The Relationship between Ritual, Personal Involvement and Travel Intention: A Study of Check-in-Travel on DouYin

This study aimed to understand the relationship among ritual, personal involvement and travel intention by examining short video clip users’ behavior. Data was gathered from 254 regular users and nonusers of Douyin, which is the most popular Chinese short video platform. Findings show that ritual is positively related to personal involvement, while both ritual and personal involvement are positively related to travel intention. Personal involvement was found to partially mediate the relationship between ritual and travel intention. Implications for destination managers, limitations and future research directions are discussed.


Introduction
Social media and Web 2.0 as technological concepts become more and more familiar with tourists. The increase of tourists prefers to use social media to search for travel information, make a travel plan and share their travel experience. Social media is reshaping our travel form (Kietzmann, Hermkens, McCarthy, & Silvestre, 2011;Xiang & Gretzel, 2010). Tiktok was a short video app, which allows users to record 15-second videos and upload them for the whole world to see. More than 500 million people globally use Tiktok monthly. In Tiktok, lots of people participate topics they are interested in by imitating the similar behavior of others.
DouYin which is Tiktok's Chinese version also has a million users monthly, and lots of users like to share their travel experience in Douyin. In Douyin's tra-

Ritual
As "Little Prince" mentioned, ritual can bring us sacredness: they are what make one day different from other days, one hour from other hours. Indeed, the ritual is initially and traditionally studied in religion and then mythology and psychology started paying attention to ritual (Belk, Wallendof, & Sherry Jr., 1989;Bell, 1992;Matthews, 2017). The earlier research has focused on the consequences of group ritual with the anthropological, sociological, and psychological domain in which ritual is been seen as sacred and most associated with religion (e.g. Goffman, 1955;Durkheim, 1912;Smith, 1890).
Recently, organizational research (Blake, 2014;Feldman, 1977) and marketing research (Ratcliffe, Baxter, & Martin, 2019;Rook, 1999) have a tendency to focus on rituals. This is, People's participation in religious group ritual activities has gradually decreased, and the sacred sense brought by religious group rituals is difficult to meet people's needs. People create rituals in their everyday life to seek a sense of sacredness (Belk et al., 1989). Everyone is a consumer, it is easy to create ritual behavior in our consumption. For example, eating Oreo should follow the American Journal of Industrial and Business Management ritual behavior of "twist, lick and soak". Actually, consumption is general and has even been explained as the needed ritual of modern life (Wright & Snow, 1980). In this period, studies focus on the benefits of ritual for individuals in daily living.
Whether we are as individuals or as members of some larger group, we are unaware of participating in ritual activities in our daily life, at home, work, and study. Additionally, from the small-scale and individuals, such as gift-giving, making tea in the morning, to wider, collectivization, such as annual festive celebrations, rituals are present in our daily experience (Kapitány & Nielsen, 2015;Sherry Jr., 1983). During this process, typologies of ritual and the causal inferences about rituals have become clear. Ritual in holiday, gift-giving, food and drink all have been verified can enhance consumption, meanwhile can bring more satisfaction for the consumer (Harrison, Reilly, & Gentry, 2010;Bradford & Sherry Jr., 2013;Tang, 2017).

Ritual in the Travel-Related Literature
Tourism is an industry with great coverage and involves multiple consumptions, such as accommodation, transportation, food, and drink, etc. Since rituals are used to explain consumer behaviors, it is reasonable to apply ritual in tourism consumption. Ritual in tourism research has been mostly associated with religion travel (Moufahim & Lichrou, 2019;Robb, 2001). Meanwhile, some scholars notice there exist rituals in tourism consumption. Graburn (2004) regards tourism as a special ritual and proposes the syllogism of tourism: "secular-sacred-secular". It has been verified that during tourism which is a kind of secular consumer behavior, tourists will appear on the way anomalous phenomena that are completely different from the secular (Graburn, 1983). But there is few studies focus on the ritual in tourist consumption, such as go to some certain destination, find the same scene, buy the same food, pose the same gestures and so on. Different scholars give different definitions of rituals based on their own research perspectives (e.g. anthropology, religion, sociology, psychology), each helps to understand the ritual structure. This leads scholars to suggest that "there is no clear criterion by which cultural anthropologists or other scholars of religion or classics determine that particular type of behavior is or is not an instance of a ritual" (Liénard & Boyer, 2006).
Accommodating past theoretical research and practical conceptualizations of ritual, we identify three key criteria for ritual: 1) a fixed, episodic sequence of behaviors; 2) symbolic meaning 3) repeat over time (Cohn, 1990;Rook, 1985;Crews & Boutcher, 1986;Rossano, 2012). First, a fixed and episodic sequence is the most and also easily characterizes ritual. For example, when we celebrate a birthday, everyone follows this sequence: buy a cake-put the candles-make a wish-blow out the candles-eat the cake. Second, by acting this birthday ritual, we are getting the rite of passage which bring the sacredness. Finally, we repeated this behavior years over years.
Based on the definition of the ritual, scholars have put forward different views on the ritual structure. One is to explain the sacredness of rituals from a reli-gious perspective (Hobson et al., 2018;Belk et al., 1989). Rook (1985) divides ritual into 4 dimensions through the perspective of consumer behavior. Recently, Collins (2004) based on "Scene" claimed another ritual structure through the perspective of micro-sociology (see Table 1). Each structure is valuable to help us understand rituals in our daily life which include tourism. Though Rook (1985)'s framework of ritual scale is broad, this structure suggests commonalities in ritual activities. It enables us to study ritual behavior from different perspectives, such as tourism perspectives. Thence we use Rook's ritual framework to measure ritual in tourism consumption.

Ritual versus Personal Involvement
Ritual appears in our daily life. Most rituals have a boundary which we can clearly know who is in the ritual and who is not (Collins 2004), this boundary can be seen in community worship or individual birthday party.
Due to the boundaries of rituals, some studies have shown that personal involvement can increase the perception of rituals, while rituals can also increase the involvement of ritual audiences (Dohle et al., 2014;Vohs et al., 2013). Ritual and personal involvement are inseparable. Sherif and Cantril (1947) were the first scholars concerned about personal involvement. Since then, involvement has received extensive attention in marketing and been defined variously. While there is no agreement has been reached on the definition of involvement, it is generally believed that involvement is people's values to something such as an issue, idea or activity (Sherif & Cantril, 1947). That is, involvement is the connection of people to a particular thing based on their past experiences, interests, and values (Zaichkowsky, 1985). In tourism, we can find clues that more and more travelers are following others to travel through social media. So, we predict that when the user using social media plan their travel, the video with more ritual behaviors will enhance their interest (which is personal involvement) about travel destinations that this video played.
Hypothesis 1: Ritual is related positively to personal involvement. Hypothesis 1a: Ritual artifacts is related positively to personal involvement. Hypothesis 1b: Ritual script is related positively to personal involvement. Hypothesis 1c: Ritual performance roles are related positively to personal involvement.
Hypothesis 1d: Ritual audience is related positively to personal involvement.

Involvement versus Travel Intention
Involvement is an important component of the travel experience (Gunter & Gunter, 1980). Involvement is widely used in the field of tourism, the relationship between involvement and sundry tourism issues are been studied. Among them, personal involvement shows a positive effect on satisfaction, loyalty, and travel intention et al. (Huang et al., 2012;Chen & Tsai, 2008;Lee et al., 2008). So, we predict that when the user using social media plan their travel or see someone travel video, the more they interest (involve) in the more travel intention they have.
Hypothesis 2: Involvement is related positively to travel intention.

Ritual versus Travel Intention
As mentioned above, in general, ritual research is still in its infancy, and the bulk of research focuses on the definition and structure of rituals. In terms of the effect of the ritual, most researches concern about the social outcomes of rituals.
Recently, the impact of rituals on individuals, especially consumers, has been concerned. From a personal perspective, Ratcliffe et al. (2019) think ritual produces certain outcomes by effect, cognition, and behavior. Ritual enhance consumption has been verified (Vohs et al., 2013). Tourism as a special consumption contains a lot of kinds of consumption, such food, drink, and traffic et al. Watson and Kopachevsky (1994) have argued that "modern tourism is best understood in the context of the commodification process and contemporary consumer culture". In fact, most of the researches in the tourism field are interesting in consumer goods and tourism consumption (Richards, 1996). So, we predict that the ritual travel video comparing to the normal one on social media will enhance user's consumption which is travel intention.
Hypothesis 3: Ritual is related positively to travel intention. Hypothesis 3a: Ritual artifacts is related positively to travel intention. Hypothesis 3b: Ritual script is related positively to travel intention. Hypothesis 3c: Ritual performance roles are related positively to travel intention.
Hypothesis 3d: Ritual audience is related positively to travel intention. American Journal of Industrial and Business Management

The Mediating Effect of Personal Involvement
We predict ritual can enhance travel intention. However, due to differences in personal characteristics, different people may feel the different extent of ritual although they see or do the same ritualized behaviors. Because involvement may play an intermediary role. When we saw something, the different people got different involvement which can influence our later attitudes or even behaviors (Slama & Tashchian, 1985). When something is consistent with our own hobbies, we would have a high involvement with it. Highly involvement brings us more positive and strong attitudes toward it. These attitudes then may influence our future plans (Sherif, Kelly, Rodgers, Sarup, & Tittler, 1973). So, we think when the user sees the same ritual video, personal involvement plays a mediating role between ritual and travel intention.
Hypothesis 4: Involvement plays a mediating role between rituals and tourism intentions.
All in all, based on previous literature and practical experience, this study builds a structural relationship between rituals, personal involvement, and travel intentions. At the same time, we divide the ritual into four dimensions according to Rook (1985)'s theory: ritual artifacts, ritual script, ritual performance roles and ritual audience.

Sample and Procedure
In Douyin, there are lots of popular travel videos. Users are willing to go to the same destination to imitate the video's context, such as purchase the same product, pose the same gesture, capture the similar video content. One of them is Xi'an "super-bowl wine". This "super-bowl wine" is consumed in a fixed sequence: 1) the wine is served in a special porcelain bowl; 2) everyone will buy one bowl of wine, after that, they will wait in line and go to the designated place for drinking; 3) in their turn, they will hold the wine above their heads first and then acquiesce to their wishes; 4) they drank the wine in one time and broke the porcelain bowl. We can see the typical ritual behaviors in this tourism consumption based on our definition of ritual before. One of the "super-bowl wine" video clip has been clicked 81.113 million times in Douyin. So, we selected the "super-bowl wine" video to measure ritual.
Responders were chosen based on a convenience sampling method. Travelers who are visiting oct loft (a cultural and creative park in Shenzhen) are our main sampling frame. We chose Shenzhen oct loft as the main research place because oct loft is one of the most popular destination in Shenzhen which attract many people to visit. Before they formally answer the questionnaire, we will first introduce the purpose of this study and show them the video of "super-bowl wine". Respondents were told to feel free to ask us questions during their do this survey. Once they have questions about the questionnaire, an explanation will be given to them which has been told to every investigator in advance. Questionnaires were collected onsite. We returned 312 questionnaires with anonymity American Journal of Industrial and Business Management and confidentiality. We eliminated 58 data missing questionnaires. Finally, we got 254 (81.4%) valid questionnaires.

Instrument
Since the original instruments we used were in English, the respondents were Chinese, we use the backward and forward translation approach to ensure the translation was authentic (Hayashi, Suzuki, & Sasaki, 1992).
To ensure the accuracy of the instrument, we conducted a pilot study. Fifty students were invited to participate. They were told any questions and confusion in the process of filling out the questionnaire can tell us at any time, such as the wording and clarity of the survey questionnaires.
Ritual experience in tourism is a perceived consumption value for tourists as consumers (Ratcliffe et al., 2019;Belk et al., 1989;Rook 1985). And we use the Perceived Value scale (Voss et al., 2003) to measure the ritual framework. The Perceived Value scale includes two dimensions (Hedonic and utilitarian) with 10 items. This instrument divides consumers' perceived value into hedonic and utilitarian. Marketing and tourism research use this instrument to measure consumer perceived value (Okada, 2005;Oh, Fiore, & Jeoung, 2007;Hosany & Witham, 2010;Grohmann, 2009;Alba & Williams, 2013).
In this study, we use this scale to measure people's attitudes toward ritual. A 5-item short form of the Personal Involvement Inventory (Zaichkowsky, 1985) was utilized to measure personal involvement. This PII scale has been widely used in marketing and tourism research (Sweeney & Soutar, 2001;Mano & Oliver, 1993;Sparks, 2007). Travel intention was measured by using the purchase intention scale (Petroshius & Monroe, 1987), which has one dimension with 3 items. The item was worded as follows: "How likely would you be to take a journey according to the popular video context in the next 12 months?", "How likely would you be to take a travel in different cities based around popular video context in the 12 months?", "How likely would you recommend your friends to take a travel to the city according to around popular video context?". All the scales were measured using Likert type scale from 1 strongly disagree through to 5 strongly agree. In this study, travel intention was a stand-in for the actual travel plan. There are two reasons. First, there are many factors that influence people making a travel plan, and this would create some sort of response bias. At the same time, in our research, we found that the person who actually travels will affect his actual willingness to travel because of certain travel experience. By asking their travel intention we can reduce questions mentioned before (Wee, Ta, & Cheok, 1995).

Data Analysis
In this study, we used the Statistical Package for Social Sciences (SPSS) for sample descriptive analysis, reliability and validity analysis, and factor analysis. These analyses are the basis for subsequent correlation regression analysis. Finally, the Process in SPSS was used to run linear regression and intermediary inspection to regulate the relationship between variables.

Responders' Profile
Of the 254 completed and valid questionnaires, 60.4% of the samples were females. Among respondents, most of their ages were under 30 years old (90.2%).
77.2% of the samples have a college or university degree. Almost three-fourth of them spent last than 5000 CNY per month. Totally the respondents' profile is basically consistent with Douyin user portraits (Penguin Intelligence, 2018).

Internal Consistency and Construct Validity
First, we use Cronbach's reliability coefficients to exam the internal consistency of each construct. Table 2  are also above the generally agreed upon lower limit of 0.70 (Hair, Anderson, Tatham, & Black, 1984). Thus, all the scales have good reliability for subsequent analysis. Table 2 also presents the correlations between ritual, personal involvement, and travel intention. Results showed that ritual artifacts, ritual script, ritual performance roles, ritual audience, and personal involvement were all correlated significantly to the travel intention in a positive direction.
Then, we run the validity analysis to verify the internal structure of the concepts in which EFA (exploratory factor analysis) and KMO (Kaiser-Meyer-Olkin) were conducted (Anderson & Gerbing, 1988). This item will be eliminated if it meets both of the following conditions: a) its corrected item-total correlation was less than 0.4; b) it has large coefficients across multiple dimensions at the same time (Byrne, 2001). Consequently, only one item from ritual artifacts (not fun-fun) were eliminated, other items are all retained. Table 3 reveals the final outcomes. The KMO of the ritual's four dimensions were rung 0.899 to 0.927.
Personal involvement (0.777) and travel intention (0.706) all have good validity which was both higher than the recommended index of 0.60 (Garson, 2001).

Hypothesis Testing of Relationships
To test the hypothetical relationships, linear regression was conducted by taking the maximum likelihood estimation method. Table 4 shows the results of the hypothesis. As hypothesized, ritual had a positive and significant effect on personal involvement (β = 0.399, t value = 6.884, p = 0.000). Hypothesis 1 was supported. Additionally, ritual artifacts, ritual script, ritual performance roles, and the ritual audience had a positive and significant effect on personal involvement. Hypothesis 1a, 1b, 1c, 1d were supported. Hypothesis 2 predicted that personal involvement was positively related to travel intention which is also been supported (β = 0.687, t value = 15.008, p = 0.000). Hypothesis 3 proposed that ritual can enhance our travel intention. Data analysis results show that this hypothesis is verified (β = 0.430, t value = 7.537, p = 0.000). Also, ritual artifacts, ritual script, ritual performance roles, and the ritual audience had a positive and significant effect on travel intention. Hypothesis 3a, 3b, 3c, 3d were supported. Ritual was hypothesized to affect travel intention through personal involvement. The direct, indirect and total effects of ritual on travel intention were 0.2211, 0.2916 and 0.5127 (0.2211 + 0.2916). Personal involvement was an important mediator variable to travel intention due to the indirect effect (0.2916) are higher than the direct effect (0.2211). However, there is still a significant correlation between rituals and travel intentions which means that personal involvement is partial rather than full mediation (Hair et al., 2006). In this part, we also tested the mediating effect of personal involvement between ritual artifacts, ritual script, ritual performance roles, ritual audience, and travel intention. American Journal of Industrial and Business Management

Mediating Effect
The results of the mediation analysis are illustrated in Table 5. Ritual performance roles (0.4291) and ritual audience (0.4081) have more potential to predict travel intentions than ritual artifacts (0.3951) and ritual script (0.3855). Based on the above evidence, Hypothesis 4 was partially supported.

Summary of Result
As tested before, the four dimensions of the ritual have a positive correlation with personal involvement and travel intention, and personal involvement can positively predict travel intention, which indicates that hypothesis 1, hypothesis 2, hypothesis 3 has all been verified. At the same time, we verified that personal involvement plays a part of the mediating role between ritual and travel intentions, which indicates that hypothesis 4 has been verified.

Discussion
The earliest studies of rituals were related to religion (Durkheim, 1912;Deflem, 1991). Since ritual was been found appear in our daily consumer life, the impact of the ritual on people's daily consumption is studied (Vohs et al., 2013;Darmody & Bonsu, 2007;Marshall, 2005;Tetreault & Robert III, 1990). However, as far as we know, the rituals in tourism have not yet received much attention. To expand the ritual research area, the current study first clarified the definition of ritual in tourism, then examined the relationship among ritual, personal involvement and travel intention based on previous research results.
In this study, we divide the ritual into four dimensions: Ritual artifacts (RA), Ritual script (RS), Ritual performance roles (RPR) and Ritual audience (RAU) base on Rook (1985)'s research. Our research results show that tourists can perceive all the four dimensions of rituals in tourism activities.
As expected, the ritual's four dimensions all had a powerful and positive relationship with personal involvement. At the same time, we found among them ritual script (RS) and ritual performance roles (RPR) were more powerful for investigating the relationship between ritual and personal involvement. When people see others' travel experiences in Douyin, the ritual can increase their involvement. That is people are more likely to find more information about destination if they feel more ritual in others' travel experiences. Especially ritual performance roles (RPR) and ritual audience (RAU) can increase tourist involvement more effectively. Consistent with other scholar's research results, the relationship between personal involvement and travel intentions in this study has also been verified (Yen & Teng, 2015;Ferns & Walls, 2012;Huang et al., 2010). That's means that tourist who is involved in the ritual are more likely to plan their trip. The findings also indicate rituals can predict not only personal involvement but also travel intentions. Further, compared with ritual artifacts and ritual scripts, ritual audiences and ritual roles perceived by people are more possible to plan their travel by watching the same travel video.
In summary, we found that ritual and personal involvement are related to travel intention. What's more, ritual enhances travel intention through personal involvement. In other words, if a tourist has highly ritual-perception, he or she is more involved in the travel destination, which is more possible to turn to travel intention finally. At the same time, the mediation effect results expose that the

Conclusion
There are several theoretical implications of this study. First, the division of ritual dimensions (ritual artifacts, ritual script, ritual performance roles, and ritual audience) in tourism activities have been verified in this study. In subsequent studies, scholars can analyze ritual behavior in tourism according to this dimension. Second, the relationship model between ritual, personal involvement, and travel intentions was verified, which enriches ritual theoretical research. By reason of personal involvement has played a part in mediation, it implies that there may be some mediators that could be included in the current model to better predict travel intention. Finally, we found some follow-up behaviors in tourism can be explained by ritual theory. In this study, we can find that part of the reason why people are willing to follow social media to travel today is that people feel a sense of ritual in the travel videos shared by others. This perception of ritual increased their willingness to travel. Based on the findings of the study, a number of practical implications can be derived. Ritual is a useful tool to enhance personal involvement and travel intention. In tourism marketing, tourism companies can create more ritual activities based on the ritual structure in this research and use social media to spread. Moreover, the positive effect of personal involvement on travel intention implies that tourism companies should provide more convenient ways for tourists to understand relevant tourism information to increase tourism involvement. For example, adding vacation destination product links to related tourism videos.
Although the results of the current study confirmed that ritual may enhance personal involvement and travel intention in the context of tourism, there are a number of limitations. First, the sample of our research is limited to Shenzhen oct loft, and the diversity of the sample needs to be further strengthened. Second, the questionnaire recycle time is almost three months. The time period is a little short. A longitudinal design in future studies would be more powerful in identifying relationships among the constructs of the current study. Third, this paper uses the perceived value scale to measure ritual, and subsequent research can continuously clarify the ritual behavior in tourism and generate a tourism ritual scale. Additionally, with the development of ritual theory, there may be more reasonable ritual structures suitable for explaining tourism behavior.