INTERACTION AMONG FELLOW CRUISE PASSENGERS: DIVERSE EXPERIENCES AND IMPACTS

Abstract

By employing a social‐psychological approach to the study of tourist experience, the present study investigated the social interaction between fellow cruise passengers and the impact of such interaction on cruise vacation experience as perceived by the customers. Results based on content analysis of interviews and virtual focus group discussions showed that cruise passengers had diverse experiences meeting fellow passengers, which were categorized into three broad scenarios. The levels of interaction among passengers generally followed Levinger and Snoek's (1972) model of relationship development, starting from unilateral impression, through surface contact (superficial and role‐bounded interaction), to increasing mutuality (relationship beyond that demanded by externally structured role prescription). Apparently, interaction among fellow passengers only entails special meaning when it exceeds the common role script for unacquainted travelers, and interactional partners identify each other as unique individuals. This exploratory study demonstrated that the social dimensional approach of studying tourist behavior was able to unfold a neglected aspect of the tourist experience, i.e., how interaction between tourists encountered in the tourism space contributed to the cruise experience to various extent.

Keywords:

• Tourist‐to‐tourist interaction
• customer‐to‐customer interaction
• cruise experience

INTRODUCTION

Leisure travel is essentially social behaviors (Fridgen, 1984). The social, interactive aspects of the tourist experience were highlighted by Stringer and Philip (1984), who argued that, from a social‐psychological point of view, tourism was rarely experienced and interpreted directly and individually, but rather through encounters with other people. Pearce (2005) recognized that in a tourism context, three types of social contact exist: tourist‐local community, tourist‐service personnel, and tourist‐tourist. However, existing socio‐psychological research on tourism was fragmentary (Stringer & Philip, 1984), and themes under investigation mainly focused on interactions or relationships between tourists and indigenous population (mostly from the disciplines of sociology and anthropology), and between tourists and service personnel (e.g., in the services marketing literature). Investigation about the interaction among tourists themselves seemingly lags far behind.

Nevertheless, the importance of tourist‐to‐tourist (t2t) interaction was highlighted by several scholars. Nettekoven (1979) stated that besides intercultural relations between the indigenous population and foreign tourists, relations among tourists themselves were equally significant, because “stereotypes of the indigenous population are formed by tourists and disseminated through intertourist communication” (p. 135). Pearce (1982) argued that the company of fellow tourists served as a buffer insulating tourists from difficulties during the travel, and other tourists played a role of mediating and shaping tourist‐host contacts.

From a broader perspective, interaction among tourists can be seen basically as interaction among customers, which is a common phenomenon in many servicescapes but remains an under‐researched area in the services marketing literature (Martin & Pranter, 1989). Customer‐to‐customer (c2c) interaction could be central, additional, or nonexistent in customers' service experience, depending on the service settings (Jones, 1995). Several studies (e.g., Grove & Fisk, 1997; Martin & Pranter, 1989; Clark & Martin, 1994) provided descriptive accounts of service settings in which c2c interaction may be a primary component of the service experience, which raises the importance of customer compatibility management (e.g., Pranter & Martin, 1991). One aspect of c2c interaction that has received relatively more attention in recent years was observable oral participation (OOP) among customers, mainly in a retail setting. (For a comprehensive review, refer to Harris, Baron, and Parker, 2000.) If c2c interaction is an issue of concern for transitory customer encounters in retail settings, it is likely to bear more in the experience of fellow tourists who share companionship en route.

Studies on c2c interaction indicated that cruise vacation may be an area where c2c interaction could exert substantial influence on customer experience. Martin and Pranter (1989) outlined the characteristics of services for which customer compatibility could be an issue of importance. For example, customers are in close physical proximity to each other, engaged in varied activities, and expected to share time, space, or service utensils with one another. Likewise, Clark and Martin (1994) made a list of customer compatibility sensitive industries, including passenger transportation, retail, accommodation, recreation, education, health, and financial services.

Figuratively called a floating hotel or floating resort, a cruise ship is a combination of accommodation, transportation, and recreation. The relatively long period of time‐sharing, the self‐contained space, and the complexity of activities both onboard and at ports of call denote that interactions among cruise passengers are expected to be of diverse varieties. In addition, normative and situational factors affect a person's openness to meet others. A situation like a vacation may increase a person's receptivity to stranger others (Levinger & Snoek, 1972). Thus, the characteristics of cruise travel apparently lend itself to an ideal setting for investigating the interaction among unacquainted customers.

By employing a social‐psychological approach to the study of tourist experience, the present study resonates with Wearing and Wearing's (1996) call for a feminist approach to studying tourist experience, focusing on the social interaction that occurs between fellow tourists encountered in the tourism space. The research objectives of the study were to construct the multiple realities of the phenomenon under study (i.e., interaction among cruise passengers) as experienced by cruise customers themselves, and to explore the impact of such interaction on cruise vacation experience as perceived by the customers. The scope of the present study was limited to the North American cruise market, because it constituted over 70% of the world cruise market (World Tourism Organization, 2003).

LITERATURE REVIEW

The notion that tourists generally expect to interact with people of the host community rather than fellow tourists may be an overgeneralization. Arguably, some tourists may prefer to interact with fellow tourists of the same national background rather than foreign hosts (Brislin, 1981, cited in Reisinger & Turner, 2003, p. 46). Similarly, Cohen (1972) commented that international tourists traveling together with others of their own culture were virtually transposed to the foreign soil in an environmental bubble of their native culture. The same phenomenon was revealed by Crompton (1979) that interacting with local people was difficult to achieve due to little common identity between the tourists and hosts, whereas interacting with other tourists on the same trip was more likely, resulting in a natural tendency to “turn inward rather than outward” (p. 419) in companionship and interaction.

While some studies (e.g., Baker & Crompton, 2000; Cole & Scott, 2004) suggested that social group interaction played an important role in determining the quality of the tourist experience, not many studies focused specifically on this issue. As Koth, Field, and Clark (1992) noted, three types of tourism activities—group tour, backpacking, and cruise—were areas in which interactions among tourists were likely to be common and “potential development of social structure might occur” (p. 105). Similarly, Sørensen (2003) commented that social interaction among fellow tourists in the same three areas provided “mobile settings for tourist cultures to unfold” (p. 9). The following presented a review of studies on t2t interaction in the three areas, where most of the literature on t2t interaction was found.

Studies on group tours often found that interaction among group members was a crucial theme in the whole tourist experience. Holloway (1981) recognized that physical proximity within the coach tour setting provided a framework for social interaction among tourists, which was seen as an essential element of the holistic tourist experience. Chan (2003) showed that group cohesion was an antecedent to customer satisfaction with tour experiences. Both Holloway and Chan highlighted the tour guides' role as catalysts that encouraged social cohesion within tour groups, which was one of the four major roles of the tourist guide conceptualized by Cohen (1985). However, these studies often focused on tour services, especially the role of the guides in the service delivery process, while t2t interaction was not their central research focus. One early exception was made by Gorman (1979), who unfolded how the camaraderie and sense of shared purpose developed during the trip contributed significantly to the tourists' travel experience. Quite a number of these studies focused on coach tours (e.g., Gorman; Holloway; Quiroga, 1990), apparently because the confined environment of the bus as well as plenty of shared time on the journey facilitated group formation and at the same time segregated the tour group from the world outside.

Backpacking was another area where t2t interaction was awarded due attention. By employing social situation analysis, Murphy (2001) revealed that social interaction among backpackers appeared to be an important factor in choosing backpacking as a means of travel, although the interactions pursued may be for either functional or social benefits. Sørensen (2003) investigated how a backpacker travel culture evolved through impromptu social interaction along the road among backpackers. Loker‐Murphy and Pearce (1995) found that both mixing with locals and with fellow backpackers were elements of the best backpacking experiences. T2t interaction was also investigated in studies on budget travelers/wanders/drifters, which overlapped with the backpacker concept. For example, Vogt (1976) analyzed the characteristics of the transient yet intense interpersonal encounters among backpackers/wanders, and Riley (1988) found friendship struck quickly among budget travelers on the road, and communication networks among travelers were a salient feature for this mode of travel.

Research systematically investigating relationships among cruise passengers was quite limited. One notable exception was made by Yarnal (2003). Through participant observation, the author illustrated that the playful cruise ship space fostered social interaction among like‐minded passengers and being part of a cruise group contributed to the repeat behavior of “old timers.” Other studies provided insights about interaction between passengers, although this topic was not their focus of investigation. For example, Foster (1986) described the relationships among a group of senior passengers on a 23‐day cruise as being friendly, informal, and agreeable; yet at the same time, superficial and guarded. For these passengers, interests in nature and the local community weighed far more heavily than the social companionship from fellow passengers. From an anthropological point of view, the author studied the cruise culture for a “short‐lived society,” whereas the personal relevance of c2c interaction and its impact on customer experience remained untouched.

Somewhat contrarily, a study by Koth et al. (1992) showed that for cruise passengers to Alaska, interests in attractions and ports decreased after visitation to the initial ports; and for the phase of the trip traveling back to the home port, customer focus shifted to entertainment and social events. The authors observed that onboard receptions and team games opened up communication channels among passengers and aided in group formation, and after a relatively long period of time, “social networks and information channels become comparatively stabilized” (p. 105). Bauer's (2001) study revealed that, for a cruise trip to Antarctica, interaction with fellow passengers was cited by 20% of the respondents as the most enjoyable activity onboard, while interaction with the crew was cited by 7% of the respondents as the most enjoyed activity.

The issue of t2t interaction has received more attention in recent years. For example, Arnould and Price (1993) found that a feeling of community with other tourists (and with guide) was one component of an extraordinary river rafting experience. Pearce (2005) provided a comprehensive review of studies on tourists' social contacts and relationships, and presented an array of tourists' multiple perspectives on other tourists, ranging from positive to neutral to negative views. The author argued for the social dimension of tourist behavior as a “greater area of scholarly interest and a focus for potential management action” (p. 134).

In summary, previous research on t2t interaction was rudimentary, and in many cases these studies were not fully devoted to the issue of t2t interaction. While the issue of t2t interaction surfaced naturally from investigations on other themes (e.g., tour guide roles, tourist‐host interaction), the actual level and nature of t2t interaction and its unique impact on tourist experience remained to be fully explored. By focusing on interaction among fellow passengers confined in a “cocoon” cruise ship for an extended period of time, the present study attempted to obtain an in‐depth understanding about interaction among tourists and its impact on tourist experience.

DATA COLLECTION

Focus groups could be used to achieve a number of different purposes. One was for preliminary exploration of a topic area about which little was known from previous research. The advantage of focus groups for such purpose lies in their efficiency and capability to solicit multiple perspectives based on interaction among participants (Bloor, Frankland, Thomas, & Robson, 2001). Efforts were made from April to June 2006 to recruit participants for face‐to‐face focus group discussions. Potential connections and sites were contacted or visited for the purpose of recruiting eligible participants (North American residents, 18 years or older, with cruise vacation experience) in the city where the authors resided. Several eligible people were found but only one agreed to participate. (A face‐to‐face interview was eventually conducted with this only volunteer.) To circumvent the practical difficulty of convening face‐to‐face focus groups with North American cruise customers, alternative methods were sought.

Virtual Focus Groups

A number of online cruise communities was available with a large repertoire of cruise customers and their comments and stories about cruise travel. The existence of active computer‐mediated support communities indicated that virtual focus group (VFG) could be a feasible data collection technique for the research question under study (Bloor et al., 2001). To test the feasibility of these online cruise communities as sources of data collection, one researcher registered on one cruise forum and successively posted two discussion topics (one of which related to the focus of the present study). Reaction from the online cruise community was quite optimistic, with more than 30 responses from May 30 to June 13, 2006. Subsequently, participants from similar cruise forums were recruited to join VFG discussions in a private forum hosted on the authors' institutional website. Efforts were made to target people who were currently active members in the public forums, appeared to be American/Canadian residents, were experienced or “information‐rich” (Patton, 1990, p. 169) cruise vacationers, or had recently taken a cruise vacation.

The transition from a public discussion board to a private one was made based on several considerations. First, participants in public online forums are usually anonymous, which makes it difficult to verify their identity and obtain any background information of relevance. Second, public forums rely on voluntary participation from the subjects involved; but without serious commitment from participants, any lengthy and in‐depth investigation would be difficult to accomplish. Third, due to the open accessibility to public forums, it would be impossible to protect participants' confidentiality as required by ethical research conduct. In line with the dominating medium of communication in existing online cruise forums, asynchronous discussion board was adopted in the present study, in which participants could logon and post their responses any time at their convenience.

The first focus group discussion started with seven participants on June 26, 2006. Two volunteers failed to register at the beginning due to technical problems, and another one quit midway. Eventually, four participants completed the first VFG discussion. The second group started with seven participants but one person disappeared after her introductory posting. The third group starting with five volunteers was less optimistic. Three participants quit due to various reasons at the opening of the discussion, and another one withdrew midway. Left with only one participant, the third VFG eventually became an online individual interview. The first two VFGs took 25 and 28 days, respectively, and the third one took 16 days. (For more detailed discussion about VFGs as the research method, refer to Huang & Hsu, 2007.)

Individual Interviews

Because the VFG participants were recruited from online cruise forums, this might set them aside from people who do not go online. Finding other sources of participants could reduce the bias resulting from a limited sample (Morgan, 1997), and data from different sources on the same research issue also served to deepen understanding of a topic (Bloor et al., 2001). Therefore, individual interviews were conducted as a complementary device, which was a good way to tap subjects' reconstruction of experiences in the past (Lincoln & Guba, 1985). To supplement the one face‐to‐face interview conducted earlier, six telephone interviews were conducted. Interviewees were recruited through referral by friends and academicians. The average duration for the seven interviews was about 38 minutes. All interviews were audio‐recorded and later transcribed.

Interview Guide

The same interview guide was used for both VFGs and individual interviews. Open questions were asked first, followed by more specific follow‐up and probing questions. This was a funnel approach introduced by Morgan (1997) in which the discussion began with a less structured approach and moved into a more structured discussion of specific questions. The focus of discussion was the most recent cruise vacation taken by participants, although previous cruise experiences were also tapped when appropriate. Participants were asked to tell their story about the fellow passengers they met (if any), and also comment on the impact of these interactions on their cruise experience.

Following the suggestion of Krueger (1998), the interview guide was pilot‐tested. A mock interview was conducted with the first author acting as the interviewer and a PhD student as the interviewee. Two other PhD students were asked to observe the interview and commented about the language (e.g., clarity of the wording, double‐barrelled and leading questions), sequence of the questions, and timing. The interview guide was subsequently reviewed by one experienced cruise customer and one cruise industry expert; and further revisions were made based on their comments.

In summary, the study was based on a small sample constituting three VFGs (with 4, 6, and 1 participants, respectively) and seven interviews. The mixed techniques of data collection provided triangulation by using different sources (online cruise community members versus people who do not go online) and methods (focus group versus individual interview), thus enabling the imperfection and bias of one source or method to be cancelled out by the strengths of another (Lincoln & Guba, 1985). In addition, asynchronous VFG discussion possesses several advantageous communicational characteristics unmatched by the traditional face‐to‐face discussion. For example, Ryan and Valverde (2005) argued that asynchronous methods give participants more time to reflect before responding; thereby the depth of the discussion could be enhanced. Such reflection (rather than mere reporting) of one's experience is essential for complex process and feelings to be given meanings and be optimally articulated (Crouch & McKenzie, 2006). At the same time, the prolonged engagement with the participants in the virtual space (ranging from 16 to 28 days) provided the researcher with the opportunity to build trust with participants and to familiarize herself with the context under study. These factors could help build the trustworthiness of an inquiry (Lincoln & Guba).

Although the sample size was small, this limitation may not seriously hamper the validity of the present study because its main purpose was to develop descriptions and explanations to provide insight and understanding of the phenomenon under study (Sobal, 2001), and upon which to formulate propositions rather than set out to verify them (Crouch & McKenzie, 2006). In such exploratory concept‐generating studies, a small sample may even bring advantage rather than weakness because it permits greater involvement of the researcher and all the emerging concepts “can be kept in the researcher's mind as a totality under investigation at all stages of the research” (Crouch & McKenzie, p. 493). There exist trade‐offs between the breadth (the number of units examined, e.g., the number of interview participants), the depth (the intensity of inquiry about a specific topic), and the scope (the broadness of coverage of substantive areas) of the extensiveness of data collection (Sobal). Similarly, Patton (1990) suggested that the “validity, meaningfulness, and insights generated from qualitative inquiry have more to do with the information‐richness of the cases selected and the observational/analytical capabilities of the researcher than with the sample size” (p. 185). In the kind of research conducted herein, what were sampled were not the individual participants, but rather “variants of a particular social setting” and “of the experiences arising in it” (Crouch & McKenzie, p. 493). The sample of the present study, albeit small, manifested cases or instances of states that covered various dimensions of the situation under investigation; namely, the varying levels of the social interaction and relationship between fellow passengers, which intersected with Levinger and Snoek's (1972) model of relationship development as discussed later. Such intersection with pre‐existing theoretical knowledge enhanced the validity of and confidence in the research findings (Crouch & McKenzie).

DATA ANALYSIS

The overall data set for analysis constituted the content of three VFGs and seven interviews, as well as responses (from 10 participants) to questions posted on the public cruise forum. Data analysis followed mainly an inductive approach, in which theoretical categories and relational propositions were arrived at by inductive reasoning based on the data (Lincoln & Guba, 1985). The general processes of data analysis followed those proposed by Glaser and Strauss (1967) and Strauss and Corbin (1998)—including open coding, followed by axial coding, and finally conceptual integration. First of all, the transcript was unitized. Namely, the text was read thoroughly and then divided intuitively into units of information or incidents for subsequent analysis. Then the constant comparative method proposed by Glaser and Strauss was used to develop concepts and categories, involving mainly three steps.

First, at the early phase of data coding, incidents were compared to other incidents to develop categories or concepts that could unify a set of incidents. Second, after the concepts and their properties were well defined, the comparison of incidents to incidents changed to comparison of incidents to concepts, because the properties provided guidelines for assigning an incident to a category. Third, concepts were compared to concepts, and some concepts were combined into higher level categories. Axial coding took place at this stage where categories were related to subcategories. A category was a phenomenon of significance to respondents, while subcategories “answer questions about the phenomenon such as when, where, why, who, how, and with what consequences” (Strauss & Corbin, 1998, p. 125), therefore fleshing out the whole story. Relationships among categories were integrated to produce a construction of the phenomenon under study. Accompanying the analytical process, memos were written to keep notes about categories and their properties, relations among categories, and new thoughts as they evolved with the analysis. The data analysis was facilitated with the NUD*IST 4 program (Qualitative Solutions and Research Ltd).

PARTICIPANTS PROFILE

A questionnaire about participants' demographic and cruise background was delivered to participants through e‐mails immediately after the completion of each interview/VFG. Table 1 shows the demographic profile of the 18 participants of the interviews and VFGs, as well as the background of their cruise vacations. The number of female participants doubled that of the male participants. Other than this, the participants' demographic characteristics mirrored the North American cruise market profile released by Cruise Lines International Association (CLIA, 2006). About two thirds of the participants had college or post‐graduate degree and were below 59 years of age, and these figures were close to the statistics (80% and 78%, respectively) from CLIA. In terms of the length of the cruise, 13 of the 18 participants had cruises from 7 to 15 nights. This reflected the fact that in the North America market, the most popular cruise sector was 6‐ to 8‐day cruises, which accounted for approximately 52.2% of the market share in 2005 (CLIA). The travel companion profile of the participants was also in line with the CLIA statistics, which revealed that the most frequent travel companions were spouses (79%), followed by friends (24%) and other family members (not spouse or children: 20%). The cruise group size and total number of cruises taken (not included in the CLIA statistics) varied widely.

Table 1. Profile of VFG and Interview Participants

Demographics	No. of People	Cruise Background	No. of People
Nationality		Length of cruise
United States	15	1–6 nights	4
Canada	3	7–15 nights	13
		16 or more nights	1
Gender
Male	6	Travel companion^a
Female	12	Spouse/Partner	12
		Friends	3
Education		Children	2
Vocational/techni‐cal	2	Other family members	6
Some college	4	No one else	1
College	5
Post graduate	7	Total number of people in group
		1 person	1
Age		2 persons	10
18–29 years	1	3–5 persons	4
30–39 years	3	6–10 persons	2
40–49 years	3	48 persons	1
50–59 years	6
60–74 years	5	Time of most recent cruise (year)
		2006	8
Ethnicity		2005	4
White	12	2004	2
Other	2	before 2004	4
Not available	4
		Total number of cruises taken
		Once	2
		2–3 times	6
		4–6 times	4
		7–9 times	2
		10 times or more	4
Note. Four participants were from the first VFG, six from the second VFG, one from the third VFG, and seven from individual interviews. ^aParticipants could choose more than one category for travel companions.

FINDINGS AND DISCUSSIONS

In the discussion of the results, an analytical paradigm consisting of conditions, actions/interactions, and consequences (Strauss & Corbin, 1998) was used as an organizing tool to integrate the explanatory power of the concepts/categories developed. The overall atmosphere of cruise vacations, though not directly of relevance, was recognized to be the condition shaping customers' behaviors onboard. The actions/interactions were the various levels of c2c interaction, while the consequences were limited to the impact of c2c interaction on the cruise vacation experience. The varying levels of c2c interaction and their corresponding impacts were summarized into three scenarios. In addition, strategies adopted by participants to cope with positive and negative c2c interactions were discussed.

In the following narrative, typical accounts by participants were cited for illustration purposes, and each citation was suffixed by a reference that designates the document name and a code for participant's name. Nationalities of the participants were appended when necessary. (Participants joining the discussion on the public forum may have nationalities other than American/Canadian.)

Overall Atmosphere on Cruise Vacations

The role of environment in shaping behavior has long‐established roots in psychology, and this general rule was applied to the commercial environment by Kotler (1974). Martin and Clark (1996) suggested that one causal condition for c2c interaction was environmental factors, which consisted of both physical and social environments. A number of factors associated with the product offerings of a cruise vacation came into play, and collectively they facilitated an overall cruise environment that was conducive to c2c interaction. Although not directly related to the central inquiry of c2c interaction, these factors created the ambiance on a cruise vacation, which set the keynote for life on a cruise ship. Since no human activity was context‐free as claimed by naturalistic inquiry (Lincoln & Guba, 1985), a description about what a cruise vacation was like may aid understanding about passenger behavior onboard.

First, a cruise is a safe mode of travel, and safety is a major issue underlying passenger behavior. All participants agreed that cruise travel was safe, especially when they were on the ship. Comments about risk‐taking were mainly associated with specific on‐shore activities (e.g., snorkelling). Several participants also compared the security issue on cruises with that on land tours or resorts, and commented the former scored far better than the latter. Apparently, the confined and excluded space of the ship added to its sense of security.

An implication of the safety issue was that passengers could explore and take up whatever activity that interested them, without worrying about the safety of themselves and their travel companions, as one participant expressed:

While you have the family on the cruise ship, they can go wherever they want around the ship. And you don't have to worry. Unlike [when] you are traveling in a foreign country… have to worry. While on the cruise ship, it is secure, much more secure. (Interview_AG)

Second, a cruise vacation was basically an all‐inclusive package, which produced two psychological impacts on the part of the customers. On the one hand, it relieved the financial burden accompanying a person while on vacation and put people more at ease because almost everything was paid for in advance. On the other hand, because virtually all basic activities and provisions were “free,” choice of what to do or eat was based on personal preferences without consideration of prices. Therefore, an all‐inclusive package helped customers to make the most out of what was offered, as illustrated by one participant: “You just can pick which restaurant you want to go to, and you didn't need to look at menus, prices and things like that … And that made everybody more at ease, as I felt” (Interview_AG).

Third, a cruise was seen as an easy vacation. For one thing, a cruise ship is a floating hotel that transports its passengers from one destination to another. The logistic advantage of this arrangement was that passengers could visit plenty of places without transfer and packing and unpacking, resulting in the frequently claimed hassle‐free vacation. For another thing, a cruise ship, especially the large kind, is a self‐contained resort with all the provisions, entertainment and amenities at arm's length. Participants appreciated the fact that everything was readily available on the ship, and having all necessities around reflected a notion of having no wants and a kind of utopian idealism. A cruise vacation creates a short period of time when people felt all of their needs were well taken care of.

The implication of an all‐inclusive package and a self‐contained cruise ship was more far‐reaching than it initially appeared. It was interesting to find out that, although purchasing a cruise vacation did involve a lot of preparation and planning, participants embraced the idea that once they were on board a ship, all mental work and decision making was minimized. Consequently they could unwind completely and enjoy what was brought to them. The psychological result of having everything readily available and planned was a worry‐free or easy vacation during which customers don't need to wrestle with how to have a good time. For example, one participant wrote:

It's a stress‐free vacation. You just need to get to the ship, after that, someone else does all the planning work. Can you imagine a vacation that involved going to 4 different destinations in a week and the amount of planning that takes? Forget it. (Online chat_benefit_JI)

Moving from the product offerings to the customer side, it seems that the customers themselves were another important factor that shaped the overall atmosphere of a cruise vacation. All participants talked about their own or fellow passengers' desire or determination to enjoy themselves, which was a salient common ground for all passengers. This kind of customer expectation (with the facilitating product offerings) acted upon itself like a self‐fulfilling prophecy, creating a happy scene of a ship full of vacationers, ready to have fun, as reflected in the following accounts:

…everybody wants to enjoy a good time. I think that's a common denominator, like everybody really just wants to make the most out of their vacation time. (Interview_WE)

Everybody is there to enjoy themselves, I think the personal barriers are down and they are more willing to talk… (Interview_RE)

In addition, because of less time constraints when on vacation, especially when during days at‐sea, passengers were willing to stop and talk. As best described by one participant:

We were on a 14‐day Panama cruise last April. Noticed how everyone you pass, meet, or mingle with all give a big hi, smile, or nod. If this was back at home…everyone is too busy, grumpy, or disturbed to bat you an eye. (Online chat_people_DE)

The result was a cohort of fellow passengers marked by diffusing friendliness—a common scenario depicted by almost all participants.

In summary, the product offerings provided by cruise lines combined with enjoyment‐preparedness from the customer side created an overall atmosphere that was conducive to c2c interaction. The shared goal of boarding the same ship to enjoy themselves elevated customers' tolerance for fellow passengers (Pearce, 2005), which in turn facilitated interaction among passengers. This was evidenced by several participants' comment that interactions among strangers on cruises were much friendlier compared with those in resorts and local communities, confirming Yarnal and Kerstetter's (2005) view of the cruise ship space as a conduit for social interaction among passengers.

Levels of c2c Interaction: Diverse Experiences and Impacts

Although the general atmosphere on cruise vacations was amiable, the experience of c2c interaction undergone by participants was wide and varied—from virtually no interaction at all, through superficial interaction, to spontaneous communication without further commitment, and to close interaction with lasting relationships. The following depiction attempted to construct the diversifying stories of c2c interaction as experienced by participants, categorized into three broad scenarios. Attempts were also made trying to delineate the impacts of different levels of c2c interaction on the cruise experience as perceived by participants.

Scenario 1: No Interaction or Superficial Interaction With Negligible Impact

Although cruises generally had a friendly environment conducive to c2c interaction, two participants reported that they had no interaction with other passengers on their most recent cruise. As one respondent wrote: “I was with my husband; we were at a small table at dinner time and really didn't have much of a relationship with the other passengers” (FG_Two_CO). Three other participants reported very limited interaction with fellow passengers on their most recent cruise. In these cases, interaction with other passengers remained at a superficial level and had no particular significance to the subject. Under such a scenario, participants saw other passengers mainly as the formal role played by others. For example, one participant reflected: “Everyone we met was really nice, and we have e‐mailed our dinner companions a few times. However, I guess we look at them (and they at us) as cruise dinner companions” (FG_One_CH).

These circumstances arose because on a cruise vacation, many activities and services were offered in group settings. Dinner and shore excursions were the most frequently cited occasions where passengers met each other. Very often, cruise lines arranged dinner settings for large tables; therefore, passengers were bound to congregate with others when they shared a table. The same thing often happened when people went on the same coach or boat for shore excursions and other activities. C2c interaction under such circumstances was unintended or may be imposed. It did not occur out of the subjects' initiation, nor did it always evolve into any significant level.

For cases with no or superficial levels of c2c interaction, the common admission from participants was that they kept more within their own group. This signaled a trade‐off between inter‐group and intra‐group interactions (i.e., interaction between passengers and interaction between travel companions). This accorded with Levy's (2005) comments that travelers with strong intra‐group ties may put forth less effort in socializing with other stranger travelers. One underlying factor shaping this scenario was the “frame of mind” brought onboard by customers. Compared with cases with higher levels of c2c interaction, these cases tended to have a “frame of mind” that oriented their attention toward their travel companions. This “frame of mind” was shaped by significant life events or situations at home that elicited immediate and strong relational needs among travel companions and in most cases motivated the specific cruise vacation at hand.

For example, one participant decided to “reward” his parents “by taking them on a cruise,” and he “got to spend some good quality time” with his father “after leaving the house so many years ago” (FG_One_MA). Another participant wrote: “The most enjoyable part was seeing the smile on my husband's face… He really needed it because his mother had passed suddenly at the age of 63” (FG_Two_CO). In another case in which both husband and wife had high‐stress jobs, the couple tended to keep to themselves a lot because “We just like to be together and do things together, and since we have so little time to do that while working…” (FG_One_CH). One participant went on her most recent cruise with a family reunion of 10, and they kept to themselves, consequently “had no situation positive or negative” with other passengers (FG_Three_ME).

Therefore, although the general atmosphere on cruises was very friendly, individuals' tendency to communicate with people outside of their own travel group was, to a large extent, dependent on their frame of mind, which keynoted their relational needs with travel companions. However, this frame of mind was of a contingent nature. By definition, it was related to specific incidents or situations in the home environment, therefore changeable with each cruise. For example, one participant who had virtually no interaction with fellow passengers on her most recent cruise also explicated that “on previous cruises I have met wonderful passengers and kept in touch with them through e‐mails and some with phone calls” (FG_Two_CO).

For participants who had a frame of mind oriented toward their travel companions, interaction with other passengers, if any, was undertaken in a half‐hearted way. Although participants still described the people they met as “nice,” “interesting,” and “friendly,” they would not use words like “friendship” or “friends.” Superficial would be the appropriate word to describe such interactions, and acquaintance was the name given to the interactional counterpart.

In summary, in Scenario 1, participants had no or very limited interaction with fellow passengers. C2c interaction, if any, remained at a superficial level, characterized by role‐bound encounters, instrumental communication, and little knowledge about the interacting partners except for their public self‐presentation. Occurrence of c2c contact was dictated by external context, and there was little concern for maintaining relationship with a particular person (Levinger & Snoek, 1972). Consequently, except for a holistic impression in the participants' mind about the generally friendly fellow passengers, these contacts did not have much substantive impact on a person during the course of his or her cruise vacation.

Scenario 2: Spontaneous c2c Interaction That Became an Autotelic Part of the Cruise Experience

In Scenario 2, although interaction still remained unintended (i.e., depending very much on the structure of the daily activities), what distinguished it from Scenario 1 was the level of communication among passengers. In Scenario 2, and also in Scenario 3, “real” communication (not mere exchange of greetings) occurred among passengers, and this was viewed as an added part of a cruise vacation. Participants frequently reported how they shared their daily cruise experience with others (about what they did or did not but wanted to do). Discussion about previous travel experiences was also the most favored topic among passengers. The following discussion about the communication among passengers applied to Scenario 3 as well, although Scenario 3 had other salient characteristics that set it apart from Scenario 2; and these are reported in the next section.

Communication among customers enhanced the cruise experience in two ways. First, information from others, particularly seasoned cruisers, was very helpful for first‐timers or those not well‐prepared. Although cruise lines often pre‐distribute brochures or lists of their daily itineraries and programs, participants often found information from these materials inadequate for them to make discernable selection, taking into consideration the numerous activities and options available. Participants valued the input from fellow passengers, especially people who had been to the places before, because such input helped them make decisions about where to go and what to see, thereby making the most out of a cruise vacation. As illustrated by one participant:

You always had a number of choices of what to do, and sometimes some of the things that they said will help me decide what I wanted to do … Because a number of them have been on similar cruises in the same area… we respected what they felt …and what they told us was a good use of time. (Interview_NA)

For some participants, they had the habit of researching and preplanning for their cruise vacation, and some of them prebooked all excursions before boarding. Once onboard, they had already made up their mind about what to do and see; the same was true for most experienced cruisers. In such cases, the functional value of the information interchange diminished, but the social value of the interaction gained prominence. Words used when talking about interaction with others included “fun,” “joy,” “interesting,” “happiness,” and “laugh.” Interaction was seen as a window through which participants obtained insights about other peoples' lifestyles, other cultures, and other places. It was a shared experience valued by most participants:

A lot of the people on the cruise had traveled a lot [more] than we had. So they were able to relate stories of other trips, which were very interesting, places we've not been [but] we wanted to go to … (Interview_NA)

You meet people from all over the world. I can't say they influence us in the least; it is just interesting to get perspectives on different lifestyles. (Online chat_people_P3)

Communication among fellow passengers at this level not only involved people's experiences and attitudes about the world, but also was combined with self‐disclosure to some extent. Sometimes, getting to know others a little bit also propelled reflection on one's own life, as exemplified by one participant's story:

…the lady then told us she wasn't working any longer because she had been in the Pentagon the day the jetliner hit it on 9/11. At that moment it made the horrible event so very real to us. …it made me grateful for all I had and that my little aches and pains were absolutely nothing compared to what she had been through…(FG_Two_GE)

Interaction with other people met en route as a source of self‐reflection was also identified by Caton and Santos (2007).

To summarize, c2c interaction in Scenario 2 was impulsive and free‐flowing, and passengers got to know others a little bit. However, interaction at this level remained spontaneous on the spot, and seldom lasted beyond the course of a cruise vacation. Remarks from one participant best portrayed this situation: “You get to know people a little bit if you wish. As far as lasting friendships go, I think that is rare” (Interview_RE). The wide range of topics of conversation among passengers was seen as an add‐on to the cruise experience, with both instrumental and autotelic functions (Holt, 1995). What distinguished Scenario 2 from the previous scenario was that the autotelic aspect of interpersonal interaction created a more enjoyable experience for participants, echoing Yarnal's (2003) comments that social interaction among passengers created playful spaces on a cruise ship. Like what Holt found, autotelic social interaction among customers became part of the fun‐making and playful experience.

Scenario 3: Close Interactions With Lasting Relationships That Became One of the Highlights of the Cruise Experience

In Scenario 3, chance and unintended c2c interactions were able to evolve into a stage in which bonds with particular others were established. Passengers usually met others when engaged in small group activities (e.g., dinner and games), and got to know others gradually during the process. When people felt they had much in common or could get along well, bonds began to develop. In other words, initial c2c interaction occurred usually not out of a person's efforts of actively seeking for socialization; it was more of a natural process in group activities. But when people found others interesting or compatible, they began to set aside time to meet each other or plan activities together, signaling a transition from unintended interactions with whomever happened to be at the same spot at the same time to intended or planned interactions with particular others. The movement from coincidental encounters to coordinated activities signified the establishment of mutual bonds or a connection between passengers, implying the development of interdependence beyond that demanded by a transient meeting (Levinger & Snoek, 1972). For example:

This last cruise is the first time we actually did planned activities with other couples we met on board. For example, planning to meet at a certain place later that night, so we can all hang out …just today I got an e‐mail from one of the London couples, asking us how we were. We are hoping to meet them somewhere in the future. (FG_One_LO)

Spontaneous and free‐flowing communication among passengers created valuable sharing experiences in both Scenarios 2 and 3. Besides, participants in Scenario 3 often reported that they became more involved in group activities as a result of meeting others, as one participant depicted: “The 3 couples we met … kept us well entertained, more than usual. We found ourselves staying out later, and going to activities we normally would not go to” (FG_One_LO). In a few cases participants reported that they tried new things or did something they had not expected to do as a result of meeting others. People also tended to get to know others much better, and interaction at this level was usually described as very “active” and “close.”

The fact that people were able to make friends changed intermittent interactions into meaningful relationships, which usually extended beyond the boundary of a cruise trip. Participants expressed a desire to meet their newly made friends again, and a few have done so afterwards. This echoed with Hinde's (1979) assertion that relationships depend not only on a series of intermittent interactions, but also indicate an expectation of continuity in future interactions. For example,

I think we will always be friends with this doctor and his wife from Carson City we had dinner with almost every night [on our last cruise] … They have already visited us … and I know that we will remain in contact for many, many years. (Interview_GO)

In terms of the impact of c2c interaction on cruise experience in Scenario 3, participants often used words like “definitely” and “absolutely” to affirm the positive contribution from the friends they made to their cruise experiences. This kind of contribution was recognized in Scenario 2 with much reservation, as testified by remarks from one participant: “I think they do enhance the voyage somewhat… But I don't think there is a major factor in it, it is an incidental situation” (Interview_RE). In Scenario 3, sociability onboard with newly made friends was often recognized as one of the highlights of participants' cruise experiences and one major benefit of cruise vacations. For instance, one participant said:

we met a couple and their daughter (they were our tablemates) on that cruise and we became friends—not aquaintances—friends. …Did all of these wonderful people we've met on cruises make a difference? Yes. Would we have had a good time without them on the cruise? Of course. But, those individual cruises would not have been the same without the interaction of those delightful people. (Online chat_people_OK)

Strategies to Positive and Negative Encounters

T2t encounter was a fact of life on cruise ships. Passengers had different expectations and experiences about it, and they used different approaches to coping with it either consciously or subconsciously. For positive interactions, the general position was that it was a plus if it happened, although most participants did not set out to seek for it. For example, “If we run into others who share common interests, that's great. If we keep in touch, even better. But if all we do is enjoy ourselves and each other, then we have accomplished what we set out to do” (FG_Two_JO).

The strategy adopted by most participants could be described as a passive “wait and see” approach, since for most people the purpose of going on a cruise was not to meet new people. Part of the reason may be that interaction with other passengers was perceived to be unpredictable, as one participant observed: “…it is an incidental situation. It's never programmed, because you just happen to meet these people” (Interview_RE). But for several participants who were more outgoing and sociable, they were more proactive toward c2c interaction. For example, one participant wrote:

If we aren't traveling with a group, we always ask for a large table for dinner and never hesitate to ask to sit with someone at the buffet if tables are full. We've met a lot of interesting people that way, but a few grumps too. (FG_Two_PA)

In particular, with the facilitating aid of the Internet and a growing cruise‐enthusiastic population, there is an increasing number of websites with dedicated discussion boards on cruise travel. New platforms (such as the one called “roll‐call” on the Cruise Critic website) appear, where people going on a cruise can find out who else will be cruising on the same ship and start to buddy up prior to departure. Often, these people (if finding compatible others online) would plan to meet and mingle onboard the ship.

As the VFG participants in the present study were actually recruited from these cruise websites, several of them did take the initiatives on “roll‐call” to make c2c interaction happen. These participants may be more sociable, but another reason may be that they were simply more experienced cruisers and knew how to plan for a better vacation experience. One participant described his experience:

We've already gotten set up with one couple for tablemates at dinner, and have made several friends in the process of getting to know each other online… It's also kind of neat that we've been able to establish a relationship with new people before we even get on the ship! (FG_Two_JO)

Strategies were also employed in the opposite direction to strengthen intra‐group interaction and avoid interruption from others. In extreme cases when inter‐personal needs with travel companions became paramount, a few participants utilized strategies that enhanced interaction within their own group, without side distraction from other passengers or activities. For example, one participant (FG_Two_CO) was traveling with her husband who was undergoing a tough time in life and requested a dinner table for two, and another participant (FG_Three_ME) had a table of 10 for her family reunion. During the whole course of the cruise, they kept pretty much to their own groups, and interaction with other passengers was rare.

Although the majority of the incidents about c2c interaction reported by participants were positive, several negative incidents were also reported. Most of the negative incidents were about deviation from code of conduct (such as rudeness, foul language, smoking, not cooperating, queue‐jumping, chair‐saving, misbehaving children) as found by Grove and Fisk (1997), or about people of grumpy character, but was seldom at a personal level. The most common strategy employed by the participants was avoidance; namely, participants took action to get away from unpleasant encounters and managed not to get caught up in them. Words like “steer away,” “avoid,” and “ignore” were widely used. One explanation for such inaction may be that customers often believe they cannot influence or control the behavior of other customers (Pranter & Martin, 1991); therefore, they are less likely to interfere directly. A typical account was: “Dealing with unpleasant people really isn't what I want to do on my vacation, so I typically just try to avoid them at all costs. You aren't going to change their minds about anything anyway” (FG_Two_PA).

Note that in a confined environment on a ship, people may run into the same persons over and over again. This was true especially on small ships. While a confined space may be conducive to c2c interaction, it also made an occasion more awkward when negative incidents arose, because the avoidance strategy was less effective in such situation, as described by one participant: “With such a small passenger list, we quickly learned who the old grumpy ones were and avoided them the rest of the trip! On a smaller boat it is kind of hard to get away from them though” (FG_Two_PA).

Direct intervention into negative incidents was rare, and it seemed that passengers only resorted to it as the last resource or when they felt that they had the obligation to do so. One participant explained that she had to respond because the other person “was really getting to” her and she “just couldn't help” (FG_Two_CO), and another participant said he had to go over and speak to a group of young men who were using foul language and getting very lousy, because there were many families on the ship (Interview_AG).

However, negative incidents apparently did not abate significantly participants' enjoyment of their cruise vacation. Passengers usually had firm determinations to have a good time and to make the most out of their cruise vacation, thus they simply won't allow such negativity from fellow passengers to spoil their vacation. To put it another way, psychologically cruise vacationers were immune to negativity from other passengers, as exemplified by the following depiction: “I just don't think that I would ever let somebody else affect my cruise vacation in a negative way. I'm there to have a good time and you can believe that I will one way or another” (FG_Two_PA).

Sometimes a person with negative personality just became a target of joking. For the more outgoing and optimistic customers, meeting all kinds of people became part of the cruise experience, although these negative incidents may not be taken so light‐heartedly by everyone. A grumpy character became sources of gossip among passengers and topic for recollection back at home. For example,

As for the one couple or I should say person on our Sept cruise that was so negative, it didn't ruin or affect our cruise but did leave a lasting memory as to how someone can be so unhappy with everything. …it did affect me to some degree, not that it bothered me as much as made me wonder how these people get along in life. (FG_Two_NI)

SUMMARY OF FINDINGS

Results of the study showed that cruise passengers had diverse experiences meeting fellow passengers, which were categorized into three broad scenarios. However, it should be noted that these scenarios were not an exhaustive list of all possible scenarios, but they were typical of the participants' accounts and were simplified categorizations for illustration purposes. The experiences of the 18 participants to the VFGs and interviews could be classified into one of the three scenarios: 5 participants fell within Scenario 1, 4 within Scenario 2, and 8 within Scenario 3. (One participant was excluded from the classification because there was not enough information about her experience of interacting with fellow passengers on her most recent cruise vacation.) In Scenario 1, passengers had virtually no or limited interaction with fellow passengers, and interaction with travel companions was recognized to be the dominating social need. Accordingly, impact from fellow passengers on the cruise experience was indiscernible. In Scenario 2, passengers usually had free‐flowing communication with a limited number of fellow passengers. They resorted to fellow passengers for confirmation of instrumental as well as expressive aspects of their experience (Pearce, 2005). The spontaneous communication among passengers became an autotelic part of the playful cruise experience and added to the enjoyment of the cruise. In Scenario 3, participants' accounts began to narrow down and target particular others with whom bonds and friendships were established. Unintended and chance interactions with whomever met on the spot turned into coordinated activities with particular others, and relationships usually extended beyond the course of the cruise. This was the stage where c2c interaction became one of the highlights of customers' cruise experiences.

The levels of interaction among passengers generally followed Levinger and Snoek's (1972) model of relationship development, starting from unilateral impression, through surface contact (superficial and role‐bounded interaction), to increasing mutuality, which means “development of interdependence beyond that demanded either by a transient meeting or by externally structured role prescription” (p. 8). Apparently, t2t interaction only entails special meaning when it exceeds the common role script for unacquainted travelers, and interactional partners identify each other as unique individuals. This was confirmed by Levy's (2005) finding that good communication among travelers was characterized by a deeper interest beyond “standard travel talk.” The natural progression of t2t interaction and the eventual cement of stable relationships embody the claim that tourist activities require distinctive types of social interaction and the forging of new relationships (Otto, 1997).

Strategies utilized by passengers to handle t2t interaction were different based on the nature of t2t interaction. Roughly, an approach‐avoidance strategy could summarize passengers' reactions to t2t interaction. For passengers who had no salient relational needs with travel companions, general reaction to positive t2t interaction was approaching, meaning they were willing to participate if it happened. Avoidance was the universal strategy for negative interactions. Interestingly, negative encounters with fellow passengers apparently did not have a significantly adverse impact on the cruise experience, partly because passengers won't allow such negativity to spoil their vacation and managed to steer away from them. This verified Ryan's (1999) claim that for holidays or vacations, there was “a strong motivation for successful and satisfactory outcomes on the part of the client” (p. 279). This motivation immunized passengers against the negativity from fellow passengers.

IMPLICATIONS AND LIMITATIONS

By focusing on participants' experience of direct interaction with fellow passengers, the present study complemented Yagi's (2001) study, which was devoted to indirect interaction among tourists (i.e., tourists' impressionistic perceptions of other tourists). In Levinger and Snoek's (1972) model of relationship development, this unilateral awareness or impression was the very beginning of a relational process. Furthermore, the study revealed generally positive interactions between fellow passengers, contrasting the largely negative dimensions of t2t encounters reported in the literature (Pearce, 2005). The social dimensional approach of studying tourist behavior (Pearce, 2005) unfolded a neglected aspect of the tourist experience, and revealed the active role played by fellow tourists in the construction of the tourist experience. The authors argued that tourist experience is the result of tourists' interactions with various physical and social elements encountered en route; a sketch of the tourist experience in which fellow tourists are invisible may not be a realistic depiction.

The study revealed that tourists did have different relational needs when on vacation, which exerted great impact on tourist experience and behavior. Further research could investigate how these relational needs surface and evolve both before and during the vacation/trip, as well as the interplay between different relational needs. For example, the tentatively proposed trade‐off between intra‐group and inter‐group interactions in the present study deserves further investigation, and this is in fact one of the conceptual schemes proposed by Pearce (2005) to expand extant understanding about tourists' social contacts.

This exploratory study revealed that positive onboard c2c interaction did contribute to the cruise experience to different degrees. This implies that cruise lines could add value to their product offerings by designing programs and services that promote positive inter‐personal interactions among customers. Cruise ship capacities vary from less than 100 to more than 4,000 passengers; large cruise ships were often blamed for being impersonal, and smaller ships were said to be more intimate by participants. Taking into consideration of Yagi's (2001) finding that indirect interaction between tourists tended to be more negative than direct interactions, cruise lines that could successfully nurture more personal and positive interactions among their passengers could create an advantageous position for themselves. Even 100 passengers on a small river cruise could be regarded as too many for meaningful inter‐personal interactions to occur (Interview_GO). In this regard, small group activities would be preferable to mass‐production of services. For example, similar to the findings of Teye and Leclerc (1998), the present study found that the most common occasion for cruise passengers to meet others was dinner time, where a small number of people could sit close to each other for a relatively extended period of time (usually 1–2 hours). Other occasions included shore excursions, games (e.g., playing bridge), and lectures. In addition, for the avoidance strategy for negative c2c interaction to be effective, there should be enough options (services, activities, or spaces) for customers to choose from. Cruise lines must allow sufficient flexibility in their program/service planning that suits customers' different relational needs, whether this is to avoid negative passengers, to arrange tables for newly made friends, or for travel companions to stay together.

The study found that cruise passengers had different social experiences when on vacation, e.g., intra‐group versus inter‐group interactions. This indicates that cruise lines must have profound knowledge about their customers so that the appropriate inter‐personal touch could be incorporated in the service experience design. In particular, the study found that the majority of the participants could be categorized into either Scenario 2 or 3, suggesting that cruise passengers generally were willing and able to communicate with each other for either instrumental or autotelic purposes. This resonates with Teye and Leclerc's (2003) finding that the most salient motivation for cruise vacation of the North American Caucasian segment was the social aspects (e.g., to be with other people and to make new friends). For the customer segment that is interested in c2c interaction, businesses could deliberately create spaces for c2c interactions, transforming the service places from modernist non‐places of goal‐oriented individuals to post‐modern communities (Aubert‐Gamet & Cova, 1999). The communion between customers as well as shared customer expertise constitutes intangible corporate equities which should be invested and utilized. Such equities are exceptionally valuable for products/services where customer interest and level of involvement are intense (e.g., cruise vacations), thereby calling for extra instrumental support and emotional exchange between fellow customers.

In addition, the present study revealed that communication between unacquainted customers occurs both in the physical space of cruise ships and the virtual space of the Internet. Currently, the link between the physical and virtual spaces was mainly initiated by the customers themselves. For example, customers acquainted online take actions to meet face‐to‐face onboard. In addition to creating virtual spaces for their customers, cruise lines should build the link between the online and offline spaces, and such strategy could greatly reinforce their efforts of building consumer communities with compatibility and identification. For example, cruise lines could offer a group discount to members of a virtual community on the same ship when they book their vacations directly with the cruise lines. Programs and social activities on an upcoming cruise journey could also be posted on the online discussion boards to attract the participation of virtual community members. Incentives (e.g., lucky draw, gifts) could be provided to encourage virtual community members to participate in an onboard get‐together to meet others face‐to‐face.

One limitation of the present study was connected with the characteristics of the sample. The VFG participants were recruited from several cruise websites and may represent a special segment of the cruise population, therefore resulting in probably biased findings. A comparison of participants in the VFGs and individual interviews, nevertheless, did not reveal significant differences both in terms of their experiences in t2t interaction and demographic background. However, four participants of VFGs were indeed very experienced cruisers, with the number of cruises taken reaching 48, 28, 20 plus, and 16 times, respectively; while none of the interview participants took more than 9 cruises. Second, the VFGs had limited number of participants due to the high attrition rate, and this constituted another limitation of the study. Including more participants could have promoted more interactions among them and thus might be able to trigger more new ideas or perspectives. Future research using VFG with an extended discussion period should over‐recruit at the beginning to account for possible midway dropouts. In addition, due to the difficulty of finding eligible participants, the sampling process did not strictly follow the theoretical sampling procedure, which could have exhausted possible ranges of the phenomenon under study and was therefore capable of unfolding the full array of multiple realities (Lincoln & Guba, 1985). Therefore, concepts and their relationships proposed in this study should be taken as tentative and needs to be tested with further empirical research.