Cultural Values, Social Status, and Chinese American Immigrant Parents’ Emotional Expressivity

Abstract

Cultural values and social status are two salient factors in the psychological experiences of immigrant families, and both have been associated with immigrant parents’ patterns of emotional expression in previous studies. The present study examined how endorsement of cultural values (collectivism and conformity) and social status were uniquely associated with immigrant parents’ emotional expressivity in the family. First-generation Chinese American immigrant parents (N = 239, 80% mothers; M = 41.31 years old) of elementary-age children reported on their endorsement of values of collectivism and conformity, their patterns of emotional expressivity in the family context, and their family income and education levels. Path analyses indicated unique positive associations between family income and all domains of parents’ emotional expressivity and negative associations between family income and parents’ endorsement of collectivism and conformity. Parents’ endorsement of collectivism was negatively associated with negative-dominant expressivity. We discuss implications of our findings for theories of culture and emotion, as well as for future intersectional approaches with Asian American populations.

Parents’ emotional processes play a central role in family functioning, development, and well-being (Eisenberg, Cumberland, & Spinrad, 1998; Morris, Silk, Steinberg, Meyers, & Robinson, 2007). Parents’ patterns of emotional expression are particularly relevant in the context of the immigrant family: they can reflect engagement in their host or heritage cultures (Chen, Zhou, Main, & Lee, 2015; Tao et al., 2012) or serve as a means of transmitting cultural norms and values regarding emotion (Camras, Shuster, & Fraumeni, 2014). Likewise, immigrant parents’ emotion-related processes may be shaped by stressors unique to processes of immigration or acculturation, such as experiences of discrimination, limitations in employment or language, or changes in socioeconomic status over the course of immigration (Beiser, Hou, Hyman, & Tousignant, 2002; Mistry, Biesanz, Chien, Howes, & Benner, 2008; Qin, 2006). Finally, immigrant parents’ emotional processes can shape aspects of children’s socioemotional development, including their self-regulatory capacities (Chen et al., 2015) and their understanding of emotions (Doan & Wang, 2010).

Despite the relevance of parental emotion in the immigrant family context, surprisingly little research has examined factors contributing to individual differences in immigrant parents’ emotional processes (Camras et al., 2014). Research to-date examining emotional processes in immigrant parents has focused primarily on the effects of broad indicators of cultural orientation, such as immigration history or acculturation, with little attention to the role of psychological mechanisms in these processes (Camras et al., 2006; Chen et al., 2015). Though previous investigations have demonstrated how variations in emotional processes are associated with endorsements of cultural values such individualism or collectivism (Matsumoto et al., 2008a; 2008b), few investigations to-date have examined these associations within immigrant parents. Likewise, despite increasing evidence for the effects of social status on individuals’ emotional processes (Destin, Rheinschmidt-Same, & Richeson, 2017; Grossmann & Huynh, 2013; Grossmann & Varnum, 2011; Kraus, Cote, & Keltner, 2010), research examining the associations between cultural values, social status, and emotion has been conducted primarily in separate lines of investigation.

To integrate these lines of research, the present study examined the unique associations between cultural values, social status, and parents’ emotional expressivity in a sample of first-generation Chinese American immigrant parents. As one of the largest, fastest-growing, and most socioeconomically-diverse immigrant groups in the United States (Taylor et al., 2012), Chinese American immigrant families provide an ideal sample in which to examine these associations; moreover, a within-group, rather than cross-cultural approach, can be an effective means of highlighting variability in social status within ethnic groups (Yoshikawa, Mistry, & Wang, 2016). Previous research with Chinese and Chinese American families has identified a number of specific cultural values relevant to emotional expression (Chen, Zhang, Chen, & Li, 2012; Luo, Tamis-Lemonda, & Song, 2013). As Chinese American immigrants vary in the degree to which they are oriented to Chinese and American cultures (Ryan, 2013), they may also vary in their endorsement of these traditional Chinese values regarding emotion. By focusing on emotional expressivity, a construct with documented associations with both cultural values and social status, the present study was able to examine how variations in cultural orientations and social stratus could be associated with individual differences within a single ethnic group.

Cultural Orientations and Emotional Expressivity in Immigrant Parents

Previous research has consistently documented cross-national differences in emotional expressivity, with members of traditionally collectivist cultures emphasizing a greater restraint or control of emotions and members of traditionally individualist cultures emphasizing more open expression of emotion (Markus & Kitayama, 1991; Matsumoto et al., 2008a; Matsumoto et al., 2008b; Oyserman, Coon, & Kemmelmeier, 2002). Applied to the family context, previous investigations have focused primarily on parents’ self-reports of emotional expressivity, their “persistent pattern or style of exhibiting facial, body, vocal, and verbal expressions that are often but not exclusively emotional in nature” in the family context (Halberstadt et al., 1999, p. 110) and have found consistent cross-ethnic differences between families identifying with cultures traditionally emphasizing collectivist values or interdependent views of self, and those emphasizing individualist values or independent views of self (Camras et al., 2006; Camras et al., 2008; McCord & Raval, 2015).

Results from cross-cultural investigations can serve as broad indicators of cultural processes; however, they are limited by the use of ethnic groups as a proxy for culture, as well as by assumptions that between-group differences can be attributed to cultural mechanisms that are not explicitly assessed (Phinney, 1996). Despite increasing attention to the advantages of within-group approaches in research with ethnic groups (Deater-Deckard et al., 2017), relatively few investigations have examined whether individual differences in emotional expression within ethnic groups can be associated with differences in cultural orientation - their affiliations and engagement with the ethnic and/or majority cultures (Tsai, Chentsova-Dutton, & Wong, 2002). Moreover, these investigations have been limited primarily to undergraduate samples and have yielded mixed results. For example, differences in cultural orientation were associated with within-group differences in emotional expression among Hmong American (Tsai, Chentsova-Dutton, Freire-Bebeau, & Przymus, 2002), Asian American (Chentsova-Dutton et al., 2007), and Mexican American college students (Soto, Levenson, & Ebling, 2005); but not among Chinese American college students (Soto et al., 2005). In the limited literature examining emotional expression and cultural orientation in immigrant parents, Mexican American mothers’ orientation to American culture was positively associated with verbal expressions of warmth (Ispa et al., 2004). Similarly, in a previous study with the current sample (Chen, Zhou, Main, & Lee, 2015), immigrant parents’ behavioral engagement in various domains of American culture (e.g., media, language, and social relationships) was positively associated with their emotional expression.

Taken together, the limited research to-date examining emotional expression within samples of immigrant parents suggests positive associations between orientation to American cultural domains and parents’ emotional expression. What is lacking in this research, however, is knowledge of the specific cultural processes through which cultural orientation shapes emotion expression. Specifically, while our previous research (Chen et al., 2015) theorized that engagement in Chinese or American cultural domains would encourage the endorsement of emotion-relevant cultural values, the endorsement of these values was not explicitly assessed. Thus, the present study builds on this previous work by assessing Chinese immigrant parents’ endorsement of emotion-relevant values and examining their associations with emotional expression.

Previous research has identified specific cultural values among Chinese families that have direct relevance to parents’ emotional expressivity. The Confucian value of self-restraint or regulation (yue 约) is viewed as a critical element in allowing one to conform to social norms (li 礼) (Hagen, 2010), and has been applied to controlling one’s emotions in the interests of the group (Chen, Zhang, Chen, & Li, 2012; Luo, Tamis-Lemonda, & Song, 2013). Similarly, the value of harmonious interpersonal relationships (he 和) can be expressed by emphasizing the collective needs of the group, rather than those of the individual (Tamis-Lemonda et al., 2008). A substantial body of research with Chinese and Chinese American immigrant families has demonstrated how these cultural values continue to be reflected in parents’ socialization practices in the family context (see Luo et al., 2013 for a review).

Social Status and Emotional Expression

While research examining relations between culture and emotion has traditionally operationalized culture using national, racial, or ethnic designations, a separate line of theory and investigation has conceptualized social class or social status as a cultural construct and examined its associations with a number of social cognitive processes, including emotional processes central to social interactions (Cohen & Varnum, 2016; Grossmann & Huynh, 2011; Kraus et al., 2011; Kraus et al., 2012). Greater access to resources such as education, income, and occupational prestige is theorized to afford greater freedom in the pursuit of individual goals, which in turn fosters more individualistic orientations to the social context (Markus & Stephens, 2017; Kraus et al., 2012). Within this framework, the reduced power associated with lower social status requires a greater monitoring and inhibition of emotional processes; by contrast, the greater power associated with higher social status affords more individual freedom in emotional processes (Keltner, Gruenfeld, & Anderson, 2003).

To-date, investigations of these theoretical models have examined associations between individuals’ social status and various emotional processes, including empathic accuracy (Kraus et al., 2010; Kraus et al., 2011), emotional attunement (Varnum, Blais, & Brewer, 2015) and pro-social emotion (Stellar et al., 2011; Piff & Moskowitz, 2017). By contrast, relatively fewer studies have directly examined associations between social status and emotional expression. Within this limited body of research, lower social status been used to explain more frequent emotional suppression among ethnic minorities in the United States compared to European Americans (Gross & John, 2003), and greater freedom in the regulation of emotional expression among upper-class, compared to lower-class, individuals (Cote, Gyurak, & Levenson, 2010; Gross & John, 1998). More recently, Park and colleagues (2013) found that higher social status was associated with higher self-reported anger among Japanese adults, suggesting that even within cultures emphasizing collectivist values and interdependent views of the self, individuals in positions of higher social status are afforded the privilege of open emotional expression.

Though they underscore the potential effects of social status on emotional expression, these theoretical frameworks have yet to be examined within the context of the family. As such, it remains unknown whether adults’ experiences of high- or low-power, which are typically conceptualized as taking place in places of employment or other settings outside the home, can also be associated with patterns of emotional expression among family members. Furthermore, within the family context, family stress models (Conger & Conger, 2002; Conger, Conger, & Martin, 2010) present an alternative mechanism for associations between family social status and parent emotion. These models suggest that the stress of economic hardship contributes to parents’ emotional distress, which may in turn be expressed through harsh child-rearing behaviors. As such, while social power frameworks would suggest that lower social status would be associated with less frequent expression of both positive and negative emotions, family stress models would predict that lower social class would be associated with less frequent expression of positive emotion, but more frequent expressions of negative emotion. Given these competing models, and in consideration of the socioeconomic changes and challenges frequently accompanying immigration and acculturation (Berry, 1998), it is critical to understand these processes among immigrant parents.

Cultural Values, Social Status, and Emotional Expression

To our knowledge, no single investigation has specifically examined how both cultural values and social status are uniquely associated with immigrant parents’ emotional expression in the family. An examination of these associations within a single ethnic immigrant group can be informed by theories of intersectionality, which emphasize the heterogeneity of experiences and processes within sociocultural groups and attribute these within-group differences to inequalities of power, resources, and opportunities (Cole, 2009; Mahalingham, Balan, & Haritatos, 2008). Applied to the current study, within-group variations in experiences associated with immigration and acculturation, such as changes in employment, education, and communication, can make both cultural values and social status salient factors in the emotional processes of Asian American parents.

Indeed, sociodemographic research suggests that within-group variations in social status have become increasingly pronounced among Chinese American immigrants over the past two decades, particularly among immigrants from Mainland China, who constitute the vast majority of the foreign-born Chinese population in the United States (Taylor et al., 2012). Rapid economic growth in China has contributed to increasing wealth, asset ownership, and education levels among higher-status immigrants to the United States from Mainland China (Keister & Aronson, 2017). For these Chinese immigrants, their path to the United States represents a “hyper-selection” process (Zhou & Lee, 2017) in that they exceed median levels of socioeconomic status not only in China, but also in the United States. These demographics of contemporary Chinese immigration over the past three decades have contributed to growing intragroup socioeconomic disparities within Chinese American immigrants, and have resulted in contrasting trajectories of social mobility. While “hyper-selected” Chinese American immigrants may find success in the mainstream labor market and reside primarily in mixed-ethnicity suburban neighborhoods, Chinese American immigrants with limited education and English skills may face challenges in income and employment, and as a result, may reside in or near urban ethnic enclaves (Zhou, 2009).

These within-group disparities in social status among Chinese American immigrants provide a unique opportunity to examine the unique associations of cultural values, social status, and parents’ emotional expressivity. Examining these constructs together in a single model not only integrates two independent lines of research, but also advances longstanding theoretical models of culture and emotion. Cross-cultural research has long attributed differences in emotional expression to contrasting cultural values. Integrating both cultural values and socioeconomic status into a single theoretical model and examining their unique contributions to emotional expressivity can refine this assumption and test whether within-group variation in emotional expressivity can be better explained by variations in social status. Furthermore, examining the associations between immigrant parents’ social status and endorsement of cultural values can advance our understanding of how these intersecting aspects of immigrant identity may be complementary or diametrically opposed.

Finally, existing theories of culture and emotion suggest that these effects of cultural values and social status are most apparent in self-reported components of emotional expressivity. As proposed by biocultural theories of emotion (Levenson, Soto, & Pole, 2007), self-reports of emotional experience are most susceptible to cultural influence, and as such are most likely to reflect cultural values and norms. By contrast, other components of emotional expressive behavior (e.g., facial expressions, autonomic system response) are less subject to individuals’ voluntary control, and as a result, are less likely to reflect cultural influence.

The Present Study

The present study had three central hypotheses. First, consistent with previous research demonstrating positive associations between endorsement of collectivist values and emotional suppression (Matsumoto et al., 2008a; Matsumoto et al., 2008b; Oyserman et al., 2002), we expected that Chinese immigrant parents’ endorsement of similar emotion-relevant values (collectivism and conformity) would be negatively associated with their self-reported expressivity in the family context. Second, consistent with previous investigations and theoretical models of social status, power, and emotional expression, we expected indicators of higher social status to be positively associated with parents’ self-reported expressivity. At the same time, in consideration of family stress models, our alternative hypothesis was that higher social status would be associated only with higher positive expressivity, and that lower social status would be associated with higher negative expressivity. Finally, in accordance with theoretical models of higher social status and individual orientations, we expected that indicators of higher social status would be associated with lower endorsement of the emotion-relevant values of collectivism and conformity.

Method

Participants

All data were collected from the second wave of an ongoing longitudinal study of 258 first-generation (i.e., born outside the United States) and second-generation (i.e., born in the United States) Chinese American immigrant children and their parents. Families were eligible for the project if: (a) the child was in first or second grade at the time of screening; (b) the child lived with at least one of her/his biological parents; (c) both biological parents identified as ethnic Chinese; (d) the child was either a first generation (born outside the U.S.) or a second generation (born in the U.S. with at least one foreign-born parent) Chinese American; and (e) the parent and child were able to understand and speak English or Chinese (Mandarin or Cantonese). Eligible families (child and one participating parent from each family) completed the 2.5- hour laboratory assessment. Of this number, 63.6% were recruited through community recruitment fairs, 17.4% through school events, and 19% were recruited through community referrals. Analyses for the present study were conducted with parents who participated in the second wave of the longitudinal study (N = 239; 80% mothers). Parents ranged age from 29.76–56.36 years old (M = 41.31; SD = 5.20 years). Within this sample, each household had an average of 2.07 children (SD = 0.72) and 2.35 adults (SD = 1.04), including the parent assessed in the present study. The average number of household members (i.e., adult and child relatives who lived with the family and had no other home) was 4.42 (SD = 1.26).

Procedure

All research procedures at both time points were approved by the Institutional Review Board at the University of California, Berkeley (CPHS Protocol Number 2010-11-2570; “The Risk and Protective Factors for Mental Health Adjustment in 1st and 2nd generation Chinese American Immigrant Children). As part of the larger study protocol, each parent and his/her child participated in a 2.5-hour laboratory assessment. After obtaining parental consent and child’s assent to participate, two bilingual interviewers led parent and child into separate rooms to administer a series of structured interviews and questionnaires. All the questionnaires and tests were administrated in the preferred language (English, Mandarin, or Cantonese) indicated by the parent at the beginning of the visit. All written measures were available in English, simplified Chinese, or traditional Chinese. For measures not used previously with Chinese-speaking populations, original instructions were forward and back-translated by bilingual research assistants and the lead investigator. The majority of parents (83.8%) completed the questionnaires in Chinese. At the end of the laboratory visit, parents were paid $50 and reimbursed for transportation, and children were given a small prize.

Measures

Emotion-relevant values.

Parents completed the Asian American Values Scale-Multidimensional (AAVS-M, Kim, Li, & Ng, 2005), a 42-item self-report instrument for assessing adherence to different dimensions of Asian values. The present study used two subscales with theoretical relevance to emotional expressivity: collectivism (“The welfare of the group should be put before that of the individual”, “One’s efforts should be directed toward maintaining the well-being of the group first and the individual second”) and conformity (“One should recognize and adhere to the social expectations, norms and practices”, “Conforming to norms provides one with identity”). Parents provided ratings on scales of 1 (“Strongly disagree”) to 7 (“Strongly agree”) for each item. The AAVS-M has been used previously with European American and Asian American samples (Park & Kim, 2008), and has demonstrated good internal consistency (alphas ranging from .79 to .89), and concurrent and discriminant validity. In the present sample, the alphas were .70 for Conformity (7 items) and .76 for Collectivism (6 items). Mean scores for each subscale were used for analysis.

Social status.

Participating parents provided current estimates of their annual household income and indicated their completed years of formal education. A majority of participating parents were either working full-time (57.1%) or part-time (12.9%), or were enrolled as full-time students (3.1%); the remaining parents did not report employment outside the home. Estimated per capita annual income ranged from $1,000-$33,333.33 (M = $11,653.57, SD = $8035.46), and parents’ years of education ranged from 2 years (some elementary school) to 20 years (graduate degree) (M = 13.27 years, SD = 2.87). Though not included in analyses, based on parents’ reports, 60.8% of children were eligible for free or reduced lunch at their schools.

Parent expressivity.

Parents completed a 34-item version of the Self-Expressiveness in the Family Questionnaire (SEFQ; Halberstadt et al., 1995), which asked parents to report the frequency with which they expressed different emotions during situations with family members. Items reflected a) negative-dominant expressivity (10 items, e.g., “Showing contempt for another’s action”); b) negative-submissive expressivity (9 items, e.g., “Going to pieces when tension builds up”), and c) positive expressivity (15 items, e.g., “Praising someone for good work”). Parents provided ratings on scales of 1 (“I rarely express these feelings”) to 9 (“I frequently express these feelings”) for each item. This measure has been used previously in research with both Chinese and Chinese American immigrant parents (Camras et al., 2008; Chen et al., 2015; Chen, Zhou, Valiente, Eisenberg, & Wang, 2011) and has shown good internal consistency and predictive validity). Internal consistency estimates of the SEFQ were .87 for Negative-Dominant Expressivity; .78 for Negative-Submissive Expressivity; and .93 for Positive Expressivity. Mean scores for each subscale were used for analysis.

Results

All variables were screened for normality prior to analyses (Table 1). As previous research has indicated unique associations between income, education and emotional processes (Kraus et al., 2010; 2011), annual per capita income and the participating parent’s years of formal education were retained as separate variables for all analyses. Finally, as gender differences in emotional expression have been demonstrated in research with both American and Chinese samples (Bagozzi, Wong, & Yi, 1999; Chaplin, Cole, & Zahn-Waxler, 2005), parent gender was included as a main variable in all analyses.

Table 1.

Descriptive Statistics of Main Study Variables

	N	Min	Max	Mean	SD	Skew	Kurtosis	Cronbach’s α (items)
Negative Dominant Expressivity	231	1.00	6.00	2.57	1.25	0.78	−0.09	0.87 (10)
Negative Submissive Expressivity	232	1.00	7.67	3.90	1.46	0.97	0.64	0.78 (9)
Positive Expressivity	232	1.00	9.00	5.70	1.86	−0.56	−0.22	0.93 (15)
Value - Conformity	234	2.00	7.00	5.53	0.80	−0.72	1.48	0.70 (7)
Value - Collectivism	228	1.50	7.00	4.71	0.97	−0.04	0.29	0.76 (6)
Participating parent’s years of education	231	2.00	20.00	13.27	2.87	−0.52	1.36
Estimated annual family income	235	5000	>100,000	48,064	39,452	0.61	−1.03
Participating parent’s years in the United States	232	2.00	43.00	14.00	7.90	0.94	0.81

Zero-order Correlations

Zero-order correlations were conducted to identify associations between demographic and key variables. Significant associations were found between parent gender and expressivity, such that fathers reported lower positive and negative expressivity compared to mothers (rs = −.17 and −.13, ps = .011 and .042, respectively). Zero-order correlations (Table 2) between main study variables provided initial support for hypotheses. All domains of parent expressivity were positively associated with both indicators of social status (rs between .22-.33; ps < .01). In partial support of hypotheses, Negative Dominant expressivity was negatively associated with values of conformity and collectivism. As hypothesized, there were negative correlations between the two indicators of social status and endorsement of all cultural values.

Table 2.

Zero-order correlations among main variables

	Negative Dominant Expressivity	Negative Submissive Expressivity	Positive Expressivity	Value -Conformity	Value - Collectivism	Estimated Annual Income	Parent Years of Education
Negative Dominant Expressivity	--	.65***	.26***	−.20**	−.19**	.31***	.30***
Negative Submissive Expressivity		--	.59***	−.10	−.05	.22**	.29***
Positive Expressivity			--	.02	−.05	.24***	.33***
Value -Conformity				--	.29***	−.31***	−.24***
Value - Collectivism					--	−.24***	−.31***
Estimated Annual Income						--	.62***
Parent Years of Education							--

Path Analysis

Path analysis was conducted to test the hypotheses that a) parents’ endorsement of cultural values of collectivism, and conformity would be negatively associated with their emotional expressivity, and b) indicators of social status (parents’ income and years of education) would be positively associated with their emotional expressivity. These associations were tested in a single path model in which endorsement of cultural values and indicators of social class were hypothesized to simultaneously predict the three domains of parental expressivity. Given associations between parent gender and positive and negative submissive expressivity, relations between parent gender and expressivity were controlled. The model was estimated with Mplus 7.3 (Muthén & Muthén, 1998–2014) using full-information maximum likelihood to handle missing data and the maximum likelihood (ML) estimator for adjustment to correct standard errors for non-normality. The raw data were analyzed.

Hu and Bentler (1999) recommended the following cutoffs as criteria for a good overall model fit: ≥ .95 for the comparative fit index (CFI) ≤ .05 for the root mean square error of approximation (RMSEA); and ≤ .08 for the standardized root-mean-square residual (SRMR) Based on these recommendations, the path model (Figure 1) fit the data well, χ² (df = 9; N = 232) = 14.82; p = .10, CFI = .98, RMSEA = .05, SRMR = .03. Consistent with our first set of hypotheses, parents’ endorsement of collectivism was associated with lower negative dominant expressivity. Consistent with our second set of hypotheses, parents’ estimated annual income was positively associated with all domains of expressivity; however, parents’ years of education not significantly associated with parents’ expressivity. Significant associations were also noted between parent gender and expressivity, in that mothers reported higher positive and negative submissive expressivity than fathers. Finally, in partial support of our third hypotheses, higher parental income, but not education, was associated with lower endorsement of both conformity and collectivism.

Figure 1.

The path-analytic model predicting emotional expressivity from cultural values and socioeconomic status. Though not displayed in the figure, parent education was included as a variable in the model. Numbers within parentheses represent standardized path coefficients. Only significant paths are shown. *** p < .001; ** p < .01; * p < .05

Consistent with power-based frameworks of SES and individualism, and emotional expression (Markus & Stephens, 2017; Keltner et al., 2003; Kraus et al., 2012), the individualistic orientations of higher-status parents in our sample may disincline them towards traditional values of collectivism and conformity; and in turn, they may be less likely to model lower emotional expressivity in the family context. Thus, though not specified in our original hypotheses, post-hoc mediation analyses were conducted to test whether endorsement of collectivism mediated the relations between income and negative dominant expressivity.. Indirect associations were tested using the bias-corrected bootstrap confidence interval (CI) approach (MacKinnon, Lockwood, & Williams, 2004); however, the indirect association was not significant (95% CI [.000, −.003]).

Discussion

Variations in emotional expression across ethnic groups have long been attributed to differences in emotion-relevant values. Findings from the present study suggest that these variations may also be observed within ethnic groups as a function of social status. Our results are consistent with recent conceptualizations of social status as a cultural identity (Destin et al., 2017; Grossmann & Huynh, 2013), and further affirm the value of intersectional approaches in examining within-group variations in immigrant identity (Mahalingham et al., 2008). Furthermore, by integrating measures of emotion-relevant cultural values and social status into a single model, our findings provide an argument for advancing theories of culture and emotion beyond simple attributions of individualism and collectivism – namely, by considering ways in which experiences of social status may also contribute to intra-group variations in emotional expressivity. Indeed, in examining potential influences on immigrant parents’ emotion, the unique associations of family income with all domains of emotional expressivity suggest that parents’ social status may play an even larger role than the endorsement of cultural values.

Although the associations between social status and cultural values and emotional expression have been observed among general adult populations (Kraus et al., 2012; Matsumoto et al., 2008a), the present study focused specifically on parents’ emotional expressivity within the context of the immigrant family. By doing so, results from the present study provide new directions for future research on the socioemotional development of immigrant families and processes of emotion socialization within the family. Family stress theories (Conger et al., 2010) suggest that lower socioeconomic status would be associated with parents’ increased expression of negative emotion and decreased expression of positive emotion; however, these associations were not observed in the present sample. Rather, our results provided support for theories attributing increased expression of both positive and negative emotion to the power and individualistic focus afforded by higher social status (Markus & Stephens, 2017; Keltner et al., 2003; Kraus et al., 2012). As such, future research can identify experiences, policies, or social structures that contribute to immigrant parents’ positions of higher or lower social status; and in turn, examine how these positions are associated with emotional processes in the family context.

The lack of support for family stress theories in the current data does not necessarily discount effects of socioeconomic stress on parental emotion; rather, higher socioeconomic stress may be reflected more consistently in observed, rather than self-reported expressions of parents’ negative emotion. Furthermore, our hypothesis regarding socioeconomic stress was based on the assumption that parents’ lower standing on objective measures of social status (i.e., educational attainment and income) would be subject to greater socioeconomic stress; however, parents’ experience of socioeconomic stress was not directly assessed. Indeed, a now well-established body of research indicates that, compared to objective indicators of SES, subjective or relative social standing may be a unique and stronger predictor of individual well-being, (Goodman et al., 2001; Operario, Adler, & Williams, 2004; Quon & McGrath, 2014). Thus, the experience of socioeconomic stress, and the resulting expression of negative emotion the family, may be more frequent among parents perceiving lower subjective social standing, rather than those lower on objective measures of SES.

The present study also focused on parents’ endorsement of emotion-relevant values that reflect traditional Confucian principles and which continue to inform current socialization practices among Chinese and Chinese American parents (Luo et al., 2013). To our knowledge, this is the first study to demonstrate that endorsement of these traditional cultural values is lower among higher-income Chinese American immigrant parents. Though parents’ emotional expressivity has been well-established as a central mechanism of emotion socialization in the family context (Eisenberg et al., 1998; Camras et al., 2014), examination of these processes within immigrant families has focused primarily on traditional indicators of acculturation and cultural orientation. Findings from the present study suggest that, in addition to these traditional indicators, constructs of social status should also be integrated in future models of emotion socialization in immigrant families.

Though these results provided overall support for our hypotheses, there were two main exceptions to the expected pattern. First, parents’ endorsement of conformity was positively associated with their positive expressivity. As operationalized in the present study, conformity represents an adherence to social expectations, norms, and practices. For participating parents who had spent a greater percentage of their life in the United States, it is possible that the mainstream culture of the United States served as the reference point for these societal norms and expectations. Previous research has indicated higher levels of parental positive emotion among American compared to Chinese parents (Camras et al., 2006; Ng, Pomerantz, & Lam, 2007), as well as higher self-reported positive emotion among Chinese American parents with higher orientation to American culture (Chen et al., 2015). Thus, it is likely that parents conforming to these American norms of parental emotion would report more frequent expressions of positive emotions in the family context.

Second, while associations between income and parental expressivity were consistent with previous theory and research on social status and emotional expression (Keltner et al., 2003; Park et al., 2013), our second indicator of socioeconomic status, parents’ level of education, was negatively associated with negative dominant expressivity, and was not significantly associated with either positive or negative submissive expressivity. On one level, these findings seem to contradict theoretical models attributing higher emotional expressivity to higher social power (Keltner et al., 2003). As such, one interpretation of our results is that for Asian American immigrants in the United States, higher levels of education do not confer the same social power as monetary wealth, and conversely, result in more frequent suppression of negative dominant emotions. Alternatively, it is likely that parents with higher levels of education, particularly those who completed their highest level of education in the United States, may be more aware of increasingly negative attitudes in the United States toward parents’ use of corporate punishment, harsh criticism, and similar expressions of negative dominant emotion. Awareness of these societal expectations may result in parents displaying negative dominant emotions less frequently in the family context, or at the very least, reporting that they occur less frequently. More broadly, the contrasting relations between parents’ emotional expressivity and two frequently-aggregated indices of socioeconomic status (i.e., income and education level) suggests that these constructs should be analyzed separately in future investigations of socioeconomic status and emotion.

Limitations

Two major limitations to the study merit mention. First, our conceptual framework is based on a causal relation between immigrant parents’ endorsement of cultural values, social status, and their emotional expressivity in the family. Though our results suggest that parents’ income and collectivist values may shape their patterns of emotional expressivity, any interpretations of directionality are limited by the study’s cross-sectional design. In addition to longitudinal investigations modeling changes in these variables over time, future research can also integrate cultural priming paradigms (see Oyserman & Lee, 2008, for a review) to test whether primes of collectivism can induce similar variations in self-reported emotion.

Second, we have argued that self-reported measures of emotional expression are most likely to reflect cultural influence, and were thus uniquely suited for the aims of our investigation. However, the study’s reliance on parent-reported measures for all key variables also makes it more susceptible to single-reporter bias. Indeed, the conclusions of the study could be strengthened by incorporating objective or corroborating data regarding income or education, or behavioral measures of conformity or collectivism.

Conclusions and Future Directions

Immigrants’ unique positions on dimensions of cultural orientation and social status may foster identification with multiple social categories, with unique effects on individual processes (Mahalingham et al., 2008; Yoshikawa et al., 2016). Indeed, results from the present study suggest that social status is a particularly salient aspect of immigrant identity that is associated with parents’ patterns of emotional expression in the family. By examining these processes in the context of the immigrant family, findings from the present study also affirm theories suggesting that the family context may be a point of origin for cross-cultural differences and similarities in emotional expression (Chasiotis, 2011). While the present study did not assess children’s own emotional expression, previous cross-cultural investigations (Camras et al., 2006), in addition to a wealth of developmental research, indicate that parents’ patterns of emotional expression serve as models for those of their children (Eisenberg et al., 1998). If culture is indeed “a unique meaning and information system, shared by a group and transmitted across generations” (Matsumoto and Juang, 2008, p. 15), the study of emotion in the immigrant family provides a unique window through which to observe this process of intergenerational transmission.