Intercultural romantic relationship quality: What is the role of accommodation?

The puzzle presented by intercultural romantic relationships

Romantic relationship quality, comprised of commitment and satisfaction, is an important factor for the well-being and stability of relationships. Partner similarity, or congruency in how two people view the world, is one of the most robust factors relating to romantic relationship quality and other romantic relationship outcomes. The literature on attraction and romantic relationships finds that similarity is positively related to attraction and satisfaction but negatively related to relationship dissolution (Davis, 1981; Morry, 2005). Similarity is theorized to reduce the uncertainty that individuals might otherwise experience when interacting with their partner (Parks & Adelman, 1983). Similarity also facilitates social interactions in ways that may allow couples to enjoy spending time with one another to a greater degree since they share similar desires and interests (Burleson et al., 1994). For these reasons, cultural difference may be among the challenges that intercultural couples have to face and may play a role in the mixed findings in the existing literature on intercultural romantic relationship quality.

One pattern in the literature indicates that romantic relationship outcomes are poorer for intercultural couples ¹ : romantic relationship quality in intercultural couples is lower (Hohmann-Marriott & Amato, 2008) and the break-up or divorce rate in nationally representative samples of intercultural couples is higher (Bramlett & Mosher, 2002; Zhang & Van Hook, 2009) when compared to intracultural couples. For example, being in an intercultural relationship has also been found to be predictive of a break-up one year later in a mid-Atlantic U.S. college student sample (Reiter & Gee, 2008) and divorce 10 years later in a nationally representative sample (Bratter & King, 2008). Yet, a second pattern indicates that romantic relationship outcomes are better for intercultural couples. At least two studies have found that interethnic couples reported similar levels of romantic relationship satisfaction when compared to intraethnic couples (Negy & Snyder, 2000; Shibazaki & Brennan, 1998), and romantic relationship satisfaction has also been found to be higher for interracial couples compared to intraracial couples (Troy et al., 2006).

The variation in the literature suggests that managing cultural difference is challenging but it may or may not impact romantic relationship quality for intercultural couples. The reasons for the mixed pattern of findings are likely complex but it is important to understand the nuances in intercultural romantic relationship experiences. One possibility is that some individuals may navigate the cultural differences of their relationships in ways that more readily facilitate higher quality romantic relationships. Accommodation is one such psychological process that is known to be beneficial for relationships and may be distinctly relevant for individuals in intercultural romantic relationships.

Accommodation

Relationship scholars have studied accommodation as a positive process that is important for relationship stability and maintenance (Rusbult et al., 1991). Accommodation is conceptualized as varying along two dimensions – the first dimension addresses the extent to which a response is constructive versus destructive. Responses that are more constructive (e.g., voicing displeasure with a behavior) are high in accommodation while responses that are less constructive – destructive (e.g., walking away in the middle of a discussion) – are low in accommodation. To successfully engage in accommodation, a partner must suppress their natural inclination to respond destructively and instead make a conscious effort to respond in a constructive, highly accommodative way. This ability to respond constructively and inhibit impulses to respond destructively is known as the accommodative dilemma, a term coined by Gaines et al. (1997), which highlights the effort that is required of individuals to respond to relationship problems in ways that help preserve the relationship.

The current study

We used an online cross-sectional self-report research design to answer our research questions. This allowed us to facilitate recruitment of a sample of individuals who were in an intercultural romantic relationship that might otherwise be challenging to recruit. Given the mixed literature, we first examined whether romantic relationship commitment and satisfaction were similar or different between the two relationship types (intercultural vs. intracultural relationships). Next, we tested whether accommodation related to romantic relationship commitment and satisfaction differently between the groups. All measures were obtained with reference to the participating partner’s actions and perceptions within their relationship.

Participants and procedure

Undergraduates at a large research university on the West Coast of the United States participated in a larger study on romantic relationships. Participants (N = 615) enrolled in the online study using the university’s social science human subjects research pool through the SONA system and were compensated for their study participation with extra credit that they could apply to a participating course of their choice. Upon enrolling in the study, participants were automatically directed to an online consent form. The sole inclusion criterion was that the participant be at least 18 years of age. Participants were directed to the online survey if they fit this criterion and consented to be in the study. Relationship and demographic questions were asked at the end of the questionnaire to ensure that self-identification did not influence responses to other study measures. All measures and procedures were approved by the University Institutional Review Board.

The final sample for the current study consisted of participants who reported being in a romantic relationship for at least 3-months (n = 343). The sample was ethnically diverse with a majority self-reporting having a Mexican (n = 75) followed by Chinese (n = 51) background (see Table 1 for full ethnic break-down). Participants self-reported whether they were in an intercultural (n = 174) or intracultural (n = 161) romantic relationship (see measures section for the exact definition). Participants ranged in winsorized-age from 18 to 29 years old (M = 21.17, SD = 2.71, Median = 21), and identified their sexual orientation as heterosexual (n = 291), bisexual (n = 24), gay/lesbian (n = 11), asexual (n = 2), and other (n = 5). Three participants reported that they preferred not to state their sexual orientation. Participants also identified their gender as women (n = 215), men (n = 117), non-binary or gender non-conforming (n = 3), and one participant reported that they preferred not to state their gender.

OPEN IN VIEWER

Table 1.

Descriptive statistics for key study variables.

Measures	Intercultural	Intracultural
Missing = 8	n (%) 174 (51.0)	n (%) 161 (47.0)
Race/Ethnicity
Southeast Asian	32 (18.40)	15 (9.30)
Mexican	29 (16.70)	46 (28.60)
Mixed-Race	28 (16.10)	16 (9.90)
Chinese	20 (11.50)	31 (19.30)
White	18 (10.30)	24 (14.90)
Middle Eastern	11 (6.30)	3 (1.90)
Other Latino	8 (4.60)	4 (2.50)
Korean	7 (4.00)	8 (5.00)
Pacific Islander	6 (3.40)	2 (1.20)
Indian	5 (2.90)	4 (2.50)
Black	3 (1.70)	4 (2.50)
Native American	2 (1.10)	0 (.00)
Japanese	2 (1.10)	2 (1.20)

	M (SD)	M (SD)
Age	20.86 (2.15)	21.25 (2.25)
Romantic Relationship Quality
Romantic Relationship Commitment*	6.32 (1.06)	6.05 (1.10)
Romantic Relationship Satisfaction*	6.04 (1.09)	5.77 (1.20)
Accommodation	5.26 (.95)	5.13 (.89)

Measures

Data analytic plan

First, we examined the descriptive statistics and bivariate correlations for demographic and key study variables. Next, hierarchical linear regressions were used to test hypotheses. These analyses were run using the PROCESS Version 3 macro model 1 in SPSS (Hayes, 2018). PROCESS calculated the association between x (predictor) and y (outcome) at each value of the categorical moderator resulting in coefficients that represented the main effects collapsing across the other main effects and an interaction term that related to the dependent variable over and above the main effects and statistical covariates. Accommodation was added as a predictor and romantic relationship commitment and satisfaction were entered as outcome variables resulting in two separate models. Relationship type was entered into each model as a moderator. We then added covariates to the base models in a sequential testing order to determine how robust our models were at predicting romantic relationship quality over and above specific demographic factors of interest. These covariates were age, gender, ethnicity, and relationship length. If the addition of a covariate changed the pattern of results, then that covariate was removed and the next was added. Gender and ethnicity were entered into each model as dummy coded variables with women and Asian as the reference groups as they were the largest groups in their respective variable. All hypotheses were pre-registered on the Open Science Framework (OSF).

Hypothesis testing

Predicting commitment

Model 1: No covariates

The association of accommodation with romantic relationship commitment was moderated by relationship type (b = −.26, p = .03; see Table 3). As Figure 1 shows, a simple slopes test revealed that the positive association of accommodation with commitment was weaker for those in an intercultural romantic relationship (b = .35, p < .001) than those in an intracultural romantic relationship (b = .60, p < .001). For both groups, having a higher endorsement of accommodation was associated with also having a higher endorsement of romantic relationship commitment, but the strength of the association was weaker for those in an intercultural romantic relationship than those in an intracultural romantic relationship.

OPEN IN VIEWER

Table 3.

Regression table predicting relationship commitment and satisfaction from accommodation, relationship type, and the interaction between the two without any covariates.

Step	Model 1 (n = 334)			Model 6 (n = 334)
	Commitment			Satisfaction
	b	SE	95% CI	b	SE	95% CI
R 2	.18***			.11***
Constant	2.96***	.46	2.06, 3.86	3.32***	.50	2.33, 4.31
Accommodation	.60***	.09	.43, .77	.48***	.10	.29, .67
Relationship type	1.55*	.62	.32, 2.77	1.20	.69	−.15, 2.55
Δ R 2	.01			.03
Interaction	−.26*	.12	−.49, −.02	−.19	.13	−.44, .07

OPEN IN VIEWER

Figure 1.

Interaction between accommodation and relationship type predicting relationship commitment (Base Model).

Model 2: Age covariate

After adjusting for participant winsorized-age, the same association and moderation held (b = −.26, p = .03; see Table 4

OPEN IN VIEWER

Table 4.

Regression table predicting commitment from accommodation, relationship type, and the interaction between the two with demographic covariates.

Step	Model 2 (n = 334)			Model 3 (n = 334)			Model 4 (n = 329)			Model 5 (n = 329)
	b	SE	95% CI	b	SE	95% CI	b	SE	95% CI	b	SE	95% CI
R 2	.19***			.20***			.21***			.27***
Constant	2.22***	.62	1.00, 3.44	2.23***	.62	1.01, 3.46	2.35*	.64	1.10, 3.61	2.30***	.60	1.13, 3.47
Accommodation	.60***	.09	.43, .77	.60***	.09	.43, .77	.58***	.09	.40, .75	.56***	.08	.40, .72
Relationship type	1.56*	.62	.34, 2.78	1.53*	.62	.31, 2.75	1.39*	.63	.16, 2.62	1.31*	.59	.14. 2.47
Δ R 2	.01			.01			.01			.01
Interaction	−.26*	.12	−.49, −.02	−.25*	.12	−.48, −.02	−.22	.12	−.46, .01	−.20	.11	−.42, .02
Age	.04	.02	−.00, −.08	.04	.02	−.00, .08	.03	.02	−.01, .07	.01	.02	−.03, .05
Men	—	—	—	−.21	.11	−.43, .02	−.19	.12	−.41, .04	−.12	.11	−.33, .10
Other gender	—	—	—	−.25	.50	−1.24, .74	−.29	.51	−1.29, .71	.07	.49	−.88, 1.03
Latino	—	—	—	—	—	—	.20	.14	−.08, .46	—	—	—
White	—	—	—	—	—	—	.29	.16	−.03, .61	—	—	—
Other ethnicity	—	—	—	—	--	—	.07	.16	−.25, .38	—	—	—
Relationship length	—	—	—	—	—	—	—	—	—	.20***	.04	.13, .27

*p < .05, **p < .01, ***p < .001.

); the association of accommodation and romantic relationship commitment was weaker for those in an intercultural romantic relationship (b = .34, p < .001) than those in an intracultural romantic relationship (b = .60, p < .001).

Model 3: Age and gender covariates

After adjusting for participant winsorized-age and gender, the same association and moderation held (b = −.26, p = .04; see Table 4); the association of accommodation and romantic relationship commitment was weaker for those in an intercultural romantic relationship (b = .34, p < .001) than those in an intracultural romantic relationship (b = .60, p < .001).

Model 4: Age, gender, and ethnicity covariates

After adjusting for participant winsorized-age, gender, and ethnicity, the association of accommodation with romantic relationship commitment was no longer moderated by relationship type (b = −.22, p = .06; see Table 4). Follow up tests revealed that the moderation held when including the dummy coded variable for Asian participants, Latino participants, and Other participants in the model (ps < .05) but not when including the dummy coded variable for European participants in the model (p = .05). To further understand these effects, we explored the moderation analyses among the non-European participants only and then among the European participants only. The interaction was marginally significant among the non-European participants (p = .07) but was largely not significant among the European participants (p = .48).

Model 5: Age, gender, and relationship length covariates

After adjusting for participant winsorized-age, gender, and relationship length, the association of accommodation with romantic relationship commitment was no longer moderated by relationship type (b = −.20, p = .08; see Table 4).

Romantic relationship quality variation between relationship types

We observed that romantic relationship commitment and satisfaction were moderately correlated with accommodation in the intercultural (r ranged from .25** to .31**) and intracultural (r ranged from .36** to .49**) samples. We also observed higher romantic relationship quality in individuals in an intercultural romantic relationship. This latter observation is contrary to most findings in the literature. One possible explanation for this finding is that people willing to expand their dating pool to outgroup members may be more likely to find a compatible partner from that wider dating pool. This would be consistent with the similarity-attraction hypothesis that indicates that perceived similarity is a more important factor for romantic relationship quality than objective similarity (Montoya et al., 2008).

Another explanation is that self-selection processes may be at play for individuals in an intercultural romantic relationship. Individuals in an intercultural romantic relationship may have personalities that are more open to experience and more accepting of difference. This factor may both draw individuals into an intercultural romantic relationship and prove helpful for navigating difference. This reasoning is consistent with studies that have found that the personality factor of openness-to-experience relates to romantic relationship satisfaction and longevity (e.g., Gattis et al., 2004); perhaps this personality factor is more important for or more prevalent among intercultural couples. We should also note that our sample was drawn from a university that is exceptionally diverse and may represent a group that is particularly open to cultural differences. If so, our sample may be self-selecting into intercultural romantic relationships and may be reasonably well prepared to manage the challenges of these romantic relationships. Indeed, it may be that individuals willing to date outgroup members may be those who have previously been exposed to diverse others and thus may have better experiences with new individuals even if they are an out-group member (Page-Gould et al., 2010).

Third, the cultural make-up of the couple may create specific experiences that are related to romantic relationship quality. Zhang and Van Hook (2009) found that couples who were made up of a Black partner and a White partner experienced relationship dissolution more frequently than couples made up of a Hispanic partner and a White partner. It could be that couples with cultural compositions that experience higher societal disapproval or have experienced historical turmoil, as is true for Black and White couples, may have more challenges to overcome. Relationship challenges of this nature may be most prevalent in regions where intercultural couple data are collected from participants who are surrounded by disapproving societies and families. Given that these data were collected from a diverse, liberal region in the United States, it is plausible that the current study included a sample of participants who were not experiencing severe amounts of disapproval related to their intercultural romantic relationship. If this is the case, it could be that the current sample included specific intercultural couples who were more likely to experience better outcomes. This latter possibility is not something the current study could assess but it should be a consideration in future studies on this topic.

Accommodation and relationship type on romantic relationship quality

Strengths & limitations

The literature on relationship processes is robust but the study of individuals in an intercultural romantic relationship remains rare. A strength of the current work is the examination of people in this kind of romantic relationship. Participants in the current study were allowed to self-identify their relationship type. This approach allowed for the inclusion of individuals that other forms of recruitment of this kind of couple might have missed such as those who perceive their partner to be culturally different, but the source of difference is not captured by U.S. census ethinc/race categories. The definition of an intercultural romantic relationship that participants were provided also included ethnic, racial, and other cultural variation that participants could use in their self-identification. Broad census ethnic/race categories may not be the only aspect that determines cultural variation and the current study aimed to capture the nuances that may arise from different aspects of culture (e.g., generation status, socio-economic status, language spoken at home, religious identity, etcetera). Lastly, the current study of a diverse sample of racial/ethnic groups provides the possibility that our sample was comprised of various intercultural couples.

The current study also has limitations. We collected self-report data that could be further enhanced with the use of behavioral or qualitative measures. The cross-sectional nature of this study also precludes causal interpretations of the findings. It also precludes our ability to determine directionality between the associations; it could be that higher romantic relationship commitment and satisfaction predict higher accommodation. The college student sample also included participants who were only in short-term relationships. Although our sample was ethnically diverse, it was likely not diverse in other characteristics that are important for psychological science. For this reason, the extent to which these findings may or may not be generalizable to gender identities, socioeconomic statuses, relationship statuses, physical and mental abilities, sexual orientations, and relationship lengths not represented in our sample is unclear. The current study was also unable to verify with certainty that participants were in an intercultural romantic relationship because we are unsure of the factors that lead individuals to identify as being in an intercultural romantic relationship. It could have been that they reported a different ethnicity from their partner, but it could have been a number of other factors as well. Therefore, we offer the following as considerations in future studies on this important topic.

Future directions

Future studies can build on the initial steps of the present work by collecting data from both members of the couple to further enrich our understanding of romantic relationship functioning using dyadic analyses and by decomposing the intercultural make-up of couples. In fact, Craig-Henderson and colleagues (2015) offer suggestions on how to classify couples based on their ethnicity and race self-classifications. Dyadic data would also allow researchers to examine the self-report data from each partner to determine which demographic factors are leading participants to self-identify as belonging to an intercultural romantic relationship. Future work should also examine whether results differ based on the demographic make-up of the couple and whether participants with specific cultural backgrounds experience a stronger association between accommodation and romantic relationship quality. For example, our sample may have included couples with specific cultural configurations (e.g., a Latino partner with a White partner, or a Black partner with an Asian partner) that influenced our results. Particular combinations of intercultural couples may be more vulnerable to experiencing outside stressors and societal disapproval. This may be the case, for example, for intercultural couples made up of partners other than cisgender heterosexual individuals. Relationship science will benefit from future studies that examine different types of couples to understand difference more broadly.

In addition, it is known that some intercultural couples may experience disapproval from friends, family, and/or society that contributes to relationship difficulties. For this reason, understanding the role of disapproval or “outside stressors” is likely to expand on our understanding of possible reasons for differences in romantic relationship quality observed between intercultural and intracultural romantic relationships. One possibility that emerged from this study is that accommodation and romantic relationship commitment may be influenced by the length of a relationship. Relationship status (e.g., dating casually, dating seriously, dating exclusively, engaged, married, domestic partnership, etcetera) may also be relevant to the findings on relationship length. One’s relationship status may be associated with romantic relationship commitment. If couples have established higher commitment, accommodation may unfold differently as well. Future longitudinal studies are needed to better understand this intriguing possibility of changes over time in accommodation as well as directionality between accommodation and romantic relationship quality. Lastly, the extent to which differences in rates of transgressions across different levels of relationship distress may be relevant but has yet to be examined in the literature.