Core Self-Evaluation, Emotional Reactivity to Interpersonal Conflict, and Subjective Well-Being: The Moderating Role of Horizontal Collectivism

Abstract

This study aims to examine a moderated mediation model wherein core self-evaluation (CSE) and horizontal collectivism (HC) interact to predict negative emotional reactivity to interpersonal conflict, and thus HC moderates the indirect effects of CSE on subjective well-being (SWB) through emotional reactivity. A short-term prospective study was conducted with 257 South Korean university students. Participants completed measures of CSE and HC, and then reported their experiences with interpersonal conflict and SWB in an online survey about two weeks later. We found that, among low HC participants, after controlling for the importance of conflict issues, individual differences in CSE predicted emotional reactivity to interpersonal conflict, and emotional reactivity was in turn negatively associated with SWB. By contrast, among high-HC participants, emotional reactivity to interpersonal conflict did not differ as a function of CSE, and the indirect effects of CSE on SWB were not significant. These findings highlight the importance of high CSE and HC values in fostering emerging adults’ resilience against the detrimental effects of interpersonal conflict in sustaining individuals’ SWB. Implications for conflict management interventions are discussed.

1. Introduction

Over the past two decades or so, research in the field of positive psychology has accumulated considerable evidence emphasizing the importance of building individuals’ well-being in various important life domains, such as mental health, longevity, and work-related outcomes [1]. Subjective well-being (SWB), defined as “people’s evaluations of their lives, which can be judgments such as life satisfaction, evaluations based on feelings, including moods and emotions” [2] (p. 1), is not only an ultimate life goal, but it also serves as a critical resource in promoting psychological wellness and sustainable living through upward spirals (i.e., happiness brings about more happiness) [3,4,5]. Thus, researchers have sought to identify individual factors that enhance SWB and to promote resilience to stressful events that threaten SWB [6]. Following this line of approach, in this study, we identify young adults’ personal resources that guard against the detrimental effects of interpersonal conflict on SWB.

Previous research has identified interpersonal conflict as one of the most common daily stressors, with the largest effects on SWB, including mental health and physical symptoms [7,8,9,10,11]. For example, in an experience sampling study, Ilies et al. [12] reported that the occurrence of interpersonal conflict was responsible for changes in employees’ negative affect. Coiro et al. [13] reported that undergraduate students’ experience of interpersonal stress was significantly associated with greater depression, anxiety, and somatization symptoms. However, relatively little research has examined the individual characteristics that contribute to resilience to interpersonal stressors, despite the fact that the negative effects of interpersonal stressors may vary depending on individual characteristics such as agreeableness and emotional regulation [12,14]. As interpersonal stress has a more detrimental impact on emerging adults’ symptoms of depression than non-interpersonal stress [15], it is important to identify emerging adults’ personal factors that protect them from the detrimental effects of interpersonal conflict and ultimately promote SWB.

Thus, this study explores the interplay between core self-evaluation (CSE) and horizontal collectivism (HC) to elucidate the individual characteristics that shape emotional distress in interpersonal conflict, which can have detrimental effects on SWB. Specifically, by drawing on previous research showing the beneficial effects of high CSE on stressors in various life domains [16,17,18,19], we first suggest that high CSE may reduce young adults’ emotional reactivity to interpersonal conflict, thus contributing to sustaining SWB. More importantly, given the recent findings about high-HC individuals’ resilience to workplace incivility [20], we propose that high HC values may offset the vulnerability of individuals with low CSE in conflict, thus attenuating the effects of CSE on SWB through emotional reactivity. Therefore, this study aims to test a moderated mediation model in which the indirect effects of CSE on SWB through emotional reactivity to interpersonal conflict change depending on the levels of HC. The findings are expected to identify individual characteristics that serve as resilient factors that help emerging adults sustain SWB while protecting from the detrimental effects of interpersonal conflict. We also seek to elucidate the functions of CSE and HC in interpersonal conflict in association with SWB.

2. Theoretical Background and Hypotheses

2.1. Core Self-Evaluation and Emotional Reactivity to Interpersonal Conflict

CSE refers to the fundamental appraisal that individuals hold about their own self-worth, competence, and capability as a person. It has been conceptualized as a higher-order construct containing four well-established constructs: self-esteem, generalized self-efficacy, locus of control, and emotional stability [21,22]. Much research has found that individual differences in CSE have strong predictive power on SWB. This is because individuals with high CSE tend to have consistently favorable views about themselves across situations, which can be particularly effective in stress processes to reduce the experience of psychological strain and maintain SWB [16]. For example, Lin et al. [17] found that CSE levels of medical students predicted SWB during their stressful clerkship trainings. Lim and Tai [18] demonstrated the buffering effects that high CSE has on the effects of family incivility on psychological distress, such that high levels of family incivility were related to increased psychological stress among individuals with low CSE, whereas this relationship was not significant among those with high CSE.

Although previous studies have detailed the role of CSE in SWB and adjustments, relatively little research has tested the effects of CSE on the emotional responses to interpersonal conflict that reduces SWB. Thus, this study investigates whether individual differences in CSE predict emotional reactivity to interpersonal conflict, and in turn, SWB. We suggest that high-CSE individuals may experience less emotional reactivity to conflict for three reasons. Firstly, high-CSE individuals tend to worry less about rejection, because their higher self-esteem and emotional stability allow them to generally feel accepted and valued by others [23], and they are thus less likely to appraise interpersonal conflict situations as threatening. By contrast, low CSE individuals are more likely to perceive conflict as threatening because their lower self-esteem and greater emotional instability can make them hypersensitive to rejection [24]. Secondly, as perceived controllability can influence the way one responds to a stressful event [25], high-CSE individuals’ perception of high controllability over their environments may prevent them from interpreting conflict situations as threatening, thus resulting in lower emotional distress than low-CSE individuals, who likely feel overwhelmed in such situations due to a perception of low controllability [17]. Thirdly, high-CSE individuals are more likely to be better at controlling their gut-level impulses in conflict situations, thus preventing downward spirals of destructive reciprocity that can exacerbate emotional reactivity [26].

Overall, compared to low-CSE individuals, high-CSE individuals are less likely to perceive interpersonal conflict as a threat; thus, it seems likely that higher-CSE individuals are more likely to experience less negative reactivity to interpersonal conflict. Furthermore, as emotional reactivity to interpersonal conflict strongly affects SWB [12,13], we also expect that individual differences in CSE would predict SWB through emotional reactivity. This leads to the following hypotheses. Moreover, given the previously discussed findings on the relationship between CSE and SWB, here, we suspect that CSE may affect SWB through other mechanisms as well. Thus, we only propose partial mediation.

Hypothesis 1.

CSE will be negatively associated with emotional reactivity to interpersonal conflict.

Hypothesis 2.

CSE will be positively associated with SWB through partial mediation of emotional reactivity to interpersonal conflict.

2.2. The Moderating Role of Horizontal Collectivism

Individualism–collectivism was initially discussed in this field as a cultural value dimension to capture significant distinctions between cultures [27]; however, it has also been used to capture individual differences within cultures [28,29]. Collectivism places greater importance on maintaining harmonious relationships with others, and in-group interests over self-interest, whereas individualism focuses on personal autonomy and gives priority to one’s own interests over those of the in-group [27,28,30]. Triandis and Gelfand [31] further distinguished between individual-level collectivism and individualism using a four-category typology: horizontal–vertical individualism and horizontal–vertical collectivism. Horizontal individualism focuses on egalitarianism and high self-reliance from groups; vertical individualism values distinction and winning competitions over others; horizontal collectivism involves the inclination for sociability, interdependence, and maintaining harmonious relationships with others; and vertical collectivism emphasizes norm conformity to social obligations and sacrificing for one’s group.

There has recently been growing interest in the adaptive functioning of collectivism. For example, Lu et al. [32] reported that collectivism positively predicted mask wearing to effectively protect against COVID-19 within the United States, and they also found higher rates of mask usage in more collectivistic countries. Meanwhile, in an experiment with German college students, Pfundmair et al. [33] demonstrated that high horizontal collectivism facilitated the mitigating effects of oxytocin on negative responses to social exclusion. Furthermore, Welbourne et al. [18] reported that high HC values attenuated the effects of workplace incivility on burnout, while VC did not show such effects. Given this background, we focus here on high HC as a key individual characteristic that may offset the vulnerability of low-CSE individuals to interpersonal conflict, as HC may be a moderator that changes the effects of CSE on emotional reactivity to conflict.

We specifically suggest that, as conflict is inevitable in social interaction, high-HC individuals are more likely to be equipped for conflict management strategies that allow them to be resilient against daily interpersonal conflict, because their happiness derives from harmonious relationships and is dependent on the happiness of others around them [27]. Previous research has suggested that, in conflict resolution, high-HC participants tend to prefer using the integrating or obliging styles which consider the other party’s needs [34]. Similarly, Aslan et al. [35] found that HC was positively associated with the use of the obliging style of conflict resolution, which focuses more on the others’ interests than one’s own self-interests. These strategies are likely to effectively attenuate the tension that arises in conflict situations. In addition, as interpersonal conflicts are not typically one-sided, high-HC individuals that are more concerned with social harmony are likely to engage in pro-relationship responses to transgressions by attempting to consider the perpetrator’s perspective as well as contextual factors [36]. Along similar lines, Kurniati et al. [37] recently reported that the motivation for a harmonious relationship is positively associated with intentions to act friendlily toward an offender and to inhibit the desire for revenge for transgressions. Thus, high-HC individuals are more likely to make pro-relationship responses toward the offender and are, therefore, less emotionally reactive to conflict. As such, pro-relationship responses are associated with the experience of lower levels of negative emotions [38].

Considering these characteristics of high-HC individuals, we suggest that high HC values might serve as a personal resource that attenuates the emotional reactivity of individuals with low CSE in interpersonal conflict, thus reducing the relationship between CSE and emotional reactivity. As a result, a high HC value may weaken the indirect effects of CSE on SWB via emotional reactivity. However, among low-HC individuals who are less resilient in interpersonal conflict, their emotional reactivity to interpersonal conflict is more likely to be shaped by their level of CSE, and it is therefore more likely to be associated with SWB. Thus, we have the following hypotheses.

Hypothesis 3.

HC will moderate the relationship between CSE and emotional reactivity to interpersonal conflict, such that the relationship will be weaker among high-HC individuals than low-HC individuals.

Hypothesis 4.

HC will moderate the indirect effects of CSE on SWB through emotional reactivity to interpersonal conflict, such that the indirect effects will be weaker among high-HC individuals than low-HC individuals.

3. Methods

3.1. Participants and Procedures

For the sample in this study, 259 students were recruited from the undergraduate psychology participant pool of a large university in Seoul, South Korea. Study participants received course credits. (These data were collected as part of a project that produced a previous publication in a Korean journal [39]; however, the current research questions and variables do not overlap with this published article. 13 foreign students participated in this study but were not included in this sample because of their language difficulties in responding to the questionnaire. In total, 257 students completed two surveys about two weeks apart. We excluded the data from one participant for careless responses, as well as from six participants identified as outliers after applying the criteria of univariate outliers (±3 SD) and multivariate outliers (Mahalanobis test). Thus, the final data set ultimately consisted of 250 participants (56.4% female). Participants completed the first survey in a laboratory session to measure CSE and HC. Then, about two weeks later, they completed the second survey online to collect data about interpersonal conflict events and SWB. Using G*Power 3.1 [40], we performed a sensitivity power analysis. The results revealed that our sample size of 250 participants attained a power of 0.82. All English-based measures were translated into Korean following Takeuchi et al.’s survey translation procedure [41].

3.2. Measures

Core self-evaluation was assessed using Judge et al.’s scale [23] on a 5-point scale (1 = strongly disagree, 5 = strongly agree). An example item is “I am confident I get the success I deserve in life” (Cronbach’s alpha = 0.83).

Horizontal collectivism was measured using Singelis et al.’s scale [30] on a 5-point scale (1 = strongly disagree, 5 = strongly agree). An example item is “My happiness depends very much on the happiness of those around me” (Cronbach’s alpha = 0.73).

Emotional reactivity was measured after participants were first asked to recall a negative interpersonal conflict event that they had experienced with a family member, friend, or anyone else over the past two weeks. After they described that event, participants then responded to items measuring their emotional reactivity to that conflict event. Following Almeida et al. [42], we measured emotional reactivity using seven items from the Affects Balance Scale [43] (e.g., afraid, helpless, irritable). Two items (irritable and angry) were eliminated from the scale because of low factor loadings (<0.20). Participants were asked to indicate their feelings at the time when the conflict event occurred by rating the degree to which they felt each of the emotion items on a 5-point scale (1 = not at all, 5 = a lot) (Cronbach’s alpha = 0.84).

Subjective well-being was measured according to the method described by Sheldon and Niemiec [44]. We used six items from the Positive Affect Negative Affect Scale (PANAS) to measure each positive (e.g., excited) and negative affect (e.g., nervous), as well as three items to measure life satisfaction (e.g., “I am satisfied with my life”). All of these were answered on a 5-point scale (1 = strongly disagree, 5 = strongly agree). An SWB score was calculated by summing the positive affect and life satisfaction scores, before then subtracting the negative affect score (Cronbach’s alpha = 0.77).

To control for potential effects of the characteristics of the particular type of interpersonal conflict, we measured the importance of the specific issue in the interpersonal conflict (sample item: “the central issue in this conflict is important to me”) as well as the relationship closeness of the other party in the conflict (sample item: “I have a close relationship with this person”) using two items for each variable. We also measured vertical collectivism, horizontal individualism, and vertical individualism using the scale of Singelis et al. [29] to control for their possible effects. The results of multiple regression analyses indicated that only issue importance had a significant relationship with the outcome variables. Thus, we included it as a control variable in the following analyses.

3.3. Data Analytic Strategy

The hypotheses were tested using regression-based path analyses in Mplus 6.0 [45,46]. To capture the significance of the indirect effects of CSE on SWB with HC as a moderator, we used a bias-corrected bootstrap confidence interval (20,000 iterations) [47], and we reported the unstandardized coefficients and confidence intervals for our effect sizes, as suggested by Pek and Flora [48]. Finally, for the interaction term, CSE and HC were mean-centered and multiplied [49].

4. Results

Before testing our hypotheses, we conducted confirmatory factor analysis (CFA) to assess the discriminant validity among CSE, HC, emotional reactivity, and SWB. The results showed that the four-factor model had reasonable overall model fit, χ²(498) = 719.51, CFI = 0.94, TLC = 0.93, RMSEA = 0.04, SRMR = 0.06. Furthermore, all factor loadings were significant. To test whether the measures of the variables were distinct, we compared this four-factor model with a single-factor model with all of the items loading onto one latent variable, χ²(515) = 2414.72, CFI = 0.46, TLC = 0.41, RMSEA = 0.12, SRMR = 0.12. The chi-squared difference test showed that the four-factor model provided a significantly better fit to the data than the single-factor model, Δχ²(17) = 1695.21, p < 0.01, supporting the discriminant validity among the four measures.

Table 1. Descriptive statistics and intercorrelations among variables.

Variables	M	SD	1	2	3	4	5	6
1. Gender	0.44	0.50
2. Issue importance	3.29	1.16	−0.05
3. Relationship closeness	4.07	1.00	−0.07	0.21 **
4. Core self-valuation	3.46	0.53	0.20 **	−0.06	0.02
5. Horizontal collectivism	3.78	0.52	−0.01	0.13 *	0.18 **	0.21 **
6. Emotional reactivity	2.52	0.98	−0.02	0.41 **	0.13 *	−0.15 *	0.01
7. Subjective well-being	3.58	1.84	−0.01	−0.11	−0.01	0.40 **	0.11	−0.31 **

* p < 0.05. ** p < 0.01.

presents the descriptive statistics and correlations for the study variables. To test Hypotheses 1 and 2, we conducted regression-based mediation analyses in Mplus, following Hayes and Rockwood’s suggestion [46]. This model had good fit to the data, χ²(1) = 0.13, CFI = 1.00, TLC = 1.04, RMSEA = 0.00, SRMR = 0.01. The results showed that CSE was negatively associated with emotional reactivity after controlling for issue importance (b = −0.24, SE = 0.12, p < 0.05, 95%CI [−0.46, −0.004]) and that emotional reactivity was negatively related with SWB (b = −0.48, SE = 0.13, p < 0.01, 95%CI [−0.72, −0.23]). CSE had direct relationships with SWB (b = 1.24, SE = 0.21, p < 0.01, 95%CI [0.84, 1.65]). The indirect effect of CSE on SWB was also significant, indicating that CSE showed a significant positive relationship with SWB via emotional reactivity (b = 0.11, SE = 0.06, 95%CI [0.01, 0.27]). Thus, Hypotheses 1 and 2 are both supported.

Following Hayes’ procedure [45], we conducted a moderated mediation path analysis in Mplus to test Hypotheses 3 and 4. This model had good fit to the data, χ²(3) = 1.20, CFI = 1.00, TLC = 1.05, RMSEA = 0.00, SRMR = 0.01. The path analysis results are displayed in Figure 1. As shown in Figure 1, the interaction between CSE and HC on emotional reactivity was significant (b = 0.53, SE = 0.21, p < 0.05, 95%CI [0.11, 0.95]). Following the procedure suggested by Aiken et al. [50], we plotted the interaction between CSE and HC to further interpret the nature of the interaction. As shown in Figure 2, the simple slope test revealed that CSE significantly predicted emotional reactivity among low HC participants (b = −0.72, SE = 0.21, 95%CI [−1.14, −0.31]), but not among high-HC participants (b = 0.35, SE = 0.27, 95%CI [−0.19, 0.89]). We also found that the indirect effect of CSE on SWB through emotional reactivity changed depending on participant’s level of HC (moderated mediation index = −0.26, SE = 0.13, 95%CI [−0.57, −0.06]). Specifically, the conditional indirect effect of CSE on SWB through emotional reactivity was significant among low-HC participants (b = 0.35, SE = 0.14, 95%CI [0.13, 0.68]), but not among high-HC participants (b = −0.17, SE = 0.15, 95%CI [−0.53, 0.06]. Thus, Hypotheses 3 and 4 are both supported.

5. Discussion

The current study shows that CSE predicted SWB through emotional reactivity to interpersonal conflict, and that the indirect effects of CSE on SWB through emotional reactivity changed depending on the level of HC. Specifically, individual differences in CSE predicted emotional reactivity and, in turn, SWB among low-HC participants but not among high-HC participants, as low- and high-CSE participants with high HC had similar low levels of emotional reactivity.

These findings have several important implications. Firstly, this study extends research on individual characteristics that may promote resilience against the detrimental effects of interpersonal conflict on SWB. Previous research has identified personal variables (e.g., reappraisal tendency) that may buffer the effects of interpersonal stress on well-being [14]. However, to our knowledge, no studies have examined the roles of CSE and HC to explore why some people sustain SWB despite experiencing interpersonal conflict. By demonstrating the relationships of high CSE and high HC with emotional reactivity to interpersonal conflict, which can impair SWB, this study elucidates the personal resources that promote resilience to interpersonal conflict.

Secondly, this study has important implications for collectivism literature. Although maintaining harmonious relationships with others is known to be critical for collectivists’ life satisfaction [27], little research has examined the role of collectivism in interpersonal conflict, which negatively affects SWB. Thus, this study advances the existing knowledge of collectivism by demonstrating the adaptive role of HC values, which may offset the vulnerabilities of having low CSE in interpersonal conflict.

Thirdly, these findings have implications for CSE literature. The present findings corroborate and extend the findings of previous studies of CSE. Consistent with previous findings in organizational research [19], our findings show a positive relationship between CSE and SWB. Furthermore, in line with earlier research emphasizing the importance of CSE in explaining how individuals react to workplace stressors [17], our results show that individual differences in CSE significantly predict emotional reactivity to interpersonal conflict. More importantly, these findings demonstrate that emotional reactivity partially mediates the link between CSE and SWB. As researchers have yet to explore the role of CSE in relation to the effects of interpersonal conflict on SWB, these findings extend previous studies on CSE by elucidating the mediational processes underlying how CSE contributes to SWB.

Finally, from a primary prevention perspective, our findings have important practical implications for training programs aimed to develop young adults’ resilience in interpersonal conflict. As interpersonal conflict is a major source of work stress [8], young adults need to be equipped with personal resources that allow them to effectively manage interpersonal conflict, so that they can have successful organizational socialization and adjustment processes and further pursue sustainable development when they begin a full-time job in an organization [50]. Our findings imply that the vulnerability of low-CSE individuals to interpersonal conflict might be offset by high HC, although previous research has suggested that low-CSE individuals are more vulnerable to various stressors [16]. Prior research indicates that high-HC individuals have a more flexible approach to conflict situations that involves considering more contextual factors as well as the other person’s perspective [35]. They may, therefore, respond to ordinary conflict in daily life less defensively and with more understanding. This tendency may help foster strong social support networks, which may in turn help them to adaptively cope with experiences of ostracism or workplace incivility [20,32]. As there are interventions that can change one’s mindset regarding stress [51], future research should seek to identify constructive conflict mindsets and response strategies that determine the resilience of high-HC individuals in interpersonal conflict, with the ultimate goal of developing a conflict mindset change training program that promotes resilience to interpersonal conflict. Further, our findings imply that when selecting employees, organizations may consider a selection system to identify and attract individuals with high CSE and high HC to increase the likelihood of employing adaptive employees who are more competent in managing interpersonal stressors.

This study has some limitations. Firstly, although we measured the outcome variables about two weeks after measuring the antecedent variables, the short-term prospective design did not allow us to draw definitive causal inferences. Future research could use experimental research designs to vigorously test the causal relationships among the variables in our model. Secondly, this study did not include agreeableness, interpersonal sensitivity, and emotion regulation, which may moderate the link between interpersonal stressors and SWB [12,14,52]. Future studies could include these variables to test whether they affect the strength of the indirect effects of CSE on SWB by moderating the link between emotional reactivity and SWB. Thirdly, the sample here consists of South Korean college students, which limits the generalizability of our findings to other cultures. The collectivistic cultural context of South Korea emphasizes the importance of maintaining high levels of closeness with in-group members. Thus, it might make HC values more salient as core cultural values, which might increase the commitment of high-HC individuals to their values and, in turn, facilitate the adaptive effects of high HC. Future research could replicate this study with US participants and South Korean participants to investigate the generalizability of our findings and to determine whether there are moderating effects of cultural differences.