The relationship among tourists' persuasion, attachment and behavioral changes in social media

TFS-18659; No of Pages 11 Technological Forecasting & Social Change xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Technological Forecasting & Social Change

The relationship among tourists' persuasion, attachment and behavioral changes in social media Namho Chung, Heejeong Han ⁎ College of Hotel and Tourism Management, Kyung Hee University, Republic of Korea

a r t i c l e

i n f o

Article history: Received 6 April 2016 Received in revised form 2 September 2016 Accepted 7 September 2016 Available online xxxx Keywords: Social media Elaboration likelihood model Reference group influence Attachment Behavioral changes Smart tourism

a b s t r a c t In recent years, individual travelers are in a situation where they have to search through more information than ever before via diverse smart devices. Social media has become an important role in dispersing travel information. Unlike other types of communication media, social media not only provides users with information, but also allows them to identify who the source of the information is. Our study found that, firstly, argument quality affected neither the informational influence nor attachment, secondly, source credibility had statistically significant impacts on all the variables of informational influence, attachment, and normative influence, thirdly, network externality also had statistically significant impacts on all the variables of informational influence, attachment, and normative influence, and finally, informational influence, attachment, and social media normative influence were predictors of travelers' behavioural changes. This study formulated a theoretical framework and empirically analyzed the travelers' behaviors using social media using the elaboration likelihood model, reference group influence theory and attachment theory. Practically, social media platforms should promote source credibility and network externality. Tourism marketing organizations should build attachments with social media users and should utilize profile and reputational reviewers. © 2016 Elsevier Inc. All rights reserved.

1. Introduction Aided by the development of the Internet, social media users are increasing. Nearly 2.3 billion people actively use social media and its number grew 10% year on year (Kemp, 2016). About one-fifth of leisure travelers worldwide have used social media to gather ideas and inspiration regarding travel destinations, hotels, vacation activities, attractions and restaurants (eMarketer, 2013b). The estimated number of social networking users was about 1.73 billion in 2013 and is expected to rise to 2.55 billion by 2017 (eMarketer, 2013a). In the Republic of Korea, an analysis revealed that 88.4% of total social media users used some profile-based services, and 40.7% used community (Korea Internet and Security Agency, 2015). According to an examination of the characteristics of such social media users, users valued ‘information acquisition and exchange,’ and ‘communication’ above other uses such as recreation, leisure, fellowship, etc. (National Information Society Agency, 2010). Such facilitation of online communities also substantially influences the facilitation of social media, which correlates with an increase in free independent travelers (FITs). Thus, prospective travelers use social media to search, share, and gather travel information from people's pre-travel, during-travel, and post-travel reviews (ParraLópez et al., 2012). ⁎ Corresponding author. E-mail addresses: [email protected] (N. Chung), [email protected] (H. Han).

Recently, social media have various types such as blogs, social networking sites (SNS), pictures or video sharing applications, and dictionary-type applications (Leung et al., 2013a; Parra-López et al., 2011). Social media also plays an increasingly important part in online information searching (Xiang and Gretzel, 2010) and online reputation. Unlike other online behaviors regarding information, social media allows travelers to search, share, organize, integrate, forward, re-share travel information and engage in potential travel (Leung et al., 2013a). In this way, we could find a social media definition from the literature, which was “a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0, and that allow the creation and exchange of User Generated Content” (Kaplan and Haenlein, 2010). Such travel information searches using social media are similar to the characteristics of travel products. In other words, because of the unique characteristics of travel products (high-involvement, experiential goods, intangibility, high purchase risk, etc.), the travel industry is an industry where information is highly important (Sheldon, 1997; Tan and Chen, 2012; Werthner and Klein, 1999). Therefore, potential tourists utilize various information sources because they recognize the importance of travel information, and, the majority of people obtained travel information from acquaintances, travel agencies, or travel books in the past. With the development of the Internet, however, today's travelers consider it more important to search for travel information on websites (Kim et al., 2007). Unlike past group travelers, individual

http://dx.doi.org/10.1016/j.techfore.2016.09.005 0040-1625/© 2016 Elsevier Inc. All rights reserved.

Please cite this article as: Chung, N., Han, H., The relationship among tourists' persuasion, attachment and behavioral changes in social media, Technol. Forecast. Soc. Change (2016), http://dx.doi.org/10.1016/j.techfore.2016.09.005

2

N. Chung, H. Han / Technological Forecasting & Social Change xxx (2016) xxx–xxx

travelers who potentially make overall decisions based on someone's reviews obtain a great deal of information necessary for travelling. Therefore, the social media become as not only an information source but also popular (Lee et al., 2011). In social media communities such as Facebook, Twitter, cafés, and blogs, people build relationships while talking about various travelrelated subjects (Lee et al., 2012). There have been many existing studies undertaken from various perspectives including participation benefits, loyalty, websites, advice and effects of Word-of-Mouth (WOM) (Arsal et al., 2008; Casaló et al., 2011; Chung and Buhalis, 2008; Kim et al., 2004; Ku, 2012; Lee et al., 2011; Leung et al., 2013b; Wang and Fesenmaier, 2004). Particularly, reviews on social media are cognitively and affectively important as a source of tourism information, which has been significantly dealt with. For example, studies in relation to information sharing through community identification (Lee et al., 2012; Qu and Lee, 2011), information search (Tan and Chen, 2012), and the intention to use information sharing (Ku, 2012) have been widely conducted. Despite this, the existing studies show that they have not sufficiently reviewed the intrinsic characteristics of social media. Unlike other travel information media thus far, social media is characterized by providing not only specific review information, but also reputational information. Travelers who use travel information in social media are influenced not only by the personal quality of travel information, but also the reputation of the reviewers that have created, generated and forwarded the information. Compared to online search behaviors before the advent of social media, online travel information searches are currently more socially connected. Therefore, unlike traditional media, social media plays a role of the new highly persuasive media that changes travelers' behavior. Furthermore, researchers should simultaneously identify the quality of information (argument quality) and the credibility of the information provider (source credibility) to ascertain the effects of social media related to travel information searches. In addition, researchers should also take into account the persuasive effects of social media. The present study intends to explain how social media changes traveler behavior during information searches. This study adopts a viewpoint that inclusively integrates the theories of the elaboration likelihood model (ELM) and reference group influence in explaining the behavior. Instead of reflecting on each particular characteristic of various social media types, the current study focuses on the general feature among social media types which is the: “creating and exchanging of user generated content” (Kaplan and Haenlein, 2010, p. 61), and tries to understand behavioral changes among prospective travelers. To this end, the following research objectives are suggested. We develop a research model that combines the theories of the ELM, attachment theory and reference group influence. We empirically test the relevance, relationship between the above three theories identified by performing an analysis using data. The results of this study provide valuable information for both academics and practitioners investigating and developing social media strategies for travel information searches. 2. Theoretical background 2.1. Elaboration likelihood model The ELM theory offers a conceptual basis for investigating attitude and persuasion (Angst and Agarwal, 2009; Li, 2013). It is a dual process theory of attitude formation and change (Petty and Cacioppo, 1986), suggesting that external information is the primary driver of attitude change, and the consequent behavioral change (Bhattacherjee and Sanford, 2006). According to this theory, persuasion can act via a central or peripheral route, where personal attributes determine the relative effectiveness of central or peripheral processes (Petty et al., 1981). When elaboration is high, the recipient will trace a central route of persuasion; however, when elaboration is low, he or she will take a peripheral route (Petty and Cacioppo, 1986).

Central route processing represents the process of elaborating on an appeal by paying attention to the quality of an argument and then evaluating the argument. Argument quality refers to an individual's perception that an argument is strong, cogent as opposed to weak, and specious (Petty and Cacioppo, 1986). In comparison, peripheral route processing describes the process of drawing conclusions from rules of thumb or reliance on heuristic cues (e.g., number of messages, number of message sources, source likeability, and source credibility) with little regard for the actual merits of an argument (Bhattacherjee and Sanford, 2006; Petty and Cacioppo, 1986). Individuals who use the peripheral route generally do not want to devote the necessary cognitive energy to elaboration, or they cannot expend the effort (Angst and Agarwal, 2009). The theoretical framework of the ELM is illustrated in Fig. 1 below. The studies of ELM in the context of the Internet suggest that individual judgments of cognitive authority are affected by heuristic cues (Rieh and Belkin, 1998; Sussman and Siegal, 2003). Moreover, Tam and Ho (2005) and Li (2013) insist that one information cue is a good criterion, and recipients base their decision-making on inference. However, prospective travelers have a huge amount of travel information via social media. Thus, peripheral cues are considered as an important factor that can affect prospective travelers' judgment. Previous studies consider argument quality related to central route processing and source credibility related to peripheral route processing by using the framework of ELM theory to explain individual attitude (Bhattacherjee and Sanford, 2006; Chen and Ku, 2012; Li, 2013; Sussman and Siegal, 2003). Bhattacherjee and Sanford (2006) indicated that the number of users (i.e., network externality) and number of times an informational message is considered as alternative peripheral cues in the context of the information technology acceptance. Argument quality refers to arguments with sufficient persuasive information. Source credibility refers to a recipient's perception of an information source as believable, trustworthy, and reliable (Bhattacherjee and Sanford, 2006; Petty et al., 1981; Sussman and Siegal, 2003) such as an online product review. In addition, network externality indicates that an increase of users will increase the value or effect of services or products (Lin and Lu, 2011; Katz and Shapiro, 1985). Social media users encounter multitudes of travel information, and they try to gather personal but reliable information. In this context, the recipients are affected by whether the information is personally accurate, whether the informant is credible and by whether the information is from social media users. Overall, current research on prospective travelers using social media has considered argument quality as a factor related to the central route. In addition, source credibility and network externality are considered peripheral cues. 2.2. Reference group influence Some existing studies considered social influences as an important factor in the formation of user behaviors. Behavioral research examples can be found in subjective norms in the theory of reasoned action (TRA), and social systems in the theory of innovation diffusion (Hsu and Lu, 2004). When social media—the present study's context—is considered, it is formed based on human relationships, and social reviews (Chu and Kim, 2011). Thus, the understanding of social influences in social media is of utmost importance in order to understand traveler behavior. Meanwhile, reference groups are defined as actual or imaginary individuals or groups that influence the standards for individual evaluations, aspirations, and behaviors (Park and Lessig, 1977). In searching travel information on social media, people exhibit certain behaviors in terms of purchasing travel products, information search methods, and product-purchasing methods under the influence of people connected with themselves: namely, reference groups. Social influences are divided into informational social influence and normative social influence (Deutsch and Gerard, 1955). Firstly, informational social influence is the evidence about reality. Informational social

Please cite this article as: Chung, N., Han, H., The relationship among tourists' persuasion, attachment and behavioral changes in social media, Technol. Forecast. Soc. Change (2016), http://dx.doi.org/10.1016/j.techfore.2016.09.005

N. Chung, H. Han / Technological Forecasting & Social Change xxx (2016) xxx–xxx

Argument Quality

Central Route

+

+

Elaboration Likelihood (e.g., Motivation, Ability)

−

3

Attitude

+

Adoption Behavior

+ Peripheral Route

Source Credibility

Fig. 1. Elaboration likelihood model framework.

influence refers to the amount of exertion required to accept the information acquired from other people. Vice versus, normative social influence, however, is the influence of other people that prompts one to conform to the positive expectations of others (Deutsch and Gerard, 1955). Normative social influence is the concept of complying with group harmony, or inducing positive evaluations of other people. The two influences generally materialize through three phases: internalization, identification, and compliance. Internalization occurs when people recognize certain information by strengthening their own knowledge at a higher level than that of their reference group's. Identification occurs while they define themselves in relation to their reference group, and adopt the opinions of others (e.g., online product reviews). Compliance occurs when people conform to the expectations of other people by taking into account rewards, rejection or hostility (Hsu and Lu, 2004). Therefore, informational social influence is the process of internalization, whereas normative social influence is the form of identification and compliance (Hsu and Lu, 2004; Kelman, 1961). Malhotra and Galletta (1999) also viewed such social influence processes as the processes of forming psychological attachment. As normative and social influences occur during information processing (Chen et al., 1996; Li, 2013; Lundgren and Prishlin, 1998; Wood, 2000), in using social media to search travel information, people who are under informational influence exhibit a greater need for the acquisition of information. Conversely, people who are under the normative influence try to adhere to the expectations of others, and gain social approval by showing certain behaviors such as using brands or products that other people consider appropriate (Chu and Kim, 2011). Consequently, given that social media is composed of human relationships, behavioral changes caused by social media can be explained by informational and normative influences. 2.3. Attachment theory According to Bowlby (1969), attachment is defined as the special mutual relationship between a child and its mother. In addition, strong emotions such as connection, affection and passion correlate with strong attachment (Mugge et al., 2010). The original definition of attachment, like that of close relationships with people, is broadly defined as an emotional bond between a person and a specific object (Hyun and Kim, 2012). When people have an attachment to a place or community, it is known as ‘placement attachment’ or ‘community attachment’ (e.g., Gu and Ryan, 2008; Lee et al., 2010; Yuksel et al., 2010). Recent studies have started to apply the broad definition of attachment, and consider two specific types of attachment (common bridge and common bond) in the online community context. These two attachments differ in several aspects (Table 1). Prentice et al. (1994) asserted that there are two types of clubs within a community. The one is a common bridge group (a topic-based group) and the other one is a common bond group (a relation-based group). According to researchers, members of a common bridge group feel more attached to the group as a whole than to their fellow group members. In opposition, members of

common bond groups feel attached to group members as well as to the group as a whole. This classification is related to individuals' motivation for engaging in groups. Ren et al. (2007) and Sassenberg's (2002) studies have been applied to various domains, including the online community environment (Fiedler and Sarstedt, 2010; Ren et al., 2012). Ren et al. (2007) explained that attachment in a community could be divided into the common bridge and common bond concepts. Common bridge (common identity) attachment functions through individual group member identities and goals; common bond attachment functions through interpersonal bonds between members. Therefore, attachment theory needs to look at the dichotomy where, on the one hand, members may share a common purpose (common bridge attachment), and on the other, members use interesting content as a means of building relationships (common bond attachment). Similarly, social media is one of the communities and thus, individuals tend to belong to certain communities with various purposes. For example, they may focus more on accumulating and interchanging their shared interests (common bridge group) or building relationships with others (common bond group) in the communities. They who belong to different groups may interchange their opinions and generate various types of contents. Such processes lead to establish new relations with others and more close to others. In other words, social media users may form an attachment through the process to achieve a common purpose or to establish a bond with other users.

3. Research model and hypotheses Based on ELM theory, attachment and reference group influence, this study proposes a research model in Fig. 2. The model suggests that the argument quality, source credibility and network externality are predictors of travel informational influence, social media normative influence and group attachment. In addition, it suggests that travel informational influence and social media normative influence are predictors of the group attachment. Lastly, it suggests that travel informational influence, social media normative influence and group attachment are predictors of group members' behavioral changes. This study also added control variables – i.e., gender, age, education, income, and occupation – to the research model. Prior studies regarding information

Table 1 Comparison between common bridge and common bond attachment. Dimensions

Common bridge attachment (bridge)

Common bond attachment (bond)

Group characteristic Attachment subject Individual's motivation

Topic-based Group Common purpose

Relations-based Group members Interpersonal bonds between members

Please cite this article as: Chung, N., Han, H., The relationship among tourists' persuasion, attachment and behavioral changes in social media, Technol. Forecast. Soc. Change (2016), http://dx.doi.org/10.1016/j.techfore.2016.09.005

4

N. Chung, H. Han / Technological Forecasting & Social Change xxx (2016) xxx–xxx

Fig. 2. Research model.

technology indicate that these variables may have the potential to affect individuals' behavior (e.g., Kim et al., 2012; Leong et al., 2013; Lu and Hsiao, 2010). Hence, this study included these variables as control variables.

diverse and highly influential. Hence, this work proposes the following hypotheses: H1. Argument quality positively impacts travel informational influence. H2. Source credibility positively impacts travel informational influence.

3.1. Argument quality, source credibility, network externality and travel informational influence Informational influence occurs as the result of critical thinking (Henningsen et al., 2003). During the processing of travel information, people who think in a cognitive manner with a focus on argument quality are inclined to evaluate messages. This leads to information collection and sharing behavior (Lee et al., 2006). Because such information collection and sharing activities affect people's existing beliefs, argument quality in social media affects the informational social influence (Li, 2013). Alternatively, the credibility of information sources can affect people's cognitive thinking if the sources are endorsed by celebrities or highly reliable experts (Bhattacherjee and Sanford, 2006). In addition, Tam and Ho (2005) claimed only some pieces of information affect users' cognitive processing based on rules of thumb. Li (2013) proved that source credibility affects users' thoughts. In other words, source credibility plays a role as a standard of cognitive judgment, such as whether travel information in social media has an informational influence. Therefore, the credibility of information sources would affect decision-making quality. Network externality indicates that the effects or values that network users acquire from goods or services increase in proportion to the increase in the number of users and alternative goods or services (Katz and Shapiro, 1985). Therefore, network externality in a social media community represents the number of the members of the community (Lin and Lu, 2011). This network externality would be important for peripheral cues, which is related to nonmessage elements of social media (Liu et al., 2012). In addition to the source credibility, the number of social media users also affects cognitive and high-elaboration process in information seeking behavior. A larger number of members in a social media community affect travel information search results because of the greater informational value of the community as a travel information search tool due to the large number of participants within that social media community. Accordingly, travel information explored inside the community also becomes widely

H3. Network externality positively impacts travel informational influence. 3.2. Source credibility, network externality and social media normative influence Normative influence occurs as the result of peripheral thinking rather than central routing (Kaplan, 1989; Henningsen et al., 2003). If people employ only a minor portion of cognitive thinking during their information processing, they rely on heuristic methods, which in turn makes them rely on the cues of credible people for their decisionmaking. According to McGuire (1968), as the characteristics of such reliable information sources are determined during the course of identification, people who rely on heuristic methods rather than cognitive thinking while searching travel information on social media sites tend to rely on the information of highly credible people. People attempt to identify credibility through reference groups or by means of information source reliability (Li, 2013). Persuasive messages with higher levels of reliability within a group tend to comply with the group's own rules, and thus, affect the normative social influence, which is the influence that leads people to live up to the positive expectations of others (Deutsch and Gerard, 1955; Li, 2013). Moreover, a higher level of network externality can result in a corresponding increase in the pressure on network users to accept opinions or information from other people due to their relationships with other members within the network (Lin and Lu, 2011). Hence, the following hypotheses are proposed: H4. Source credibility positively impacts social media normative influence. H5. Network externality positively impacts social media normative influence.

Please cite this article as: Chung, N., Han, H., The relationship among tourists' persuasion, attachment and behavioral changes in social media, Technol. Forecast. Soc. Change (2016), http://dx.doi.org/10.1016/j.techfore.2016.09.005

N. Chung, H. Han / Technological Forecasting & Social Change xxx (2016) xxx–xxx

3.3. Argument quality, source credibility, network externality and attachment In several studies, argument quality, source credibility, and network externality have all been regarded as a crucial determinant of use intention (e.g., Ayeh et al., 2013; Bhattacherjee and Sanford, 2006; Li, 2013; Lin and Lu, 2011). Furthermore, these factors increase frequency of use in social media. Social media is a tool based on the relationships between users that forms a group identity (Kietzmann et al., 2011); thus, these factors would respectively affect attachment to social media and others. Prospective tourists try to acquire travel information using external sources such as other people, travel agencies, books, and the Internet (Gursoy and McCleary, 2004; Kim et al., 2007; Xiang et al., 2008). In particular, information provided by other users in social media is considered an external source (Kaplan and Haenlein, 2010; Xiang and Gretzel, 2010). Social contacts are crucial when people use others as an external source; thus, the quality of information will lead to increased interaction between members in a virtual community (Chen, 2007). In exploring travel information, a better quality argument (i.e., rich cognitive information) in the information of social media will increase interactions among social media members as they question or share respective travel information. Additionally, the peripheral routes of persuasive messages have a larger influence on social interaction or human affection (Bhattacherjee and Sanford, 2006). In social media, the credibility of information sources is expressed via profiles, photos or messages. This personal information is a factor related to bond-based attachment because reciprocal affection is created by self-disclosure in online communities (Fiedler and Sarstedt, 2010; Ren et al., 2012). In addition, source credibility is an important determinant of attitude in social media (Ayeh et al., 2013). Thus, highly credible travel information sources leads to users' positive attitudes towards social media. This social media reputation in particular builds a positive, bridge-based attachment. Here, the stronger credibility of an information source enables frequent interactions among community members who are engaged in questioning the identity of information providers, and sharing travel information. Users will eventually develop a psychological attachment (both between people in a specific travel group and between specific groups). Existing studies identify the relationship between network externality and attachment (Lin and Lu, 2011). Ellison et al. (2007) measured the intensity of social media use in terms of the number of social media “friends (people connected with me)” and usage time. This means that a higher level of network externality creates a closer bond between the network and the user, which results in an increase in the intensity of use. In addition, Zhao and Lu (2012) insisted that the perception of network size has a positive influence on connectedness in a micro-blog. In addition, Fiedler and Sarstedt (2010) noted that network externality is the cause of attachment. Hence, the following hypotheses are proposed: H6. Argument quality positively impacts attachment. H7. Source credibility positively impacts attachment. H8. Network externality positively impacts attachment.

3.4. Travel informational influence, social media normative influence and attachment Social influences are considered an important element in the formation of user behaviors (Henningsen et al., 2003; Hsu and Lu, 2004; Li, 2013). In addition, Malhotra and Galletta (1999) viewed users' social influence processes as the processes of forming attachment. This is based

5

on their opinion that the processes of compliance, identification, and internalization that comprise social influence processes eventually lead users to continue to use the information system, thereby creating attachment. Among travelers who use social media to search travel information, those who are under a greater informational influence will form a stronger level of attachment. This is because social media users with a greater informational influence are likely to form an attachment by exchanging various types of travel-related information while using social media. In addition, people with a stronger normative influence are inclined to follow their groups or endeavor to receive the positive evaluations of other people, and therefore, their attachment will be further strengthened. Hence, the following hypotheses are proposed: H9. Travel informational influence positively impacts attachment. H10. Social media normative influence positively impacts attachment. 3.5. Travel informational influence, social media normative influence, attachment and behavioral changes Behavioral changes are defined by social media users' modification of their behaviors in relation to the value of social media (Bagozzi and Dholakia, 2002; Qu and Lee, 2011). Furthermore, in the social media environment where people are linked together, people linked to me (reference group) affect my behavioral standards (Park and Lessig, 1977). In searching for travel information on social media, people are influenced by those linked with them, and thereby modify their own behavior when it comes to purchasing travel products, information search methods, and product purchasing methods. Stronger attachments to a social media community, and among its users, would result in a higher frequency of substantial behavioral changes. Hence, the following hypotheses are proposed: H11. Travel informational influence positively impacts behavioral changes. H12. Social media normative influence positively impacts behavioral changes. H13. Attachment influence positively impacts behavioral changes. 4. Research methodology 4.1. Instrument development As presented in Fig. 2, the measurement model was assessed using the first-order model or the second-order model. Argument quality, source credibility, network externality, travel informational influence, social media normative influence and behavioral changes were all measured by the first-order model. The remaining inherent variable, i.e., attachment, was measured with sub-constructs, using a secondorder model. Five argument quality and source credibility items were adopted from the previous study (Ha and Ahn, 2011). Three network externality items were adopted from previous research (Lin and Lu, 2011). In addition, four travel informational influence items related to social media were drawn from previous research (Huang et al., 2010). In addition, three social media normative influence items were drawn from the work of Kim and Chan (2007). Five bridge attachment items were adopted from the previous study (Chang and Zhu, 2012). In addition, five bond attachment items were adopted from the previous study (Chang and Zhu, 2012; Fiedler and Sarstedt, 2010). Finally, four behavioral change items were adapted from the previous research (Qu and Lee, 2011). This study adopted multi-measurement items for each construct in order to overcome the limitations of a single item because a single item is usually too specific to capture all the attributes of a

Please cite this article as: Chung, N., Han, H., The relationship among tourists' persuasion, attachment and behavioral changes in social media, Technol. Forecast. Soc. Change (2016), http://dx.doi.org/10.1016/j.techfore.2016.09.005

6

N. Chung, H. Han / Technological Forecasting & Social Change xxx (2016) xxx–xxx

construct, and is likely to have a high rate of measurement error. All these items were measured on a 7-point Likert scale ranging from strongly disagree (1) to strongly agree (7). 4.2. Data collection This study employed a Korean Internet survey firm to collect data. The firm has a nationwide panel of 999,911 online respondents from which representative samples are selected. Their standard procedures use Korean resident registration numbers matched against personal passwords to verify the identity of the panelists included in each sample. The Internet survey was conducted in June 2012. Online questionnaires were sent to 17,894 potential respondents who had been randomly chosen from the panel of the Internet survey firm. Confirmation of email addresses was received from 7013 respondents and 6936 respondents connected to the questionnaire. For screening, the survey firm selected those who had joined at least one social media for sharing travel information during the past year. At this point, respondents answered the questionnaire with one social media type they often use in mind. With this procedure, 824 questionnaires were collected. After checking for outliers, we verified all questionnaires to ensure that the data sets could be coded for analysis, and after checking for outliers, all 632 questionnaires were coded for analysis (76.7%). The characteristics of the respondents are as follows in Table 2. The respondent gender ratio was male 310 (49.1%) and female 322 (50.9%). The under 30 years old age group had the largest proportion (207, 32.8%), followed by those 30 ~ 39 years old (200, 31.6%) and those 40–49 years old (118, 18.7%). The percentage of respondents who had completed two years of college education or received higher degrees was 75.5%. The following

Table 2 Demographic characteristics of respondents. Characteristics Gender Age

Education

Occupation

Income

Type of using device

Using frequency per week

Amount of time spent per searching

Total

Male Female Under 30 30–39 40–49 Over 50 High school 2 year college University Graduate school Student Office worker Services Technician Professional Self-employed Civil servant Homemaker Other Less than 1 million won 1 million won to 1.99 million won 2 million won to 2.99 million won 3 million won to 3.99 million won 4 million won to 4.99 million won More than 5 million won Desktop/laptop Smart phone Tablet PC Less than 1 time 2 times 3 times 4 times More than 5 times Less than 30 min 30 min to 1 h 1 h to 2 h More than 2 h

percentages indicate the proportions of monthly income levels of the respondents in the sample: 2 million won to 2.99 million won (22.0%), 3 million won to 3.99 million won (16.5%), more than 5 million won (16.1%), 1 million won to 1.99 million won (16.0%), 4 million won to 4.99 million won (15.0%) and less than 1 million won (14.4%). A majority (54.0%) used desktop or laptop for searching travel information using social media, followed by smart phone (271, 42.9%) and tablet PC (20, 3.2%). The following percentages indicate the proportions of weekly usage numbers of the respondents in the sample: 2 times (28.6%), more than 5 times (25.6%), 3 times (24.8%), less than 1 time (14.4%) and 4 times (6.5%). The largest proportion of respondents (53.8%) reported spending 30 minutes to 1 hour per search. 5. Data analysis and results For the statistical analysis, AMOS 18, a maximum likelihood-based structural equation modeling (SEM) software, was used. As presented in Fig. 2, the measurement model was assessed using the first-order model or the second-order model. Argument quality, source credibility, network externality, travel informational influence, social media normative influence and behavioral changes were all measured by the first-order model. The remaining inherent variable, i.e., attachment, was measured with sub-constructs, using a second-order model. The sub-constructs for behavioral changes are bridge and bond. The SEM used a two-step hybrid method by specifying a measurement model in the confirmatory factor analysis (CFA) and testing a latent structural model developed from the measurement model (Kline, 2005). In addition, the maximum likelihood estimation (MLE) was employed to test our research model because MLE is a process that “iteratively improves parameter estimates to minimize a specified fit function” (Hair et al., 2010, p. 614). 5.1. Confirmatory factor analysis

Frequency

%

310 322 207 200 118 107 155 89 332 56 109 232 25 51 56 43 8 85 23 91 101 139 104 95 102 341 271 20 91 181 157 41 162 116 340 132 44 632

49.1 50.9 32.8 31.6 18.7 16.9 24.5 14.1 52.5 8.9 17.2 36.7 4.0 8.1 8.9 6.8 1.3 13.4 3.7 14.4 16.0 22.0 16.5 15.0 16.1 54.0 42.9 3.2 14.4 28.6 24.8 6.5 25.6 18.4 53.8 20.9 7.0 100.0

We assessed the constructs for convergent validity and discriminant validity via CFA using AMOS 18. In CFA, the measurement model is revised by dropping items that share a high degree of residual variance with other items. We dropped nine items that shared a high degree of residual variance. The χ2 fit statistic was 406.657 with 247 degrees of freedom (χ2/d.f = 1.646) (p = 0.000). The goodness-of-fit index (GFI) is 0.950, the adjusted goodness-of-fit index (AGFI) is 0.935, the normed fit index (NFI) is 0.959, the comparative fit index (CFI) is 0.983, the root mean square error of approximation (RMSEA) is 0.032, and the standardized root mean square residual (RMR) is 0.030. All statistics supported the overall measurement quality given the number of indicators (Anderson and Gerbing, 1992). Convergent validity was checked using three other criteria. First, the standardized path loadings must be statistically significant and greater than 0.6 (Bagozzi and Yi, 1988). Second, the composite reliability (CR) and the Cronbach's α for each construct must be larger than 0.7. Third, the average variance extracted (AVE) for each construct must exceed 0.5 (Fornell and Larcker, 1981). As shown in Table 3, the standardized path loadings were all significant and greater than 0.6. The CR and the Cronbach's α for all constructs exceeded 0.7. The AVE for each construct was greater than 0.5. Therefore, the convergent validity of the constructs was supported (Bhattacherjee and Sanford, 2006, p. 815). The discriminant validity of the measurement model is checked by comparing the square root of the AVE for each construct with the correlations between that construct and other constructs. If the square root of the AVE is greater than the correlations between that construct and other constructs, then this indicates discriminant validity (Fornell and Larcker, 1981). As shown in Table 4, the square root of the AVE for each construct exceeded the correlations between that construct and the other constructs. Therefore, discriminant validity was established. In addition, this study checked for common method bias by conducting a Harman's single-factor test in SPSS (Podsakoff et al.,

Please cite this article as: Chung, N., Han, H., The relationship among tourists' persuasion, attachment and behavioral changes in social media, Technol. Forecast. Soc. Change (2016), http://dx.doi.org/10.1016/j.techfore.2016.09.005

N. Chung, H. Han / Technological Forecasting & Social Change xxx (2016) xxx–xxx

7

Table 3 Results of convergent validity testing.a Constructs and variables

Loadings

Argument quality I think that the travel information in social media was: AQ1: Ambiguous – (neutral) - Definited AQ2: Incomplete – (neutral) - Complete AQ3: Inaccurate – (neutral) - Accurate AQ4: Inconsistent – (neutral) - Consistent AQ5: Untimely – (neutral) - Timely Source credibility SC1: The person who wrote the travel information in social media is knowledgeable. SC2: The travel information in social media is written by an expert.d SC3: The person who wrote the travel information in social media is trustworthy. SC4: The person who wrote the travel information in social media is reliable. Network externality NE1: I think a good number of people use Social media. NE2: I think most people are using Social media. NE3: I think there will still be many people joining Social media.d Travel informational influence II1: I am using social media to obtain travel information. II2: I am using social media because contributions by other members help me to make the right travel decisions. II3: I am using social media to benefit from other's experiences before I make travel decisions.d II4: I am using social media to find advice and solutions for my travel problems. Social media normative influence NI1: My friends in social media think that I should manage my online image here. NI2: My friends in social media think that I should present myself here. NI3: Many people in social media think that it is important to manage our online images here. Attachment (2nd order factor) Bridge BR1: Interacting with people at social media makes me want to try new things. BR2: Interacting with people at social media makes me interested in what people unlike me are thinking. BR3: Interacting with people at social media makes me feel like a part of a large community.d BR4: Interacting with people at social media makes me feel connected to the bigger picture. BR5: I am willing to spend time to support general social media activities. BR6: At social media, I come into contact with new people all the time. Bond BO1: There are several people at social media I trust to solve my problems.d BO2: If I needed an emergency loan of 100 thousand won, I know someone t social media I can turn to. BO3: There is someone at social media I can turn to for advice about making very important decisions. BO4: I know several people at social media well enough to get them to do anything important. BO5: I feel very close to the other members of social media.d Behavioral changes BC1: The way I search for information about travel product/services has changed as a result of my being in social media.d BC2: The social media has influenced my behaviour in some ways, such as what things I buy. BC3: Where I buy travel products and service has changed as a result of my being in social media.d BC4: The social media has influenced how I go about buying things. a b c d

CRb

AVEc

α

0.840

0.571

0.837

0.857

0.672

0.836

0.920

0.851

0.919

0.831

0.622

0.828

0.909

0.769

0.909

0.863

0.558

0.860

0.922

0.797

0.917

0.812

0.685

0.809

– 0.792 0.848 0.732 0.634 0.624 – 0.906 0.898 0.927 0.918 – 0.826 0.817 – 0.719 0.903 0.862 0.866

0.747 0.693 – 0.794 0.779 0.717 – 0.850 0.915 0.912 – – 0.773 – 0.879

χ2 = 406.657, d.f = 247 (χ2/d.f = 1.646), p = 0.000, GFI = 0.950, AGFI =0.935 NFI = 0.959, CFI = 0.983, RMSEA =0.032. Composite reliability. Average variance extracted. The item was deleted after confirmatory factor analysis.

2003). This test contains exploratory factor analysis and decides whether a single factor holds a majority of the variance (Podsakoff et al., 2003). The results of principal component analysis using Varimax rotation indicated that the first factor explained only 34.8 percentage of the total

variance. This result indicates that there is no problem of common method bias. Furthermore, this study used the variance inflation factors (VIFs) for the multicollinearity problem and carried out a linear regression analysis. Hair et al. (2010) stated that the cutoff value of VIFs is 10

Table 4 Correlation and descriptive statistics. Construct

Correlation of constructs 1

1. Argument quality 2. Source credibility 3. Network externality 4. Informational influence 5. Normative influence 6. Attachment (Bridge) 7. Attachment (Bond) 8. Behavioral changes

0.756 0.465⁎⁎ 0.263⁎⁎ 0.242⁎⁎ 0.195⁎⁎ 0.287⁎⁎ 0.183⁎⁎ 0.218⁎⁎

2 0.820 0.334⁎⁎ 0.382⁎⁎ 0.352⁎⁎ 0.464⁎⁎ 0.290** 0.269⁎⁎

3

0.923 0.575⁎⁎ 0.285⁎⁎ 0.456⁎⁎ 0.185⁎⁎ 0.340⁎⁎

4

0.789 0.400⁎⁎ 0.553⁎⁎ 0.269⁎⁎ 0.453⁎⁎

5

0.877 0.562⁎⁎ 0.429⁎⁎ 0.552⁎⁎

6

0.747 0.499⁎⁎ 0.579⁎⁎

7

0.893 0.445⁎⁎

Mean

S.D.

4.39 4.37 5.14 5.26 4.60 4.93 4.68 4.69

0.95 0.97 1.10 0.95 1.07 0.89 1.45 1.15

8

0.828

Note. The diagonal elements in boldface in the “correlation of constructs” matrix are the square root of the average variance extracted (AVE). For adequate discriminant validity, the diagonal elements should be greater than the corresponding off-diagonal elements. ⁎⁎ p b 0.01.

Please cite this article as: Chung, N., Han, H., The relationship among tourists' persuasion, attachment and behavioral changes in social media, Technol. Forecast. Soc. Change (2016), http://dx.doi.org/10.1016/j.techfore.2016.09.005

8

N. Chung, H. Han / Technological Forecasting & Social Change xxx (2016) xxx–xxx

and more than 10 of VIFs have the concern about multicollinearity. In analysis, seven variables were used as independent variables and behavioral change was used as a dependent variable. The test revealed that all VIFs were below the 10 and its ranges were 1.302 to 2.195, indicating there is no evidence of multicollinearity. 5.2. Hypothesis testing First, this study examined the assumptions of SEM. According to Hair et al. (2010), these procedures included the checking the variable-tosample ration, normality, linearity, no extreme communalities, and sampling adequacy. The variable-to-sample ratio was 1 to 18.6 and met the Nunnally (1978)’s standard. The value of Kaiser-MeyerOlkin's measure of sampling adequacy was 0.897 and Bartlett's test of sphericity index had a significant with a p value of 0.01 (Kaiser, 1974). Extracted commonalities of all measurement items were ranged from 0.600 to 0.915, which indicated that there are no evidences of extreme multicollinearity or strong linear combinations. The 8% of nonredundant residuals with absolute values was greater the 0.05 and their numbers were 26. The criterion value of nonredundant residuals with absolute values is below 50% and the model is satisfied with the criterion. The result of the measurement model analysis suggested that the measurement model should have some adjustments for this study and then, the original model was analyzed. In terms of goodness of fit, the original model was good except that the level of the chi-square statistic was not suitable for the standard. Analysis of modification indices and estimated path coefficients between latent constructs suggested that eliminating the items AQ1, SC2, NE3, II3, BR3, BO1, BO5, BC1, and BC3 would improve the model. Thus, the model was modified accordingly, becoming a revised model (Table 5, Fig. 3). The χ2 statistic fit was 672.961 with 362 degrees of freedom (p = 0.000). The GFI is 0.934, the AGFI is 0.915, the NFI is 0.937, the CFI is 0.970, and the RMSEA is 0.037. These multiple indicators suggest that the revised model has a good fit, thus, justifying further interpretation. The squared multiple correlations (R2: coefficient of determinant) for the structural equations for travel informational influence, social media normative influence, attachment and behavioral changes are shown in Fig. 3. For travel informational influence, 50.6% of the variance is explained by the direct effects of argument quality, source credibility and network externality. In addition, 19.2% of the variance in social media normative influence is explained by the direct effects of source credibility and network externality. Additionally, 66.1% of the variance in attachment is explained by the direct effects of argument quality, source credibility, travel informational influence and social media

normative influence. Finally, 45.2% of the variance in behavioral changes is explained by the direct effects of travel informational influence, social media normative influence and attachment. Table 5 presents the standardized parameter estimates. Hypotheses H1, H2 and H3 address the structural relationships among argument quality, source credibility, network externality and travel informational influence. The argument quality (H1) did not have a positive effect on the travel informational influence (β = −0.041, t-value = −0.879, n.s.), thus invalidating H1. Nevertheless, source credibility had a positive effect on the travel informational influence (β = 0.263, t-value = 5.301) with statistical significance at the p b 0.001 level; thus, this result supports H2. In addition, H3 is supported by the significant, positive impact of network externality on travel informational influence (β = 0.589, t-value = 13.660, p b 0.001). Hypotheses H4 and H5 address the structural relationships among source credibility, network externality and social media normative influence. Source credibility had a positive effect on intention to social media normative influence (β = 0.319, t-value = 6.790) and was statistically significant at the p b 0.001 level; thus, H4 is supported. Additionally, H5 was supported by the significantly positive effect of network externality on social media normative influence (β = 0.205, tvalue = 4.694, p b 0.001). Hypotheses H6, H7, H8, H9, and H10 address the structural relationships among argument quality, source credibility, network externality, travel informational influence, social media normative influence and attachment. Argument quality (H6) did not have a positive effect on attachment (β = 0.021, t-value = 0.466, n.s.), thus invalidating H6. Nevertheless, source credibility had a positive effect on attachment (β = 0.229, t-value = 4.343) with statistical significance at the p b 0.001 level; thus, this result supports H7. H8 is supported by the significant, positive impact of network externality on attachment (β = 0.104, t-value = 2.038, p b 0.05). H9 was supported by the significantly positive effect of travel informational influence on attachment (β = 0.294, t-value = 4.950, p b 0.001). Additionally, the significant positive impact of social media normative influence on attachment supports H10 (β = 0.439, t-value = 8.794, p b 0.001). Finally, H11, H12 and H13 address the relationships among travel informational influence, social media normative influence, attachment and behavioral changes. Travel informational influence has a positive effect on behavioral changes (β = 0.236, t-value =3.959) at the p b 0.001 level; thus, this result supports H11. Social media normative influence has a positive effect on behavioral changes (β = 0.184, t-value = 3.016) and is statistically significant at the p b 0.01 level, supporting H12. In addition, H13 is supported by the significant positive impact of

Table 5 Standardized structural estimates and tests of the main hypotheses. Hypothesis

Path

Estimates (t-value)

Results

H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13

Argument quality → TII Source credibility → TII Network externality → TII Source credibility → SNI Network externality → SNI Argument quality → Attachment Source credibility → Attachment Network externality → Attachment TII → Attachment SNI → Attachment TII → Behavioral changes SNI → Behavioral changes Attachment → Behavioral changes

−0.041 (−0.879) 0.263 (5.301) 0.589 (13.660) 0.319 (6.790) 0.205 (4.694) 0.021 (0.466) 0.229 (4.343) 0.104 (2.038) 0.294 (4.950) 0.439 (8.794) 0.236 (3.959) 0.184 (3.016) 0.376 (4.356)

Not supported Supported Supported Supported Supported Not supported Supported Supported Supported Supported Supported Supported Supported

R2 TII: SNI: Attachment: Behavioral changes:

0.506 0.192 0.661 0.452

Note. TII: Travel Informational Influence, SNI: Social media Normative Influence.

Please cite this article as: Chung, N., Han, H., The relationship among tourists' persuasion, attachment and behavioral changes in social media, Technol. Forecast. Soc. Change (2016), http://dx.doi.org/10.1016/j.techfore.2016.09.005

N. Chung, H. Han / Technological Forecasting & Social Change xxx (2016) xxx–xxx

9

Fig. 3. Results of SEM Analysis χ2 = 672.961, d.f = 362 (χ2/d.f = 1.859), p = 0.000, GFI = 0.934, AGFI = 0.915, NFI = 0.937, CFI = 0.970, RMSEA = 0.037.

attachment on behavioral changes (β = 0.376, t-value = 4.356, p b 0.001). This study also considered the effects of demographic variables such as gender, age, education, income, and occupation. The results revealed that there were no significant effects of respondents' characteristics. Therefore, these control variables did not have any significant relationships with behavioral changes in social media. 6. Discussion and conclusions The present study established a theoretical model on how the argument quality and source credibility of social media, which is a type of persuasive media, and network externality, which denotes the size of social media, influence the formation of attachment and the behavioral changes of travelers. The study then performed an empirical analysis. Particularly, the study examined the role of mediation that informational and normative influences play in searching travel information through attachment in social media. The results of the analysis revealed some interesting facts such as types of information, causal relationship the types of information and attachment, and traveler's behavioral change. Firstly, argument quality, which indicates the quality of the information provided by social media, affected neither the informational influence nor attachment (H1 and H6 were not supported). This may imply that argument quality as a fundamental cue of central routing (Bhattacherjee and Sanford, 2006; Filieri and McLeay, 2014; Li, 2013; Petty and Cacioppo, 1986; Shu and Scott, 2014) has assumed a relatively lower level of impact compared with social source credibility or network externality. The result of our study, in contrast, is not consistent with another finding from a study of Li (2013). Li (2013) found that argument quality embedded in a persuasive message had a strong influence on informational social influence in the context of organizational information technology usage. However, this finding could not be applied to the context of tourism information because a tourism product is a non-refundable, one time experience.

Therefore, travelers make their decision very carefully to decrease the risk of purchasing a tourism product in a planning stage (Kim et al., 2007; Tan and Chen, 2012). In this way, someone's review affected seriously travel behavior (Jensen, 2012; Jun et al., 2007). Like the results of this study, we argue that people will depend more on peripheral clues (i.e., credibility of the information source and diverse networks), which enables travelers to rely on simpler judgment from reputational reviewers, rather than focus on the quality of professional information. Likewise, people who obtain travel information on social media are more greatly influenced by who posted the information or how many people are registered in the social media community rather than the information itself. Secondly, source credibility had statistically significant impacts on all the variables of informational influence, attachment, and normative influence. It is relatively easy to identify who provides the review information on social media by browsing profiles and photos. Easily identifiable information sources are more useful, and allow people to form emotional reactions and cognitive judgments about the information. These results are important elements that affect people's attitudes in the community, and are in line with existing studies that considered the credibility of an information source (e.g., Li, 2013; Tam and Ho, 2005). Of course, the results of this study show that the credibility of the information is critical when seeking travel information in social media. Thus, people who would like to post their travel information on social media should prove their identity by a profile and photos, which leads to have social influences and to form an attachment toward other users. Thirdly, network externality also had statistically significant impacts on all the variables of informational influence, attachment, and normative influence (H3, H5, and H8 were supported). In particular, the informational influence exhibited a relatively strong impact with β = 0.589, which suggested that the number of members in a social media community can be an important social influence. This finding indicates that the size of social media users exerts informational and normative

Please cite this article as: Chung, N., Han, H., The relationship among tourists' persuasion, attachment and behavioral changes in social media, Technol. Forecast. Soc. Change (2016), http://dx.doi.org/10.1016/j.techfore.2016.09.005

10

N. Chung, H. Han / Technological Forecasting & Social Change xxx (2016) xxx–xxx

influence and creates a space that they form an attachment with others. Fourthly, while both informational and normative influences had statistical significance on attachment (H9 and H10 was supported), the normative influence was revealed to be more significant. While this is contrary to the result of Li's (2013) study where informational influence was greater than normative influence, it could be an interesting result suggesting that in the domain of social media, normative influence is greater than informational influence. That is, attachment is formed by the results of travel informational influence and the compliance with other users and thus, maintaining these influences is important in social media. Fifthly, in using social media, travel informational influence, social media normative influence, and attachment exhibited a statistically significant positive influence on traveler's behavioral changes (H11, H12 and H13 were supported). In particular, the attachment exhibited a relatively strong impact with β = 0.376, which suggested that attachment in social media can be proved. These results also indicate that social influences and attachment have a profound effect on the behaviors of travelers. If travel marketers want to change or affect travelers' behaviors, they need to pay keen attention to the formation of attachment and social influences. To sum up the above results: in searching and utilizing travel information via social media, the influences of source credibility-normative influence, network externality, and attachment were revealed to be greater than that of argument quality-information influence. The theoretical contribution of this study is found in the fact that the processes of travel information search using social media were empirically analyzed by combining the ELM, reference group influence theory and attachment theory. This study is the first to show how these combined theories can be used to gain insights into travelers' behavioral changes using social media for travel information searching. Social media, which is related to travel information search, is recognized as an important travel information source (Xiang and Gretzel, 2010). In the tourism area, existing research has focused on WOM (Words-ofMouth) and information source credibility (Leung et al., 2013b). When using social media as a travel information source as many travelers do, it is necessary to examine the social relationship and affection within social media because social media is a persuasive media that affects behavioral changes. Based on travel information itself and source elements (information quality and source credibility), which are provided by social media and social networking elements of social media (e.g., network externality), there were limited studies about the influences of users' affection, other people's influences on travel information search and related behavioral changes. In order to explain this phenomenon, this study formulated a theoretical framework that could be able to explain tourists' behavioral changes from the perspective that social media is a source to provide the persuasive message with the overarching viewpoints of ELM, social influence theory, and social capital theory. The findings offer practical information for tourism operators with regard to social media. First, the results show that source credibility and network externality have an effect on the travel information influence, social media normative influence and attachment; thus, there is a need in social media to promote source credibility and network externality in preference to argument quality. While tourism related companies and DMOs (destination management organizations) are currently posting relevant travel information, they need more review response by interacting with users and driving marketing promotion. Therefore, travel agencies, DMOs, and users who want to utilize social media as marketing tools for providing travel information, should utilize profile and reputational reviewers that clearly identify an individual profile reflected by the information source and network externality. Second, this study examined whether the informational and normative influence are determinants of attachment and behavioral changes. Both social influences play a role in the formation of attachment and behavioral changes. As expected, other users' positive informational and normative influences lead to many interactions in social media. This stimulates the formation of attachment. In addition, individuals under

these influences change their behaviors in social media. Therefore, tourism marketers should take notice of informational and normative influence in social media and based on this, create marketing plans for promoting their destinations or tourism products. Finally, given the strong effect of attachment on the behavioral changes, tourism-marketing organizations should pay closer attention to building destination attachment with social media users. Therefore, travel marketers can make their information social in order to form the affection through everyday conversation professionally as well as personally with social media users. In addition, they need to manage their online reputation. The limit of this study is that it did not fully reflect the motivation of travelers who use the information provided by social media. This is due to the importance social media users place in character. Additionally, various motives represent different aspects of their behavioral changes that include: whether they search travel information just before actually traveling; whether they search for information about the places they would like to visit in the future; whether they simply enjoy travel information; whether they enjoy communicating with people who share strong homophily in travel itself as opposed to travel information. Therefore, future studies will be required to perform an additional analysis that reflects such individual characteristics. Another limitation is that this study only focuses on social media users in South Korea. Although our samples are adequate to solve our research questions, our findings are difficult to generalize to different cultures. Future studies should examine our proposed research model to validate our results in a different cultural context.

Acknowledgements This work was supported by the National Research Foundation of Korea Grant funded by the Korean Government (NRF-2013S1A3A2043345).

References Anderson, J.C., Gerbing, D.W., 1992. Assumptions and comparative strengths of the twostep approach. Sociol. Methods Res. 20 (3), 321–333. Angst, C.M., Agarwal, R., 2009. Adoption of electronic health records in the presence of privacy concerns: the elaboration likelihood model and individual persuasion. MIS Q. 33 (2), 339–370. Arsal, I., Backman, S., Baldwin, E., 2008. Influence of an online travel community on travel decisions. Information and Communication Technologies in Tourism 2008. Springer Verlag, Vienna, Austria, pp. 82–83. Ayeh, J.K., Au, N., Law, R., 2013. “Do we believe in TripAdvisor?” Examining credibility perceptions and online travelers' attitude toward using user-generated content. J. Travel Res. 52 (4), 437–452. Bagozzi, R.P., Dholakia, U.M., 2002. Intentional social action in virtual communities. J. Interact. Mark. 16 (2), 2–21. Bagozzi, R.P., Yi, Y., 1988. On the evaluation of structural equation models. J. Acad. Mark. Sci. 16 (1), 74–94. Bhattacherjee, A., Sanford, C., 2006. Reviewed influence processes for information technology acceptance: an elaboration likelihood model. MIS Q. 30 (4), 805–825. Bowlby, J., 1969. Attachment and Loss, Volume 1: Attachment. Hogarth Press and the Institute of Psycho-Analysis, London. Casaló, L.V., Flavián, C., Guinalíu, M., 2011. Understanding the intention to follow the advice obtained in an online travel community. Comput. Hum. Behav. 27 (2), 622–633. Chang, Y.P., Zhu, D.H., 2012. The role of perceived social capital and flow experience in building users' continuance intention to social networking sites in china. Comput. Hum. Behav. 28 (3), 995–1001. Chen, I.Y., 2007. The factors influencing members' continuance intentions in professional virtual communities—a longitudinal study. J. Inf. Sci. 33 (4), 451–467. Chen, C.D., Ku, C.S., 2012. Building member's relationship quality toward online community from the elaboration likelihood model perspective. Proceedings of the Pacific Asia Conference on Information Systems (PACIS) 2012. Chen, S., Schechter, D., Chaiken, S., 1996. Getting at the truth or getting along: accuracy versus impression-motivated heuristics and systematic processing. J. Soc. Psychol. 71 (2), 262–275. Chu, S.C., Kim, Y., 2011. Determinants of consumer engagement in electronic word-ofmouth (eWOM) in social networking sites. Int. J. Advert. 30 (1), 47–75. Chung, J.Y., Buhalis, D., 2008. Web 2.0: a study of online travel community. Information and Communication Technologies in Tourism 2008. Springer Wien, New York, pp. 70–81. Deutsch, M., Gerard, H.B., 1955. A study of normative and informational social influences upon individual judgment. J. Abnorm. Soc. Psychol. 51 (3), 629–636.

Please cite this article as: Chung, N., Han, H., The relationship among tourists' persuasion, attachment and behavioral changes in social media, Technol. Forecast. Soc. Change (2016), http://dx.doi.org/10.1016/j.techfore.2016.09.005

N. Chung, H. Han / Technological Forecasting & Social Change xxx (2016) xxx–xxx Ellison, N.B., Steinfield, C., Lampe, C., 2007. The benefits of facebook “Friends:” social capital and college students' use of online social network sites. J. Comput.-Mediat. Commun. 12 (4), 1143–1168. eMarketer, 2013a. Social networking reaches nearly one in four around the world. http:// www.emarketer.com/Article/Social-Networking-Reaches-Nearly-One-Four-AroundWorld/1009976 (Accessed 25.02.14). eMarketer, 2013b. In Asia-Pacific, social media inspires travelers. http://www.emarketer. com/Article/Asia-Pacific-Social-Media-Inspires-Travelers/1009605 (Accessed 25.02.14). Fiedler, M., Sarstedt, M., 2010. Influence of community design on user behaviors in online communities. ICIS 2010 Proceedings, Paper 112. Filieri, R., McLeay, F., 2014. E-WOM and accommodation an analysis of the factors that influence travelers' adoption of information from online reviews. J. Travel Res. 53 (1), 44–57. Fornell, C., Larcker, D.F., 1981. Evaluating structural equation models with unobservable variables and measurement error. J. Mark. Res. 18 (1), 39–50. Gu, H., Ryan, C., 2008. Place attachment, identity and community impacts of tourism—the case of a Beijing hutong. Tour. Manag. 29 (4), 637–647. Gursoy, D., McCleary, K.W., 2004. An integrative model of tourists' information search behavior. Ann. Tour. Res. 31 (2), 353–373. Ha, S., Ahn, J., 2011. Why are you sharing others' tweets?: The impact of argument quality and source credibility on information sharing behavior. ICIS 2011 Proceedings, Paper 4. Hair Jr., J.F., Black, W.C., Babin, B.J., Anderson, R.E., 2010. Multivariate Data Analysis. 7th edition. Prentice Hall, Upper Saddle River, NJ 07458. Henningsen, M.L.M., Henningsen, D.D., Cruz, M.G., Morrill, J., 2003. Social influence in groups: a comparative application of relational framing theory and the elaboration likelihood model of persuasion. Commun. Monogr. 70 (3), 175–197. Hsu, C.L., Lu, H.P., 2004. Why do people play on-line games? An extended TAM with social influences and flow experience. Inf. Manag. 41 (7), 853–868. Huang, Y., Basu, C., Hsu, M.K., 2010. Exploring motivations of travel knowledge sharing on social network sites: an empirical investigation of U.S. college students. J. Hosp. Mark. Manag. 19 (7), 717–734. Hyun, S., Kim, I., 2012. Identifying optimal rapport-building behaviors in inducing patron's emotional attachment in luxury restaurants. J. Hosp. Tour. Res. http://dx. doi.org/10.1177/1096348012451458. Jensen, J.M., 2012. Shopping orientation and online travel shopping: the role of travel experience. Int. J. Tour. Res. 14 (1), 56–70. Jun, S.H., Vogt, C.A., MacKay, K.J., 2007. Relationships between travel information search and travel product purchase in pretrip contexts. J. Travel Res. 45 (3), 266–274. Kaiser, H.F., 1974. An index of factorial simplicity. Psychometrika 39 (1), 31–36. Kaplan, M.F., 1989. Task, situational, and personal determinants of influence processes in group decision making. In: Lawler, E.J. (Ed.)Advances in Group Processes vol. 6. JAI Press, Greenwich, CT, pp. 87–105. Kaplan, A.M., Haenlein, M., 2010. Users of the world, unite! The challenges and opportunities of Social Media. Bus. Horiz. 53 (1), 59–68. Katz, M.L., Shapiro, C., 1985. Network externalities, competition and compatibility. Am. Econ. Rev. 75 (3), 424–440. Kelman, H., 1961. Processes of opinion change. Public Opin. Q. 25 (1), 57–78. Kemp, S., 2016. Digital in 2016. Retrieved from http://wearesocial.com/uk/specialreports/digital-in-2016. Kietzmann, J.H., Hermkens, K., McCarthy, I.P., Silvestre, B.S., 2011. Social media? Get serious! Understanding the functional building blocks of social media. Bus. Horiz. 54 (3), 241–251. Kim, H.W., Chan, H.C., 2007. Why people pay for digital items? Presentation desire of online identity. PACIS 2007 Proceedings. Kim, W.G., Lee, C., Hiemstra, S.J., 2004. Effects of an online virtual community on customer loyalty and travel product purchases. Tour. Manag. 25 (3), 343–355. Kim, D.Y., Lehto, X.Y., Morrison, A.M., 2007. Gender differences in online travel information search: implications for marketing communications on the internet. Tour. Manag. 28 (2), 423–433. Kim, M.J., Chung, N., Lee, C.K., Kim, J.M., 2012. Do loyalty groups differ in the role of trust in online tourism shopping? A process perspective. J. Travel Tour. Mark. 29 (4), 352–368. Kline, R.B., 2005. Principles and Practice of Structural Equation Modeling. Guildford Press, New York. Korea Internet & Security Agency, 2015. 2015 Survey on the Internet Usage. Ku, E.C.S., 2012. Distributed fascinating knowledge over an online travel community. Int. J. Tour. Res. http://dx.doi.org/10.1002/jtr.1895. Lee, M.K.O., Cheung, C.M.K., Sia, C.L., Lim, K.H., 2006. How positive informational social influence affects consumers' decision of internet shopping? Proceedings of the 39th Hawaii international conference on system sciences. Paper 115 Lee, C.K., Kang, S., Reisinger, Y., 2010. Community attachment in two rural gaming communities: comparisons between Colorado gaming communities, USA and Gangwon gaming communities, South Korea. Tour. Geogr. 12 (1), 140–168. Lee, H., Law, R., Murphy, J., 2011. Helpful reviewers in TripAdvisor, an online travel community. J. Travel Tour. Mark. 28 (7), 675–688. Lee, H., Reid, E., Kim, W.G., 2012. Understanding knowledge sharing in online travel communities: antecedents and the moderating effects of interaction modes. J. Hosp. Tour. Res. http://dx.doi.org/10.1177/1096348012451454. Leong, L.Y., Ooi, K.B., Chong, A.Y.L., Lin, B., 2013. Modeling the stimulators of the behavioral intention to use mobile entertainment: does gender really matter? Comput. Hum. Behav. 29 (5), 2109–2121. Leung, D., Law, R., van Hoof, H., Buhalis, D., 2013a. Social media in tourism and hospitality: a literature review. J. Travel Tour. Mark. 30 (1-2), 3–22.

11

Leung, X.Y., Bai, B., Stahura, K.A., 2013b. The marketing effectiveness of social media in the hotel industry: a comparison of facebook and twitter. J. Hosp. Tour. Res. http://dx.doi. org/10.1177/1096348012471381. Li, C.Y., 2013. Persuasive messages on information system acceptance: a theoretical extension of elaboration likelihood model and social influence theory. Comput. Hum. Behav. 29 (1), 264–275. Lin, K.Y., Lu, H.P., 2011. Why people use social networking sites: an empirical study integrating network externalities and motivation theory. Comput. Hum. Behav. 27 (3), 1152–1161. Liu, Z.L., Liu, L., Li, H., 2012. Determinants of information retweeting in microblogging. Internet Res. 22 (4), 443–466. Lu, H.P., Hsiao, K.L., 2010. The influence of extro/introversion on the intention to pay for social networking sites. Inf. Manag. 47 (3), 150–157. Lundgren, S.R., Prishlin, R., 1998. Motivated cognitive processing and attitude change. Personal. Soc. Psychol. Bull. 24 (7), 715–726. Malhotra, Y., Galletta, D.F., 1999. Extending the technology acceptance model to account for social Influence: theoretical bases and empirical validation. Proceedings of the 32nd Hawaii International Conference on System Sciences, pp. 1–11. McGuire, W.J., 1968. Personality and susceptibility to social influence. In: Borgatta, E.F., Lambert, W.W. (Eds.), Handbook of Personality Theory and Research. Rand McNally, Chicago. Mugge, R., Schifferstein, H.N.J., Schoormans, J.P.L., 2010. Product attachment and satisfaction: understanding consumers' post-purchase behavior. J. Consum. Mark. 27 (3), 271–282. National Information Society Agency, 2010. 2010 Yearbook of Information Society Statistics. National Information Society Agency. Nunnally, J.C., 1978. Psychometric Theory. 2nd ed. McGraw-Hill, New York, NY. Park, C.W., Lessig, V.P., 1977. Students and housewives: differences in susceptibility to reference group influence. J. Consum. Res. 4 (2), 102–110. Parra-López, E., Bulchand-Gidumal, J., Gutiérrez-Taño, D., Díaz-Armas, R., 2011. Intentions to use social media in organizing and taking vacation trips. Comput. Hum. Behav. 27 (2), 640–654. Parra-López, E., Gutiérrez-Taño, D., Díaz-Armas, R.J., Bulchand-Gidumal, J., 2012. Travellers 2.0: motivation, opportunity and ability to use social media. In: Sigala, M., Christou, E., Gretzel, U. (Eds.), Social Media in Travel, Tourism and Hospitality. Ashgate, Farnham, Surrey, pp. 171–187. Petty, R.E., Cacioppo, J.T., 1986. Communication and Persuasion: Central and Peripheral Routes to Attitude Change. Springer-Verlag, New York. Petty, R.E., Cacioppo, J.T., Goldman, R., 1981. Personal involvement as a determinant of argument-based persuasion. J. Pers. Soc. Psychol. 41 (5), 847–855. Podsakoff, P.M., MacKenzie, S.B., Lee, J.Y., Podsakoff, N.P., 2003. Common method biases in behavioral research: a critical review of the literature and recommended remedies. J. Appl. Psychol. 88 (5), 879. Prentice, D.A., Miller, D.T., Lightdale, J.R., 1994. Asymmetries in attachments to groups and to their members: distinguishing between common-identity and common-bond groups. Personal. Soc. Psychol. Bull. 20, 484–493. Qu, H., Lee, H., 2011. Travelers' social identification and membership behaviors in online travel community. Tour. Manag. 32 (6), 1262–1270. Ren, Y., Kraut, R.E., Kiesler, S., 2007. Applying common identity and bond theory to design of online communities. Organ. Stud. 28 (2), 377–408. Ren, Y., Harper, F.M., Drenner, S., Terveen, L., Kiesler, S., Riedl, J., Kraut, R.E., 2012. Building member attachment in online communities: applying theories of group identity and interpersonal bonds. MIS Q. 36 (3), 841–864. Rieh, S.Y., Belkin, N.J., 1998. Understanding judgment of information quality and cognitive authority in the WWW. Proceedings of the ASIS Annual Meeting 35, pp. 279–289. Sassenberg, K., 2002. Common bond and common identity groups on the Internet: attachment and normative behavior in on-topic and off-topic chat. Group Dyn. Theory Res. Pract. 6 (1), 27–37. Sheldon, P.J., 1997. Tourism Information Technology. CAB International, Oxford. Shu, M., Scott, N., 2014. Influence of social media on Chinese students' choice of an overseas study destination: an information adoption model perspective. J. Travel Tour. Mark. 31 (2), 286–302. Sussman, S.W., Siegal, W.S., 2003. Informational influence in organizations: an integrated approach to knowledge adoption. Inf. Syst. Res. 14 (1), 47–65. Tam, K.Y., Ho, S.Y., 2005. Web personalization as a persuasion strategy: an elaboration likelihood model perspective. Inf. Syst. Res. 16 (3), 271–291. Tan, W.K., Chen, T.H., 2012. The usage of online tourist information sources in tourist information search: an exploratory study. Serv. Ind. J. 32 (3), 451–476. Wang, Y., Fesenmaier, D.R., 2004. Towards understanding members' general participation in and active contribution to an online travel community. Tour. Manag. 25 (6), 709–722. Werthner, H., Klein, S., 1999. Information Technology and Tourism: A Challenging Relationship. Springer, Vienna. Wood, W., 2000. Attitude change: persuasion and social influence. Annu. Rev. Psychol. 51, 539–570. Xiang, Z., Gretzel, U., 2010. Role of social media in online travel information search. Tour. Manag. 31 (2), 179–188. Xiang, Z., Wöber, K., Fesenmaier, D.R., 2008. Representation of the online tourism domain in search engine. J. Travel Res. 47 (2), 137–150. Yuksel, A., Yuksel, F., Bilim, Y., 2010. Destination attachment: effects on customer satisfaction and cognitive, affective and conative loyalty. Tour. Manag. 31 (2), 274–284. Zhao, L., Lu, Y., 2012. Enhancing perceived interactivity through network externalities: an empirical study on micro-blogging service satisfaction and continuance intention. Decis. Support. Syst. 53 (4), 825–834.

Please cite this article as: Chung, N., Han, H., The relationship among tourists' persuasion, attachment and behavioral changes in social media, Technol. Forecast. Soc. Change (2016), http://dx.doi.org/10.1016/j.techfore.2016.09.005