Interaction quality and satisfaction: An empirical study of international tourists when buying Shanghai tourist attraction services

Abstract

The main purpose of this article is to examine the relationship between the quality of interaction with tourist attraction services and satisfaction. The model proposes that interaction quality (IQ) affects both satisfaction with tourism services and the overall tourist satisfaction (OTS) with the destination. In addition, the link between the OTS and the intention to revisit the destination is assessed. Data were collected from 165 international tourists visiting Shanghai tourist attractions and analyzed through PLS-SEM using second-order hierarchical modeling. IQ is seen as a multidimensional construct, including aspects such as language, nonverbal behavior, attitude, and expertise. The results confirm the direct effect of IQ on service satisfaction and the OTS as well as its indirect effect on revisit intention. The overall tourist’s satisfaction does directly impact the intention to revisit the destination. For international tourists visiting Shanghai, IQ constitutes an essential part of their experience not only with services rendered but also with the destination itself.

Keywords:

• interaction quality
• service satisfaction
• overall tourist satisfaction
• revisit intention
• Shanghai tourism services
• PLS-SEM

PUBLIC INTEREST STATEMENT

During the last three decades, Shanghai has experienced a rapid rise as a tourist destination in Asia. However, while the city has much to commend it, there is still a lot of work to be done to make it tourist friendly in its services. This article analyses how interaction quality during a service encounter impacts customers’ service satisfaction, the overall tourist satisfaction with the destination, and the revisit intentions. The findings show that interaction quality during a service encounter at Shanghai tourist attraction services is crucial when determining international tourists’ level of satisfaction with both services and destination in general. Thus, destination managers should create employees’ training programs to share how to deliver an excellent interaction with a foreign tourist, and together with the local government develop integrated marketing strategies to improve the image of the city, and motivate overseas tourists to revisit Shanghai. Aspects such as language accommodation, employees’ attitude, and nonverbal cues are vital aspects of a tourism service interaction.

1. Introduction

Shanghai is seen as a cosmopolitan metropolis, a big city where modernism and traditionalism converge. It is seen as one of the safest cities in Asia and internationally recognized for having one of the best public transportation systems in the world. It is also a city where it is easy for tourists to speak English and find the exotic and mystical culture they are looking for. However, while all this is true, there is still a lot of work to be done to make it tourist friendly. This is especially true during intercultural service encounters where the employee is Chinese and the customer is foreign.

According to the China Institute of City Competitiveness,¹ Shanghai was the most competitive city of China in 2017, followed by Hong Kong, Shenzhen, and Beijing. In another study from the Center for China and Globalization (CCG) and Ctrip,² Shanghai has also been selected as the most popular city for overseas tourists to visit, ahead of Beijing. This may be, probably, because if well it is true that Beijing is the political center of the Asian giant, the city suffers from heavy pollution which may detract tourists. Indeed, the number of tourists visiting Beijing decreased by 5.8% last year. When visiting Shanghai, tourists find that not only the air is cleaner, the subway is less crowded, it has the highest skyscrapers of China and it also many foreign tourists view it as the New York of Asia. In addition, the Shanghai Tourism Bureau³ shows that by the end of 2017 the city received 8.730.111 of foreign tourists, which is more than 4 millions international tourists who visited Beijing,⁴ increasing by 2.18%. At the same time, the statistics demonstrate that the inbound tourist market is changing. The rate of tourists coming from countries such as South Korea has decreased in a 19.13% while countries such as Australia, Thailand, Canada, and the United States have considerably increased on 14.41%, 16.30%, 15.73%, and 5.94%, respectively.

Additionally, the report of the CCG and Ctrip points out that overseas tourists coming to China are mainly for business, although, potential new tourists showed more interest in leisure and sightseeing. The research also confirms that foreign tourists are highly satisfied with service transportation, however, aspects such as exorbitant prices of some services, food hygiene, and other tourism experiences are still a problem. Overseas tourists still complain about employees’ attitude, nonverbal behaviors, and/or satisfaction with some tourism services when visiting the city. Hence, it is important to conduct research and develop strategies to tackle these problems and keep increasing the number of international tourists visiting the city.

Since tourism is an amalgam of service industries (Fuchs & Weiermair, 2003, p. 7), tourists will typically purchase and consume a whole range of services, which together make up the holiday or vacation experience (Weiermair, 2000, p. 398). Furthermore, it is essential to pay attention to what tourists (customers) value during service encounters. This will help service providers to attract more customers and deliver a service that exceeds expectations, enabling tourists to think about taking another trip to Shanghai and recommend the city as a tourism destination.

Thus, this article considers interaction quality (IQ), service satisfaction (SS), and the overall tourism satisfaction with the destination as important sources of a city competitive advantage. Past literature asserts that aspects such as visitor satisfaction, service quality, employees’ attitude and skills, prices, level of visitor safety, visual appeal, and others resources determine a destinations’ competitiveness. In addition, the service encounter is seen as the face-to-face interactions between the customer and service provider in a service setting (Lloyd & Luk, 2011, p. 177). The overall tourism satisfaction with the destination is seen as the sum of small episodes of positive experiences during the trip, which includes not only the attributes of the destination such as nature, people, and city life, but also the whole amalgam of services (transportation, accommodation, dinning, visitor attractions), products, and interactions during the trip. In contrast, satisfaction with a tourism service is seen as an episode of the overall tourist satisfaction (OTS) with the destination, where customers evaluate all the service attributes and provide an enjoyable level of consumption-related fulfilment (Oliver, 1997, p. 13).

Our aim is to examine the relationship between the quality of interaction with tourism services and satisfaction. The model proposes that IQ affects both satisfaction with tourism services and the OTS with the destination. In addition, the link between the overall tourist satisfaction and the intention to revisit the destination will be assessed. By doing so, we expect to help managers to improve their service, paying special attention to IQ as a key driver of satisfaction.

2. Literature review

2.1. Interaction in the literature

To live is to live in constant interaction. Hence, nothing in this world operates in isolation. Many authors have studied the interaction as a process (e.g., Abrahamsen, 2016; Blumer, 2012; Bourdieu & Wacquant, 1992; Jensen, 1998; Rafaeli, 1988; Spencer-Oatey, 2000; Wagner, 1994; West & Turner, 2010) from different angles such as human–computer connection, sociolinguistics, communication, physics, human information behavior, branding, firm performance, and consumer culture theory (Arnould & Thompson, 2005; Fyrberg & Jüriado, 2009; Homburg, Wieseke, & Bornemann, 2009; cited in Grönroos & Voima, 2013; p. 140). Thus, offering a wide spectrum to services marketing where services have been defined as “a series of interactions between participants, processes and physical elements” (Tax & Stuart, 1997, p. 107), or to relational marketing (Grönroos, 1982, 1984; Grönroos & Voima, 2013; Gummesson, 2002), interactive marketing (e.g., Echeverri & Skålén, 2011 ; Grönroos, 1982, 2011), or experiential marketing, where interactions have been studied mostly in the context of buyer–seller relationships (Snehota & Hakansson, 1995), and experiences have been said that come from interactions (Ramaswamy, 2011, p. 195), and therefore, interactions become the locus of value creation (Prahalad & Ramaswamy, 2004, p. 12).

However, it is hard to portray what an interaction episode looks like, when it starts or finishes, since a moment of interaction is part of a continuum of facts, and therefore, what we see is the continuation of things from the past (Abrahamsen, 2016; p. 467; Håkansson etal., 2009, p. 36). Abrahamsen (2016) states that these interaction episodes are all interconnected. For this reason, what occurs in one relationship cannot be seen as an isolated element but quite the opposite. The author argues that it is the interpretation of interactions rather than reality itself that forms the basis for interaction (p. 468). In a service context, Echeverri and Salomonson (2017) state that during a service encounter, both the employee and the customer participate actively in a dialogue process, and it is here where creation and destruction of value can take place.

Within the marketing literature, several studies have been done on interactions, whether through brand communities (Schau, Muñiz, & Arnould, 2009), technology applications (Racherla, Babb, & Keith, 2011), or face-to-face (Echeverri & Skålén, 2011; Lloyd & Luk, 2011; Solomon, Surprenant, Czepiel, & Gutman, 1985). For example, Sheth makes a relevant distinction between two interaction dimensions: the content versus the style of communication (Solomon etal., 1985, p. 101); the content of communication entails suggesting, offering, promoting, or negotiating a set of product-specific utilities and their expectations (Sheth, 1976, p. 382); the latter dimension recognizes the centrality of ritualistic behavior patterns in shaping the outcome of the buyer/seller interaction (Solomon etal., 1985, p. 101) which includes task-oriented, tradition-oriented, and people-oriented style (for more, see Sheth, 1983). Bailey etal. (2001) recognize the emotional elements of the interaction through 14 prepositions and posit that understanding the emotional aspects of an ongoing interaction is related to the concept of emotional intelligence (Goleman, 1995; Salovey & Mayer, 1990), especially the dimensions of empathy and self-regulation (p. 9), and Albrecht, Hattula, Bornemann, and Hoyer (2016) assert that employees’ emotions may influence customers’ behaviors during the customer service interaction through the conscious or unconscious induction of behavioral attitudes (Schoenewolf, 1990, p. 705). Thus, interaction can be seen as the mutual relation where actors have an effect upon each other, whether positive or negative and can be carried out through different channels.

2.2. IQ and its dimensions

Face-to-face or direct interaction plays an important role in any service setting. It is through interaction that customers and employees get to know each other for a short period of time and are able to leave a certain impression on both sides. Therefore, customer–employee direct interaction is no more than the relationship where actors can have an effect upon each other, whether positive or negative. Brady and Cronin (2001) express that because services are inherently intangible and characterized by inseparability, the interpersonal interactions that take place during service delivery often have the greatest effect on service quality perceptions (p. 38) and therefore, on the perceived experience value of the service and the OTS.

Grönroos (1982, 1984) and Parasuraman etal. (1988) are some of the first authors to suggest the term IQ. Grönroos (1982) asserts that when designing a customer–employee interaction, it is imperative to acknowledge all the quality components (p. 33). In the same article, Grönroos posits that within a service encounter, the manner in which the employee behaves and communicates with the customer is crucial for the perception the customer gets of service (p. 38). Chahal (2010) finds that dimensions such as attitude, friendliness, helpfulness, and responsiveness influence the patient–physician interaction (Hau etal., 2017, p. 255). In the same line, Choi & Kim (2013) assert that certain aspects of IQ such as politeness, friendliness, sensitivity, and empathy are considered critical in driving customer satisfaction (p. 191). Gaur etal. (2011) examine how patients’ loyalty and confidence in their doctors are influenced by doctors’ interaction behavior, specifically exploring dimensions such as listening, explaining, and perceived competence. Their study supports the findings of other authors who claim that listening and explaining (e.g., Chandon, Leo, & Philippe, 1997; Ramsey & Sohi, 1997), together with competence (e.g., Chandon, Leo, & Philippe, 1997; John, 1991; Parasuraman etal., 1985, 1988), are important aspects of any interaction, and therefore affect customers’ level of satisfaction. In addition, Wu, Tsai, Hsiung, and Chen (2015) explore the relationship between service competencies of frontline employees and service quality. The findings show that frontline employees’ service competences positively impact customer perceptions of service quality.

On the other hand, some aspects have also been analyzed by referring to IQ dimension. For example, Lloyd and Luk (2011) consider facial expressions, body language, tone of voice, and language as a manifestation of service manner. The authors argue that both facial expressions and body language should communicate qualities such as patience, helpfulness, politeness, passion for the job, and cheerfulness (p. 188). In a qualitative study, Echeverri and Salomonson (2017) find six general behavior practices present in a customer–employee direct interaction, such as mood expressing, caring, connecting, responding, substantializing, and embedding. Each of them include some reciprocal activities (explaining, adjusting, small talk, paying attention, knowledge gaining, and so on). The research findings guide services’ managers on how to assess their interactions with customers.

Holmqvist (2011) discovered that most consumers are prepared to pay more in all service encounters to interact with a service provider who uses the customer’s first language, rather than one who does not (p. 188). Balaji etal. (2017) demonstrate that language divergence has a negative impact on customer assessment of service interactions and strength of relationship with the service provider (p. 213), whereas Kraak and Holmqvist (2016) propose that service providers could benefit from language convergence no matter what language the customer speaks (p. 3). Racherla etal. (2011) posit that interaction face-to-face allows simultaneous observations of multiple informative cues including body language, facial expressions, and the tone of voice (p.143). Puccinelli etal. (2013) find out that an employee capability to read dissimilar types of nonverbal behaviors impacts service’ quality impressions (p. 369).

Hence, aspects such as nonverbal behavior (Ekinci & Riley, 2003; Gabbott & Hogg, 2000; Lloyd & Luk, 2011; Puccinelli etal., 2013), employee’s attitude (Brady & Cronin, 2001; Chahal, 2010; Ma & Dubé, 2011; Ranjan, Sugathan, & Rossmann, 2015), perceived employee’s expertise, or knowledge (Brady & Cronin, 2001; Crosby etal., 1990; Spake & Megehee, 2010; Wu etal., 2015) are some of the important factors enabling interaction in past studies. These are dimensions that we are going to use in our study.

2.3. SS and OTS

For more than four decades, satisfaction has been the center of attention in marketing or services literature, countless researchers have tried to assess and define customer satisfaction (Armario, 2008; Babin & Griffin, 1998; Hansemark & Albinsson, 2004; Hunt, 1977; Neal, Sirgy, & Uysal, 1999; Oliver, 1980, 1996, 1999; Pawitra & Tan, 2003; Prebensen & Rosengren, 2016; Spreng, Mackenzie, & Olshavsky, 1996; Westbrook & Oliver, 1981; Woodruff & Gardial, 1996; Yi, 1991). For example, Oliver (1980) asserts that satisfaction is “a function of an initial standard and some perceived discrepancy from the initial reference point” (p. 460). Pawitra and Tan (2003) affirm that tourist satisfaction can be accomplished by measuring the difference between predicted and perceived service (p. 400). According to Armario (2008), tourist satisfaction with a destination includes not only the tourist’ expectations before and during the trip but also the customer’s perception of the services. Prebensen and Rosengren (2016) state that “satisfaction is the result of the consumer’s evaluation of the experience value derived from the experiences at various service providers through the experience process” (p. 118). Other authors argue that overall satisfaction and attribute satisfaction are not the same (Oliver, 1993). The former can be defined as the evaluation of the consumer’s subjective satisfaction coming from an examination of attribute performance (Oliver, 1993, p. 421), the latter constitutes an aspect of satisfaction with single elements of a product or service that compose the experience, such as accommodation, weather, natural environment, and social environment (Lounsbury & Hoopes, 1985; Pizam & Ellis, 1999; cited in Chi & Qu, 2008, p. 626).

When studying SS at different stages of a trip and the OTS, Neal etal. (1999) assert that the tourism SS and its determinants must be a priority when determining satisfaction with travel or tourism trip experiences. These services are seen not only within the destination, but also during the other phases of a travel such as pretrip, enroute, and return trip. The authors confirm that satisfaction with tourism trip services together with trip reflections, which include perceived freedom from control, perceived freedom from work, involvement arousal, mastery, and spontaneity, play a significant role in determining overall satisfaction with tourism experiences. In a similar line, Chen, Huang, and Petrick (2016) hypothesize tourism satisfaction as a mediator between tourism recovery experience and overall life satisfaction. The authors argue that individuals who experience something new during the trip are more likely to be satisfied with their vacation. In another study regarding to satisfaction with services, Lu etal. (2015) find out that for guests, satisfaction emerges when: (1) the value of service received is equal to or greater than the price they paid, and (2) when the services meet customers’ expectations, thereby, IQ would impact the customer level of satisfaction with the service. In addition, Su, Swanson, and Chen (2016) examine customer satisfaction as a construct of relationship quality and test the full mediating role of satisfaction between service quality and subjective well-being. The investigators state that greater personalized services mean a greater level of satisfaction and lower probability of switching to competitors.

Despite the difficulty to conceptualize and operationalize satisfaction (Dmitrović etal., 2009, p.120), the literature reveals that the level of tourist satisfaction with a particular trip is the result of different factors (Peter & Olson, 1996; cited in Armario, 1996, p. 367). Therefore, we can assert that within a tourism destination SS is an episode of satisfaction where the customer (tourist) subjectively evaluates his/her expectations with what she/he receives. However, OTS is the sum of all small episodes of positive experiences, and hence, small episodes of satisfaction that include not only services at various stages of a trip, but also the subjective evaluation of the individual destination’s attributes. As the literature shows, the OTS with the destination is associated with individual components of the destination (e.g., Danaher & Arweiler, 1996; Hsu, 2003; Mayer, Johnson, Hu, & Chen, 1998; Ross & Iso-Ahola, 1991; cited in Chi & Qu, 2008).

2.4. Revisit intentions

Authors have linked revisit intention (RI) not only with satisfaction (Baker & Crompton, 2000; C. Chen & Chen, 2010; C. F. Chen & Tsai, 2007; Chi & Qu, 2008; Cronin, Brady, & Hult, 2000; Kozak & Rimmington, 2000; Oh, 1999; Petrick, Morais, & Norman, 2001; Ryu, Han, & Jang, 2010; Su etal., 2016; Williams & Soutar, 2009), but also with a memorable tourism experience (Agapito, Pinto, & Mendes, 2017; Barnes, Mattsson, & Sørensen, 2016; Zhang, Wu, & Buhalis, 2017), emotions (Han, Back, & Barret, 2009), novelty seeking (Shawn & Feng, 2007), quality of tourist shuttles (Loi etal., 2017), perceived attractiveness (Um, Chon, & Ro, 2006), and destination image (De Nisco, Mainolfi, Marino, & Napolitano, 2015; Kim & Lee, 2015; Stylos, Vassiliadis, Bellou, & Andronikidis, 2016; Tosun, Bora, & Fyall, 2015). Thus, for many tourism destinations, repeat visitors constitute a desired market segment, because they tend to stay longer at a destination, tend to be more satisfied as they experience and realize expectations (Zhang, Wu, & Buhalis, 2017, p. 4).

Intention to revisit refers to tourists’ plans to visit the same site or destination again (Cole & Scott, 2004; cited in Loi etal., 2017, p. 117), providing the most precise prediction of a decision to revisit (Stylos etal., 2016, p. 42). Um etal. (2006) posit that tourists who intent to revisit the city for first time may be motivated by the performance of the destination as a whole due to their initial stay. The authors determine that OTS mediates the relationship among aspects such as perceived value for money, perceived quality of service, perceived attractiveness, and RI. In another study, Shawn and Feng (2007) measure RIs on short-, mid-, and long-term bases, the study shows satisfaction as the direct predictor of the short-term RI. The authors claim that travel destinations can stimulate tourists’ RI by maximizing traveler satisfaction (p. 587). In addition, Han, Back, and Barrett (2009) discover a strong relation between satisfaction and RI of customer regarding to restaurants. The researchers state that RI can be conceptualized as an asserted likelihood to revisit in the absence or presence of a positive attitude toward the provider (p. 564).

However, to our knowledge, despite the existence of countless studies linking satisfaction and RI, acknowledging that the more satisfied a tourist is with the tourism destination, the higher the probability that this tourist return to the destination, just few of them assess the indirect effect of IQ on RIs through the mediating role of overall tourism satisfaction.

Hypotheses

H₁: Interaction quality during a service encounter has a direct effect on the satisfaction with the service.

H₂: Interaction quality during a service encounter has a direct effect on the overall satisfaction with the destination.

H₃: Service satisfaction directly affects the overall tourist satisfaction with the destination.

H₄: The overall tourist satisfaction with the destination directly affects the revisit intentions.

3. Methodology

3.1. Research design

A second-order reflective–formative hierarchical model of SEM, following a two-stage approach, was used. IQ is seen as a higher-order formative construct formed by four reflective first-order constructs (language, nonverbal behavior, attitude, and expertise), while SS, OTS, and RI are reflective constructs. A reflective indicator is an observed variable implying that the construct causes the measurement of the indicator variables (Hair etal., 2016, p. 13), whereas formative indicator is a variable measuring an assumed cause of or a component of a latent construct (Lowry& Gaskin, 2014; cited in Rahman etal., 2017, p. 5).

Data analysis for this study has been done using SPSS 21 and SmartPLS 3.2. 7 (Ringle, Wende, & Becker, 2015). For this, the procedures given by Hair etal. (2016) are followed. The authors oriented that there are different ways to treat formative and reflective measurement. When evaluating the extended or measurement model, internal consistency, convergent, and discriminant validity need to be assessed for reflective constructs whereas convergent validity, collinearity between indicators, significance, and relevance of outer weights need to be evaluated. Finally, coefficients of determination (R²), predictive relevance (Q²), size and significance of path coefficients, f² effect size, and q²effect size (Hair etal., 2016) are computed when evaluating the structural model.

3.2. Survey instrument

The survey questionnaire used for this study consists of four sections: 1. demographic information, 2. aspects related to IQ (13 items), 3. satisfaction with visitor attraction services (5 items), and 4. OTS with the destination (4 items) and RI (1 item). For sections two and three, respondents were asked to answer on a 5-point Likert scale, from 1 (completely disagree) to 5 (completely agree), for each of the statements given. In section four (5 items), items 1, 2, and 3 were also measured on a 5-point Likert scale, from 1 (completely disagree) to 5 (completely agree); item four was also assessed on a 5-point Likert scale, from 1 (very dissatisfied) to 5 (very satisfied), and item five being from 1 (far below expectations) to 5 (greatly exceeded expectations). RI was measured with one item, also with a 5-point Likert scale, from 1 (definitely not) to 5 (definitely). Table 1 shows the measurement items of each section were adapted from different authors.

Table1. Constructs and items used in the model

Construct	Indicator	Question	Adapted from
Language (LA)	rLA1	The service employee accommodates my language preference	Kraak & Holmqvist, 2016, interviews Holmqvist, Van Vaerenbergh, & Grönroos, 2014
	rLA2	The service employee has good language skills
	rLA3	The service employee uses positive language
Nonverbal behavior (NVB)	rNV1	Voice is not boisterous	Lloyd & Luk, 2011 Jung & Yoon, 2011
	rNV2	Listens carefully to what I have to say
	rNV3	Converses at a proper speed
Attitude (AT)	rAT1	Is friendly and respectful	Brady & Cronin, 2001 Lloyd & Luk, 2011
	rAT2	Is willing to help me
	rAT3	Behaves gracefully
Expertise (KN)	rKN1	She/he knows the job very well	Brady & Cronin, 2001 Lloyd & Luk, 2011
	rKN2	She/he is able to answer my questions quickly
	rKN3	She/he understands that I rely on his/her knowledge to meet my needs
	rKN4	She/he possesses the expertise to analyze customers’ problems
Service satisfaction (SS)	SS1	The cost of this tourist service is reasonable and well worth it	Dagger etal., 2007; Neal etal., 1999, 2007
	SS2	The service experience is well organized and therefore problem-free
	SS3	This service made the trip a richer experience for me
	SS4	The quality of this tourism service has been
	SS5	My feelings toward this tourist service are very positive
Overall tourist satisfaction (OTS)	OTS1	In general, I was pleased with the experience quality of this tourism destination	Neal etal., 1999, 2007 Prebensen, Kim, & Uysal, 2015 Um etal., 2006 Um etal., 2006
	OTS2	In general, this trip helped me to rest and feel fully relaxed
	OTS3	Overall, based on the experiences you had in Shanghai, how satisfied are you with this tourism destination?
	OTS4	To what extent Shanghai meet your expectations as a tourism destination?
Revisit intention (RI)	RI1	How likely would you return to Shanghai in the next five years for pleasure travel?

3.3. Sampling and data collection

This study was carried out in Shanghai, known as the economic and financial center of China. The target population consists of international tourists visiting Shanghai from August to October. Specifically, 251 questionnaires were sent via Survey Monkey and others were distributed face-to-face, some of the questionnaires were dropped at different tourist places around Shanghai in order and collected one week later. From 251 questionnaires distributed, only 189 were received back, 24 of the 189 questionnaires were rejected for being incomplete, so that 165 questionnaires were finally used for the analysis, which yielded an overall response rate of 65.74%.

4. Results

4.1. Demographic findings

Table 2 summarizes the demographic profile of the respondents. Females represent 55.8% of the population studied, of which 21.8% and 15.2% are between 18–25 and 26–35 years old, respectively, whereas the higher percentage of males coming to Shanghai (15.2%) are between 26–35 years. Within education level, women portray a higher level of education with 28.5% having a bachelor’s degree and 16.4% master’s degree. However, when it comes to occupational level, males have a higher employment level in positions such as Director/Manager (9.7%) and self-employed (7.9%), while 16.4% of women are employee. The percentage of females (26.7%) visiting Shanghai for first time is higher than males (18.2%), same as with repeat visits. It is remarkable to note that these differences might be result of a higher number of females (92) visiting the city comparing with males (73). Regarding with nationality, the data show most of the visitors came from North America (f = 54; % = 32.7) and Europe (f = 52; % = 31.5) followed by Asia (f = 27; % = 16.4), Africa (f = 23; % = 13.9), South America (f = 6; % = 3.6), and Australia and Oceania (f = 3; % = 1.8). The three main native languages were English (32.1%), Spanish (13.9%), and French (10.3%).

Table2. Demographic data

Demographic profile	Groups	Frequency	Percentage
Gender	Female	92	55.8
	Male	73	44.2
Age	18–25	55	33.3
	26–35	55	33.3
	36–45	31	18.8
	46–55	6	3.6
	56–65	9	5.5
	Below 17	9	5.5
Occupational position	Student	63	38.2
	Employee	47	28.5
	Director/Manager	23	13.9
	Self-employed	21	12.7
	Homemaker	9	5.5
	Retired	1	0.6
	Unemployed	1	0.6
Trip reasons	Sightseeing/Leisure	61	37.0
	Visit friends/relatives	33	20.0
	Study	22	13.3
	Conference/Business	17	10.3
	Work	14	8.5
	Other	13	7.9
	Study|Sightseeing	5	3.0
Educational level	Bachelor’s degree	77	46.7
	Master’s degree	52	31.5
	Doctorate degree	13	7.9
	College	12	7.3
	High school	11	6.7
Number of visits	First time	74	44.8
	Repeat visit	91	55.2
Travel party	Alone	66	40.0
	Family/Relatives	41	24.8
	Friend(s)	26	15.8
	Spouse	20	12.1
	Tour group	8	4.8
	Business associate(s)	4	2.4

Among the respondents, bilinguals’ native speakers were present, especially from Singapore, Germany, Canada, England, Colombia, France, Ukraine, and the United States. Some of the respondents also asserted their double nationality, for example, Cuban-American.

4.2. Measurement model

First, the first-order reflective constructs were analyzed following the steps given by Hair et al. (2016). Table 3 shows the individual reliability of each indicator as well as the average variance extracted (AVE), Cronbach’s alpha (CA), and composite reliability (CR) of the constructs. In order to assess the discriminant validity, Fornell-Larcker criterion (Table 4) and the Heterotrait-monotrait criterion (HTMT) were evaluated. When assessing the HTMT (Table 5), the higher value was 0.87, although the more conservative threshold value is 0.85, however, the value is still below 0.90, as Hair et al. (2016) express, a HTMT value above 0.90 suggests a lack of discriminant validity (p. 119). Thus, we can conclude that all the values obtained comply with the rule of thumbs. Figure 1 visualizes the proposed measurement model.

Table 3. Results summary for reflective measurement models

		Convergent validity		Internal consistency reliability
Latent variable	Indicators	Loadings	AVE	CA	CR	Discriminant validity	t Value	p Value
		> 0.70	> 0.50	0.60–0.90	0.60–0.90	HTMT confidence interval does not include 1	> 1.65 (10%) > 1.96 (5%) > 2.57 (1%)	< 0.05
LA	rLA1	0.805	0.696	0.781	0.873	Yes	24.400	0.000
	rLA2	0.883					52.143	0.000
	rLA3	0.813					15.233	0.000
NVB	rNV1	0.853	0.736	0.820	0.893	Yes	35.797	0.000
	rNV2	0.874					33.488	0.000
	rNV3	0.846					30.406	0.000
AT	rAT1	0.900	0.822	0.892	0.933	Yes	47.690	0.000
	rAT2	0.919					66.006	0.000
	rAT3	0.902					59.224	0.000
KN	rKN1	0.826	0.711	0.864	0.908	Yes	23.431	0.000
	rKN2	0.850					34.369	0.000
	rKN3	0.838					27.133	0.000
	rKN4	0.858					30.169	0.000
SS	SS1	0.733	0.658	0.870	0.906	Yes	15.089	0.000
	SS2	0.838					29.884	0.000
	SS3	0.794					22.624	0.000
	SS4	0.857					34.801	0.000
	SS5	0.829					25.409	0.000
OTS	OTS1	0.790	0.678	0.842	0.894	Yes	21.484	0.000
	OTS2	0.824					25.196	0.000
	OTS3	0.879					43.044	0.000
	OTS4	0.798					26.747	0.000

Table 4. Fornell-Larcker criterion

	AT	KN	LA	NVB	OTS	RI	SS
AT	0.907
KN	0.735	0.843
LA	0.598	0.679	0.835
NVB	0.722	0.716	0.701	0.858
OTS	0.457	0.474	0.417	0.407	0.824
RI	0.216	0.215	0.240	0.216	0.519	1.000
SS	0.659	0.709	0.602	0.579	0.522	0.263	0.811

Table 5. HTMT

	AT	KN	LA	NVB	OTS	RI
AT
KN	0.83 [0.733, 0.907]
LA	0.71 [0.541, 0.846]	0.82 [0.708, 0.910]
NVB	0.84 [0.752, 0.908]	0.84 [0.718, 0.937]	0.87 [0.712, 0.984]
OTS	0.52 [0.348, 0.660]	0.55 [0.352, 0.715]	0.51 [0.323, 0.672]	0.48 [0.304, 0.633]
RI	0.23 [0.078, 0.395]	0.23 [0.078, 0.393]	0.27 [0.111, 0.432]	0.23 [0.083, 0.401]	0.56 [0.433, 0.676]
SS	0.73 [0.613, 0.832]	0.81 [0.687, 0.895]	0.72 [0.567, 0.845]	0.68 [0.524, 0.803]	0.60 [0.429, 0.736]	0.28 [0.125, 0.442]

Second, the formative measurement model (IQ) was assessed by observing collinearity, outer weights, and significance of the indicators. As can be seen on Table 6, although the higher value of the variance inflation factor (VIF) is of 3.521, collinearity does not reach critical levels in any of the formative constructs (Hair et al., 2016, p. 146) being all the values below the threshold value of 5; hence, we determined that collinearity is not a problem in our model. Since some of the indicator’s outer weights values are too small (0.092, 0.093, 0.098, and 0.099), the recommendations given by Hair et al. (2016) who assert that “when an outer weight is nonsignificant but its outer loading is high (i.e., above 0.50), the indicator should be interpreted as absolutely important but not as relatively important. In this situation, the indicator would generally be retained” (Hair et al., 2016, p. 148).

Table 6. Results summary for formative measurement models

Formative constructs	Formative indicators	Construct Validity			Collinearity
		Outer weights (w)	Outer loadings	95%BCa confidence interval	Outer VIF	t Value	Significance
		Values different from zero	> 0.50	Cannot contain the value zero	< 5	> 1.65 (10%) > 1.96 (5%) > 2.57 (1%)	(p < 0.05)?
IQ	LA1	0.093	0.677	[0.569, 0.758]	1.866	17.045	Yes
	LA2	0.098	0.725	[0.620, 0.800]	2.328	19.168	Yes
	LA3	0.092	0.675	[0.506, 0.796]	1.870	9.651	Yes
	NV1	0.103	0.769	[0.696, 0.827]	2.327	19.315	Yes
	NV2	0.102	0.766	[0.673, 0.831]	2.416	15.770	Yes
	NV3	0.099	0.745	[0.654, 0.810]	2.173	19.539	Yes
	AT1	0.106	0.783	[0.703, 0.840]	3.246	18.003	Yes
	AT2	0.107	0.797	[0.720, 0.848]	3.521	19.676	Yes
	AT3	0.109	0.809	[0.747, 0.857]	3.355	20.895	Yes
	KN1	0.101	0.736	[0.623, 0.820]	2.186	19.901	Yes
	KN2	0.100	0.738	[0.638, 0.808]	2.235	21.208	Yes
	KN3	0.101	0.767	[0.676, 0.827]	2.610	18.300	Yes
	KN4	0.110	0.822	[0.754, 0.873]	2.859	21.191	Yes

4.3. Structural model

To test the proposed hypotheses, the structural model was evaluated following the criteria given by Hair et al. (2016); hence, collinearity, path coefficients, level of determination coefficient, predictive relevance, and the size effect of each construct incident on each endogenous construct (Gallarza, Arteaga, Del Chiappa, & Gil-Saura, 2015, p. 145) were calculated by bootstrapping 5000 samples. Bias-corrected and accelerated bootstrap, two-tailed testing, and significance level of 0.05 were also used. Figure 2 illustrates the structural model results.

Figure 1. Proposed extended model.

Figure 2. PLS analysis result.

Collinearity was assessed following the VIF. The higher value was 2.159, still under the threshold value. The coefficient of determination for the three endogenous constructs is high and moderate, with values of 0.537 (53.7%), 0.304 (30.4%), and 0.269 (26.9%) for SS, the OTS, and RI, respectively. The Stone-Geisser criterion or the predictive relevance was also determined using the blindfolding procedure (Hair et al., 2016, p. 202). All the calculated values are above zero, suggesting that the model has predictive relevance for a certain endogenous construct (Hair et al., 2016, p. 207), namely IQ is relevant when predicting SS (52.4%), the OTS (28.3%), and RIs (25.8%).

The effect sizes (f²) area measure of the impact of each incident construct for the explanation of each endogenous construct (Gallarza et al., 2015, p. 146). Table 7 shows that IQ makes a large contribution to SS as well as OTS on RI with values of 1.159 and 0.369, respectively, in contrast, the contribution of constructs such as IQ → OTS and SS → OTS is rather small. In addition, path coefficients were assessed. The results show IQ → SS path coefficients are high comparing with the other paths of the model. Nevertheless, for a 5% significance level, we found that all the relationships are significant, and therefore, all the hypotheses (H₁, H₂, H₃, H₄) are supported.

Table 7. Results of the structural model

Path	Path coefficients (β)	Effect size (f^2)	95% BCa confidence intervals	t Value	p Value	Hypothesis
Threshold values	Values between −1 and +1	Lower than 0.02 = no effect, 0.02 = small, 0.15 = medium, 0.35 = large	Cannot contain the value zero	> 1.65 (10%) > 1.96 (5%) > 2.57 (1%)	< 0.05
IQ → SS	0.733	1.159	[0.657, 0.794]	17.389	0.000	Accepted
IQ → OTS	0.261	0.045	[0.088, 0.423]	2.574	0.010	Accepted
SS → OTS	0.331	0.073	[0.173, 0.489]	3.413	0.001	Accepted
OTS → RI	0.519	0.369	[0.415, 0.610]	8.853	0.000	Accepted

Finally, since we are not only interested in the direct relationships, the indirect relationships of the model were estimated. As Table 8 shows, all the relationships are significant at a 5% level.

Table 8. Indirect effects of the structural model

Path	Path coefficients (β)	95% BCa confidence intervals	t Value	p Value	Significance
Threshold values	Values between −1 and +1	Cannot contain the value zero	> 1.65 (10%) > 1.96 (5%) > 2.57 (1%)	< 0.05
IQ → OTS	0.243	[0.129, 0.373]	3.285	0.001	Yes
IQ → RI	0.261	[0.180, 0.358]	4.830	0.000	Yes
SS → RI	0.172	[0.089, 0.271]	3.106	0.002	Yes

5. Discussion

The study has examined to what extent IQ during a service encounter impacts SS and the OTS as well as some of the indirect effect, taking into consideration the services provided to international tourists when visiting Shanghai visitor attractions.

First, the results of this study show how important is the knowledge of the service personnel for a customer during a service encounter, having expertise making the strongest link with IQ (t = 30.986), followed by attitude (t = 23.038), nonverbal behavior (t = 22.078), and language (t = 18.157). However, when it comes to indirect effects of IQ dimensions on the endogenous constructs, attitude and nonverbal behavior are the key drivers of SS (t_AT = 18.644, t_NV = 17.934), the OTS (t_A = 6.655, t_NV = 6.560), and RIs (t_A = 5.007, t_NV = 4.905). These findings are in line with Solomon et al. (1985, p. 9), who posited that the attitude, behavior, and skill of service employees affect what a customer evaluates as a satisfactory encounter (cited in Brady & Cronin, 2001, p. 38), as well as with the research of Ranjan et al. (2015), where attitude was found to positively affect customer satisfaction. The results also support the work of Wu et al. (2015) where both interpersonal and professional competences, which include aspects such as attitude, communication skills, expertise, and understanding of customers’ needs, impact the relationship between the customer and the employee.

Second, the results showed that IQ has a direct effect not only on the satisfaction with the service, but also on the overall satisfaction with the destination (H_1, H₂). Although IQ had a higher impact on SS (t = 17.389) than on the OTS (t = 2.574), service provider should pay attention to all the IQ dimensions affecting satisfaction, especially if the customers are from different cultures which is the case in our study. Being aware of what exactly international tourists want and perceived a service encounter will help service employees to provide a better service and make the customer wanting to return. The study reaffirms past research where IQ is seen as a good predictor of satisfaction (Brady & Cronin, 2001; Joon Choi & Sik Kim, 2013; Ranjan et al., 2015), and ended up seeing the importance of measuring satisfaction by both SS and the OTS.

Third, being in line with previous research (Neal et al., 1999, 2007), our research found that satisfaction with services is an antecedent of the OTS (H₃), although it presented a small effect size (0.073). However, we consider this relationship significant since p values (0.001) and t values (3.413) are both significant.

Finally, it was confirmed that the OTS with a destination is a key driver of RIs (H₄) being in line with past studies (Han et al., 2009; Shawn & Feng, 2007; Um et al., 2006), facilitating not only revisit to the destination but also a visit to a neighboring destination (Shawn & Feng, 2007, p. 588). Indirectly, RI can be also determined by IQ and SS.

Thus, the findings of this study pointed out the importance of IQ in tourism to explain SS, the OTS, and RI whether direct or indirectly. Tourists not only care about their solo experience on the destination they visit, but also to what and how tourism services providers can offer them, to what extent employees are willing to help them, listen attentively to what the customer is asking for, language skills and behavior.

6. Managerial implications

The results of this study offer several managerial implications. First, the study suggests that service managers should provide a high level of IQ in order to increase satisfaction with the service, especially aspects such as attitude, nonverbal behavior, language, and expertise of the employee. Satisfaction with services may not only influence the overall satisfaction with the destination but also the motivations of tourists to revisit a destination.

In particular, visitor attraction managers should create training programs, whether online or in-person, to educate their employees on how to resolve a conflict during a service encounter, discuss how international tourists communicate using nonverbal cues and how to read those as well as encourage them to improve their attitude toward the customer, motivate their employees to learn a second language (English, Spanish, or French), and teach them effective communication skills. Good interpersonal communication skills can help service providers’ employees to better grasp what a customer wants and deliver the right information/message at the right moment. Cross-cultural understanding will allow employees to avoid conflicts during a service encounter. For example, nonverbal cues in China highly differ from the West, thus, using hand gestures to express numbers during a service encounter may cause misunderstandings if the employee is not aware of these cultural differences. Besides, hiring employees with an enthusiastic, cheerful, caring, and polite attitude toward the customer will allow customers to have a good impression of the company. Through training programs as well as e-learning courses, companies can improve employees’ satisfaction and engagement, which in turn, increase the productivity.

Second, we advise the local government to partner with local tourism business and tourism operators to develop destination management plans to ensure both satisfaction with tourism services and the overall tourism satisfaction with the destination as well as joint marketing strategies to improve the image of the city and stimulate RIs. Since the local government is by default the one who is in charge of regional marketing activities as well as all the regulations and policies regarding to the tourism industry, small pamphlets can be written and distributed among the different touristic sites, providing tourists with relevant information about some basis of Chinese language, essential nonverbal cues, information about the prices of the visitor attractions, recommendations on cultural, recreational and scenic amenities to visit, information on what visitor attractions are accessible to people with physical disabilities and essential telephone numbers. These pamphlets are to be distributed in places such as hotels, airports, and tourism information centers. This strategy would increase the overall tourism satisfaction and a positive destination image.

Third, both local government and tourism companies could share information about foreign tourists needs, wants, and complaints, which can be used to improve tourism policies and regulations linked to tourism services. At the same time, the tourism bureau can conduct an annual survey to explore the quality of the interaction during service encounters, SS, overall tourism satisfaction, and RIs in order to enhance the services in the tourism and hospitality industry. Additionally, the ministerial council for tourism can establish partnerships not only with local governments but also academicians to conduct research on how to approach inbound tourists from different regions around the globe. This will help to know much better the tourist market visiting China and how service employees should treat each of these groups in order to guarantee greater levels of satisfaction and RIs.

Lastly, we encourage frontline employees to pay attention to their attitude toward customers, stay patient, attentive, and always willing to help them. They should observe the language customers use to describe their needs or wants, the tone of voice, body language, and facial expressions. Creating a comfortable atmosphere for good face-to-face interactions in the tourism industry in Shanghai is not difficult but it does require managers to prepare their personnel for offering good service experience as well as to keep them motivated. Making every service encounter authentic and unique should be a priority for tourism agencies to increase tourists’ level of satisfaction and behavioral intentions, which in turn, would attract more foreign tourists and increase the profitability of the company.

7. Theoretical implication

By understanding how IQ not only affects SS, but also the overall satisfaction with the destination, might help tourists decide whether to visit again Shanghai or not. Destination tourism managers would better know how to improve their marketing strategies and training of their employees as well as deliver a service that makes tourists not just satisfied but loyal. Moreover, from a theoretical perspective, the present work validated past studies by examining IQ as a multidimensional construct and included not only dimensions such as attitude and expertise, which have been measured before, but language and nonverbal behavior. In this way, the study offered a more complete analysis of the quality of the interaction during an intercultural service encounter. Thereby, IQ was assessed as a reflective–formative higher-order construct through PLS-SEM.

Additionally, the current work built on previous research that showed the link between satisfaction with the general experience of the trip and satisfaction with tourism services. Accordingly, the research demonstrated that the overall tourism satisfaction with the destination is influenced, either directly or indirectly, by the satisfaction tourists may have with tourism services and the quality of their interaction with employees during a service encounter, which confirms the importance to measure satisfaction as a specific construct, especially in the tourism industry, in order to provide better experiences.

Further, the study validated that RIs are not only affected by the overall tourism satisfaction with the destination but also by SS. It also showed that tourists’ RIs can be indirectly influenced by the quality of the interactions with tourism services employees.

Thereby, the research confirms the importance of the IQ as an essential part of a tourism experience that allows tourists to be satisfied or dissatisfied, not only with the tourism services they encounter but also with the destination itself. Thus, the study promoted a better understanding of what international tourists in Shanghai think about the quality of their interactions with tourism services employees, helping managers to develop more accurate services’ strategies and attract not only first-timer visitors but also tourists who are coming to Shanghai for second, third time, or more.

8. Future research direction

Future researchers can study specifically, how each of these IQ dimensions during a trip affects not only SS and the overall satisfaction, but also life satisfaction. In addition, the direct relationship between IQ and RIs could be examined. Other works might explore how demographic factors such as nationality, language proficiency, and occupation shape customers’ perceptions of a service encounter.

Further, since it is well-known that customers with western cultural background rely more on tangible cues than do Asian customers (Mattila, 1999; Zhao & Lin, 2014), the role of language and nonverbal behavior in intercultural service encounters could be assessed. Future studies could also examine to what extent IQ impacts tourists’ destination image and the perceived experience value. Lastly, further research projects may explore the elements or attributes forming the OTS and create a practical instrument to help destination managers to measure the level of the OTS.

Acknowledgments

We would like to thank the reviewers for their valuable comments and suggestions, which helped us to improve the article.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Lilian Consuelo Mustelier-Puig

Lilian Consuelo Mustelier Puig is a Ph.D. scholar at Glorious Sun School of Business and Management, Donghua University, China. She has also been an Assistant Professor at the University of Havana, Cuba. She is interested in tourism research, consumer behavior, service marketing, information behavior, sociolinguistics, and statistical methods.

Amna Anjum

Amna Anjum is a Ph.D. scholar at Glorious Sun School of Business and Management, Donghua University, China. She has experience both in educational and corporate sector and has well-served in areas of telecom, educational, banking, consultancy, chamber of commerce. Her area of interest is Business Management, Human Resource Management, and Organizational Behavior.

Xu Ming

Dr. Xu Ming is the director of the Department of Tourism Management and Innovation at Glorious Sun School of Business and Management in Donghua University, China. He also serves as a professor in Shanghai Advanced Finance Institute, Panyapiwat Institute of Management, Thailand, and Novi Sad University, Serbia. He works as a consultant in the fields of service and tourism management.

Notes

1. China Institute of City Competitiveness: http://www.china-citynet.com/yjh/en/fyphb.asp.

2. Center for China and Globalization: http://en.ccg.org.cn/who-is-travelling-to-china-ccg-new-research-gives-insights/.

3. Shanghai Municipal Bureau: http://lyw.sh.gov.cn/lyj_website/HTML/DefaultSite/lyj_xxgk_lytj_2017/2018-02-01/Detail_139443.htm.

4. Beijing Municipal Bureau of Statistics: http://tjj.beijing.gov.cn/tjsj/yjdsj/ly/2017/201801/t20180116_390970.html.