centered terms in follow-up Mplus analyses. Missing data patterns were first examined using independent samples paired t-tests (Enders, 2010). We found no differences in missingness among predictors and outcomes of interest in terms of family income, age, education, or proportion of life in the U.S. Missing data were then addressed using full-information maximum likelihood (FIML) estimation assuming missingness at random, which uses all available data to estimate model parameters (Enders).

Results

Zero-Order Correlations

Correlations showed that higher Spanish use was related to higher ACV, familism and traditional gender roles, whereas greater English use was related to lower endorsement of both ACV and traditional gender roles (see Table 2). ACV was positively related to familism and traditional gender roles (positive correlations between American and Mexican values dimensions were also seen in Knight et al., 2010). Husbands’ and wives’ endorsement of language use and cultural values were positively associated, with correlations small in magnitude, suggesting that their ratings are somewhat independent.

Table 2.

Correlations among Husband and Wife Cultural Variables and Couple Demographic Characteristics

	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19
1) English (H)	1
2) English (W)	.40	1
3) Spanish (H)	−.53	−.33	1
4) Spanish (W)	−.33	−.57	.50	1
5) ACV (H)	−.13	−.16	.16	.10	1
6) ACV (W)	−.09	−.12	.09	.15	.21	1
7) Familism (H)	−.09	−.01	.18	.07	.42	.06	1
8) Familism (W)	.02	−.07	.05	.16	.14	.52	.12	1
9) TGR (H)	−.18	−.19	.12	.14	.57	.14	.32	.19	1
10) TGR (W)	−.20	−.18	.16	.13	.07	.53	.00	.40	.13	1
11) Partner Warmth (H)	.10	.06	.12	.03	.00	.07	.19	.12	−.02	−.02	1
12) Partner Warmth (W)	.05	.05	.04	.14	.01	.08	.08	.22	.05	.09	.39	1
13) Relationship Quality (H)	.21	.10	−.10	−.01	−.02	.07	.15	.16	−.03	.04	.49	.41	1
14) Relationship Quality (W)	.10	.07	−.05	.08	−.04	.02	.01	.18	−.02	.01	.35	.69	.44	1
15) Prop. Life in U.S. (H)	.44	.36	−.51	−.43	−.12	−.10	−.04	−.02	−.14	−.16	−.04	−.03	.12	.06	1
16) Prop. Life in U.S. (W)	.43	.57	−.45	−.53	−.18	−.22	−.07	−.10	−.20	−.23	−.02	−.13	.10	.06	.67	1
17) Family Income	.26	.31	−.25	−.24	−.23	−.12	−.17	−.15	−.26	−.19	−.15	−.08	.01	.02	.39	.47	1
18) Mean Age	.00	−.07	.01	−.02	.14	.08	.10	.04	.10	.08	−.07	−.12	.07	−.10	.08	−.02	−.02	1
19) Mean Years Education	.36	.39	−.30	−.26	−.30	−.17	−.16	−.17	−.36	−.22	−.01	−.04	.08	.05	.22	.35	.36	−.08	1

Note: (r > .18, p < .001; r > .14, p < .01; r > .10, p < .05). H = Husband; W = Wife. ACV = American Cultural Values; TGR = Traditional Gender Roles.

Covariates

We next tested full regression models using predictors and covariates. Separate models were tested for (a) spouses’ nativity and (b) proportion of life in the U.S. to examine multiple ways of controlling for U.S. exposure. Results demonstrated the same pattern results across methods in almost all cases; results are presented from models controlling for proportion of life in U.S., except in the two cases of divergent results in which both methods are presented. Two model covariates had a consistent pattern of associations with specific relationship outcomes. Higher family income was related to lower husband report of wife warmth in all models (b = −0.02, p < .05), while higher mean couple age was related to lower wife report of husband warmth (b = − 0.02, p < .05). Additionally, wife proportion of life in the U.S. was negatively related to perception of husband warmth (b = −0.48 to −.75, all ps < .05) in all but the Spanish use model. Covariates are not presented in tables due to space constraints.

Dyadic Models

Acculturation Dimensions

Husband ACV moderated the association between wife ACV and her perception of husband warmth (see Table 3). At low to mean levels of husband ACV there was no relation between wife ACV and her perception of husband warmth; however, at high levels of husband ACV, wives’ endorsement of ACV was positively associated with her perception of husband warmth (b = 0.21, p = .03) (See Figure 1). This result suggested that American cultural value similarity is beneficial when both members of the marital dyad are high in ACV. Similarly, husband ACV moderated the association between wife ACV and wife report of relationship satisfaction. There was a marginally significant association between wife ACV and relationship satisfaction only at high levels of husband ACV (b = 0.16, p < .07), and further interaction probing indicated that this association became significant only when husbands were at least 1.5 SD above the mean in ACV (b = 0.21, p = .04).

Table 3.

Results from Dyadic Regression Models: Husband and Wife Relationship Quality Regressed on Cultural Predictors and Covariates

Predictor	Partner Warmth				Relationship Satisfaction
	Husband Report		Wife Report		Husband Report		Wife Report
	b (SE)	95% CI	b (SE)	95% CI	b (SE)	95% CI	b (SE)	95% CI
Husband English	0.11 (.05)	[.01, .21]*	0.08 (.05)	[−.03, .19]	0.13 (.05)	[.04, .22]**	0.05 (.05)	[−.06, .16]
Wife English	0.06 (.04)	[−.03, .24]	0.13 (.04)	[.04, .23]**	0.01 (.04)	[−.07, .09]	0.02 (.05)	[−.07, .11]
H x W Interaction	0.04 (.05)	[−.06, .13]	0.00 (.05)	[−.11, .11]	0.00 (.05)	[−.10, .10]	0.08 (.05)	[−.02, .17]
Husband ACV	−0.02 (.07)	[−.16, .11]	0.01 (.08)	[−.14, .16]	−0.02 (.06)	[−.14, .09]	−0.01 (.06)	[−.13, .11]
Wife ACV	0.10 (.07)	[−.04, .23]	0.07 (.08)	[−.08, .22]	0.09 (.07)	[−.04, .22]	0.04 (.07)	[−.09, .17]
H x W Interaction	0.17 (.13)	[−.08, .43]	0.33 (.14)	[.06, .61]*	0.16 (.12)	[−.07, .39]	0.26 (.12)	[.02, .50]*
Husband Spanish	0.18 (.06)	[.08, .29]**	−0.02 (.06)	[−.13, .10]	−0.06 (.04)	[−.14, .03]	−0.05 (.05)	[−.15, .06]
Wife Spanish	0.03 (.06)	[−.08, .15]	0.12 (.07)	[−.01, .26]^	0.07 (.05)	[−.03, .17]	0.17 (.06)	[.05, .29]**
H x W Interaction	0.16 (.06)	[.05, .27]**	0.04 (.06)	[−.08, .16]	0.09 (.08)	[−.05, .23]	0.19 (.06)	[.07, .30]**
Husband Familism	0.35 (.10)	[.16, .54]***	0.14 (.11)	[−.07, .35]	0.23 (.08)	[.07, .39]**	0.02 (.09)	[−.15, .19]
Wife Familism	0.19 (.10)	[.00, .37]^	0.44 (.11)	[.23, .66]***	0.25 (.10)	[.06, .44]***	0.35 (.10)	[.16, .55]***
H x W Interaction	0.13 (.32)	[−.49, .76]	0.13 (.29)	[−.44, .70]	0.26 (.28)	[−.29, .82]	0.09 (.26)	[−.42, .59]
Husband TGR	−0.03 (.04)	[−.11, .06]	0.04 (.05)	[−.07, .13]	0.00 (.04)	[−.09, .08]	0.00 (.04)	[−.08 .08]
Wife TGR	−0.03 (.05)	[−.13, .06]	0.07 (.05)	[−.03, .17]	0.05 (.04)	[−.03, .12]	0.03 (.04)	[−.06, .11]
H x W Interaction	0.06 (.07)	[−.07, .19]	−0.03 (.07)	[−.17, .11]	0.02 (.06)	[−.09, .13]	−0.08 (.06)	[−.20, .04]

Note: Covariates for each model [family income, mean age, education (years), and wife and husband proportion of life in U.S.] are not shown in the table due to space constraints. ACV = American cultural values; TGR = Traditional Gender Roles.

^***p < .001;

^**p < .01;

^*p < .05;

^{^}p < .08.

Figure 1.

Interaction between Wife and Husband American Cultural Values Predicting Wife Report of Husband Warmth.

There were no main or interactive associations between husband and wife English use and marital relationship outcomes. We found a positive association between husband English use and husband report of wife warmth and relationship satisfaction, suggesting that husbands who endorsed higher levels of English use perceived their marital relationship as more positive. We also found a positive association between wife English use and her report of husband warmth, but no associations between either spouses’ English use and wife relationship satisfaction.

Enculturation Dimensions

Husband Mexican cultural values (familism and traditional gender roles) did not moderate the associations between wife Mexican cultural values and positive relationship qualities. As husbands’ and wives’ familism values increased, they reported greater partner warmth and relationship satisfaction. However, husbands’ and wives’ traditional gender role values were unrelated to their report of positive marital quality. Husband Spanish use moderated the relation between wife Spanish use and husband report of wife warmth. There was no relation between wife Spanish use and husband report of wife warmth at low and medium values of husband Spanish use; however, at high levels of husband Spanish use there was a significant positive association between wife Spanish use and husband report of wife warmth (b = 0.16, p = .03)(see Figure 2). We also found that husband Spanish use moderated the association between wife Spanish use and wife relationship satisfaction. There was no relation between wife Spanish use and her report of relationship quality at low levels of husband Spanish use; however, greater wife Spanish use was associated with higher positive relationship quality at medium (b = 0.20, p = .001) and high (b = 0.30, p < .001) levels of husband Spanish use.

Figure 2.

Interaction between Wife and Husband Spanish Use predicting Wife Report of Relationship Satisfaction.

Individual Models

Husbands’ Cultural Orientation

Husband Spanish use moderated the effects of husband English use on his report of wife warmth (see Table 4). Specifically, husband’s English use was positively associated with his perception of his partner’s warmth at low levels of husband Spanish use (b = 0.32, p < .001), and this association decreased at mean (b = 0.22, p < .001) and high levels (b = 0.12, p = .05) of Spanish use. This suggested that husband English use minimally influenced perceived wife warmth for husbands high in Spanish use, but greater husband English use had a stronger link with perceived wife warmth as his Spanish use decreased. Moreover, husband familism values moderated the association between husband ACV and husband report of wife’s warmth, with the opposite pattern as language use. There was no association between ACV and warmth at high levels of familism values (b = −0.01, p = .91), with an inverse association at mean (b = −0.16, p = .03) and low levels of familism (b = −0.30, p = .002)(see Figure 3). This suggested that husbands’ American values do not influence perception of partner warmth when husbands report high levels of familism values. However, for husbands at or below the mean of familism, there is an increasing strength of association between higher familism values and lower perceived partner warmth. The interaction between husband familism values and ACV as related to perceived relationship satisfaction was marginally significant (p = .06) and followed the same pattern as that of perceived partner warmth. Husband traditional gender role values were not related to positive relationship quality suggesting that husbands’ gender role values do not influence positive marital quality. Greater husband English use was related to higher relationship satisfaction, which supports evidence from the dyadic results in accentuating the importance of husband English use.

Table 4.

Results from Individual Regression Models: Husband and Wife Positive Relationship Quality Regressed on Cultural Predictors and Covariates

Predictor	Husband Report				Wife Report
	Warmth		Relationship Satisfaction		Warmth		Relationship Satisfaction
	b (SE)	95% CI	b (SE)	95% CI	b (SE)	95% CI	b (SE)	95% CI
English Use	0.25 (.06)	[.13, .36]***	0.14 (.04)	[.05, .23]**	0.16 (.08)	[.01, .31]*	0.20 (.06)	[.09, .33]**
Spanish Use	0.22 (.05)	[.12, .32]***	0.01 (.05)	[−.08, .10]	0.20 (.05)	[.10, .29]***	0.09 (.04)	[−.01, .17]
English x Spanish	−0.14 (.06)	[−.26, −.03]*	0.03 (.07)	[−.11, .17]	0.03 (.06)	[−.09, .14]	−0.06 (.05)	[−.16, .04]
ACV	−0.16 (.07)	[−.30, −.02]*	−0.11 (.07)	[−.23, .02]	−0.07 (.08)	[−.24, .09]	−0.07 (.07)	[−.21, .07]
Familism	0.51 (.11)	[.30, .73]***	0.34 (.10)	[.15, .54]*	0.49 (.12)	[.25, .73]***	0.38 (.11)	[.18, .59]***
ACV x Familism	0.50 (.18)	[.13, .84]**	0.29 (.15)	[−.01, .58]^	−0.27 (.22)	[−.69, .16]	−0.29 (.19)	[−.66, .08]
ACV	0.02 (.08)	[−.14, .17]	−0.02 (.07)	[−.14, .11]	0.08 (.09)	[−.09, .25]	0.06 (.07)	[−.08, .20]
TGR	−0.04 (.05)	[−.14, −.06]	0.00 (.05)	[−.10, .10]	0.06 (.06)	[−.05, .17]	0.02 (.05)	[−.07, .11]
ACV x TGR	0.08 (.08)	[−.07, .24]	0.06 (.07)	[−.08, .20]	−0.22 (.10)	[−.42, −.02]*	−0.11 (.09)	[−.28, .07]

Note: Covariates for each model [family income, mean age, education (years), and wife and husband proportion of life in U.S.] are not shown in the table due to space constraints. ACV = American cultural values; TGR = Traditional Gender Roles.

^***p < .001;

^**p < .01;

^*p < .05;

^{^}p < .08.

Figure 3.

Interaction between Husband American Values and Familism Values Predicting Husband Perception of Wife Warmth.

Wives’ Cultural Orientation

Wife Spanish use was positively associated with her report of partner warmth and relationship satisfaction. In addition, wife English use was positively related to partner warmth but not relationship satisfaction. In terms of interactions, we found that wife traditional gender role values moderated the association between ACV and her report of partner warmth (p = .04). Probing the interaction suggested that wives’ ACV and report of partner warmth were not associated at high, medium or low values of traditional gender roles, and only evidenced a positive association (b = .31, p = .05) when she was at least 1.5 SD below the mean in traditional gender roles values (see Figure 4). This indicated that wives’ decreased endorsement of ACV is associated with decreased perception of partner warmth only for wives’ with very low endorsement of traditional gender role values. On the other hand, wife familism values did not moderate the association between ACV and wife relationship satisfaction.

Figure 4.

Interaction between Wife American Values and Traditional Gender Role Values Predicting Wife Report of Husband Warmth.

Discussion

The present study provides an initial empirical foundation for investigating dyadic similarity and discrepancy of spouse cultural orientation in relation to positive marital relationship qualities among Mexican-origin couples. Results from our dyadic models suggest that wives’ greater American Cultural Values (ACV) were associated with her increased report of positive marital quality when her spouse is also high in ACV. Likewise, wives’ greater Spanish use was associated with husbands’ greater perception of wife warmth and wives’ greater relationship satisfaction when her husband was also more fluent in Spanish. This may provide partial support for the acculturation gap hypothesis in the marital dyad in that cultural differences between partners in the marital relationship are associated with less positive marital quality when one partner strongly endorses a cultural dimension and the other does not. This may also signify that cultural similarity is important if ACV or Spanish use is strongly endorsed within the dyad, while similarity is less important for couples that weakly endorse that particular domain. Similar to the negative effects of acculturative family distancing on parent-child relationships (Hwang & Wood, 2009), spouse cultural differences may be related to a breakdown in communication and a sense of distance between married partners. Cultural similarity between spouses may promote feelings of love, trust, and marital friendship, and may facilitate immigrant couples’ ability to successfully navigate relationship demands in the U.S. receiving context.

The current study also supplemented prior research by examining interactions between individual spouses’ acculturation and enculturation levels in relation to positive marital quality. Results demonstrated that endorsement of traditional Mexican cultural values are important for husbands’ and wives’ perception of positive marital qualities, although in the individual models familism and traditional gender role values operated differently for husbands and wives. For husbands low in familism values, there was a sharp decrease in relationship quality as American cultural values increased, and husbands high in familism values reported higher levels of positive marital quality regardless of American cultural values endorsement. This provides additional evidence that familism may facilitate positive family relations, and point to detrimental effects of increasing American cultural values when not balanced with traditional familism values.

For wives, there was a positive association between American cultural values and perceived husband warmth only at very low (i.e., less than −1.5 SDs below the mean) levels of traditional gender role values. This suggests that wives with lower levels of American values are more likely to perceive their husbands as less warm when they (i.e., wives) do not evidence traditional gender role values. These results may imply that, among wives who are less acculturated in terms of American cultural values, traditional gender role values may facilitate a more favorable perception of their partner’s behavior in the context of more traditional partner sex roles. These findings provide additional evidence that unique cultural domains may differentially influence relationship quality across husbands and wives (e.g., Parke et al., 2004), which warrants additional investigation of gender-moderating effects such as gender role expectations and contextual demands (e.g., employment).

Our findings also suggested that increased English use is associated with higher perceived partner warmth and relationship satisfaction. This is somewhat surprising given previous evidence that increased linguistic acculturation is linked with disrupted marital dynamics (e.g., Negy & Snyder, 1997; Parke et al., 2004). These findings may reflect a strength of the current study compared to prior research, in that we modeled unique associations of English use controlling for Spanish language use. Husbands’ greater English language use may reflect greater acculturation, which may help husbands to navigate demands in the community in the marital relationship. On the other hand, husband’s report of wife warmth was lowest for husbands who endorsed low levels of both English and Spanish use (perhaps those considered marginalized in Berry’s acculturation styles). These men may have the most difficulty navigating contextual demands (e.g., occupational and in the broader community), placing added strain on the marital relationship. Interestingly, wife English use and Spanish use were both influential for her report of positive marital quality, which accentuates unique benefits of fluency in each language. Perhaps Mexican-origin wives fluent in both languages are more adept at navigating environmental demands and simultaneously maintaining heritage language within the family, which then enhances the quality of the marital relationship.

We found an interesting pattern of associations for cultural predictors based on reporter. In most cases, the dyad member’s report of their language use and values predicted their own report of relationship quality, but not that of their spouse. This pattern of associations may mean that individual changes in language use and cultural values are linked with changes in the meaning of particular relationship behaviors for that individual; people may view their relationship differently, and have different expectations for relationship behaviors at different levels of acculturation and enculturation. Future research should address unique reporter effects to understand how cultural orientation may influence the perception of relationship qualities.

Overall, the current results accentuate the importance of assessing bidimensional cultural orientation (acculturation end enculturation), across multiple cultural domains, including cultural practices and values. Our findings indicated that spouses’ cultural orientation may dynamically influence perception of positive marital qualities. Spouse fluency in English and Spanish both appear to be important factors associated with marital quality. Spouse American cultural value similarity appears to be related to higher positive marital quality, however, marital quality may be negatively affected by American value acculturation for husbands endorse lower levels of familism. Furthermore, our results suggested that traditional gender roles generally do not appear to play a notably influential role as related to positive marital quality, although there was some evidence that wives’ traditional gender roles may be associated with greater partner warmth for wives that indicate low endorsement of American cultural values.

Although the current study focused on language use, a possible limitation is that we also did not also address language of survey administration (English vs. Spanish), which has been shown to influence survey responses (Luna, Ringberg, & Peracchio, 2008). In addition, our measure of cultural values relied solely on the Mexican American Cultural Values Scale (Knight et al., 2010), which may have limitations such as the high correlation between Mexican and American subscales and overlap between familism items (e.g., “it is important for family members to show their love and affection to one another.”) and our measure of relationship warmth. Other limitations include the reliance on cross-sectional data, and limited generalizability given our sample was comprised only of married couples. Furthermore, the sample was comprised largely of first generation immigrants and did not include a large group of 2nd or 3rd generation couples that might possibly present with greater cultural discrepancy. Our findings are most relevant to first-generation Mexican-origin population in the U.S. (40% = 1^st generation; Grieco, 2010), which may be most likely to experience struggles with rapidly shifting cultural values associated with immigration. Future research should investigate whether differences or similarities in values or language use may be associated with spouse acculturative stress, which has also been implicated as a risk factor for disrupted marital processes (e.g., Leidy, Parke, Cladis, Coltrane, & Duffy 2012).