The affective impact of sightseeing bus tour experiences: using Affective Events Theory (AET) to examine length-of-stay and electronic word-of-mouth

ABSTRACT

This study investigates the key components and influences of positive affect and electronic word-of-mouth (eWOM) on tourist visits at two developing urban destinations, namely Birmingham, United Kingdom and Valencia, Spain. These two data collection sites yielded evidence gathered from 615 and 627 sightseeing bus tourists, respectively. Through the analytic lens of Affective Events Theory (AET), data were examined, and results verify the significant mediating role of affect in two regards: (1) tourists’ decision to extend their visits and (2) eWOM of sightseeing bus tour experiences. The moderating role of past sightseeing experiences in these relationships was also supported by the data analysis. This paper further strengthens the role of affect in tourism management scholarship as well as expands AET from the work-setting into the tourism context thus marking a new research trail. Practical implications for tourism destination management organizations (DMOs) are also discussed.

KEYWORDS:

• Affective Events Theory (AET)
• sightseeing bus tour
• urban destinations
• tourist visit extension
• eWOM
• destination management organization (DMO)

Introduction

Behavioural intention models are commonly used to predict tourists’ behaviour. Since the majority of published research indicates that intention to revisit a destination and tourist recommendations to others are two strong proxies to predict actual return (Jiang & Balaji, 2021; Stylos & Bellou, 2019), then the investigation of the factors such as tourists’ past experiences; the role of affect upon individuals’ future intentions and plans; and the ubiquitous electronic word-of-mouth (eWOM) communication of their experience and intention to revisit a destination is critical. Better understanding of these factors can aid planning, designing and implementation of successful tourism destination products/services (Bornhorst et al., 2010; Correia et al., 2015). Furthermore, immediate actions that tourists might take to extend and enrich an ongoing visit have been scarcely investigated. Better understanding not only via post-tour reflection/feedback provided by tourists, but even more importantly, the tourists’ ongoing affective responses to the tour experience at hand can expand the scope of the managerial considerations and approaches of future behaviour in terms of revisit optimization.

Sightseeing bus tours have grown over the last years because of the increase of short breaks (Ghanem & Shaaban, 2021). Worldwide enterprises such as City Sightseeing which is in more than 130 locations and Gray Line which in about 700 locations are responding to the call for an expanded model of touristic choices/options. Additionally, tourists increasingly desire to gain more holistic experiences of their chosen destinations.

For the purposes of this study, a ‘city tour’ is defined as guided transportation of a group within a destination for a short period of time which aims to obtain a glance of a particular destination or an overall impression of the destination’s attractions/activities. According to Antón et al. (2017), Carreira et al. (2014) and Jomnonkwao and Ratanavaraha (2016), the impact of sightseeing bus tour experiences on the overall tourist experience while visiting urban destinations has not been fully examined. However, the study by Ross and Iso-Ahola (1991) analysed the motives and satisfaction of a day tour in Washington, DC. The main motive of their study was to gain ‘general knowledge’ by examining ‘social interactions’. Recently, Larsen et al. (2021) argued that having limited time for visiting a destination and/or having preliminary cursory overview of a destination narrows a tourist’s options and decision-making when choosing the most attractive site to visit. Other prior literature suggests that perceived attractiveness is a key indicator of revisit intention (Um et al., 2006).

Notwithstanding previous studies which sought to outline the effects of various tourism product components on the intention to revisit a destination, such as transportation, accommodations, and receptivity of local residents, a broader array of factors can be explored. This seems particularly essential in light of the growing interface of palm-held technology and eWOM. The link between preliminary perceived attractiveness of a destination offered by a sightseeing tour beforehand and then subsequent key tourist response of their actual experiences can impact future revisits, extend the current length-of-stay and/or prompt posting eWOM content. Better knowledge of these emerging decision-making drivers and the magnitude of their impact is not only timely but can be immediately practicable.

Although revisit intentions have been assessed with WOM or eWOM as output variables (Kim et al., 2009; Liu & Lee, 2016), the option of extending current visitation has not been examined. Since positive emotional experiences influence eWOM generation in the case of hotels (Serra-Cantallops et al., 2020), it seems logical to investigate whether sightseeing tours which are often packaged and/or promoted with hotel stays, might also generate eWOM.

Other research literature suggests that cognitive and affective images predict behavioural intentions, as found in Afshardoost and Eshaghi (2020), as well as in Stylos et al. (2016). Typically, tourists’ affective states have been addressed as part of the cumulative tourist experience at a destination (Hosany, 2012; Loureiro et al., 2021). However, effects on tourists’ behavioural intentions associated with specific tourism activities such as bus tour sightseeing have not been adequately researched. To further understand these relationships, tourists’ affect towards bus tour sightseeing experiences is incorporated as an explanatory mechanism in our hypothetical model.

Having taken all these issues into consideration, the overarching aim of this study is to clarify the role of bus tour sightseeing experience components on extending the tourists’ current visit and on their eWOM communications. The analytical framework adopted will be Affective Events Theory (AET), a well-established theory and well-suited for this study. Specifically, the AET generic framework has been properly adapted to the study context and has been further enhanced with a context-specific variable (i.e. past sightseeing experience). There are three objectives of the study. First, to analyse the indirect role of affect regarding bus tour sightseeing experiences upon two outcome behaviours: the decision to extend a current visit and posting content via eWOM. Second, to study the direct influence of bus tour affect on future tourist plans/behaviours, namely eWOM on bus tours and current visit extension. Third, to provide a theoretical explanation based on the AET (Weiss & Cropanzano, 1996) to not only better embed the affective evaluation of the events related to the city bus tour experiences with an evaluation/feedback plan, but also to broaden the utility of the theory beyond workplace settings.

The AET (Weiss & Cropanzano, 1996) postulates that events elicit an affect response in people, which in turn influences their attitudes and behaviours. Heretofore, evidence gathered, and findings derived from research using Affective Events Theory come mainly from workplace settings especially in regard to job satisfaction (a positive affect). Despite the wide recognition of the role of emotions in theorizing consumers’ decision making, Affective Events Theory has not attracted broad attention with the exception of Härtel, McColl-Kennedy, and Lyn McDonald (1998, p. 430), who stated, ‘Although the theory [AET] is specifically intended to help understand employee attitudes and behaviors, its process description should generalize across settings’. This study offers findings into the types of touristic events that may give rise to affect and contribute to the academic research corpus by extending Affective Events Theory from organizational settings into the realm of consumer behaviour.

Theoretically, to the best of our knowledge, there is no study that has yielded evidence on the role of a sightseeing bus tour experience on tourists’ behavioural intentions in the moment. Almost entirely, research regarding bus tour sightseeing experiences has conceptualized the tourists’ experiences as a determinant of future visit intentions. In this study, the potential for current visitation extension is investigated as a new behavioural output as well as the investigation of eWOM which is oftentimes very much ‘in the moment’. Additionally, the extension of Affective Events Theory, as mentioned earlier, goes beyond the workplace setting to shed light upon tourists’ attitudes and behaviours. For the practitioner, this study may provide a better understanding of tourists’ experience while having an on-board bus tour sightseeing experience. Such rich insights can help practitioners effectively incorporate and further invest time and attention to the details and desires of bus tour sightseeing in their destination offerings.

Theoretical background

Weiss and Cropanzano (1996) introduced Affective Events Theory to illuminate the role of exogenous influences on individuals within their work settings. Events in one’s work life might include important happenings, a particular place, and/or a special time that notably influences one’s affect. Such events oftentimes trigger significant changes in one’s attitude or behaviour (Lazarus, 1991). The experience of an event goes through a cognitive evaluation process with regards to: (a) the relevance of this event to individual’s well-being and (b) the significance of this event. To the extent that a certain event is considered to be relevant and significant as well as assessed as favourable, then individuals are likely to experience positive affective states. On the other hand, relevant and significant but unfavourable events tend to elicit negative affective states (Cropanzano & Dasborough, 2015; Weiss & Cropanzano, 1996).

Affective Events Theory goes beyond the specific events and their elicited emotions and affective states. The theory focuses on the role of emotion and evaluative judgment in the relationship between an individual’s perceived experiences and their subsequent intentions or actual behaviours (Weiss & Cropanzano, 1996). The reaction to an event, whether considered good or bad, is based upon the attainment of one’s personal goals and values (Frijda, 1986; Lazarus, 1991). Therefore, as Rosen et al. (2009) assert, recognizing the impact of emotions, mood and affect on attitudes and behaviours may help guide organizations towards the creation of positive work events and avoidance of negative ones.

Affective Events Theory was originally conceptualized to identify employees’ affective responses to workplace events with the initial focal construct being job satisfaction (Mignonac & Herrbach, 2004; Wegge et al., 2006). Since behavioural models from corresponding mainstream theories such as Theory of Reasoned Action, Theory of Planned Behaviour and Motivation Theory have rarely exceeded an explanatory power of more than 50% (Stylos & Bellou, 2019), perhaps Affective Events Theory can yield a better understanding of consumer attitudes and behaviours within the domain of tourism management.

The rationale for extending Affective Events Theory into the tourism management is built upon the following points. In regard to well-being, research studies have long recognized the role of vacation for an individuals’ wellness. In fact, Dolnicar et al. (2012) revealed that vacations are so important for individuals that they should form a distinct domain in the measurement of quality-of-life. While on vacation, individuals engage in various activities, all of which are part of the satisfaction sought via travelling (Kelly, 1985). Vacations are comprised of different travel activities, which are both relevant and significant to the satisfaction that emerges from visiting a destination. A tourism activity, as any special experience, is a very specific event with specific characteristics and a certain time-duration, which may change the way a visitor views a travel destination. Additionally, a tourism activity, similar to events in one’s everyday workplace, goes through an assessment process and yields a subjective evaluation as being relatively good or bad. A tourism activity may be considered in regard to attainment of an individual’s particular goals and/or personal values. Evaluating the operation of value chains of distinct events/activities, whether workplace bound or within a touristic activity, can be related to individual behaviours in the given setting (Wu et al., 2020). Thus, we argue that tourism activities, in this case bus tour sightseeing experiences, influence tourists’ affect which will furthermore influence a tourist’s attitudinal and/or behavioural response.

Sightseeing in general – vs – bus tour sightseeing

Tourism destinations include a vast array and diverse set of attractions. Tourists visiting a destination seek a cognitive and affective engagement of multiple stimuli to which they are exposed (Bigné et al., 2008; Stylos et al., 2017). In general, sightseeing from a macro-perspective is a key aspect of the overall tourism experience. Sightseeing can potentially encapsulate most of the well-known aspects of a destination in a nice, packaged version, but also reveal hidden gems of a particular place that warrant visitors’ attention. A variety of sightseeing options can greatly increase the attractiveness of an urban destination as tourists may select ones that fit best their personal preferences, travel plans and affordability of available sightseeing activities (Huang et al., 2014). Diverse types of sightseeing are available to tourists depending on the tourism destination and the available resources to support sightseeing services. These especially tailored sightseeing tours activities would include walking sightseeing tours (either guided or self-guided), cycling tours, motorized tours (via land vehicles, boats, helicopters) and even more recently – virtual reality sightseeing.

A sightseeing bus tour offers a more micro-perspective albeit meaningfully still connected to the overview purpose of the main tour attractions at tourism destinations. A bus tour is especially tailored and is particularly well suited for both city-breaks and lengthier vacations, especially in urban destinations. Sightseeing bus tours offer a convenient, organized and flexible way to explore key city landmarks in an abbreviated period of time (Ashworth & Page, 2011). As the tourism industry recovers from COVID-19 pandemic lockdowns and extraordinary measures imposed across the travel and tourism industry, bus tour sightseeing activities can greatly contribute to the enhancement of the tourist experience. The revitalization of domestic and localized tourism economies can look towards bus tour sightseeing as a viable option.

Carreira and his colleagues (2013 and 2014) provide a useful framework for an initial examination of the transport-related dimensions based on their extensive literature review on transport and customer experience. As a result, they provide a set of seven dimensions for measuring the bus travel experience that they tested with 1226 passengers in mid-distance bus trips. The sightseeing bus tour experience, according to these authors, is based on a variety of components, including individual space, information provision, staff’s skills, social environment, vehicle maintenance, off-board facilities and ticket-line service. Although this scale has been extensively applied to public utilitarian transportation research, no study has used it for hedonic transportation research. The composite of all these elements forms an overall experience; therefore, the 29-item scale proposed by Carreira et al. (2014) was adopted for this study for measuring passengers’ bus travel experience in urban settings. Given that the bus tour sightseeing experience, like nearly all touristic activities, has a specific duration and characteristics, it can be conceptualized overall as an event that produced emotions and influences an individuals’ affect. Therefore, the following hypothesis is offered:

H₁: Bus tour sightseeing experience has a positive effect on bus tourists’ affect.

The moderating role of past sightseeing bus experience

The key role of prior visit experiences having an impact upon tourists’ attitudes and behaviours towards a destination has been widely acknowledged in the literature (Lehto et al., 2004). Even earlier, Alba and Hutchinson (1987) associated familiarity with previous experience accumulated by the consumer as a proxy construct to tourists’ knowledge and disposition towards a destination as did Maestro et al. in their later (2007) study. Typically, the literature on tourism familiarity has focused on the tourist destination (Bigne et al., 2019; Tan & Wu, 2016) not so much on tourists’ affect before, during and/or after a bus sightseeing tour. Broadening the concept of familiarity to include tourists’ attitudes, emotions and general affect more fully regarding services when examining bus sightseeing tour experiences can enrich and inform scholars and practitioners alike.

Scholars have shown that destination familiarity plays a moderating role in the relationships between destination images as well as attitudinal states and tourists’ future behavioural intentions (Sanz-Blas et al., 2019). Also, previous research has identified the moderating role of familiarity and prior experiences on the relationship between destination experience and tourists’ satisfaction (Huang et al., 2010). Further extending these arguments to include the relationship between sightseeing bus tour experiences and tourism emotions, it is expected that:

H₂: Past sightseeing bus experiences moderate the relationship between current bus tour sightseeing experience and bus tour affect, such that the relationship will be stronger for bus tour passengers with high past sightseeing bus experiences versus those with low past sightseeing bus experiences.

Current visit extension

Length-of-stay is a critical issue for all aspects of sales, marketing, financial planning and operations. Length-of-stay influences tourism expenditures at the destination and have been analysed extensively (Alen et al. 2014; Gokovali et al., 2007; Wang et al., 2018). The literature identifies four types of variables that impact length-of-stay: sociodemographic, lifecycle stage, travel motivations and travel characteristics (Soler et al., 2018). The literature has focused on explaining the drivers of length-of-stay prior to a visit. For example, Wang et al. (2018) considered antecedents, such as transportation mode, number of companions, information channels, WOM and the destination status, as tourist considerations weighed even before the trip takes place.

However, only a few studies have addressed the determinants of extending the length-of-stay at a destination (Wang et al., 2018). Others, such as Barros and Machado (2010) and Adongo et al. (2017) have argued that tourist experiences during a tour might trigger a decision to actually expand the length-of-stay during that ongoing trip. These scholars contend that variables that might influence the length-of-the visit during a visit have not been adequately addressed. The argument put forth that factors that transpire concurrent with a sightseeing bus tour may trigger visit extension is explored in this study.

Electronic word-of-mouth (eWOM)

The spread of the Internet and social media has transformed the way that individuals communicate their everyday experiences. Under this new reality afforded by technology, word-of-mouth has been largely replaced by eWOM (Filieri & McLeay, 2013; Minazzi, 2015). This includes e-mail, discussion forums, blogs, business sites and a whole host of social media options (Vilpponen et al., 2006). eWOM has been defined as ‘any positive or negative statement made by potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet’ (Hennig-Thurau et al., 2004, p. 39). eWOM allows individuals to interact easily and quickly with large numbers of individuals (Dellarocas, 2003). Although the importance of examining tourists’ eWOM has been acknowledged (Babic Rosario et al., 2020; Ruiz-Mafe et al., 2020), its investigation in various tourism settings and contexts remains underwhelming (Sotiriadis 2017).

The mediating role of sightseeing bus tour affect

According to Gullestad (2005), ‘affect represents an immediate evaluation of the environment, informing us about our experience of ourselves in the world and providing a basis for action’ (p. 28). Affect represents a conscious and subjective aspect of an individual’s emotions. Within the tourism setting, affect refers to the emotional responses during a travel experience (Russell, 1979). For this study, tourist’s affect towards the bus sightseeing tour draws on the conceptualization of brand affective evaluation by Alcántara-Pilar et al. (2018).

Affect has been receiving increasing attention within the tourism literature (Prayag et al., 2013). Tourists are increasingly seeking pleasurable experiences and positive emotional expectations (Bigné et al., 2005; Hosany, 2012). Evidence suggests that tourists’ affective states are significantly influencing their satisfaction, revisit intentions, travel intentions and loyalty (Bigné et al., 2005; Loureiro et al., 2021). Yet, there is limited evidence on the role of affect at a specific destination level. Among the few exceptions, Yuksel et al. (2010) found that positive emotional and cognitive links with a tourism destination will lead to a sense of loyalty to the place. Given that intention to extend a current visit is reasonably considered an indicator of satisfaction with the destination and loyalty to stay, then an expectation that positive tourists’ affect will increase tourists’ intention to extend their stays at a particular destination seem plausible. Therefore, this study hypothesizes that:

H₃: Sightseeing bus tour affect enhances tourists’ intention to extend current visits at the destination.

It has also been shown that tourists’ affective states have a positive impact on intention to communicate their experiences while at a destination (Hosany et al., 2017). Regarding eWOM in particular, Verhagen, Nauta and Feldberg (2013) posit that (negative) eWOM can be regarded as the outcome of an unpleasant emotional release. As the authors posited,

an explanation for this relationship comes from the theory of social sharing (Rimé, 2009), which states that people want to communicate their emotions openly with others as a way to arouse empathy, to get help and support, to get social attention, or to strengthen social ties. (Verhagen et al., 2013, p. 1433)

In addition, recent insights by Ladhari (2007) suggest that emotions (both positive and negative) influence the possibility, the type and the amount of WOM produced (Nyer, 1997; White & Yu, 2005). In tourism, Serra-Cantallops et al. (2020) have also confirmed the key role of eWOM is influenced by guests’ emotional/affective states within a hotel context. Hence, the next hypothesis formed is:

H₄: Sightseeing bus tour affect enhances tourists’ eWOM communication of their bus tour sightseeing experience.

Not surprisingly, emotions are triggered by the bus sightseeing tour experience and the type of emotion is dependent upon the evaluation of this event or tourist activity a more so satisfactory/positive or unsatisfactory/negative (Prayag et al., 2017). Thus, Affective Events Theory enables researchers to discretely examine sightseeing bus tour experiences as associated with short-term events related to viewing each touristic attraction/activity during the trip versus an entire sightseeing bus tour as an overall singular event. Therefore, sightseeing tours may be dissected into finer touristic events that can influence bus tour passengers’ affective states. In turn, this may encourage them to change their travel plans by extending their length-of-stay, and/or share their opinion about the destination over the internet. Taken together, in consonance with Affective Events Theory, we hypothesize that:

H_1–3: Sightseeing bus tour affect mediates the relationship between bus tour sightseeing experience and tourists’ intention to extend the length of their current visit at the destination.

H_1–4: Bus tour affect mediates the relationship between bus tour sightseeing experience and tourists’ eWOM communication of the experience.

Conceptual model development

Following from the hypotheses building provided above, Figure 1 depicts the conceptual model of this study to be put under examination and further analysis.

Figure 1. The hypothesized model under investigation.

Methodology

Contexts for the study

Two European cities were chosen as representatives of medium sized, developing urban tourism destinations: Birmingham/United Kingdom and Valencia/Spain. Large destinations were avoided for two reasons: such destinations tend to have more previous visits that might bias future visits; and major urban destinations also tend to offer experiences that address numerous visitation motives. On the other hand, emerging urban destinations have relatively lower number of repeating visitors and more specific motives for tourist visits. Birmingham is the second-largest city of the UK with 1.15 million inhabitants and 1.11 million inbound tourists in 2019 (VisitBritain, 2020). Valencia is the third-largest city of Spain with an urban population of 1.5 million inhabitants and 922,000 international visitors in 2019 (VLC, 2020). Notwithstanding both destinations have deep cultural backgrounds and histories, Birmingham is a city-break destination, while Valencia targets vacationers. Hence, these two urban destinations exhibit different destination profiles and attractiveness.

Furthermore, Birmingham and Valencia were selected because their city councils have recently started investing heavily in the advancement of their respective local tourism destination products/services. The city councils are strongly supportive of new tourism destination strategies. Both cities are developing their urban destinations with an aim to substantially enter the lucrative tourism sector. These two destinations offer hop-on/hop-off sightseeing, which is the most international form of city tour services. Typically, tourists hop-off several times, and therefore this type of bus-tour facilitates administration of surveys while they wait to board the next bus.

Tourists visiting these two destinations were asked to complete a survey questionnaire. The research team employed the intercept sampling technique: while being on-boarded or waiting for the next bus after their hop-off, survey-collectors gathered data from passengers who had already formed perceptions about their on-board experiences.

Methodological procedures

The content validity and reliability of the measurement items were established through a number of research procedures both before and during data-collection. This was done in order to moderate any repercussions that might emerge from potential response set or response style biases. First, the questions were mixed for about half of the survey questionnaires (McLafferty, 2003); additionally, some questionnaire item scales were reversed (Linacre, 2002). Moreover, particular attention was paid to lessen any situational pressure upon participants while completing the survey (McNeeley, 2012). Lastly, care was exercised to ensure similar data-collection conditions were applied at both field research settings.

A set of measures were undertaken to address commonplace issues associated with survey-based research such as minimizing sampling, participant coverage, measurement and non-response errors (Billiet & Matsuo, 2012). For instance, field research studies can avoid random sampling error by collecting large data samples (Groves & Heeringa, 2006). Also, coverage error was avoided by ensuring that all respondents were sightseeing bus passengers. Furthermore, all participants in the study were informed about the aim and content of the research beforehand. Also, the survey/data-gathering process was conducted by the same group of field researchers in order to minimize systematic biases (Paluck & Cialdini, 2014). To ensure that responses collected from the two field studies were highly compatible, the measurement instrument employed in both cities has been offered both in English and Spanish. The Spanish version was translated from the original English survey instrument by a professional translation service, and then to ensure that both versions of the questionnaire expressed the same information, the Spanish version was double-back translated by two other qualified translators (Fotiadis et al., 2021). Subsequently, ambiguous wording was avoided in all measurement scales, so that acquiescence response bias would be controlled (Krosnick, 1999). An extra point was included in Likert scales to facilitate any indecisive tourists to be more accurate in their responses, thus contributing to minimizing the effect of midpoint responses (Weems & Onwuegbuzie, 2001).

Finally, two techniques were employed to assess the potential effect of common method bias on the results: (a) Harman’s one-factor test and (b) the Common Latent Factor (CLF) technique.

Measures

The selection of measures of the variables in this study was determined by appropriately matching the measurement instruments/items with the context of this study.

The ‘bus sightseeing experience’ construct was measured through 29 measurement items and followed the corresponding factorial structure provided in Carreira et al. (2014). Bus tour passengers were asked to indicate their responses on a seven-point Likert scale, ranging from ‘1 = strongly disagree’ to ‘7 = strongly agree’, ‘0 = I cannot answer’ was added as an extra response option in order to lessen/avoid false neutral responses (Weijters et al., 2008).

The ‘past sightseeing bus experience’ was adopted from Weaver et al. (2007) five-item scale and adapted to the current setting. All values provided from the responses were then standardized before data analysis.

The ‘bus tour affect’ emerging from bus tour sightseeing was measured by adapting three items of brand affect (‘I feel good when I use this … ’, ‘This … makes me happy’ and ‘This … gives me pleasure’) developed by Chaudhuri and Holbrook (2001). These items were measured on a seven-point-Likert scale, anchored with ‘1 = strongly disagree’ and ‘7 = strongly agree’, and ‘0 = I cannot answer’ as an extra point for respondents being unsure of their emotions for the particular setting.

‘e-Word of Mouth communication’ on sightseeing bus tours was measured via four items found in Brüggen et al. (2011) again on a seven-point Likert scale, with anchors of ‘1 = strongly disagree’ and ‘7 = strongly agree’. Again, a ‘0 = I cannot answer’ extra point applied.

‘Current visit extension’ was measured on a unidimensional basis; tourists were asked to talk about their plans in response to the following statement: ‘I will extend my current visit to Valencia as a result of having great experience while touring by sightseeing buses’. A seven-point semantic differential scale anchored with ‘1 = extremely unlikely’ and ‘7 = extremely likely’, and with the ‘0 = I cannot answer’ included to record tourists’ plans more accurately.

Sampling

Two field research data gathering campaigns took place in Birmingham/UK and Valencia/Spain where bus tour organizations offer hop-on/hop-off sightseeing services. Data collection took place between September and November 2019. The a-priori sample size considerations via power analysis (G*Power software) for both pen-and-paper surveys, suggested minimum samples of 612 responses for both surveys sites (in Birmingham and Valencia the effect size = 0.5; a = 0.05; power = 0.95; df = 935; critical χ²= 1007.248).

Respondents were approached and invited to participate in the field research study either while actually being on-board which allowed researchers to capture the tourists’ experiences in a best possible way or when waiting for the next bus after their hop-off stops. A self-administered paper-based questionnaire was distributed after the first half of the entire bus tour route. In Birmingham, a total of 733 tourists were asked to participate and 651 agreed, yielding an acceptance rate of 88.81%. From these, 627 usable questionnaires were collected in Birmingham. The final response rate was 85.53%, consisting of 44.6% male and 55.4% female respondents. The age group of 20-to-39-year-old and those coming under the 40-to-59 years age band were almost equally represented in the sample (42% and 42.3%, respectively). In Valencia, 720 individuals were asked to participate and 633 agreed to participate yielding 87.91% acceptance rate. Finally, 615 usable questionnaires were fully completed and usable in Valencia, yielding a final response rate of 85.41%. The sample size exceeded the minimum recommended 612 responses. Concerning the Valencia sampling demographics, 47.8% of the respondents stated they are males and 52.2% indicated female. The 40-to-59-year-old respondents represented the largest age group in the Valencia sample (44.3%), followed by the 20-to-39-year-old passengers (32.5%).

Data analysis

In advance of testing the hypotheses, some preliminary data analysis enhanced the quality of the database; missing values analysis (MVA) was conducted for each dataset before proceeding with structural equation modelling (Hair et al., 2010). Results suggest that missing values follow a random pattern in both samples, i.e. χ_Bir²= 7630.58, df = 7476, Sig_Bir. = 0.104 and χ_Val²= 7880.36, df = 7709, Sig_Val. = 0.085 for the Birmingham and Valencia datasets, respectively (Little, 1988). The univariate normality values of the data were supported, as both skewness and kurtosis were found within proposed limits for all indicators (Mertler & Vannatta, 2004), ranging from −0.895 to 0.297 and −0.979 to 0.964 for Birmingham and, −0.940 to 0.456 and −0.941 to 0.987 for the Valencian sample, respectively. Cronbach’s alpha coefficients of all factors and corresponding dimensions fell between 0.768 and 0.953 for the Birmingham dataset, and between 0.771 and 0.952 for the Valencian one, exceeding the minimum standard for reliability of 0.70 for every construct (Nunnally & Bernstein, 1994). Structural equation modelling (SEM) has been implemented to analyse data and specify the factors that significantly influence tourists’ behavioural intentions. Specifically, covariance-based structural equation modelling (CB-SEM) has been employed, as the primary objective of the study is to estimate the factor-based model; the goal here is theory testing, i.e. the applicability of an adapted version of AET theory, and the estimation of proposed effects in the context of travel and tourism.

Results

Building on a step-by-step argumentation for each hypothesis (direct effects, moderation and mediation) separately, hypothesis testing was conducted via CB-SEM and was processed with AMOS 25 software. This way the suggested model relationships were examined concurrently in the context of urban bus tour sightseeing and visitation. First, reliability and validity of all constructs were examined. Second, confirmatory factor analysis was applied to examine the factorial structure of the measurement model and to check whether all indicators have an item loading >0.50 for explaining latent constructs according to Janssens et al. (2008). Six observed variables were eliminated in both datasets in this step: IS3, IS5, IP4, IP5, VM5 and VM6 (see Appendices 1 and 2). Last, the statistical significance of the hypothesized relationships and the predictive power of the model were assessed by estimating the variance of the endogenous constructs explained.

The proposed model fits the datasets gathered from sightseeing bus tour passengers at both Birmingham and Valencia according to the evidence. All relevant criteria are fulfilled with regards to fit indices for both measurement and structural models (see Table 1). Furthermore, convergent and discriminant validity tests, as well as composite reliability and average variance extracted, support the factorial structure of the proposed model (see Appendices 1 and 2). Specifically, CR estimates of composite reliability are higher than 0.70 for all constructs for both Birmingham (higher than 0.753) and Valencia (higher than 0.844), respectively. Regarding convergent validity, average variance extracted (AVE) values are higher than 0.50 for all latent constructs, i.e. with minimum estimates being 0.513 and 0.547 for Birmingham and Valencian samples, respectively. Finally, with regard to discriminant validity the maximum shared squared variance and the average shared squared variance values are smaller than average variance extracted (i.e. AVE > MSV and AVE > ASV) in all latent constructs and for both urban sightseeing samples. Additionally, the square root of AVE between the different pairs of factors was found in both samples to be greater than the estimated correlation of the factors, supporting discriminant validity of the proposed structures included measurement models, as per Appendices 1 and 2.

Table 1. Fit indices for measurement and corresponding structural models.

Fit indices	Measurement model		Structural model		Criteria
	Birmingham	Valencia	Birmingham	Valencia
χ^2/df	2.526 for p < .001	2.190 for p < .001	2.879 for p < .001	2.310 for p < .001	<3
CFI	.923	.934	.919	.930	>.90
TLI	.918	.922	.913	.919	>.90
RMSEA	.046	.045	.048	.047	<.07 (CFI > .90)
SRMR	.0489	.0516	.0567	.0538	<.08 (CFI > .92)

Note: χ²/df: chi-square normed; CFI: comparative fit index; TLI: Tucker Lewis index; RMSEA: root mean square error of approximation; SRMR: standardized root mean residual.

The values of multivariate and univariate normality were also estimated for both datasets via AMOS. Mahalanobis distance was 155.881 and 111.938 for Birmingham and Valencia, respectively. Comparisons with the corresponding chi-square critical value (χ²= 163.11, df = 584, α = 0.05) showed that distance values were smaller than the critical one in both cases, therefore there was no problem with multivariate outliers (Pallant, 2020). Furthermore, multivariate kurtosis was estimated with Mardia’s coefficient (Mardia, 1970), yielding m.k. = 700.636 with a critical ratio of 135.612 for the Birmingham sample and m.k. = 364.283 with a critical ratio of 69.382 for Valencia. In both cases, these values are smaller than the cut-off point value 1520 derived from the p (p+2) formula (Bollen, 1998), where p = 38 is the final number of observed variables in the model. In conclusion, the multivariate normality assumption stands for both cases of the sample data distributions.

In regard to common method bias, Harman’s one-factor test has indicated the total variance extracted (TVE) by one factor is much lower than the recommended threshold of 50% in either the Birmingham dataset (TVE = 32.938%) or in the Valencian one (TVE = 31.073%) (Podsakoff et al., 2003). Common method variance was also examined through a common latent factor (CLF) that was included in the measurement model to assess the variance explained by the common latent factor (Eichhorn, 2014). The chi-squared difference tests in both cases revealed there are no significant differences between the unconstrained and the fully constrained models (Δχ²_Bir = 21.825, df = 15, and p = .112 > .05; and Δχ²_Val= 22.318, df = 15, and p = .099 > .05 for the datasets originating from Birmingham and Valencia, respectively). Subsequently, these tests provide a strong indication that any common method bias effects may not critically distort the findings resulting from these datasets.

Loadings show that all the seven dimensions of bus tour experience, as proposed by Carreira et al. (2014) are reflecting the construct of bus tour sightseeing experience. The results strongly suggest significant effects/standardized regression weights (St. RW) of higher influence exerted from Information provision (St. RW = .637 and .881 with p < .001, for Birmingham and Valencia, respectively) and Ticket Services (St. RW = .635 and .780 with p < .001, respectively), Staff’s Skills (St. RW = .696 and .570 with p < .001, respectively), and somewhat lower influence of Individual Space (St. RW = .438 and .532 with p < .001, respectively) and Social Environment (St. RW  = .526 and .466 with p < .001, respectively). Health issues were not explicitly addressed in such scale. Also, ‘socialize with other people’ (4.71 and 3.38) and ‘seat cleanliness’ (5.69 and 4.47) items indicated how social distancing and cleanliness would impact the experience in Birmingham and Valencia, respectively (Table 2).

Table 2. Structural model results for Birmingham and Valencia.

Regression paths			City	St. RW	S.E.	C.R.	p
Bus tour affect	⇓	Bus tour sightseeing experience	Birmingham	.625	.06	10.833	<.001
			Valencia	.483	.07	7.422	<.001
Bus tour affect	⇓	Past sightseeing bus experience	Birmingham	.118	.04	2.816	.005
			Valencia	.096	.04	2.058	.040
Bus tour affect	⇓	PSBE_x_ΒΤSΕ	Birmingham	−.128	.06	−2.940	.003
			Valencia	−.021	.08	−.397	.691
e-WOM	⇓	Bus tour affect	Birmingham	.635	.05	12.700	<.001
			Valencia	.621	.07	9.142	<.001
Current visit extension	⇓	Bus tour affect	Birmingham	.314	.05	6.280	<.001
			Valencia	.263	.05	5.600	<.001

Note: e-WOM: e-word-of-mouth; PSBE: past sightseeing bus experience; BTSE: bus tour sightseeing experience; St. RW: standardized regression weight; S.E.: standard error; C.R.: critical ratio; p: p-value.

Regarding the regression weights of the core paths examined, all three hypothesized direct effects were found to be highly significant for both samples, as shown in Figures 2 and 3. The results demonstrate a positive relationship between bus tour sightseeing experience and bus tour affect for both destinations, as expected in H₁. Since the results refer to two developing urban destinations, this finding adds value to the adoption of sightseeing bus tours as a useful component of the overall destination’s product portfolio.

Figure 2. Structural model results of Birmingham sample.

Figure 3. Structural model results of Valencian sample.

The relationship between bus tour experience and bus tour affects serves as a starting point of the visit to a destination as well as the associated emotions elicited from the sightseeing experience while on-board. This relationship is pivotal for underpinning subsequent behaviours of extending the current visit (H₃) and generating eWOM (H₄). Indeed, the sightseeing bus tour can be considered as a driver/motivator of future tourist behaviours such as extending the current visit and eWOM activity through the elicited positive affect (Η_1–3 and Η_1–4, respectively). The regression values of the effects exerted from bus tour affect show a higher influence on eWOM (β = .635 and .621 with p < .001, for Birmingham and Valencia, respectively) than those influencing a possible extension of the current visit to a destination (β  = .314 and .263 with p < .001, for Birmingham and Valencia, respectively). This difference might be attributed to personal restrictions for extending the visit for those who already had standardized visitation plans with predetermined/fixed length-of-stay or other subsequent tine-bound bookings to other places.

Finally, the hypothesized moderating effect emerging from the influence of past sightseeing bus experience on the relationship between the current bus tour sightseeing experience and bus tour affect is negative and strongly significant in the case of Birmingham (β = −.128 with p  = .003). This means that past sightseeing bus experience serves as a buffering variable to the respective direct effect. However, the same effect appears to be non-significant for the Valencia case (β = −.021 with p = .691). In the Valencia case, past sightseeing bus experience seems to function as a control variable as indicated by the barely significant effect resulted from data analysis (β =−.096 with p = .04 < .05).

Overall, the study’s proposed model has good predictive power explaining 42.1% and 24.3% of the variance of bus tour affect for Birmingham and Valencia, respectively (see Table 3). Moreover, 40.4% and 38.5% of eWOM variance was explained, which shows the strong predictive ability of the hypothesized model for both samples. Additionally, the model predicts tourists’ plans for an extension of their visit in Birmingham and Valencia by 9.8% and 6.9%, respectively, based solely on their bus tour sightseeing experience. These are no negligible figures at all, as they represent the explanatory values of the effect of a sole tourism activity on tourists’ behaviour.

Table 3. Squared multiple correlation values (R²) of endogenous latent variables.

Endogenous latent variables	R^2
	Birmingham	Valencia
Bus tour affect	.42	.24
e-WOM	.40	.39
Current visit extension	.10	.07

Discussion

According to Ryan (1999), enhancing travelling with new experiences is crucial for building a holistic aspect of a tourism product/service. Yet, although transportation and tourism are indispensably linked according to Lew and McKercher (2006), and that bus tour sightseeing in urban destinations is found throughout the world, up-to-date, detailed knowledge of their interdependent relationship is extremely limited. With this in mind, the present research study sheds light upon the bus tour sightseeing experience in developing urban destinations through scrupulous research and has detailed its influence on tourists’ decision-making. Although sightseeing bus tours of urban destinations shows great potential for growth, their attractions and distinct offerings may not be well-known to the public at large. Bus tour sightseeing operators and services can proactively, creatively and positively contribute to making these destinations more appealing to potential visitors with the insights gained from this research.

Specifically, the present study focused on the impact of various dimensions of bus tourists’ experience such as individual space, information provision, staff’s skills, social environment, vehicle maintenance, off-board facilities and ticket line service (Weaver et al., 2007) upon the tourists’ willingness to extend their current visit. Also, this study enlightened the potency of the spread of bus tourists’ positive evaluation of travel experience through eWOM communications. This was accomplished largely by adopting/adapting the Affective Events Theory (Weiss & Cropanzano, 1996) to the context of sightseeing bus tour experiences. Until this study, the AET had been widely used within the organizational behaviour domain but not in the consumer behaviour domain. The AET in this study centred upon the mediating role of emotions (i.e. affect) in the relationship between bus tour sightseeing experiences and the behavioural intentions under investigation. Additionally, a context-specific variable (past sightseeing bus experience) has been considered as a moderator in the relationship between sightseeing bus tour experience and affect. These relationships were studied in two developing urban destinations, Valencia/Spain and Birmingham/United Kingdom.

The evidence garnered from these two locations shows that bus tour sightseeing experience triggers tourists’ positive affect which in turn enhances their intention to extend their visit and to spread comments about the destination via electronic media eWOM. It appears that bus tour sightseeing serves as a starting point of a visitation experience and can wield considerable influence on tourists’ reactions. In terms of extending current visit, the key role of affect in tourists’ decision-making process aligns with previous studies (Castañeda, Rodríguez-Molina, Frías-Jamilena, & García-Retamero, 2020; Lee et al., 2018). As for tourists’ intention to engage in positive eWOM about the destination, our results also reinforce those reported in similar research from across the tourism and hospitality industry. For example, evidence from hotels suggests that positive emotions may enhance eWOM (Serra-Cantallops et al. 2020). In restaurants, favourable affective states lead to certain eWOM recommendations and make individuals more open to follow the advice provided, being higher in positive–negative than with negative–positive sequences of comments (Ruiz-Mafe et al., 2020). The same likely holds true for sightseeing bus tour experiences according to the findings of this study.

Concerning the role of past sightseeing bus tour experience, Baloglu (2001) noted that previous experience reflects the level of familiarity one has with an activity which will inevitably influence one’s evaluation of that activity. Given that tourists’ familiarity with the bus sightseeing experience may vary, our expectation was that tourists’ positive affect that stemmed from the bus tour sightseeing experience would be strengthened by the tourists past sightseeing bus tour experiences.

Data analysis indicated that past bus tour sightseeing experience may dampen (not strengthen) the effect of bus sightseeing experience on bus tour affect, perhaps being better informed and thus more critical. In both samples in this study, the moderating effect was significant only in the case of Birmingham/UK. Comparing the moderating effects of the past sightseeing bus tour experience for the two samples collected in these urban destinations, revealed there is not a statistically significant difference (t-statistic = 1.604, for p = .109 > .05). Since no significant differences in past bus tour sightseeing experience between samples were found, this divergence could be attributed to the type of previous bus sightseeing experiences or to the type of destination. Indeed, both destinations differ in their positioning, as this is reflected in their websites. Rich and diverse cultural attractions are featured in Birmingham/UK’s city promotion, while Valencia/Spain focuses on the Mediterranean lifestyle and ‘the good life’ with high number of tourist-night attractions and a relatively low number of total attractions (Lascu et al., 2018).

Conclusions and implications

Conclusions

By extending the Affective Events Theory (Weiss & Cropanzano, 1996) into the touristic consumer domain with its emphasis upon affect/emotions (Bagozzi et al., 1999; Roseman, 1991), this study demonstrates the critical role of affective reactions to different life experiences or events. By using the Affective Events Theory in analysing tourism reactions beyond the workplace, shows its suitability for future researchers. Data from Birmingham and Valencia indicated that bus tour sightseeing experience generates tourists’ positive affect which further increases their visit intention positive electronic word-of-mouth with respect to the destinations. All in all, bus tour sightseeing is a vital aspect of the visitation experience, exerting significant effects on tourists’ communications, and playing a key role in tourists’ decision-making processes and further vacation planning.

Theoretical implications

This study empirically tested two behavioural outcomes of interest in the broader tourism industry. First, the extension of a current visit at any destination is crucial for the successful performance of the destination operators by stimulating additional revenue while visitors are still there. Moreover, the value of a positive relationship between affect and visit intention lies in the moment of the experience and might be considered an opportune occasion for up-scaling/selling. Most tourism destination efforts have focused on attracting future visits to the same destination, instead of extending the current visit. The idea of extending versus repeating visits, supported by our data, reveals the feasibility of adopting loyalty schemes during consumption in the tourism domain. Furthermore, previous literature is sparse in identifying factors that may influence a tourist’s decision to extend an ongoing, visit at a destination. This study shows that the experience and affect generated by a single tourist bus tour sightseeing activity has a significant effect and considerable explanatory power (of about 10%) upon tourists’ plans to extend a current visitation.

Second, literature on eWOM formation primarily concentrates on a set of factors identified by Babic Rosario et al. (2020) such as altruism, social value from interaction, hedonic benefits, impression management and identity formation, balance restoration, venting, retribution and economic incentives. However, the idea of generating eWOM communication and relevant recommendations such as these from the affective states related to a tourist activity, in this case the bus tour sightseeing experience, was a theoretical conceptualization that had not been previously investigated.

Overall, the implications for theory are twofold: identifying and modelling the effects on both the selected tourist activity as well as the overall visitation plans at a tourism destination. This study highlighted two different perspectives of how tourists’ positive experience from a certain sightseeing activity not only enhances tourists’ positive intentions towards the specific activity – but also helps in better theorizing its effect on tourists’ wider visitation plans.

As a caveat, this study was conducted before the worldwide COVID-19 pandemic and therefore did not explicitly address how the pandemic situation would affect the bus experience. However, two items of the scale of bus tour sightseeing experience may be related to a pandemic situation, such as ‘socialize with other people’ and ‘seat cleanliness’. It might be expected that both, socializing and cleanliness, may exert a stronger influence circumstance prevalent at both these destinations around the time of this publication date. Therefore, space, social distance and preventative measures enacted during the COVID-19 pandemic should be integrated as a part of the bus tour sightseeing experience.

Practical implications

For practitioners, the outcomes of this research should be useful for Destination Management Organizations (DMOs). The findings of this study clearly demonstrate the importance of running bus tour sightseeing services in developing urban destinations. The findings also suggest that sightseeing bus tours can play a key role in the destination’s tourism products/services since positive eWOM communication about a current bus tour service and can influence tourists’ consideration of extending their visit.

Since the most relevant, salient sightseeing bus tour experiences usually take place in the first hours of the visitation to a destination and continue to accumulate thereafter, then keen attention to the ongoing service is an important strategic function. Active engagement and enhanced attention towards shaping tourists’ overall attitudes and intentions towards the destination can yield positive benefits. This includes revising initial visitation plans to spend more time at a specific destination. With this knowledge in hand, DMOs can set reasonable goals for lengthening tourists’ stays. Perhaps, a local network of bus sightseeing services would be an asset not only for the business per se, but also for the development of an emerging urban tourist destinations.

Bus tour sightseeing is typically based on a hop-on/hop-off system that allows visitors to choose an attractive place at the right time based upon their preferences and any emerging interests. The COVID-19 pandemic has necessitated the closing of many indoor tourist attractions; therefore, a bus tour sightseeing operation might reschedule routes to include open attractions, or even postpone/stagger the hop-off exchanges in order to prevent over-crowding at attractions. Such an agile approach could be implemented in close contact with health authorities. Unpredictable changes in routes or stops may result in managerial complexity but not an impossibility. Further research should address if such changes occur then what is the influence upon sightseeing tour bus experiences.

This paper also indicated that sightseeing bus experiences prompt eWOM generation, particularly through favourable affective states. As Stangl et al. (2020) posited, introducing affective components in augmented reality mobile applications can enhance entertainment and enjoyment of both app-usage as well as any associated tourism activity. Therefore, DMOs should facilitate usage of digital apps during visitations. For example, this could be actualized through relevant social media tourism-related communities and smartphone downloadable applications that are integrated into urban tourism communication campaigns. It is also anticipated that tourists will continue the ever-growing use of posting user-generated content (UGC) via platforms (TripAdvisor, Instagram, Snapchat, Weibo, Facebook, YouTube) about different attractions seen during their bus tours. Widely posted and viewed comments could reinforce individuals’ destination image amongst visitors, non-visitors, but importantly amongst prospective travellers.

The travellers’ bus tour sightseeing experience as expressed through images can be curated and shared with partner organizations, such as Curata, in order to improve overall marketing performance (Salem & Twining-Ward, 2018). Moreover, if posts (text, pictures and videos) were be managed properly via Digital Asset Management tools such as Acquia, Libris, WebDam, then the entire destination would be promoted, and not just the key landmarks and attractions of destinations earmarked on the bus tour itinerary.

The sightseeing bus tour experience, in UGC content paired with the bus tour affect revealed in text, photos and videos via sentiment expressions, point to exactly how bus sightseeing businesses and urban DMOs can utilize UGC to advance destination management performance indicators. This would be to the benefit of sightseeing companies as well as the overall market positioning of the destinations. This approach could also improve social listening and reputation management such as ReviewPro app to trigger communication with potential customers/passengers by monitoring sightseeing and destination sentiments. This would assist bus tour organizations to achieve increased numbers of followers, rankings and overall marketing performance.

Limitations and suggestions for future research

As with all empirical research, this study has limitations which may lead to opening new research avenues. As Pirola-Merlo et al. (2002) noted, appraisals of an event produce the experience of discrete emotions (such as joy or anger). Yet, the adoption of the positive–negative affect typology inevitably hinders the identification of specific emotions (Lerner & Ketlner, 2000). True. In this sense, this study has addressed the overall sightseeing bus experience without identifying the attractions that cause the most appealing elements of the bus tour experience. Future research could address both; for instance, examine the linkage between specific attractions and specific emotions and affect. Then, if any, consider the order of visits to particular attractions in light of their positive affect/emotions that might be stimulated. Future studies can explore whether differences in eWOM formation are based upon sociodemographic characteristics and/or by individual versus group visitors (family members). From a methodological viewpoint, future studies may apply an alternative CB-SEM approach via a multigroup analysis (MGA), which could provide useful comparisons between different tourism destinations on the basis of the sightseeing experience effect.

Additionally, other antecedents such as place attachment, satisfaction, travel party, social norms can be included in subsequent empirical research to explain more of the variance in eWOM and revisit intentions. Also, future studies can replicate this study to different urban settings and/or geographical locations around the world. And, as suggested earlier, the effects of the COVID-19 pandemic deserve further research in two directions: (i) how preventive measures of COVID-19 are/were impacting on-board experiences; (ii) how managers face(d) changes in routes and stops. In closing, it might be of interest to analyse future behaviours subsequent to the trip, such as revisits to the same destination and repeated usage of sightseeing buses in the same destination.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Spanish Ministry of Science and Innovation [grant number PID2019 111195RB-I00/AEI/10.13039/501100011033].

Notes on contributors

Nikolaos Stylos

Nikolaos Stylos is a Senior lecturer/Assoc. Professor of Marketing at University of Bristol, UK. His research focuses on tourism marketing and decision-making with smart technologies.

Enrique Bigné

Enrique Bigné is a Professor of Marketing in the School of Economics, at the University of Valencia, Spain. His research interests are advertising and tourism marketing destinations.

Victoria Bellou

Victoria Bellou is a Professor of Management at the Department of Economics, University of Thessaly, Greece. Her main research interests include employee behaviours and tourism management.

Appendices

Appendix 1. Construct reliability and validity measures of the measurement model for Birmingham

Table

	CR	AVE	MSV	ASV	eWOM	Bus tour sightseeing experience	Bus tour affect
eWOM	0.950	0.826	0.397	0.394	0.909
Bus tour sightseeing experience	0.753	0.513	0.450	0.421	0.626	0.716
Bus tour affect	0.961	0.891	0.450	0.424	0.630	0.671	0.944

Note: CR: composite reliability, AVE: average variance extracted, MSV: maximum shared squared variance, ASV: average shared squared variance.

Appendix 2. Construct reliability and validity measures of the measurement model for Valencia

Table

	CR	AVE	MSV	ASV	eWOM	Bus tour sightseeing experience	Bus tour affect
eWOM	0.933	0.777	0.372	0.330	0.882
Bus tour sightseeing experience	0.844	0.547	0.287	0.253	0.536	0.740
Bus tour affect	0.934	0.825	0.372	0.296	0.610	0.468	0.908

Note: CR: composite reliability, AVE: average variance extracted, MSV: maximum shared squared variance, ASV: average shared squared variance.