The Effects of Cultural Self-Construal and Self-Disclosure on Attachment and Relationship Satisfaction in SNS

Abstract

The purpose of the research was to determine the effects of cultural self-construal of users in social networking sites (SNS) on the way they perceive self-disclosing activities, attach to brand community, and satisfy with brand relationship. With the growth of social media, it has brought diverse users from different countries in a single platform enabling people from divergent cultures to communicate and be influenced by each other shaping unique self-construal. Hence, we used newly developed cultural self-construal as an antecedent of self-disclosure. Data were collected from 460 female adults. The results showed that the perceptions of self-disclosure are the key antecedents of relationship advancement, which are differently influenced by cultural self-construal. The way users perceive self-disclosing activities positively influenced brand community attachment, and subsequently brand relationship satisfaction. The results provided valuable insights for practitioners by presenting the importance of using brand community in SNS as a long-term branding channel.

Keywords

consumer behavior digital communication Korea marketing satisfaction

With the rapid growth of social media, some fashion companies have started to form brand communities in social networking sites (SNS) to communicate directly with consumers (E-News Today, 2016). However, even though SNS are more useful in managing relationships with consumers (S. Cho & Lee, 2015), most companies use SNS as a simple marketing channel to increase profits by posting advertisements (E-News Today, 2016). The commercialization of communities has caused users to drastically decrease sharing activities of a private nature. In fact, overall sharing decreased 5.5% between mid-2014 and mid-2015 on Facebook (Efrati, 2016). A more severe problem was the decrease of personal postings, which fell 21% during the same period. According to social penetration theory (Altman & Taylor, 1973), self-disclosure, or the sharing of private information, is required to form a relationship among individuals. Hence, the decrease in personal postings may hinder sincere communication and relationship development. It may lead consumers to use communities as a one-time information channel rather than a long-term relational channel. Fewer than 100 members were reported as active users of more than 30% of Facebook’s 1,900 Fan Pages, necessitating a strategic approach for companies’ SNS brand communities (Choi, Lee, & Jang, 2013).

Consumers make different trade-offs between the rewards and costs of self-disclosure, perceiving it as either useful or risky (Lowry, Cao, & Everard, 2011). Thus, the way consumers perceive self-disclosure can play a key role in developing emotional connection and advancing relationships. Moreover, in order to make consumers core fans of SNS brand communities and develop long-term relationships, it is imperative to understand which propensity (i.e., perceiving self-disclosure as either risk or reward) leads them to self-disclose easily.

Generally, collective behaviors observed on SNS can be interpreted as the interactions between users’ cultural inclinations and the inherent characteristics of SNS (Yoo, Kim, & Moon, 2012). However, because users influence and are influenced by others on SNS, their experiences may differ based on their internal values (S. Cho & Han, 2013). Users with various national backgrounds and values actively communicate on SNS in this contemporary society; in other words, SNS users from diverse cultural backgrounds can communicate and influence each other by shaping their unique cultural self-construal differently, even though they belong to the same cultural area (Fiore, 2008). According to Han and Shin (1999), diffusion of the Western way of thinking has gradually increased individualistic characteristics in Eastern countries. In this way, it is difficult to define cultural inclination as a binary concept, such as collectivism versus individualism. Thus, we used Inumiya, Han, Lee, Lee, and Kim’s (2007) cultural self-construal measure, which is able to reflect the changing nature of SNS users since cultural inclinations can be reflected in self-concept.

Also, because the extent of self-disclosure varies depending on the situational context, SNS users’ cultural self-construal may influence their perception of self-disclosure (Myung & Jang, 2012). Furthermore, users’ attitudes toward a specific object or person can be predicted based on the expectation they have toward the results (Ledbetter et al., 2011; Myung & Jang, 2012). Hence, if they have a constructive expectation toward self-disclosure, they could have a positive attitude toward brand communities. Prior researchers found a significant influence of self-disclosure on community attachment and a positive influence of community attachment on brand relationship satisfaction (Jung, 2013; S. Lee, Yang, & Kwon, 2013). However, they paid little attention to the changes in cultural self-construal, even though the internal cultural values of users can be changed through SNS.

The main purpose of this study was to examine the influence of cultural self-construal on individuals’ perception of self-disclosure in SNS. In addition, we aimed to examine the mediating effect of self-disclosure on the relationship between cultural self-construal and brand community attachment, as well as brand relationship satisfaction.

SNS and Social Brand Community

SNS, or web-based interactive services, enable individuals to participate in disclosing and sharing activities, to produce and manage information, and even to form communities based on common interests (Zaglia, 2013). In SNS, users form various types of communities as they can easily exchange their opinions (E. Lee, Kim, & Ahn, 2013). Likewise, some fashion firms form brand communities in SNS, called “social brand communities,” to capture opportunities presented by this new type of online network. According to DMC Media (2015), Fan Pages, a type of social brand community on Facebook, show the highest usage rates (59.8%) in Korea, and nearly 90% of global fashion companies from luxury brands like Burberry to national brands like Beanpole use Fan Pages as a way to build strong relationships with consumers. As firms place their brands into a unique social network setting by presenting them in SNS online communities, consumers are more likely to be active in expressing and disclosing their affection and preferences toward specific brands, which is critical for the development of a relationship between consumers and a firm’s brands (Schau & Gilly, 2003). Individuals have a greater tendency to self-disclose in online venues like SNS (B. Chen & Marcus, 2012). On SNS, one of a variety platforms that enable computer-mediated communications that tend to be social, users can be “noisy, public, and even unmanageable” in comparison with typical consumers, who are more likely to be “silent, isolated, and invisible” (Zaglia, 2013, p. 216). Moreover, communities on SNS allow users to socially identify themselves with brand communities rather than the network (Zaglia, 2013). A community can be viewed as a set of people who share certain distinctive identities (Zhang, Jiang, & Carroll, 2010), and users can determine self-reflected answers to the question of “who am I” from their social interaction experiences. Thus, users’ self-disclosure in social brand communities can be regarded as the behavior for self-expression, and this space can play a role in disclosing one’s preferences or views to others (Posey, Lowry, Roberts, & Ellis, 2010).

Theoretical Framework

Social Penetration Theory

According to social penetration theory (Altman & Taylor, 1973), individuals who initiate interpersonal communication tend to disclose their self differently due to the variations in the degree of intimacy between them. For example, the breadth and depth of personal information that individuals disclose can differ. More importantly, before disclosing their self, they assess the amount and nature of rewards and costs of interpersonal activities (Y. Chen & Nakazawa, 2009).

To be specific, reciprocity, awareness, and intimacy are the key benefits of self-disclosure since the listed benefits can reward people by providing emotional support (comfort) as well as enjoyment with enhancing connectedness and self (Lowry et al., 2011). For instance, people may gain rewards in the form of reciprocity, such as self-disclosure of others in return or by receiving positive responses (Tang & Wang, 2012). On the other hand, people may suffer from costs in the form of increased vulnerability and risks when they disclose too much, resulting in the invasion of privacy (Lowry et al., 2011). According to Lowry, Cao, and Everard (2011), the trade-offs between rewards and costs are “the tension between the need for connection and protection in interpersonal communications and relationships” (p. 168). Those who expect more rewards from interaction may intentionally disclose personal information to others (Y. Chen & Nakazawa, 2009). This enables people to socially approach and penetrate the inner core of personality rather than the superficial layers of an individual, in which relatively visible and public personal information can be retrieved without significant interaction (Altman & Taylor, 1973). However, they may find it very difficult to reach the deeper and intimate layers of an individual, which reflect their true personality, because disclosing the inner core increases risks such as vulnerability and social desirability (Tang & Wang, 2012). The only way to penetrate the inner core is the advancement of relationships, which can take place through gradual self-disclosure. As such, self-disclosure is the key to socially penetrating another’s mind (B. Chen & Marcus, 2012). Although SNS enable people to present the self actively through self-disclosure (Zaglia, 2013), the disclosed information could be publicly shared and recorded on SNS (B. Chen & Marcus, 2012). Hence, the way users assess the rewards and costs of self-disclosure may be contingent upon how they perceive the SNS in terms of the degree to which they are prone to social influence.

Cultural Self-Construal

Cultural self-construal is a self-concept built on culture and is used as a framework to explain a variation in psychological processes and social behaviors in a sociocultural context (Markus & Kitayama, 1991). Earlier researchers suggested that not only individual beliefs but also traditions and culture influence how people interpret their experiences and how they respond to others in a given situation (Krasnova, Veltri, & Gunther, 2012). Likewise, culture has a strong interactive influence on one’s cognitive, emotional, and social functions (S. Kim, Yang, & Kwon, 2013). Cultural dimensions such as collectivism–individualism are strongly relevant to the degree of social influence, which is “the degree to which an individual’s belief, attitude, and behavior are influenced by others in own environment” (Posey et al., 2010, p. 184). Hence, it explains an individual’s inclination toward reciprocity, which is the main benefit of self-disclosure and the key to social penetration (Kankanhalli, Tan, & Wei, 2005). According to Markus and Kitayama (1991), the degree of connectedness–separateness toward social context varies across individuals. Individuals who desire to be separated from the social context try to master their own environment since they focus on the self and tend to be private and internal, whereas those who desire to be connected to social context try to cooperate with the environment because they focus more on the collective and tend to be public (B. Chen & Marcus, 2012). In this respect, it is natural to expect that cultural self-construal may influence the way users perceive self-disclosure and present themselves; the degree of connectedness–separateness influences how they assess rewards and costs (e.g., connection vs. protection) associated with self-disclosure (Y. Chen & Nakazawa, 2009; Krasnova et al., 2012; Myung & Jang, 2012). Prior researchers (B. Chen & Marcus, 2012; Posey et al., 2010) suggested that people who are interdependent are prone to social influence; thus, they perceive benefits of self-disclosure, such as reciprocity, on SNS, which is more collectivistic compared with an in-person environment. However, because of the risk associated with social desirability, these individuals tend to modify and monitor self-disclosure by matching their behavior to the environment to avoid hurting the self. People with independent inclinations are less prone to social influences; thus, they are unlikely to be affected by reciprocity unless self-disclosure would benefit the self in either an online or in-person situation.

Traditionally, cultural self-construal was primarily classified within the dichotomy of independent self and interdependent self (Markus & Kitayama, 1991). However, such a classification may not reflect the changing cultural self-construal since people today tend to have a positive attitude toward self-disclosure and actively express themselves (Schau & Gilly, 2003). In places like SNS where users with various cultural backgrounds communicate with each other, an individual’s cultural self-construal can be influenced by other users who hold different cultural values. Thus, an individual’s degree of connectedness–separateness toward social context may vary even in the same cultural area because there is a complementary relationship rather than a trade-off relationship between connectedness and separateness (S. Lee, 2008). Inumiya et al. (2007) suggested that both the psychological distance established toward the social context (e.g., connectedness–separateness) and the way individuals bring attitudinal differences to the social relationship (e.g., self-esteem) can shape the cultural self-construal. They integrated previous research and developed a new cultural self-construal classification based on criteria known to explain different psychological aspects of people in the same cultural area, which are presented as follows.

Objective self-construal

Individuals with objective self-construal recognize themselves as beings who are connected to social context as well as peripheral beings who embrace and accept the influence of others (Inumiya et al., 2007). They are more likely to perceive reciprocity as the benefit of self-disclosure due to their interdependent inclinations (Posey et al., 2010). They can perceive useful aspects of self-disclosure because they see it as the starting point of reciprocal communication (Kankanhalli et al., 2005). In line with this argument, reciprocity is the primary benefit of self-disclosure for individuals with interdependent self-construal (B. Chen & Marcus, 2012). On the other hand, due to their low level of self-esteem, individuals holding objective self-construal recognize the self as an objective and as peripheral to a primary being (Inumiya et al., 2007). Hence, they are sensitive to others’ appraisal and expectations and are vulnerable to the denial of others, which could hinder self-disclosure (Inumiya et al., 2007). According to Campbell et al. (1996), individuals with low self-concept clarity tend to show a high level of public self-consciousness. Due to the high level of public self-consciousness as well as the low level of self-esteem, they tend to adjust and monitor self-disclosure in order to match their behaviors to the social environment (B. Chen & Marcus, 2012). Yook (2014) suggested that individuals who are highly conscious of others tend to tailor their actions in order to increase similarity with others rather than disclosing their true self. On the other hand, they are able to reconcile rewards and costs by not revealing their true self (B. Chen & Marcus, 2012; Murray et al., 2009). Even though individuals with low self-esteem could acknowledge the risks of self-disclosure, their inherent tendencies may prevent them from perceiving these risks; they may not be in pursuit of maintaining and enhancing self-esteem like individuals with high self-esteem are because they tend to cope with the risks, enabling rewards to override the costs (Crocker & Park, 2004). However, they may still perceive risks associated with disclosing private opinions to others on SNS because they tend to use others’ opinions to evaluate the validity of their own, which makes it difficult for them to develop in-depth relationships (Choi et al., 2013).

Hypothesis 1a: Objective self-construal will positively influence perceived usefulness of self-disclosure.

Hypothesis 1b: Objective self-construal will positively influence perceived risks of self-disclosure.

Subjective self-construal

Individuals who have subjective self-construal recognize themselves as beings who are connected to social context as well as central beings who exert social influence on others (Inumiya et al., 2007). Similar to individuals with objective self-construal, they have interdependent inclinations concerning its conceptual origin (Inumiya et al., 2007). They are also quite prone to social influences and responses such as appraisal and denial. Hence, they can perceive self-disclosure as useful because they desire reciprocity as well (B. Chen & Marcus, 2012; Kankanhalli et al., 2005; Posey et al., 2010). However, since they recognize their self as a primary being due to high self-esteem, unlike individuals with objective self-construal, they comparatively could be more vulnerable to others’ negative responses (Inumiya et al., 2007). Individuals with high self-esteem tend to maintain and enhance their self-esteem as well as achieve self-actualization by getting social approval, which enables them to find desirable internal characteristics of self (Choi et al., 2013; Crocker & Park, 2004). Thus, they may perceive risks associated with self-disclosure to a greater extent because not only do they use others’ opinions to evaluate the validity of their own, but they also may find it harmful to their self if they do not receive positive responses from others (Choi et al., 2013; H. Kim & Han, 2004). To sum it up, individuals who have subjective self-construal may perceive the usefulness of self-disclosure if they expect rewards like positive responses or reciprocal disclosure. At the same time, they can perceive the risks associated with self-disclosure if they do not receive positive responses, since they try to maintain and enhance their self.

Hypothesis 1c: Subjective self-construal will positively influence perceived usefulness of self-disclosure.

Hypothesis 1d: Subjective self-construal will positively influence perceived risks of self-disclosure.

Autonomous self-construal

Individuals holding autonomous self-construal, which is congruent with a mutually independent self, recognize themselves as beings who are disconnected from social context (Inumiya et al., 2007). They are very sensitive to internal attributes and objectives; hence, they do not strive to maintain relationships with others and are not prone to social influence (Inumiya et al., 2007). They tend to be interested in their own specific problems or interests due to the high degree of private self-consciousness and thereby express self only when it could aid them to achieve internal goals (Bae, 2007). Such individualistic inclinations may guide them to perceive usefulness to a lesser extent because they are unlikely to perceive the benefits of reciprocity derived from self-disclosure (B. Chen & Marcus, 2012; Posey et al., 2010). Certainly, they could still exploit the benefits by acquiring additional information on their interests if this could aid them to solve their own problems and achieve their own goals, presuming self-disclosure to be a tool to satisfy personal needs (Kang & Park, 2011; B. Lee, Han, Yi, & Lee, 2013). However, because they have strong individualistic inclinations and high self-esteem, there may be no reason to disclose self because they could achieve personal goals by validating their internal attributes without other external information (Markus & Kitayama, 1991). They would perceive self-disclosure as useless because, mastering the environment, they tend to refer to their own internal thoughts and feelings rather than others’ (B. Chen & Marcus, 2012). People who are independent are not significantly influenced by the environment unless doing so would greatly benefit the self. In addition, they would perceive risks from other members in SNS to a lesser extent, apart from the privacy risks coming from SNS providers (e.g., secret sharing). They self-disclose spontaneously for their own internal purpose rather than for seeking positive responses from others (Krasnova et al., 2012; Myung & Jang, 2012). Independent individuals are reluctant to disclose their self because they have a clear self-concept and do not value the development of the relationship on SNS (Campbell et al., 1996; Posey et al., 2010).

Hypothesis 1e: Autonomous self-construal will negatively influence perceived usefulness of self-disclosure.

Hypothesis 1f: Autonomous self-construal will negatively influence perceived risks of self-disclosure.

Self-Disclosure

Self-disclosure is an exchange of personal information, emotion, and thoughts on issues such as personal situation, characteristics, and incidents in the past and future (S. Kim et al., 2013). In the online context, it can be a voluntary and intentional communication of personal thoughts, feelings, and experiences with others (Posey et al., 2010). Self-disclosure is an important element in forming relationships in a social context like SNS because it enhances familiarity and intimacy among SNS users (Y. Chen & Nakazawa, 2009; Posey et al., 2010). Moreover, self-disclosure is necessary to form an attachment toward a specific object or person (S. Cho & Lee, 2015), which illustrates its mediating role. However, self-disclosure does not always mean a close and satisfactory relationship with others because it is a multidimensional construct (S. Cho, 2010). Perception of the rewards and costs associated with self-disclosure varies across individuals; thus, differential assessments of rewards and costs influence how people perceive and make self-disclosure and build relationships with others (Altman & Taylor, 1973; S. Cho, 2010). In this manner, it is possible to predict individuals’ relational attitudes toward the object or person based on how they assess the rewards, costs, and the expected results associated with self-disclosure (Ledbetter et al., 2011). Thus, perceived usefulness can be viewed as the positive expectation toward the rewards associated with self-disclosure, whereas perceived risk can be considered the negative expectation toward the costs resulting from self-disclosure on SNS (Shin, 2004).

Perceived usefulness of self-disclosure

SNS could perform a positive function when users self-disclose as a way of rewarding themselves (e.g., getting social approval from others; S. Kim et al., 2013). Perceived usefulness of self-disclosure on SNS could enhance the interaction with other users and with brands. It enables users to develop relationships as they satisfy their needs to express themselves and solve their problems (Ledbetter et al., 2011; E. Lee, Kim, & Ahn, 2013; Shin, 2004). Since consumers may relate to brands in ways that resemble the social ties of an interpersonal relationship (Nobre, Brito, & Becker, 2010), developing a strong relationship between consumers and brands on SNS could be equivalent to the process of penetrating the inner parts of consumers’ personalities, such as value and emotion, through self-disclosure (B. Chen & Marcus, 2012). Prior researchers found that SNS users who perceive the usefulness of self-disclosure tend to express positive emotion, such as closeness and attachment toward a person or object, because it could satisfy both self-expression and relational needs (S. Cho & Lee, 2015; Ledbetter et al., 2011; G. Lee & Lee, 2014).

Hypothesis 2a: Perceived usefulness of self-disclosure will positively influence brand community attachment.

Perceived risk of self-disclosure

Self-disclosure on SNS can perform a negative function if users fear and perceive the risk of not getting approval from others regarding the validity of their thoughts (J. Kim & Park, 2009; Shin, 2004). In other words, users may perceive risks if others are not in favor of the disclosed private information. In this case, users could avoid expressing self or hide their emotion and disclose only a superficial level of self. Hence, perceived risk can be an avoidance factor for the interaction with brands and can hinder the relationship development between consumers and brands on SNS (Falk & Riel, 2013). Hence, we assume:

Hypothesis 2b: Perceived risk of self-disclosure will negatively influence brand community attachment.

Brand Community Attachment

Brand community attachment is a relationship-based concept, defined as the emotional connectedness formed by the interaction between users and brand community (G. Lee & Lee, 2014). Individuals desire to use a brand community in order to be a part of the brand because they feel an attachment to the process of maintaining and developing their self by projecting themselves to brands (S. Cho & Lee, 2015). The social brand community is a place where they express the self and form relationships through interactions; thus, the social brand community can be an object of attachment (G. Lee & Lee, 2014). It is important to ensure the brand community attachment in order to maintain a long-term relationship with users because it enhances brand relationship satisfaction and emotional connectedness (Jung, 2013; Lee et al., 2013). Attachment to a brand community allows users to have psychological stability and even motivates them to widen their understanding of brands (S. Cho & Lee, 2015). It makes users strive to maintain and develop a relationship to be connected to an object of attachment (Lee et al., 2013). Hence, we assume:

Hypothesis 3: Brand community attachment will positively influence brand relationship satisfaction on SNS.

As discussed above, cultural self-construal influences the degree of self-disclosure in terms of how people perceive self-disclosure, and self-disclosure influences brand community attachment; thus, we assume the mediating role of self-disclosure in the relationship between cultural self-construal and brand community attachment because self-disclosure is mandatory to develop a relationship (Altman & Taylor, 1973). Thus, we propose:

Hypothesis 4: Self-disclosure mediates the relationship between cultural self-construal and brand community attachment in SNS.

Brand Relationship Satisfaction

Firms should suggest a meaningful relationship with consumers in psychological, social, and cultural contexts through brands if they want to build a long-term relationship between consumers and their brands (Heo, 2007). Social network, as well as culture, affects consumers’ attitude and behavior toward brands; hence, the satisfaction with a brand relationship can occur through the interaction within a sociocultural context (Nobre et al., 2010). Brand relationship satisfaction, a subordinate dimension of the relationship quality, is a useful concept that can measure qualitative aspects of a consumer–brand relationship such as satisfaction and maturity (Heo, 2007; Jung, 2013). Since the social brand community is a place for managing and developing a relationship between users and brands, it can explain the qualitative strengths of consumer–brand relationships.

Method

Data Collection and Sample

We collected data between July 15 and October 30, 2016, from 460 Korean SNS users who are part of social brand communities that cover topics about fashion brands, such as product reviews. In the case of Facebook, the largest growth in users was observed in the Asia-Pacific region between 2016 and 2018; thus, it was meaningful to use Korean subjects, one of the primary sources of this growth. Facebook is forbidden in China, which otherwise would have been a prominent source of the growth (Dogtiev, 2018). We e-mailed web operators of social brand communities in order to get their permission to post an anonymous Google Docs survey link on SNS (e.g., Facebook). A total of 460 users participated in the survey, but we used 446 responses for analysis because 14 subjects did not complete the survey.

As shown in Table 1, the majority of respondents were female (97.1%). About 76% of the respondents were in their 20s, followed by respondents in their 30s (11.0%). The respondents could relatively represent the general demographic profile of SNS users since the majority of global SNS users consist of people between the ages of 18 and 29 (87%), with a higher usage rate among women (83%) compared to men (75%; Dogtiev, 2018). Also, 72.4% of respondents had a bachelor’s degree, followed by respondents with a graduate degree (25.3%). The annual income of the majority of respondents (88.6%) was less than 24 million Korean Won (equivalent to US$20,000). Respondents indicated they used at least one SNS, and their popular SNS were Facebook (84.1%), Instagram (13.9%), and Twitter (2.2%).

Table 1.

Demographic Profiles.

Demographic Variables	Frequency (N)	Percentage
Gender
Female	433	97.09
Male	13	2.91
Age
18–29	339	76.01
30–39	49	10.99
40–49	58	13.00
More than 49	0	0.00
Education
High school diploma	0	0.00
Some college, no degree	10	2.24
Bachelor’s degree	323	72.42
Graduate degree	113	25.33
Annual income (Won)
Less than 24 million	395	88.56
24–42 million	43	9.64
42–60 million	8	1.79
60–90 million	0	0.00
More than 90 million	0	0.00
Total	446	100

Measures

We used established scales in the extant literature and adapted them to social brand community context to measure the investigated constructs. We used a 7-point Likert scale ranging from definitely not” to definitely. Measuring the three types of cultural self-construal, we adapted the scale developed by Inumiya et al. (2007), which uses a total of 18 items: 6 items for each construct of objective, subjective, and autonomous cultural self-construal. We measured self-disclosure in terms of perceived usefulness (e.g., “Do you think it is helpful to talk your private opinion in the SNS?”) and perceived risk (e.g., “Do you think it is risky to talk your private opinion in the SNS?”), using a total of 8 items of the modified Disclosure-Expectation Measurement Scale used in Shin (2004). We measured brand community attachment (e.g., “I emotionally feel attachment to the social brand community that I use”) based on the 5 items from the Website Attachment Measurement Scale used in Lee and Lee (2014), and brand relationship satisfaction (e.g., “The relationship I have with the social brand community that I use is satisfactory”) based on the 4 single-dimension measurement items used in Jung (2013).

Results

Instrumental Validation

We used principal component analysis and varimax methods for exploratory factor analysis in order to test the validity of cultural self-construal measurement items. We found that the measure is suitable for factor analysis because the Kaiser–Meyer–Olkin measure of sampling adequacy was .829, which is greater than .6, and the variances are the same across samples because Bartlett’s test of sphericity is significant (χ ² = 3089.186, df = 153, p = .000). We selected items with a factor loading value above .6 and extracted all three factors—objective, subjective, and autonomous self-construal—as displayed in Table 2.

Table 2.

Exploratory Factor Analysis Results (Cultural Self-Construal).

Item	Factor 1: Subjective Self	Factor 2: Autonomous Self	Factor 3: Objective Self	Communality
1. It is important for me to have control over all situations like a leader	.696			.491
2. I can make others to follow my will	.688			.493
3. I am self-assertive and make effort to be outstanding among other people	.758			.664
4. I want to have power to lead others	.729			.551
5. I have innate aptitude to influence others	.720			.561
6. I often feel the necessity to being the central person who exerts power	.702			.501
1. I like to hang around searching for new things		.762		.710
2. I am comfortable with making decision by myself		.844		.604
3. I know the way to be live happily by myself		.860		.665
4. It is important for me to be independent and self-reliant		.682		.606
5. Personal enjoyment and satisfaction are the most important criteria for whatever I do		.567		.440
6. I try to solve problem by myself rather than count on help from other people		.757		.739
1. I am a follower rather than a leader			.821	.604
2. I feel discomfort when I become remarkable among others			.760	.720
3. I want to be a supporter rather than a leader in a meeting			.798	.750
4. I feel comfortable with being asked to do something rather than make others to do something			.627	.756
5. I usually follow other’s opinion when the opinions conflict			.772	.517
6. I usually follow the decision of my friends even if it is different from mine			.691	.605
Eigen value	2.821	2.813	2.108
Cumulative variance (%)	22.959	41.645	67.269
Variance explained (%)	22.959	18.686	25.624

Measurement Model

We performed confirmatory factor analysis (CFA) to test the model fit. After dropping 3 items (i.e., OS4, PRS1, and BRS4) that possessed low factor loadings, we assessed the validity of constructs by estimating the confirmatory measurement model. We found that the model was plausible and fit to the data based on the following CFA results: χ ² = 1233.204 (df = 443/p = .000), goodness-of-fit index (GFI) = .847 (>.9), standardized root mean square residual (SRMR) = .058 (<.08), incremental fit index (IFI) = .903 (>.9), comparative fit index (CFI) = .903 (>.9), and root mean square error of approximation (RMSEA) = .063 (<.08). Except for GFI, one of the absolute fit indices that represent how closely the model comes to replicate the observed covariance matrix (Hooper, Coughlan, & Mullen, 2008), the model resulted in a satisfactory fit to the data. Next, we tested the convergent validity and the discriminant validity of measurement models. The indicators of each measurement model and their corresponding standardized regression weights (λ) were significant, with values ranging from .572 to .924 (Table 3). Also, the Cronbach’s α values of all constructs were greater than .70, composite reliability values were greater than .70, and average variance extracted values (AVE) were greater than .50. This indicated that the measurement models were internally consistent, and the instrument substantially measured the constructs, implying convergent validity. The discriminant validity was supported because it can be achieved when the AVE of the construct is greater than the square of correlations (R ²) between the respective constructs (Fornell & Larcker, 1981; Table 4). A multicollinearity problem was nonexistent, as the correlations between all latent variables range from .003 to .354, which is below .8.

Table 3.

CFA Results.

Construct	Item	λ Factor Loading	CR	Cronbach’s α	AVE
OS	OS1	.823***	.886	.823	.662
	OS2	.640***
	OS3	.709***
	OS5	.714***
	OS6	.602***
SS	SS1	.577***	.888	.809	.665
	SS2	.643***
	SS3	.808***
	SS4	.655***
	SS5	.712***
	SS6	.582***
AS	AS1	.664***	.900	.805	.716
	AS2	.768***
	AS3	.839***
	AS4	.619***
	AS5	.799***
	AS6	.703***
PUS	PUS1	.829***	.926	.873	.798
	PUS2	.866***
	PUS3	.754***
	PUS4	.732***
PRS	PRS2	.823***	.915	.849	.776
	PRS3	.881***
	PRS4	.818***
BCA	BCA1	.802***	.957	.940	.865
	BCA2	.852***
	BCA3	.877***
	BCA4	.924***
	BCA5	.893***
BRS	BRS1	.911***	.966	.940	.912
	BRS2	.910***
	BRS3	.922***

Note. ***p < .001. AS = autonomous self-construal; AVE = average variances extracted; BCA = brand community attachment; BRS = brand relationship satisfaction; CFA = confirmatory factor analysis; OS = objective self-construal; PUS = perceived usefulness of self-disclosure; PRS = perceived risk of self-disclosure; SS = subjective self-construal.

Table 4.

AVE Values and Squared Correlation of Each Latent Variable.

	OS	SS	AS	PUS	PRS	BCA	BRS
OS	.662
SS	.125	.665
AS	.026	.119	.716
PUS	.082	.000	.016	.798
PRS	.000	.063	.003	.051	.776
BCA	.008	.000	.001	.089	.003	.865
BRS	.037	.000	.000	.113	.003	.697	.912

Note. The AVE values are in boldface. The squared correlations (R ²) of all constructs are on the off-diagonal. AS = autonomous self-construal; AVE = average variances extracted; BCA = brand community attachment; BRS = brand relationship satisfaction; OS = objective self-construal; PUS = perceived usefulness of self-disclosure; PRS = perceived risk of self-disclosure; SS = subjective self-construal.

Structural Model

We verified the structural model based on the instrument validation result using AMOS 22. The overall model fit was satisfactory as χ ² was 1074.837 (df = 446, p = .000), and the fit indices were within acceptable ranges (SRMR = .062, IFI = .909, CFI = .909, RMSEA = .056). Then, we tested hypotheses using maximum likelihood estimation. The results are shown in Figure 1.

Figure 1.

Results of structural model equation model analysis.

With respect to perceived usefulness of self-disclosure, as displayed in Table 5, objective and subjective self-construal positively influenced perceived usefulness, and autonomous self-construal negatively influenced perceived usefulness, supporting Hypothesis 1a, Hypothesis 1c, and Hypothesis 1e (β = .304, p < .001; β = .120, p < .05; β = −.118, p < .05, respectively). Regarding perceived risk of self-disclosure, there was no significant influence of objective self-construal on risk perception, rejecting Hypothesis 1b (β = .072, n.s.), whereas there was a positive influence of subjective self-construal on perceived risk, supporting Hypothesis 1d (β = .268, p < .001). Lastly, there was no significant influence of autonomous self-construal on perceived risk, rejecting Hypothesis 1f (β = −.006, n.s.). In addition, there was a positive influence of perceived usefulness on brand community attachment, supporting Hypothesis 2a (β = .319, p < .001). This indicates that those who perceive the usefulness of self-disclosure tend to show greater attachment to the brand community. Contrary to our expectation, there was a positive relationship between perceived risk and brand community attachment, which rejects Hypothesis 2b (β = .115, p < .05). Confronting risks associated with self-disclosure, users may regulate self differently depending on the degree of their self-esteem, which can differ by distinct cultural self-construal. Also, we found that brand community attachment positively influenced brand relationship satisfaction on SNS, supporting Hypothesis 3 (β = .799, p < .001). Since there was no significant influence of perceived risk on brand relationship satisfaction (β = .035, n.s.), those who perceive risks in self-disclosure would form brand community attachment first (β = .107, p < .01) before experiencing brand relationship satisfaction.

Table 5.

Path Coefficient Between Latent Variables.

Hypothesis	Path	β	SE	CR	p	Result
H1a	Objective → Usefulness	.304***	.079	5.310	.000	Accept
H1b	Objective → Risk	.072	.083	1.353	.176	Reject
H1c	Subjective → Usefulness	.120*	.073	2.255	.022	Accept
H1d	Subjective → Risk	.268***	.089	4.702	.000	Accept
H1e	Autonomous → Usefulness	−.118*	.061	−2.264	.023	Accept
H1f	Autonomous → Risk	−.006	.070	−.112	.911	Reject
H2a	Usefulness → Attachment	.319***	.066	6.150	.000	Accept
H2b	Risk → Attachment	.115*	.055	2.312	.021	Reject
H3	Attachment → Satisfaction	.799***	.039	19.728	.000	Accept

*p < .05. ***p < .001.

Then, the mediation effect was analyzed with the bootstrapping method. First, the influence of each cultural self-construal on brand community attachment was examined by controlling the mediators. Objective (β = .321, p < .001), subjective (β = .224, p < .01), and autonomous self-construal (β = −.169, p < .05) showed significant influences. After confirming the significance of influence, Hypothesis 4 was examined. Subjective self-construal influenced perceived usefulness (β = .237, p < .001) and risk (β = .123, p < .05) of self-disclosure, while objective and autonomous self-construal influenced perceived usefulness (β = .222, p < .01; β = −.168, p < .05). Also, perceived usefulness (β = .323, p < .001) and risk (β = .134, p < .05) of self-disclosure influenced brand community attachment. However, objective, subjective, and autonomous self-construal did not directly influence brand community attachment (β = .055, n.s; β = .019, n.s.; β = .019, n.s), indicating the complete mediation. Thus, the indirect influence of cultural self-construal on brand community attachment through the mediation of self-disclosure can be found, supporting Hypothesis 4. Specifically, the indirect effects of subjective and autonomous self-construal on brand community attachment were about .544 (p < .001), .288 (p < .001), and .399 (p < .001).

Discussion and Conclusion

The results of the present study have implications for both academic research and managerial decisions regarding the effective utilization of social brand communities. Academically, prior researchers primarily explained the different communication behaviors on SNS with the use of a binary cultural classification, which cannot reflect the changing cultural internal value of consumers on SNS. Moreover, the reasons behind the difference in a degree of community attachment and subsequent relationship satisfaction on SNS have not been fully examined, despite the importance of self-disclosure in advancing relationship. In this respect, we provide academic implications by presenting empirical grounds for the significant influence of cultural self-construal and self-disclosure on relationship development through the lens of social penetration theory. In addition, we can draw practical implications from this study for SNS marketers who strive to encourage users to create and share content, to activate the interaction on SNS, and to maintain long-term relationships with the brand. It enables practitioners to formulate optimized strategies for engaging SNS users in self-disclosure by understanding how people differently assess the rewards and costs associated with self-disclosure.

Specifically, it is critical for practitioners to formulate strategies that maximize the relational advantages for users holding objective self-construal, and to promote easier and safer interactions to make these users reveal the true self when disclosing their self to build an intimate relationship. Socially responsive followers tend to disclose their self in return for others’ disclosure. Thus, they only perceive usefulness, and not the risk of self-disclosure despite being prone to social influence. They could avoid risks by modifying their disclosure in a preferred way through reconciling competing rewards and costs without hurting self due to their low self-esteem (Alhidari, Iyer, & Paswan, 2015; B. Chen & Marcus, 2012).

Meanwhile, self-disclosure itself could invoke users with subjective self-construal to attach to communities regardless of what they perceived. Thus, it would be desirable to provide a community environment, which vitalizes self-expression of these users because they could be the initiators of active social interactions. Similarly, opinion leaders on SNS, who recognize themselves as leaders and tend to be socially responsive to others (Lee et al., 2013; Posey et al., 2010), would express their self by taking the risks of negative responses while anticipating greater benefits due to high self-esteem (Cavallo, Fitzsimons, & Holmes, 2009).

Lastly, users with autonomous self-construal tend to be bystanders or cherry pickers because they disclose their self only when it could internally benefit them (Lee & Lee, 2014). Thus, firms should provide either quality brand information or enjoyment, which could satisfy the personal needs of these users. Consistently, subjects with autonomous self-construal did not perceive risk because they tend to be indifferent to interpersonal activities, which validate self-disclosure, due to their clear self-concept and insensitivity to others (B. Chen & Marcus, 2012). To sum it up, considering that the perceptions of self-disclosure were the key that enabled attachment to brand communities, it is imperative to enable users to gain more rewards through disclosing self, reflecting the distinct needs of users with different cultural self-construal.

Despite the meaningful implications that we draw from this study, there is a room for further improvement. First, considering that intimate relationships can be developed based on the gradual revealing of the inner core of personality, it is desirable to use various scopes of self-disclosure in terms of amount, breadth, or depth in order to examine qualitative differences among users’ self-disclosure on SNS. Second, there is a limitation in the data source. Most survey respondents in this study are Korean SNS users who are predominantly female and aged in their 20s and 30s, which could limit the generalizability of the findings to consumer groups with different demographic and geographic backgrounds. Thus, it would be meaningful to conduct further studies with different consumer groups to obtain generalizable results.

Footnotes

Acknowledgments

I express my deep sense of gratitude to Dr. Sun-Jin Hwang, professor at SungKyunKwan University, for encouraging me to put efforts on this paper and for her valuable guidance, keen interest, and encouragement at various stages of development process of paper.

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.

ORCID iD

Hanna Lee

References

Alhidari

Iyer

Paswan

(2015). Personal level antecedents of eWOM and purchase intention, on social networking sites. Journal of Customer Behavior, 14, 107–125.

Altman

Taylor

D. A.

(1973). Social penetration: The development of interpersonal relationships. Oxford, England: Holt, Rinehart & Winston.

Bae

(2007). The relationship of forgiveness with IPC locus of control and subjective, objective, autonomous self-construal (Unpublished Master’s thesis). Korea University, Seoul, Korea.

Campbell

J. D.

Trapnell

P. D.

Heins

S. J.

Kartz

I. M.

Lavallee

L. F.

Lehmann

D. R.

(1996). Self-concept clarity: Measurement personality correlates and cultural boundaries. Journal of Personality & Social Psychology, 70, 141–156.

Cavallo

Fitzsimons

Holmes

(2009). Taking chances in the face of threat: Romantic risk regulation and approach motivation. Personality & Social Psychology Bulletin, 35, 737–751.

Chen

Marcus

(2012). Students’ self-presentation on Facebook: An examination of personality and self-construal factors. Computers in Human Behavior, 28, 2091–2099.

Chen

Nakazawa

(2009). Influences of culture on self-disclosure as relationally situated in intercultural and interracial friendships from a social penetration perspective. Journal of Intercultural Communication Research, 38, 77–98.

Cho

(2010). Cross-cultural comparison of Korean and American social network sites: Exploring cultural differences in social relationships and self-presentation (Unpublished Doctoral dissertation). Rutgers University, New Brunswick, NJ.

Cho

Han

(2013). The analysis of the use of SNS and changes in individual’s social relationship. Gwacheon, Korea: Korea Information Society Development Institute.

10.

Cho

Lee

(2015). Effects of brand self-disclosure and intimacy on communication in Facebook. The Korean Journal of Advertising, 26, 95–117.

11.

Choi

Lee

Jang

(2013). The impact of public self-consciousness, narcissism, and brand concept on brand fanpage share intention. Korea Marketing Review, 28, 65–86.

12.

Crocker

Park

(2004). The costly pursuit of self-esteem. Psychological Bulletin, 130, 392–414.

13.

DMC Media. (2015). 2015 digital lifestyle analysis report of Korean digital consumers. Retrieved from www.dmcmedia.co.kr

14.

Dogtiev

(2018, May 4). Facebook revenue and usage statistics 2018. Business of Apps. Retrieved from www.businessofapps.com/data/facebook-statistics/

15.

Efrati

(2016, April 7). Facebook struggles to stop decline in original sharing. The Information. Retrieved from www.theinformation.com/facebook-struggles-to-stop-decline-in-original-sharing?shared=eeb2ca

16.

E-News Today. (2016, April 10). Facebook, KaKaostroy, ached by commercial contents. Retrieved from http://www.enewstoday.co.kr/news/articleView.html?idxno=514025

17.

Falk

Riel

(2013). Cultural differences in user privacy behavior on social networking sites (Unpublished Master’s thesis). Jönköping University, Jönköping, Sweden.

18.

Fiore

(2008). The digital consumer. Clothing & Textiles Research Journal, 26, 177–190.

19.

Fornell

Larcker

D. F.

(1981). Evaluating structural equation models with unobservables and measurement error. Journal of Marketing Research, 18, 375–381.

20.

Han

Shin

(1999). A cultural profile of Korean society: From vertical collectivism to horizontal individualism. The Korean Journal of Psychology, 13, 293–310.

21.

Heo

(2007). Relationship process between consumer and ebrand. The Journal of Advertising & Promotion Research, 18, 111–141.

22.

Hooper

Coughlan

Mullen

M. R.

(2008). Structural equation modelling: Guidelines for determining model fit. Electronic Journal of Business Research Methods, 6, 53–60.

23.

Inumiya

Han

Lee

Kim

(2007). Development of a cultural self-construal scale. Korea Society of Psychology, 21, 17–34.

24.

Jung

(2013). The effect of characteristic of Facebook fanpage on WOM (Unpublished Master’s thesis). HanYang University, Seoul, Korea.

25.

Kang

Park

(2011). Motivational antecedents of social shopping for fashion and its contribution to shopping satisfaction. Clothing & Textiles Research Journal, 29, 331–347.

26.

Kankanhalli

Tan

C. Y.

Wei

(2005). Contributing knowledge repositories: An empirical examination. Management Information Systems Quarterly, 29, 113–143.

27.

Kim

Han

(2004). The effects of depressive vulnerability and self-consciousness on interpersonal relationship in undergraduates. Korean Journal of Counseling & Psychotherapy, 16, 277–294.

28.

Kim

Park

. (2009). The study on the type of self-disclosure of leaders and effects. The Korean Academic of Management, 2, 281–303.

29.

Kim

Yang

Kwon

(2013). Effects of online/offline self-disclosure on the sense of community and happiness. Cybercommunication Academic Society, 30, 5–42.

30.

Krasnova

Veltri

Gunther

(2012). Self-disclosure and privacy calculus on social networking sites. Business & Information Systems Engineering, 3, 127–135.

31.

Ledbetter

A. M.

Mazer

J. P.

DeGroot

J. M.

Meyer

K. R.

Mao

Swafford

(2011). Attitudes toward online social connection and self-disclosure as predictors of Facebook communication and relational closeness. Communication Research, 38, 27–53.

32.

Lee

Han

Lee

(2013). Exploratory study of underlying dimensions of use motive in social media and marketing implications. Korea Marketing Review, 28, 87–108.

33.

Lee

Kim

Ahn

(2013). Effects of brand self-disclosure and user social connectedness on response to Facebook brand fanpages. International Journal of Contents, 13, 60–71.

34.

Lee

(2014). The effects of self-concept clarity and self-disclosure on attachment to social networking sites and satisfaction with life. Korea Journal of Business Administration, 27, 1845–1861.

35.

Lee

(2008). Differences in college students’ separation-individuation according to private self-consciousness and public self-consciousness. Korean Journal of Counseling & Psychotherapy, 20, 1087–1106.

36.

Lee

Yang

Kwon

(2013). How attachment style affects relational quality? The Korean Journal of Social & Personality Psychology, 27, 107–125.

37.

Lowry

P. B.

Cao

Everard

(2011). Privacy concerns versus desire for interpersonal awareness in driving the use of self-disclosure technologies. Journal of Management Information Systems, 27, 163–200.

38.

Markus

Kitayama

(1991). Culture and self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224–253.

39.

Murray

Holmes

Aloni

Pinkus

Derrick

Leder

(2009). Commitment insurance: Compensating for the autonomy costs of interdependence in close relationships. Journal of Personality & Social Psychology, 97, 256–278.

40.

Myung

Jang

(2012). The relationships between cultural self-construal and seeking professional help. Journal of Human Understanding & Counseling, 33, 181–198.

41.

Nobre

Brito

Becker

(2010). Brand relationships: A personality based approach. Journal of Service Science and Management, 3, 206–217.

42.

Posey

Lowry

P. B.

Roberts

T. L.

Ellis

T. S.

(2010). Proposing the online community self-disclosure model. European Journal of Information Systems, 19, 181–195.

43.

Schau

H. J.

Gilly

M. C.

(2003). We are what we post? Self-presentation in personal web space. Journal of Consumer Research, 30, 385–404.

44.

Shin

(2004). Relative importance of approach factors and avoidance factors in predicting professional help seeking behavior (Unpublished Master’s thesis). Pusan National University, Pusan, Korea.

45.

Tang

Wang

(2012). Self-disclosure among bloggers: Re-examination of social penetration theory. Cyberpsychology, Behavior, & Social Networking, 15, 245–250.

46.

Yoo

Kim

Moon

(2012). Exploring cross-cultural differences in self-presentation and self-disclosure in SNSs. Journal of Advertising & Promotion Research, 1, 77–118.

47.

Yook

(2014, April). The effect of SNS self-disclosure and assessment of online comments of others on impression formation. Paper presented at the conference of the 2014 Conference of Korean Women’s Association for Communication Studies, Seoul, Korea.

48.

Zaglia

M. E.

(2013). Brand communities embedded in social networks. Journal of Business Research, 66, 216–223.

49.

Zhang

Jiang

Carroll

(2010). Social identity in Facebook community life. International Journal of Virtual Communities & Social Networkings, 2, 66–78.