Abstract
This study explored the tourism experience cocreation process from the tourist-to-tourist interaction (TTI) perspective using a mixed method approach. Two categories of TTIs (verbal and nonverbal TTIs) and the dimensions of verbal TTIs (tourism information sharing and self-disclosure) were first identified through a qualitative study. Afterwards, the mechanism of how TTIs can influence tourist engagement was empirically examined. The findings showed that each type of TTI had a significant positive impact on tourists’ sense of closeness. Both tourism information sharing and nonverbal TTIs were found to positively influence tourists’ sense of control. In addition, sense of closeness and sense of control were found to significantly enhance tourist engagement. Furthermore, nonverbal TTIs were found to moderate the mediation effect of sense of closeness in the relationship between tourism information sharing and tourist engagement.
Keywords
Introduction
The focus of tourism experience has undergone a fundamental shift in recent decades from staged entertainment to cocreation (Binkhorst and Den Dekker 2009). This shift has resulted in the belief that satisfactory tourism experiences cannot be achieved without active participation and engagement of tourists (Mkono 2013; Campos et al. 2015). Specifically, instead of being delivered to tourists (Vargo and Lusch 2008), tourism experiences are typically derived from the process in which tourists construct their own meanings and narratives (Binkhorst and Den Dekker 2009). As a result, tourists are now described as cocreators of their experiences instead of being passive sightseers (Prebensen, Vittersø, and Dahl 2013).
However, despite the burgeoning number of studies that have examined experience cocreation in the field of tourism (e.g., Z. Lin, Chen, and Filieri 2017; Zhang, Meng, and So 2020; Prebensen, Kim, and Uysal 2016), most of them have focused on the interface between tourists and tourism businesses/service providers by examining issues related to tourist engagement in design, production, and consumption of tourism experiences (e.g., Eraqi 2011; Sugathan and Ranjan 2019). Although tourist-to-tourist interaction (TTI) has been found to be an important component of the tourism experience cocreation process (Prebensen and Foss 2011), not much is known about how tourists cocreate their own experiences in collaboration with other tourists (Melvin, Winklhofer, and McCabe 2020). Thus, this study aims to explore the effects of TTIs on the tourism experience cocreation process by using Campos et al.’s (2015) conceptualization of experience cocreation process.
As suggested by Campos et al. (2015), the tourism experience cocreation process can be conceptualized as having two critical components, namely, tourist engagement (both physical and mental) in activities and interpersonal interactions. This framework has been used in several previous studies, with these two dimensions playing a parallel role (Antón, Camarero, and Garrido 2018; Campos et al. 2016). Nevertheless, the level of interpersonal interactions among tourists during the tourism experience cocreation process can influence their level of engagement in experience cocreation (Minkiewicz, Evans, and Bridson 2014). This suggests that rather than playing a parallel role, interpersonal interactions might be a critical determinant of tourist engagement (Campos et al. 2015). Therefore, this study aims to investigate whether TTI is a critical determinant of tourist engagement by examining the impact of TTIs on tourist engagement.
The only study that has explored the impact of TTI on tourist engagement, to the best of the researchers’ knowledge, examined the effect of only one type of TTI, that is, self-disclosure, on tourist engagement (H. Lin et al. 2019), but failed to examine the impacts of other types of TTIs even though other types of TTIs may also play a role in influencing tourist engagement. For instance, Prebensen and Foss (2011) argued that tourists can be better engaged in tourism activities if they are able to gain activity-specific knowledge from more experienced tourists. Furthermore, H. Lin et al. (2019) reported that self-disclosure among tourists affects tourist engagement via the mediating roles of perceived intimacy and perceived cohesion. However, both perceived intimacy and perceived cohesion are rooted in interpersonal relationships. As suggested by self-determination theory (Deci and Ryan 2000), human beings have three basic psychological needs, with the need for relatedness emphasizing the importance of person–person relationships and both the need for autonomy and the need for competence emphasizing the importance of person–environment relationships. Thus, integrating tourists’ need for person–person relationships (i.e., sense of closeness) and for person–environment relationships (i.e., sense of control) is likely to provide new insights into revealing the influencing mechanism of TTI on tourist engagement.
Furthermore, TTI has been found to be a multidimensional construct, with different dimensions resulting in different outcomes (Rihova et al. 2018). Nevertheless, there is a discord among the small number of relevant studies in regard to the dimensions of TTI. Most studies have either proposed many dimensions with little abstraction (e.g., Rihova et al. 2018), thereby hindering the empirical examination of their effects, or have focused on only some of the facets (e.g., Choo and Petrick 2014). In addition, several studies conceptualized TTI using the quantity and quality perspectives (e.g., J. Huang and Hsu 2010; Millán, Garzon, and Navarro 2016), neglecting the content of TTI. It is hence believed a comprehensive conceptual framework of TTI is needed to thoroughly investigate the effects of TTI on the experience cocreation process.
Accordingly, this study seeks to delineate the experience cocreation process among tourists by examining the impact mechanism of various types of TTIs on tourist engagement. First, a multidimensional conceptualization of TTI is developed based on the findings of a qualitative study. Afterward, drawing on self-determination theory, the present study proposes that three types of TTIs influence tourist engagement via sense of closeness and sense of control. A moderated mediation analysis is also employed to assess whether nonverbal TTIs moderate the indirect effects of verbal TTIs (via sense of closeness and sense of control) on tourist engagement.
Literature Review
The concept of cocreation refers to a participative process in which multiple actors collaborate and invest such resources as effort, time, and ability to create products or services collectively (Payne, Storbacka, and Frow 2008). Originating from the service-dominant logic, experience cocreation has become a core topic in marketing and service research (Jaakkola, Helkkula, and Aarikka-Stenroos 2015). Studies have examined its conceptualizations (e.g., McColl-Kennedy, Cheung, and Ferrier 2015), classifications (e.g., Verleye 2015), antecedents (e.g., Cova, Dalli, and Zwick 2011), and issues related to the cocreation process (e.g., Cabiddu, Lui, and Piccoli 2013; Kristensson, Matthing, and Johansson 2008).
In the tourism field, since tourist experience has been found to be a collective and cocreated phenomenon (Helkkula, Kelleher, and Pihlström 2012), both practitioners and tourists have been concerned with generating satisfactory experiences through cocreation (Sugathan and Ranjan 2019). A few studies have investigated the tourism experience cocreation process (e.g., Binkhorst and Den Dekker 2009; Campos et al. 2015; Minkiewicz, Evans, and Bridson 2014). However, most of them were conceptual and/or exploratory, mainly focused on related conceptualizations and classifications (e.g., Binkhorst and Den Dekker 2009; Minkiewicz, Evans, and Bridson 2014). These studies have reported a high correlation between tourism experience and cocreation. For example, Campos et al. (2015) argued that tourists are active participants in their experience creation, which helps them achieve various life goals. Regarding the relevant actors, although research has pointed out that tourism experience cocreation involves a set of relationships among tourists, service providers, and other stakeholders (Sfandla and Björk 2013), the majority of previous studies have focused on the impacts of interactions between tourists and tourism businesses/providers on the cocreation process (e.g., Buonincontri et al. 2017; Cabiddu, Lui, and Piccoli 2013; Mathis et al. 2016).
Tourist-to-tourist interaction and tourism experience cocreation
Understanding the role of tourists in the process of cocreation is potentially a significant impetus to tourist experience research as issues related to tourists’ self-development/self-expression during their travels have been receiving an increasing attention from scholars (Binkhorst and Dekker 2009). The socially dense nature of tourism (Rihova et al. 2013) makes TTI a significant component of experience cocreation (Rihova et al. 2018). Therefore, the examination of the impacts of TTI on tourism experience cocreation is believed to be important in aiding the understanding of the tourism experience cocreation process (Reichenberger 2017). However, little is known about how tourists cocreate their experiences and the effects of interactions among tourists on the cocreation process (Rihova et al. 2015).
Several studies have examined TTI in the context of tourism. For example, Rihova et al. (2018) identified 18 types of TTIs through a qualitative study, and J. Huang and Hsu (2010) found that the quality and the quantity of TTI exert different impacts on tourism experience. Further, Choo and Petrick (2014) suggested that TTIs are realized through status and information. Despite these attempts, research on the parsimonious dimensionality of TTIs remains insufficient since most previous studies have identified a large number of TTI types, which can overwhelm the conceptualization attempts of this concept. Furthermore, the large number of TTI types identified in previous studies makes it hard to place all those TTI types within a single conceptual model that examines their relationships with other constructs, thus hindering further empirical explorations. Moreover, definitive conclusions regarding how various types of TTI contribute to tourism experience cocreation remain scarce. Therefore, this study first explores the dimensionality of TTI in an attempt to identify conceptually sound and parsimonious subdimensions of TTI in study 1 and then empirically examines its impacts on tourist engagement and tourism experience cocreation in study 2.
Study 1: Classification of TTI
Methods
A qualitative study was first conducted to identify the dimensions of TTI. Data for the qualitative study were collected through 16 in-depth interviews. Those TTI dimensions identified through the qualitative study laid the foundation for the quantitative study that would examine how various dimensions of TTIs affect tourist engagement. Interviewees were identified using a snowball sampling technique. To help ensure comprehensive and accurate coverage of TTIs during travel, interviewees were selected from individuals who travel at least twice a year and had traveled within the last six months before the interview. The initial four interviewees were identified from the researchers’ social networks. These four were asked to recommend their acquaintances who met the interview criteria. Six interviewees were recommended. During the interviews, these six were also asked to make recommendations. Six more individuals were recommended. A total of 16 interviews were conducted. However, no new information emerged from the 14th, 15th and 16th interviews. Therefore, the interview process was terminated after the 16th interview. The last three interviews were not included in the analysis since no new information emerged from those interviews. Among the 13 interviewees, there were 6 men and 7 women, all of whom were between the ages of 23 and 55. Approximately 85% of them held a bachelor’s degree or higher. They were located in multiple cities, including Chengdu, Shanghai, Guangzhou, Shenzhen, and Hangzhou. The interviews were conducted either face-to-face or via WeChat, all lasting between 30 and 60 minutes. All interviews were audio-recorded. Then, these recordings were transcribed into a text document for further analysis.
This study adopted a semistructured interview method. The interview questions were identified based on the literature review and expert panel discussions. The interviews were designed to gain a comprehensive understanding of the interviewees’ perceptions, attitudes, or opinions toward TTI. The interviewees were asked the following questions: Do you interact with other tourists when you travel? Under what circumstances do you interact with other tourists? Did you interact with other tourists during your last trip? How did you interact? Please talk about one of your most memorable interaction experiences with other tourists. Additionally, questions that aimed to gather personal information such as age, education, etc. were also asked.
Results
The interview data were analyzed using a thematic analysis. Thematic analysis is a qualitative analytical method that is used to identify, analyze, and report the themes in the data (Braun and Clarke 2006). Following the six steps of the thematic analysis process (i.e., familiarizing oneself with the data, generating initial codes, searching for themes, reviewing themes, defining and naming themes, and producing the report), this study identified two categories of TTIs, namely, verbal and nonverbal TTIs. Two dimensions of verbal TTIs, that is, tourism information sharing and self-disclosure, were also identified (as shown in Table 1). The tourism information sharing dimension refers to verbal TTIs involving tourism-related information sharing, such as sharing information on attractions, traveling, and accommodations. Self-disclosure refers to verbal TTIs that involve sharing self-related information, that is, personal details, opinions, or feelings about oneself, such as information on work, education, and family. Nonverbal TTIs involve the exchange of thoughts and feelings via nonverbal behaviors between tourists, such as eye contact, facial expressions, and physical contact.
Results of the Thematic Analysis.
Study 2: Empirical Examination of the Impact of TTI on Tourist Engagement
Tourist Engagement and Tourism Experience Cocreation
Tourist engagement in tourism activities refers to tourists’ physical and mental participation in specific activities while traveling (H. Lin et al. 2019). To date, research on tourist engagement has mainly focused on developing measurement scales (Taheri, Jafari, and O’Gorman 2014), identifying its antecedents such as authenticity (Bryce et al. 2015) and involvement (Organ et al. 2015), and examining its effect on tourists’ evaluation of outcomes such as satisfaction and loyalty (Rasoolimanesh et al. 2019). Although previous studies have identified the relevance between tourist engagement and tourist experience cocreation, arguing that tourist engagement and interpersonal interactions are two indispensable components of tourist experience cocreation (Campos et al. 2015), empirical explorations of the role tourist engagement plays in the cocreation process remain scarce. As suggested by the extant literature, tourists’ level of engagement in activities helps determine their overall satisfaction with their tourism experiences (Taheri, Jafari, and O’Gorman 2014). Therefore, examining the impact of TTIs on tourist engagement may shed light on how tourists cocreate experiences collectively with each other.
Hypotheses Development
Self-determination theory suggests that human beings have three basic psychological needs: the need for relatedness, the need for autonomy, and the need for competence (Deci and Ryan 2000). The need for relatedness refers to individuals’ need to establish emotional bonds with others. The need for autonomy involves individuals’ need for free will in determining the activities they would like to engage in based on their own desires. The need for competence is defined as individuals’ need to feel that they are capable of controlling their environment and the activities they engage in (Van den Broeck et al. 2016). The need for relatedness reflects individuals’ expectations for connecting with others (Deci and Ryan 2000), that is, interpersonal relationships. By contrast, the need for autonomy and competence, respectively manifesting individuals’ sense of free will and sense of ability to control their environment (Greguras and Diefendorff 2009), describe individuals’ expectations for their relationship with the environment. Since the social environment can provide opportunities for meeting individuals’ basic needs for relatedness, autonomy, and competence (Gagné and Deci 2005), this study argues that TTI, as an important element of social environment, can satisfy tourists’ needs for relationships with fellow tourists and with the environment.
As closeness is a core characteristic of interpersonal relationships (Dubois, Bonezzi, and De Angelis 2016), the concept of sense of closeness can act as a basis for examining interpersonal relationships (Srivastava and Singh 2010). For example, previous studies have employed sense of closeness to represent the relationships between customers and employees (Liu, Guo, and Lee 2011; Mechinda and Patterson 2011) and between tourists and hosts (Wang et al. 2019). Thus, this study adopted sense of closeness to reflect tourists’ subjective perception of their relationship with fellow tourists. Previous studies have used sense of control to determine individuals’ ability to influence and control their environment (Alloy, Clements, and Koenig 1993). Therefore, this study also employs the concept of sense of control to describe tourists’ perception of their relationship with the environment.
Tourists usually travel to places that are away from their daily environment, resulting in the absence of their familiar social networks such as family, friends and colleagues. Thus, traveling may pose challenges for satisfying their need for relatedness, which has been found to be a motivation for seeking interactions with other tourists while traveling (Crompton 1979). As suggested by social exchange theory, interpersonal interactions can promote the development of interpersonal relationships (Cropanzano and Mitchell 2005). Thus, TTI can serve as an important medium for satisfying tourists’ need for interpersonal relationships.
The tourism information sharing dimension of verbal TTIs involves verbally sharing tourism-related information, such as talking about the history and culture of destinations, travel strategies, lodging options, etc. Research has indicated that information/knowledge sharing can be regarded as a type of social support (W. W. Ma and Yuen 2011), which is the manifestation of affective feelings shared between two subjects during social relationships (Gottlieb and Bergen 2010). Therefore, tourism information sharing can serve as an effective medium for building social relationships between tourists (W. W. Ma and Chan 2014), which can yield a sense of closeness to each other. In addition, self-disclosure, the verbal revealing of self-related information, has also been found to effectively foster interpersonal relationships (Altman and Taylor 1973). Through self-disclosure, tourists’ uncertainty about each other can be reduced, thereby improving their liking behavior (Berger and Calabrese 1974). Self-disclosure can also result in the perception that their true self is recognized, understood, and cared for by others (Reis and Shaver 1988), leading to the perception of closeness.
Hypothesis 1a: Tourism information sharing positively relates to sense of closeness.
Hypothesis 1b: Self-disclosure positively relates to sense of closeness.
Nonverbal TTIs can convey cues, information, and meanings that would not be otherwise communicated, thereby affecting interpersonal relationships. Nonverbal cues can help tourists form an impression of others and assess their trustworthiness. For example, kinesics, such as smiling, nodding, and handshake, can foster feelings of reliability, sincerity, and warmness, thereby increasing the perceived trustworthiness of other tourists (Burgoon et al. 2014). Sundaram and Webster (2000) suggested that nonverbal interactions such as eye contact may be viewed as an indicator of people’s willingness and enthusiasm about being close to each other, which can help develop and maintain trustworthiness and sincerity in interpersonal relationships. Hence, this study proposes that nonverbal TTIs predict sense of closeness.
Hypothesis 2: Nonverbal TTI positively relates to sense of closeness.
The possibility of various potential risks in an unfamiliar tourism environment may influence tourists’ perception of how much control they may have while traveling (Lepp and Gibson 2003). Furthermore, tourists may feel that they have little control over their social environment when they are constrained to interact with others (Su et al. 2017). Therefore, interpersonal interactions have been considered to be an important source of control (Rothbaum, Weisz, and Snyder 1982). The present study accordingly argues that TTI can facilitate tourists’ sense of control.
A lack of knowledge about destinations, tourism activities, and traveling can result in the perception of risk and insecurity, which can impair tourists’ perception of their sense of control while traveling (Lepp and Gibson 2008, 2003). As suggested by Bandura (1969), individuals can gain a sense of control by interacting with those who are equipped with adequate knowledge and familiar with issues related to traveling. Thus, this study argues that information sharing among tourists can enhance their level of knowledge about destinations, activities and traveling itself, yielding an improved ability and a sense of control, to deal with the uncertainties present in the environment. In addition, self-disclosure can arouse emotional reactions, making tourists feel excited and pleased about their travel experiences (Tanford and Jung 2017). These positive emotions can effectively improve their sense of control (Namasivayam 2003). Thus, self-disclosure between tourists is also proposed to be related to sense of control.
Hypothesis 3a: Tourism information sharing positively relates to sense of control.
Hypothesis 3b: Self-disclosure positively relates to sense of control.
During social interactions, one’s emotions can trigger corresponding emotions in others (Pugh 2001) via nonverbal interactions (Hatfield, Cacioppo, and Rapson 1994). Service studies have suggested that appropriate nonverbal cues, such as smiling and eye contact, can exert significant positive impacts on customers’ emotions (Hazlett and Hoehn-Saric 2000; Jung and Yoon 2011). Emotions can also guide individuals’ information processing and affect their judgments (Clore, Gasper, and Garvin 2001). Thus, this study expects that nonverbal positive TTI will lead to positive emotions and evaluations of tourists’ relationship with their environment, resulting in a higher sense of control.
Hypothesis 4: Nonverbal TTI positively relates to sense of control.
As aforementioned, tourist engagement involves both mental and physical participation. Research has indicated that a lack of belongingness can distract and consume people’s attention (Hansen and Mossberg 2013). Since closeness and warmth of interpersonal relationships have been found to be critical determinants of belongingness perceptions, sense of closeness is expected to enhance tourists’ mental engagement. Moreover, sense of closeness is usually accompanied by a psychological connection that allows people to have similar behaviors and feelings (Cwir et al. 2011), leading to a high degree of cooperation (Batson et al. 2002). Since tourism experiences are products of a cocreation process (Campos et al. 2015), sense of closeness can encourage a tourist to engage in various activities, both physically and mentally, with other tourists.
Hypothesis 5: Sense of closeness positively relates to tourist engagement.
According to self-determination theory, satisfying individuals’ need for control stimulates their intrinsic motivations (Ryan and Deci 2017), which are the driving forces of their behaviors. Consequently, after gaining a sense of control of the environment, tourists may be motivated to collaborate with others and participate in tourism activities, thus enhancing tourist engagement. As suggested by the resource theory, tourist engagement consumes their personal resources (Prebensen, Vittersø, and Dahl 2013). Since tourists’ perception of their ability to control and influence the environment acts as a personal resource (Hobfoll et al. 2003), the current study proposes a positive relationship between sense of control and tourist engagement. Furthermore, as TTI can have significant impacts on tourists’ sense of closeness and sense of control, sense of closeness and sense of control are likely to mediate the effects of the three types of TTI on tourist engagement as presented in Figure 1.
Hypothesis 6: Sense of control positively relates to tourist engagement.
Hypothesis 7a/b: Sense of closeness (a) and sense of control (b) mediate the impact of tourism information sharing on tourist engagement.
Hypothesis 8a/b: Sense of closeness (a) and sense of control (b) mediate the impact of self-disclosure on tourist engagement.
Hypothesis 9a/b: Sense of closeness (a) and sense of control (b) mediate the impact of nonverbal TTI on tourist engagement.

Conceptual model.
Nonverbal interactions not only directly influence individuals’ judgments and evaluations but also interact with their verbal interactions (Sundaram and Webster 2000). During a TTI, nonverbal interactions provide confirmation and reinforcement of the information communicated verbally. If one’s nonverbal expressions (communications) convey a contradictory message, others may question the sincerity and truthfulness of the verbal information given (Sundaram and Webster 2000). In situations where there is a conflict between verbal and nonverbal messages, individuals are more likely to view the nonverbal message as the true one because of the belief that controlling facial expressions and other nonverbal behaviors is more difficult for people compared to controlling the verbal expressions (Darioly and Mast 2013). This belief results in an attitude that nonverbal expressions reflect the true nature of a message. If there is a doubt about the truthfulness and sincerity of what an individual communicates, people will also take into consideration the individual’s nonverbal behaviors before making their final judgments (Dethmers 2017).
The communicator’s persuasive power can be enhanced by a set of nonverbal cues such as eye contact, gestures, proper tone, etc. (Mehrabian and Williams 1969). Accordingly, nonverbal interactions can greatly influence the effect of verbal interactions. If tourists feel kindness and warmness through nonverbal interactions (e.g., facial expressions) while verbally interacting, they are more likely to believe the authenticity, truthfulness, and sincerity of the verbal information they receive and to exhibit positive emotions toward the person who communicates the information (Grebelsky-Lichtman 2014). As a result, the impacts of verbal TTI (i.e., tourism information sharing and self-disclosure) on sense of closeness and sense of control can be strengthened by nonverbal TTIs. As proposed by hypotheses 7a/b and 8a/b, sense of closeness and sense of control mediate the relationships between tourism information sharing, self-disclosure, and tourist engagement. This study, thus, argues that the mediation effects are positively moderated by positive nonverbal TTI, as presented in Figure 2.
Hypothesis 10a/b: Nonverbal TTI moderates the mediation effects of sense of closeness (a) and sense of control (b) in the relationship between tourism information sharing and tourist engagement.
Hypothesis 11a/b: Nonverbal TTI moderates the mediation effects of sense of closeness (a) and sense of control (b) in the relationship between self-disclosure and tourist engagement.

Moderated mediation model.
Methods
Measurement
A structured questionnaire was used to collect data for model testing. Four items were used to measure tourism information sharing, three of which were derived from Choo and Petrick’s (2014) research. In these items, the phrases “my companion(s)” and “the farm” were replaced by “other tourist(s)” and “the destination” respectively (e.g., other tourist(s) educated me about the destination). The fourth item, “other tourists and I shared travel experience with each other,” was developed from the qualitative research in study 1.
Self-disclosure was assessed with four items (e.g., I intimately disclose who I really am, openly and fully in my conversation) developed by H. Lin et al. (2019).
Nonverbal TTI was measured by four items. These items were borrowed from Jung and Yoon’s (2011) instrument, with the phrase “the employee” replaced by “other tourists” (e.g., other tourists had polite smiles).
Sense of closeness was assessed with three items adapted from Wang et al.’s (2019) scale of perceived psychological closeness, with the subject “the host of this rental” changed to “other tourists” (e.g., other tourists seemed to be similar to me).
Four items from Lachman and Weaver’s (1998) study were adopted to assess sense of control. Items were slightly modified to fit the context of this study by adding the phrase “during my travel” to the original statement to emphasize the trip they were recalling (e.g., I could do just about anything I really set my mind to during my travel).
Tourist engagement was measured by three items that were adopted from S. Huang and Choi’s (2019) scale, which were slightly modified by changing the phrase “during this cruise trip” to “during that trip” (e.g., I actively participated in my favorite activities during that trip).
Responses were measured on a seven-point Likert-type scale anchored by 1 (strongly disagree) to 7 (strongly agree). Additionally, a screening question (Have you traveled in the past six months) was placed at the beginning of the questionnaire to identify qualified respondents. Since the survey was conducted in China, this study followed the standard translation and back-translation procedures to ensure the survey materials were accurately translated from English to Chinese. The respondents were instructed to complete the survey based on their latest travel experience. Before conducting the main survey, a pilot study was conducted to validate the measurement scale. Afterwards, the main study was carried out to test the relationships proposed in the conceptual model.
Pilot study
Sampling
Data for the pilot study were collected at a university in southwest China. Several lecturers were asked to distribute paper questionnaires to their students in their classes. The students who completed the surveys received a small gift (i.e., a bookmark) as a reward. In total, 140 valid responses were obtained. The demographic profile of the respondents is presented in Table 2. Of the 140 students, there were 63 men (45.0%) and 77 women (55.0%). The majority of the respondents (97.9%) were between the ages of 18 and 25 years. Most respondents (96.4%) were undergraduate students.
Respondents’ Profiles of the Pilot Study.
Exploratory factor analysis
First, the distribution of the data was examined. Results showed that the absolute values of the skewness and kurtosis of all items were less than 2 (Hair, Ringle, and Sarstedt 2011), suggesting a normal distribution of the data. Afterwards, an exploratory factor analysis (EFA) was conducted on the pilot study data. A principal components analysis approach with a Varimax rotation was adopted and the eigenvalues-greater-than-one rule was used to determine the number of factors. The KMO value was 0.864, suggesting that the data set was suited for EFA. Bartlett’s test of sphericity provided a significant χ2 value of 2547.614 (p < 0.001, df = 231). Because of issues of low factor loading and cross loadings, 4 items were eliminated. The remaining 22 items were used to identify six factors. Those six factors explained 81.543% of the variance, and corresponded to the six constructs included in the conceptual model. The results of the EFA are presented in Table 3.
EFA Results.
Main study
Sampling
Data for the main study were collected in June and July 2019, via Sojump, China’s largest online survey platform. Sojump has been used widely for data collection in previous tourist studies (e.g., Shi, Gursoy, and Chen 2019; Shi, Gong, and Gursoy 2020). Sojump’s consumer panel includes millions of registered members with a wide range of demographic and geographic backgrounds (Shi et al. 2020). Data were collected only from those panel members who had traveled in the past six months. In all, 398 valid responses were collected.
Demographic profile of the respondents is presented in Table 4. Across the entire sample, 187 of respondents were male (47.0%) and 211 were female (53.0%). Most respondents were between the ages of 18 and 34 (75.9%). A large portion of respondents held a bachelor’s degree (44.5%) or an associate degree (39.4%). In regard to occupation, most were employed at various businesses (73.6%). The distribution of annual household income was relatively uniform. The proportions of respondents with an annual household income below 30,000 RMB, between 30,000 and 79,999 RMB, between 80,000 and 149,999 RMB, between 150,000 and 199,999 RMB, were respectively 18.4%, 17.6%, 23.9%, and 18.8%, while the remaining 21.3% of the respondents’ annual household income was above 200,000 RMB.
Respondents’ Profiles of the Main Study.
Measurement model
Using Mplus 7.4, the main study employed a covariance-based structural equation modeling (CB-SEM) approach to analyze the data. Before the CB-SEM analysis, the normality assumption was checked. All items had absolute skewness and kurtosis values less than 2 (Hair, Ringle, and Sarstedt 2011). The CB-SEM analysis was carried out in two steps. First, the measurement model was assessed by a CFA analysis to examine the reliability and validity of the constructs. Values of goodness-of-fit indices for the measurement model suggested a good model fit: χ2/df = 1.74 (χ2 = 337.302, df = 194, p < 0.001); root mean square error of approximation (RMSEA) = 0.043, standardized root mean square residual (SRMR) = 0.042; comparative fit index (CFI) = 0.959, Tucker–Lewis index (TLI) = 0.952. As presented in Table 5, factor loadings of all constructs were between 0.658 and 0.820, indicating good internal consistency. The average variance extracted (AVE) values of the study constructs were greater than 0.50, suggesting adequate convergent validity (Fornell and Larcker 1981). All constructs’ composite reliability (CR) values were higher than 0.70, indicating that all measurement items had satisfactory reliability. Additionally, as presented in Table 6, the square roots of all constructs’ AVEs were greater than their corresponding correlation coefficients, indicating good discriminant validity (Hair et al. 2014).
Results of the Measurement Model.
Note: CR = composite reliability; AVE = average variance extracted; TIS = tourism information sharing; Sd = self-disclosure; NI = nonverbal tourist-to-tourist interaction; SCL = sense of closeness; SCO = sense of control; TE = tourist engagement.
Results of Discriminant Validity.
Note: The diagonal elements (in bold) are the square roots of all constructs’ average variances extracted.
Structural model
The model fit of the structural model was also satisfactory: χ2/df = 2.14 (χ2 = 376.127, df = 198, p <0.001); RMSEA = 0.048, SRMR = 0.048; CFI = 0.949, TLI = 0.941. Furthermore, most hypotheses were supported (Table 7). Results showed that all three types of TTI positively impact sense of closeness (tourism information sharing: β = 0.294, p < 0.01; self-disclosure: β = 0.338, p < 0.001; nonverbal TTI: β = 0.270, p < 0.05), supporting hypotheses 1a, 1b, and 2. However, although both tourism information sharing (β = 0.380, p < 0.01) and nonverbal TTI (β = 0.309, p < 0.05) exerted significant influence on sense of control, supporting hypotheses 3a and 4, self-disclosure was not significantly related to sense of control (β = 0.036, p>0.05). Thus, hypothesis 3b was rejected. In addition, hypotheses 5 and 6 were supported since both the sense of closeness (β = 0.322, p < .01) and the sense of control (β = 0.615, p < .001) were found to increase tourist engagement.
Results of the Structural Model.
Mediation effects
The PROCESS macro can provide coefficients and confidence intervals to assess the mediation, moderation, and moderated mediation effects (Hayes 2013). Results produced by the PROCESS macro have been found to have higher reliability than the results produced by traditional approaches such as Baron and Kenny’s approach (1986) (MacKinnon, Fairchild, and Fritz 2007). Thus, this study used the PROCESS macro for SPSS with bootstrapping (with 5,000 resamples) to test the mediation and moderated mediation effects. As revealed in Table 8, hypotheses 7a and 7b were both supported since both sense of closeness (coefficient = 0.0879, SE = 0.0278, 95% CI [0.0357, 0.1438]) and sense of control (coefficient = 0.1831, SE = 0.0347, 95% CI [0.1182, 0.2541]) mediated the relationship between tourism information sharing and tourist engagement. Self-disclosure had a significant indirect effect on tourist engagement via sense of closeness (coefficient = 0.2244, SE = 0.0314, 95% CI [0.1632, 0.2859]), supporting hypothesis 8a. Moreover, nonverbal TTI had a significant indirect impact on tourist engagement via sense of closeness (coefficient = 0.0944, SE = 0.0339, 95% CI [0.0260, 0.1595]) and sense of control (coefficient = 0.1957, SE = 0.0379, 95% CI [0.1254, 0.2743]), supporting hypotheses 9a and 9b.
Results of Mediation Effects.
Note: SE = standard error; CI = confidence interval; ULCI = upper limit of CI; LLCI = lower limit of CI; TTI = tourist-to-tourist interaction.
Moderated mediation effects
Results suggested that the effect of tourism information sharing on tourist engagement via sense of closeness was positively moderated by nonverbal TTI as the moderated mediation effect was significant (coefficient=0.0265, SE = 0.0143, 95% CI [0.0004, 0.0562]), supporting hypothesis 10a. Specifically, as presented in Table 9, the mediation effect was stronger when there was higher level of nonverbal TTI. However, the interaction effect of tourism information sharing and nonverbal TTI on tourist engagement through sense of control (coefficient = 0.0089, SE = 0.0259, 95% CI [–0.0433, 0.0606]), and the interaction effect of self-disclosure and nonverbal TTI on tourist engagement through sense of closeness (coefficient = 0.0239, SE = 0.0265, 95% CI [–0.0316, 0.0708]) were not significant, rejecting hypotheses 10b and 11a.
Results of Moderated Mediation.
Note: Tourism Information Sharing * Nonverbal TTI – Sense of Closeness – Tourist Engagement. TTI = tourist-to-tourist interaction; SE = standard error; CI = confidence interval; ULCI = upper limit of CI; LLCI = lower limit of CI.
Conclusions
Discussion
The findings of the qualitative study (study 1) reveal that TTIs can be classified into two categories, namely, verbal and nonverbal TTIs. This classification is consistent with previous research, which contended that interactions involve both verbal and nonverbal exchange of information, emotions, and feelings between the interacting parties (Ballantyne and Varey 2006). This study further identifies two types of verbal TTIs, that is, tourism information sharing and self-disclosure. This is also in line with existing research findings that the content of communications between customers is either product-related or self-related, which can achieve different goals (Chen 2017).
SEM results revealed that both verbal TTI types have significant positive impacts on sense of closeness, supporting Yoo et al.’s (2014) argument that verbal interactions can promote interpersonal relationships. The findings are also consistent with studies that have suggested both task-oriented and social-oriented communications are important elements in facilitating the development of relationships (Burgoon and Hale 1987) since tourism information sharing represents a task-oriented interaction, while self-disclosure is a social-oriented interaction. Additionally, the findings of this study indicate that nonverbal TTIs can also enhance tourists’ sense of closeness. This may be explained by the fact that nonverbal behaviors can convey certain conventional information during social interactions. For instance, eye contact during social interactions often results in positive reactions that can enforce the feeling of closeness, similarity, participation, arousal, etc. (Burgoon and Hale 1987).
The findings also revealed that both tourism information sharing and nonverbal TTIs positively influence sense of control, while no significant association was found between self-disclosure and sense of control. The nonsignificant effect of self-disclosure on sense of control can be explained by the fact that self-disclosure is about information related to tourists themselves, with no effect on improving tourists’ familiarity with and predictability of the environment. On the other hand, tourism information sharing refers to the exchange and sharing of tourism-related information about destinations, tours, travel strategies, etc. with other tourists. Mastering more information can improve people’s predictability of the environment (Baronas and Louis 1988). Hence, tourism information sharing can serve as an important medium for enhancing tourists’ sense of control. In addition to information sharing, appropriate nonverbal behaviors can also convey positive emotional messages (Chartrand and Lakin 2013) that can improve tourists’ positive emotions, which increase their perception of sense of control.
It was also found that sense of closeness had a positive impact on tourist engagement. This can be explained by Islam, Rahman, and Hollebeek’ (2017) proposition that positive interactions among customers can keep them active and engaged in collective activities. The present study also revealed that sense of control can significantly enhance tourist engagement, which supports the findings of previous studies that the higher the sense of control is, the more willing individuals are to participate in relevant activities (Tucker et al. 2012).
This study also provides empirical evidence for the mediation effects of sense of closeness and sense of control in the relationships between the three types of TTI and tourist engagement. As stated earlier, sense of closeness and sense of control are both basic psychological needs of human beings. The unfamiliarity and strangeness of the tourism world, reflected in two aspects, namely, “people” and “environment,” are likely to weaken tourists’ sense of closeness and sense of control, which may increase the perception of risks associated with traveling (Lepp and Gibson 2003). The perception of risk can reduce tourists’ enthusiasm and willingness to participate in activities (Seabra, Abrantes, and Kastenholz 2014). However, TTI can enhance tourists’ engagement and willingness to participate in tourism activities through increasing their sense of closeness and sense of control.
Finally, this study reveals that nonverbal TTIs moderate the mediation effect of sense of closeness in the relationship between tourism information sharing and tourist engagement. Prior research suggests that individuals’ positive emotions can be enhanced by appropriate nonverbal interactions (Chartrand and Lakin 2013), thereby improving their internal states and evaluations. Accordingly, the mediation effect of sense of closeness in the relationship between tourism information sharing and tourist engagement is stronger when nonverbal TTIs are high (vs. low). However, nonverbal TTIs are unlikely to moderate the mediation effect of sense of control in the relationship between tourism information sharing and tourist engagement. This may be because tourism information sharing is relatively objective, tourists’ sense of control of the environment and thus tourist engagement can be increased as long as they obtain tourism-related information from other tourists. Additionally, the mediation effect of sense of closeness in the relationship of self-disclosure and tourist engagement cannot be moderated by nonverbal TTIs either. This finding can be explained by the fact that self-disclosure allows people to get to know each other quickly, reducing their dependence on nonverbal ques when making judgments.
The data for this study were collected before the outbreak of COVID-19, which may limit the generalizability of the above findings during the pandemic. For instance, the impact of tourism information sharing on tourists’ sense of control may become weaker because the latter is likely to be impaired by perceived risks of interacting with others because of the COVID-19 pandemic. Additionally, to prevent the spread of COVID-19, tourists are encouraged and even forced to wear facemasks and are asked to maintain a social distance, reducing the likelihood and influence of nonverbal TTIs. Specifically, tourists cannot accurately capture other tourists’ facial expressions. However, COVID-19 will most likely eventually recede, and tourists will be able to interact with other tourists both verbally and nonverbally again, since social interaction is a critical component of the tourism experience cocreation process (Bernstein et al. 2018). Thus, this study argues that, to a large extent, the results pertaining to how TTIs can influence tourist engagement will again be applicable to the tourism setting when life returns to the normal after the COVID-19 pandemic recedes.
Theoretical Implications
This study provides insights into the question of how tourists cocreate their experiences. Previous literature has acknowledged the significance of the tourism experience cocreation process (Sugathan and Ranjan 2019), with TTI being identified as a critical component of cocreation (Rihova et al. 2018). Nevertheless, compared with the cocreation processes between tourists and businesses/employees, much less is known about the cocreation process from the TTI perspective (Melvin, Winklhofer, and McCabe 2020). This study responds to Campos et al.’s (2015) call for examining the mutual influence of social interaction and active engagement in activities. The findings show that TTI affects tourists’ mental and physical engagement in activities via the mediating roles of sense of closeness and sense of control.
The findings of this study also advance TTI research in the following three ways. First, this study deconstructs the concept of TTI by including three dimensions, contributing to the theoretical development of the TTI literature. Since the identification of unique and distinct dimensions of a construct can lead to novel findings (MacInnis 2011), the TTI taxonomy identified in this study can facilitate future research on revealing the complex relationships between TTI and other constructs, thus expanding the TTI literature. Second, this study incorporates TTI into the tourism experience cocreation research, identifying the mechanism of how tourists cocreate experiences through interacting with each other. This study is also a response to Rihova et al.’s (2018) call for investigating the roles of specific types of TTIs by identifying the differing effects of the three types of TTIs on experience cocreation. Third, most studies on social interactions have regarded verbal and nonverbal interactions as parallel dimensions (Z. Ma and Dubé 2011; Mattila and Enz 2002), rarely considering their interaction effects. This study finds that nonverbal TTIs moderate the mediation effect of sense of closeness in the relationship between the one type of TTI, that is, tourism information sharing, and tourist engagement, enriching the theoretical understanding of TTI.
This research also contributes to the tourism engagement literature by identifying TTI as one of its antecedents and recognizing the mediating role of sense of closeness and sense of control, thereby linking TTI to tourist engagement, which is a response to Melvin, Winklhofer, and McCabe’s (2020) appeal that tourist engagement research needs to take other tourists into consideration.
Practical Implications
This study also provides some practical implications for tourism service providers. First, since TTI is categorized into three distinct types, that is, tourism information sharing, self-disclosure, and nonverbal TTI, with differing effects, service providers should facilitate these three types of TTI differently based on the goals and needs each type of TTI is likely to satisfy. Specifically, the findings of this study show that each type of TTI is effective in enhancing sense of closeness. Therefore, to improve tourists’ sense of closeness, service providers can initially identify tourists who are familiar with the destination, or have rich travel experiences, or are knowledgeable about traveling, and then guide them to share their knowledge with other tourists. Additionally, they can take several measures to promote self-disclosure among tourists, such as designing game sessions of self-introductions, offering other interactive games or tourism activities that need collaboration and interactions among tourists. Moreover, they may need to remind tourists to show positive nonverbal behaviors to each other, such as making eye contact, smiling, and maintaining a proper interpersonal distance. However, since self-disclosure was found to have no impact on sense of control, businesses should only enhance sense of control by promoting tourism information sharing and through positive nonverbal TTI among tourists.
Second, tourism service providers should not only pay attention to tourists’ needs for products or services but also recognize and try to meet tourists’ psychological needs as social beings. This study finds that TTI positively affects tourist engagement through satisfying tourists’ needs for sense of closeness and sense of control. Tourists’ sense of closeness and sense of control may be challenged when they are in an unfamiliar destination. This can limit their engagement in tourism activities, which can negatively affect their experience and satisfaction. Tourism service providers should be aware of this and take actions to improve tourists’ sense of closeness and sense of control by providing opportunities that can help tourists increase their knowledge and familiarity with the destinations, activities, and other tourists.
Tourism service providers can develop strategies to make tourists more familiar with other tourists. For instance, they can make a booklet that includes basic information about all group members and distribute it to every member before a trip. This can help group members become familiar with each other before personally interacting, which can promote a sense of closeness. Tourism service providers can also develop activities to strengthen tourists’ knowledge about the destination and traveling, and, thus, enhance their sense of control over the environment. For example, tourism service providers can provide tourists with brochures about the destination in order to improve their knowledge. Tourism service providers could also use incentives such as providing small rewards to those who can correctly answer certain simple questions relevant to the content of the brochure or games to motivate tourists to read the brochure.
Limitations and Future Research
This study has several limitations. First, although this study followed the saturation principle during the interview process, the geographical distribution of the sample may not be wide enough because of limited accessibility. Future research can include tourists from different types and sizes of cities and even from the countryside, to obtain more generalizable results. Second, the respondents in this study were Chinese, which may limit the generalizability of the findings. Future studies may want to validate the conceptual model in other cultural contexts and countries. Third, TTIs in this study were limited to positive ones. Future research should integrate both positive and negative TTIs into the same framework in order to further expand the TTI knowledge and literature. Finally, the relationships examined in this study may have changed during the COVID-19 pandemic. Future research is encouraged to explore what changes may have occurred during this period.
Footnotes
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 disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by The National Social Science Foundation of China (No. 20XGL010).
