Abstract
The purpose of this study was to test the moderating effects of adult attachment in the association between extroversion and subjective well-being in Korea. A total of 334 students recruited from four universities in Korea completed self-reported questionnaires regarding extroversion, attachment, and subjective well-being. The results showed that the innate extroverted orientation of university students affected positive emotions only when the level of attachment avoidance was low. In contrast, although university students had an extroverted orientation, their extroversion did not have a significant influence on positive emotions when their level of attachment avoidance was high. The findings of this study suggest that high levels of attachment avoidance may prevent for extroverts from experiencing positive emotions.
Introduction
Subjective Well-Being (SWB) is not only the purpose of individual life, but also one of important indicators of measuring quality of life (Diener & Suh, 1997). The OECD National Happiness Index and the United Nations’ World Happiness Report are representative surveys of personal life evaluations across countries (Tofallis, 2020). Unfortunately, Korea's OECD National Happiness Index ranked 33st of 40 countries as of 2020 (OECD, 2020).The United Nations’ World Happiness Report also maintains a below-average level of achievement for Korea every year (Helliwell, Huang, & Wang, 2017). These indicators suggest that the level of SWB in Korean needs to be improved.
The level of SWB among Korean university students is also a serious issue. According to the National Statistical Office in Korea, 20- to 24-year-old university students are more likely to feel stress in their general life, including their school and work life, than teenagers aged 13 to 19 (Statistics Korea, 2017). The SWB of Korean university students is threatened by high tuition fees, interpersonal problems, and employment stress against the highest unemployment rate in Korea (Joo & Kim, 2016). The level of SWB among university students who will lead Korean society in the future is an important issue for Korean society and families, extending beyond the problems of any single individual (S. J. Lee & Ahn, 2005).
Researchers in positive psychology have examined the factors that affect an individual’s SWB. Among a variety of factors, there is a consensus among many researchers that innate traits and interpersonal relationships are crucial factors of SWB (Costa & McCrae, 1980; Diener et al., 1984, 1999; Lyubomirsky, 2001). In particular, the “big five” dimensions of personality traits which is the prominent model of personality structure among researchers in this field (See Digman, 1990, for a review) have been found to be closely associated with SWB (Anglim et al., 2020; Costa & McCrae, 1980; DeNeve & Cooper, 1998; Diener et al., 1999; Hentschel et al., 2017; Steel et al., 2008; D. C. Zhang & Renshaw, 2020). Big five dimensions of personality refer to personality traits such as nervousness, extroversion, openness, friendliness, and integrity (Goldberg, 1981). These five factors were determined to be strongly influenced by genetic factors in several twin studies and found to remain fairly stable in adulthood (McCrae & Costa, 1990, 1994).
Steel et al. (2008) conducted a meta-analysis based on 249 prior studies to identify the relationship between the five factors of personality and SWB. As a result, extraversion and neuroticism were identified as important personality factors predicting SWB. In particular, researchers have shown that neuroticism is closely related to negative emotions, whereas extroversion is associated with positive emotions (Argyle, 2001; Cheng & Furnham, 2001; DeNeve & Cooper, 1998; Lauriola & Iani, 2017; Steel et al., 2008). McCrae and Costa (1990) also suggested that extroversion is particularly associated with positive emotions and SWB due to the temperamental characteristics of sociability, tempo, energy, and social involvement. In a very recent meta-analysis study, Anglim et al. (2020) also found that extroversion was the strongest predictor of well-being based on analysis of HEXACO model. In addition, some Korean studies have highlighted extroversion as a crucial predictor (E. K. Lee & Lee, 2005; Mo et al., 2013). Based on findings on the strong associations between characteristics of extroversion and SWB, this study specifically focuses on extroversion among the big five personality factors in exploring its association with SWB.
Extraversion is the tendency to pursue a relational stimulus that is contingent on interacting with others (Costa & McCrae, 1992). Those with strong extroversion tend to be psychologically oriented toward the external world and are highly sociable. Indeed, the innate traits may provide and lead the certain conditions and environments which affects SWB (Tkach & Lyubomirsky 2006). For example, due to the sociability of extroversion, highly extroverts may be socially engaged with people and feel positive emotions and reported high levels of SWB (Steel et al., 2008). In other words, positive characteristics in the relationships of people with high extroversion play an important role in their SWB.
On the other hand, interpersonal factors were found to be as important as innate traits in the area of positive psychology. Among interpersonal relationships, attachment is the fundamental relational representation that forms the source of human relations (Bowlby, 1982). Attachment theory has typically provided a provocative framework for studying the detrimental effects of early separation or loss of parenting on socioemotional development and psychopathology (Bowlby, 1982; Shaver & Mikulincer, 2014). Attachment researchers suggest that there are two orthogonal dimensions of attachment: anxious and avoidant attachment (Brenna et al., 1998; Mikulincer & Shaver, 2012, 2016).
Two different styles of attachment are associated with qualitatively different dynamics of emotions in close relationships (Cassidy & Kobak, 1988). Specifically, people with high levels of anxious attachment are more likely to have intense close relationships characterized by demanding and controlling behaviors (Mikulincer & Shaver, 2012). They tend to have a feeling of safety and competence when they are close to people (Mikulincer et al., 2003). In contrast, people with high levels of avoidant attachment are more likely to deny attachment needs and emotional involvement with other people by ignoring and avoiding close relationships and their emotions (Mikulincer & Shaver, 2016). Indeed, the two different insecure attachment styles might be risk factors of SWB in reflecting an obsession with close relationships or an avoidance of close relationships. Based on the qualitatively distinct characteristics of anxious and avoidant attachment (Cassidy & Kobak, 1988), these two different dimensions of attachment might provide differential contexts and filters for expectations and beliefs of interpersonal relationships, thus ultimately affecting different emotions and SWB.
Through available, responsive, and consistent interactions with an attachment figure, secure attachment provides positive emotions, such as a sense of security and relief, while decreasing distress and fear (Main et al., 1985). In contrast, insecure attachment offers negative emotions through unavailable and inconsistent interactions with an attachment figure. In this sense, attachment theory proclaims that the process by which human beings develop and organize positive and negative emotions is fundamentally social and relational (Sroufe, 1996; Thompson & Goodvin, 2005).
Many researchers have empirically examined and confirmed an association between attachment and SWB (Girme et al., 2018; Mónaco et al., 2019; Moreira et al., 2020; Shaver & Mikulincer, 2008). In meta-analysis conducted in Korea, attachment was the strongest predictor of SWB among relational variables, such as attachment, social relationships, communication, and parenting attitudes (Kim, 2016). In addition, in the early days, researchers argued that secure attachment, per se, is a crucial predictor of SWB (e.g., La Guardia et al., 2000; Simpson, 1990; Torquati & Raffaelli, 2004). Recently, many researchers have highlighted the psychological mechanisms and contexts of how attachment influences SWB within diverse contexts and samples (Chung & Choi, 2014; Mónaco et al., 2019; Sirois, Millings, & Hirsch, 2016; X. Zhang et al., 2016).
Given that the interactions between innate human temperament and environmental qualities form psychological reactions in humans ( Kwon, 2008) and that both extroversion and attachment are closely associated with SWB, it is necessary to explore the interactive effects of extroversion and attachment on SWB. Although some studies have tried to integrate the effects of personality and attachment on SWB (Demir, 2007; Lu, 1999; Moreira et al., 2020; Wilkinson & Walford, 2001), they did so by examining the contribution of each personality and/or attachment style to SWB. They have not explored the interactive effects of personality and attachment on SWB. Given that the personality trait known as the set point of happiness is hard to change (Lyubomirsky et al., 2005), the moderating role of attachment in the relationship between extroversion and SWB is of particular interest in this study. Thus, the purpose of this study is to investigate the moderating effect of attachment on the relationship between extroversion and SWB among Korean university students.
Methods
Participants
The participants in this study were recruited from four universities located in Daegu and Gyeongbuk Province, South Korea. Convenience sampling was used in this study. The eligibility of criteria included attending university in Korea and having an experience of romantic relationships. We distributed 350 surveys, but a total of 334 questionnaires were used in the analysis, except for 16 questions that were excluded due to missing values or no responses. There were 137 male students (41.0%) and 197 female students (59.0%). There were 72 freshmen (21.6%), 75 sophomores (22.5%), 89 juniors (26.6%) and 98 seniors (29.3%). The average age of university students was 21.11 years old and the standard deviation was 2.20.
Measurements
Extroversion
Extroversion was measured by Goldberg’s (1999) International Personality Item Pool (IPIP) which originally measured 5 elements of personality. The IPIP was translated and validated by Yoo et al. (2004) in Korea. In this paper, only the extroversion scale was selected among the five sub-factors of IPIP. The extroversion scale consists of 10 items which reflect the propensity to pursue human-related stimuli to interact with others (e.g., “I tend to start a conversation first” and “I am the one who leads the mood in the meeting.”). The items are based on a five-point Likert scale that ranges from “not at all” (1 point) to “very much agree” (5 points). The higher scores indicate higher levels of extroversion. The internal consistency coefficient reported in this study is .86.
Adult attachment avoidance and anxiety
Adult attachment was measured by the Experience of Close Relationships-Revised (ECR-R) (Fraley et al., 2000). This scale consists of 18 attachment avoidance items and 18 attachment anxiety items. Attachment avoidance indicates discomfort with and avoidance of close relationship (e.g., “I am nervous when partners get too close to me”). Attachment anxiety indicates the fear of rejection (e.g., “I get angry when other people don’t give me the affection and support that I want to get”). The item is measured by a seven-point Likert scale ranging from “not at all (1 point)” to “very much agree” (7 points). The internal consistency coefficients of attachment avoidance and attachment anxiety in this study were .93 and .91, respectively.
Subjective well-being (SWB)
SWB was measured using the Concise Measure of Subjective Well-Being (COMOSWB) developed and validated by Suh and Koo (2011). The COMOSWB consists of nine items, with each set of three items measuring cognitive satisfaction (personal, relational, collective aspect) and the frequency of positive (joyful, happy, peaceful) and negative (irritated, negative, helpless) emotional experiences. Each item was assessed on a seven-point Likert scale. The internal consistency of life satisfaction, positive emotions, and negative emotions were .74, .83, and .69, respectively.
Covariates
In the preliminary analysis, the results of t-test and ANOVA showed that there were no significant effects of gender and grades on SWB. Thus, gender and grades were not used as covariates in main analysis.
In order to control for other personality, we used neuroticism as a covariate in the main analysis based on the findings that neuroticism was investigated as a strong predictor of SWB (Steel et al., 2008). Neuroticism was measured by IPIP which also examined extroversion in this study. The neuroticism scale consists of 10 items on the 5 point Likert scale. The internal consistency coefficient reported in this study is .89.
Procedure and analysis
This study was approved by the Institutional Review Board of Yeungnam University in May 2017. We randomly sampled universities in South Korea and contacted professors to explain the purpose and method of this study. The participants were university students who listened to the professors’ descriptions of the study and agreed to participate in the research. We distributed the questionnaires from May 2017 to June 2017 to four universities in South Korea. We explained the aim of this study, the protection of confidentiality, and voluntary participation to the participants.
The collected data were analyzed using SPSS 22.0 for reliability, descriptive statistics, and the relationship between study variables. In order to verify the moderating effects of adult attachment on the relationship between extroversion and SWB, we used PROCESS, a SPSS macro program developed by Hayes (2013). PROCESS is a useful statistical method for testing moderating and mediating models (Hayes, 2013). To test two different styles of attachment, we analyzed attachment anxiety and avoidance in separate models. In order to determine the tendency toward a conditional effect, the significance at the mean and ± 1 SD points was verified and the significance region was confirmed by the Johnson-Neyman method.
The objectives and hypothesis
The aim of this study is to investigate the interactive effect of attachment on the relationship between extroversion and SWB among Korean university students. Based on the previous research which attachment avoidance and attachment anxiety are qualitatively different (Cassidy & Kobak, 1988), we hypothesize that attachment avoidance and attachment anxiety would play a different moderating role in the associations between extroversion and SWB.
Results
Descriptive statistics and correlations of study variables
The mean score for extroversion was 3.18 (SD = 0.72). The mean scores for attachment avoidance and attachment anxiety were 3.41 (SD = 0.89) and 3.21 (SD = 1.01), respectively. The mean scores for life satisfaction, positive emotions, and negative emotions were 4.99 (SD = 0.98), 4.68 (SD = 1.16), and 3.95 (SD = 1.22), respectively.
Pearson's correlation coefficient of study variables is presented in Table 1. Based on the analysis, all the variables of this study showed a significant correlation.
Correlations between extroversion, anxious and avoidant attachment, life satisfaction, positive emotions and negative emotions.
*p < .01.
The moderating effect of attachment avoidance on the association between extroversion and subjective well-being
There was no significant moderating effects of attachment avoidance on the associations between extroversion and life satisfaction as well as the associations between extroversion and negative emotions. However, attachment avoidance moderated the relationship between extroversion and positive emotions, B = −.19, p < .05, 95% CI [−.33, −.04] (Table 2), based on 10,000 bias-corrected bootstrap samples. The results thus partially supported the moderating effect of attachment avoidance on the relationship between extroversion and SWB.
Moderating effect of attachment avoidance on the association between extroversion and subjective well-being.
p < .05. **p < .01. ***p < .001.
In order to provide a more specific examination, the conditional effects are examined according to the attachment avoidance level (Table 3). The effect of extroversion on positive emotions in the group with a low level of attachment (-1SD) was .34 (LLCI > 0). Since the group with a low level of attachment avoidance did not contain zero, which meant that there was no statistical significance between the lower limit value (LLCI) and the upper limit value (ULCI) within the 95% confidence interval, the moderating effect of the group with a low level of attachment was significant (see Figure 1). However, the group with a high level of attachment avoidance did contain a “0” between the LLCI and the ULCI, so the moderating effect was not significant These results suggest that extroversion affects positive emotions in a group with low levels of attachment avoidance, whereas the effect of extroversion on positive emotions is not significant in a group with high levels of attachment avoidance. More specifically, we examined the Johnson-Neyman test results to determine the significant areas of conditional effects. As a result, the conditional effect was found to be significant when the attachment avoidance was −.55 or less. This result indicates that when the attachment avoidance level is lower than −.55, the effect of extroversion on positive emotions is significant.
Conditional effect of attachment avoidance between extroversion and positive emotions.
p < .05.

Association between extroversion and positive emotions for the level of attachment avoidance.
The moderating effect of attachment anxiety on the association between extroversion and subjective well-being
The results of PROCESS revealed that there were no moderating effects on the associations between extroversion and any of SWB (Table 4). Specifically, although extroversion and attachment anxiety independently influenced life satisfaction, positive emotions, and negative emotions, the interaction between extroversion and attachment anxiety was not statistically significant, including ‘0’ between the LLCI and the ULCI in the 95% confidence interval.
Moderating effect of attachment anxiety on the association between extroversion and subjective well-being.
p < .05, **p < .01, ***p < .001.
Discussion
Although both extroversion and attachment have been investigated as core predictors of SWB, there has been a lack of research exploring the interactive effects of extroversion and attachment on SWB. The purpose of this study was to examine whether adult attachment has moderating effects on the relationship between extroversion and SWB among Korean university students.
First of all, extroversion showed a positive correlation with life satisfaction and positive emotions, and a negative correlation with negative emotions. The results supported the findings which documented the association between extroversion and SWB (Anglim et al., 2020; Costa & McCrae, 1980; Diener et al, 1999; S. J. Lee & Ahn, 2005; Mo et al., 2013; Schimmack et al., 2002; D. C. Zhang & Renshaw, 2020). The findings of the current study were also consistent with the study conducted by E. K. Lee and Lee (2005), who showed a significant correlation between extroversion and life satisfaction, positive emotions among Korean university students. Based on the results of this study and previous studies, extroversion is closely associated with SWB.
Attachment anxiety and attachment avoidance showed negative correlations with life satisfaction and positive emotions, and positive associations with negative emotions in the current study. The findings were consistent with the significant association reported between secure attachment and SWB (Chung & Choi, 2014; Girme et al., 2018; La Guardia et al., 2000; Mónaco et al., 2019; Moreira et al., 2020;; Torquati & Raffaelli, 2004). The results also supported the finding that the attachment anxiety and attachment avoidance of college students has a significant effect on the cognitive and emotional aspects of SWB (Min & Shin, 2007; Oh & Park, 2011). Taken together, these findings indicate that not only a personality but also attachment are closely associated with SWB. Therefore, it is suggested that both genetic and relational factors should be considered in order to better understand SWB and to prepare better intervention.
The findings from the current study revealed significant interactive effects between attachment avoidance and extroversion on positive emotions among Korean university students. This study found that extroversion positively affected positive emotions, particularly when university students had low levels of attachment avoidance. In contrast, extroversion didn’t have a significant influence on positive emotions when university students had high levels of attachment avoidance. Based on the close associations between extroversion and SWB (Anglim et al., 2020; Costa & McCrae, 1980; Diener et al., 1999), individuals with high extroversion are expected to have high positive emotions. However, even extroverted individuals could not experience positive emotions when they had high levels of attachment avoidance. These findings suggest that high levels of attachment avoidance may prevent for extroverts from experiencing positive emotions. This finding is supported by previous theory and studies that showed attachment avoidance are closely associated with avoidance of social contexts and suppression of emotions (Bowlby, 1980; Cassidy & Kobak, 1988).
In particular, the moderating effects of attachment avoidance on the association between extroversion and positive emotions may be related to the context in which extroverted people experience positive emotions and the dynamics of relationships and emotional strategies of attachment avoidance. Specifically, extroverts respond more strongly to affirmative emotions and are more sensitive to positive rewards (Larsen & Ketelaar, 1991; Lucas et al., 2000). In addition, extroverts experience positive emotions through social interactions (Moskowitz & Cote, 1995). However, people with high attachment avoidance tend to deny intimate relationships and minimize their emotions (Shaver & Mikulincer, 2016). Therefore, even extroverted people may have fewer opportunities to experience positive emotions by avoiding social interactions when they had high levels of attachment avoidance. In addition, high levels of avoidance attachment may restrict the ability to feel or express emotions, thereby depriving extroverted people of the opportunity to experience positive emotions. Although a lot of research has documented extroversion as a crucial indicator of positive emotions (Anglim et al., 2020; Costa & McCrae, 1992; Steel et al., 2008; D. C. Zhang & Renshaw, 2020), the results of this study contribute to the literature in revealing the specific contexts for extraverts that high levels of attachment avoidance may be an obstacle for expressing positive emotions.
On the other hand, the moderating effect of attachment anxiety was not significant in relation to extroversion and any SWB. The main effect of extroversion and attachment anxiety on SWB was significant. These results suggest that, regardless of the level of attachment anxiety, extroversion affected life satisfaction, positive emotions, and negative emotions. This finding may be related to the results of research (J. H. Lee, 2005; Mallinckrodt & Wei, 2005) in which attachment anxiety and attachment avoidance are qualitatively different styles in terms of emotions and interpersonal relationships (Cassidy & Kobak, 1988). Although individuals with high attachment anxiety feel anxious due to a fear of damaging relationships, unlike those with attachment avoidance, they tend to get close to other people, so they maintain and value relationships (Feeney & Noller, 1990). In other words, attachment anxiety does not deprive these individuals of opportunities for social interaction or of feeling and expressing emotions, unlike attachment avoidance. As a result, extroversion and attachment anxiety may independently affect SWB. In other words, regardless of the degree of attachment anxiety, extroversion can have an impact on SWB. The results of this study are noteworthy in showing the different effects of attachment avoidance and attachment anxiety on the association between extroversion and positive emotions.
Limitations and future research directions
The limitations of this study and suggestions for subsequent research are as follows. Attachment avoidance and attachment anxiety were measured by using self-report questionnaires, rather than the Adult Attachment Interview (George et al.,1985/1996; Main et al., 1985). In the subsequent studies, it would be better to use the AAI, which is a more validated and sophisticated tool for measuring attachment avoidance and attachment anxiety. Second, in this study, only extroversion, which has a great impact on SWB, was considered as a personality factor. It is necessary to examine the interactive effects of all five personality factors and attachment on SWB in subsequent studies. Finally, attachment anxiety did not show a moderating effect on the association between extroversion and SWB. However, there is a lack of extent research to interpret and support the results. Thus, research is necessary to explore qualitatively different psychosocial mechanisms of attachment avoidance and attachment anxiety in the context of SWB.
Treatment implications
Given the interactive effects of attachment avoidance on the associations between extroversion and positive emotions, education and counseling programs are effective for students with high attachment avoidance. In the programs, the students could understand how attachment avoidance affects emotions and interpersonal relationships, and to realize that attachment avoidance is depriving them of the opportunity to experience positive emotions. These suggestions are supported by the findings that group counseling programs had positive effects on improving interpersonal relationships for university students with high attachment avoidance in Korea (Chang & Yon, 2009). In addition, specialists should be aware and inform the moderating role of attachment avoidance to general university students and thus help them pay attention to their quality of attachment avoidance to prevent them from feeling positive emotions. Check-ups of attachment avoidance would be helpful to improve SWB for university students. Lastly, it would be useful to clarify how attachment avoidance works in the specific social context.
Conclusion
The findings contribute to the literature of positive psychology by revealing that a high level of attachment avoidance may prevent for extroverts from experiencing positive emotions. The results of the current study contribute to helping university students to feel positive emotions by recognizing the interactive effects of their attachment and extroversion. The current study extends the knowledge in the literature on SWB by examining interactions between innate human temperament and relational quality. More research is needed to explore specific contexts in which attachment avoidance interferes innate extroversion.
Footnotes
Author’s note
This paper is part of the first author’s master’s thesis.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article. This work was supported by 2019 Yeungnam University Research Grant.
