Data Loading...

gunung kinabalu Flipbook PDF

gunungkinabalu

150 Views
101 Downloads
FLIP PDF 529.41KB

Tourism Management Perspectives 33 (2020) 100581

Contents lists available at ScienceDirect

Tourism Management Perspectives journal homepage: www.elsevier.com/locate/tmp

The role of co-creation experience in forming tourists' revisit intention to home-based accommodation: Extending the theory of planned behavior

T

⁎

Bo Meng , Mengxia Cui Department of Tourism Management at Shanxi University, Address: No. 92, Wucheng Road, Taiyuan 030006, Shanxi Province, PR China

A R T I C LE I N FO

A B S T R A C T

Keywords: Home-based accommodation Co-creation experience Theory of planned behavior (TPB) Revisit intention Co-creation behavior

This study aims to provide a theoretical framework explaining how tourists' revisit intentions to home-based accommodations are formed. It does so by inserting constructs related to co-creation experience (experiencescape, perceived value, and memorability) into the theory of planned behavior (TPB) and by considering the moderating eﬀect of co-creation behavior. The study results (n = 413) from structural equation modeling (SEM) indicated that the extended TPB has better predictive power than the original TPB. The ﬁndings also showed that all proposed paths were signiﬁcant, identiﬁed the role of co-creation experience in generating intention, and veriﬁed the mediating role of study variables. Furthermore, the invariance test indicated that tourists' cocreation behavior played a moderating role between experiencescape and customers' internal factors (perceived value, memorability, and attitude). Theoretical and practical implications are discussed.

1. Introduction Home-based accommodations include various types of lodging establishments, such as commercial homes, homestays, and small-business accommodations (Lynch, 2005). This unique style of accommodation has emerged as a new trend in the hospitality industry, as its co-creation characteristic can satisfy modern customers' demands for experiential consumption (Hassanli, Gross, & Brown, 2016; Pine & Gilmore, 1998). Therefore, generating a co-creation experience—in order to meet customers' increasing experiential consumption and to boost their revisit intention—is considered a competitive and eﬀective strategy for hospitality professionals (Han, 2015; Meng & Choi, 2018). Increasing individuals' revisit intentions is an important goal for any hospitality business (Tanford, Raab, & Kim, 2012). Previous research has been conducted to explain the formation of travelers' behavioral intentions from the perspective of generating co-creation experiences (e.g., Chathoth, Ungson, Harrington, & Chan, 2016; Mathis, Kim, Uysal, Sirgy, & Prebensen, 2016; Paulauskaite, Powell, Coca-Stefaniak, & Morrison, 2017). In particular, the use of experiencescape (e.g., Meng & Choi, 2018; Mody, Suess, & Lehto, 2017; Pizam & Tasci, 2019), perceived value (e.g., Morosan & Defranco, 2016; Prebensen & Xie, 2017), memorability (e.g., Campos, Mendes, Valle, & Scott, 2017; Hung, Lee, & Huang, 2016), and co-creation behavior (e.g., Buonincontri, Morvillo, Okumus, & Niekerk, 2017; Su et al., 2016) has been repeatedly examined by tourism and hospitality researchers. Therefore, the co-

⁎

creation experience and its key constructs (experiencescape, interaction behavior, perceived value, and memorability) are believed to be a critical process, increasing customers' revisit intentions to home-based accommodations (Prebensen, Vitterso, & Dahl, 2013). In a previous literature review, co-creation experience was highlighted by customers' interaction with the service provider and active participation (Campos et al., 2017). The concept of co-creation experience is at the core of Service Dominant Logic (SDL) by Vargo and Lusch (2008). According to SDL, services should be placed instead of products at the center of the economic exchange and then, the customer should be a value creator and should interact with the service organization and/or with other customers (Prebensen et al., 2013). In terms of home-based accommodation organizations, the process of co-creation experience generation starts from the interaction between customers and the elements of the experiencescape, such as the physical environment (e.g., layout, equipment, local culture and food, etc.) and the interpersonal dimensions (e.g., C2C interaction, hospitality of employee, etc.) (Pizam & Tasci, 2019). Then, the customer becomes a producer, creating value with more beneﬁcial service and memorable experiences (Chathoth et al., 2016; Mody et al., 2017; Prebensen et al., 2013). As such, a higher evaluation of the experiencescape would contribute to a higher level of perceived value (i.e., value co-created by customer and experiencescape), which further contributes to individuals' memorability. Thus, the co-creation experience under a context of home-based accommodation could be understood as a

Corresponding author. E-mail address: [email protected] (B. Meng).

https://doi.org/10.1016/j.tmp.2019.100581 Received 10 March 2019; Received in revised form 12 July 2019; Accepted 23 September 2019 2211-9736/ © 2019 Elsevier Ltd. All rights reserved.

Tourism Management Perspectives 33 (2020) 100581

B. Meng and M. Cui

value, memorability, and attitude). Overall, in order to provide theoretical developments and practical strategies, the present study aims to develop a framework that explains customers' behavioral intentions concerning home-based accommodations by extending TPB. While a variety of home-based accommodations are on the rise, information is still lacking on why customers choose these new forms of accommodations; how co-creation experiences are shaped; and how these experiences become memorable and inﬂuence the individuals' revisit intentions. In order to ﬁll these research gaps, the present study was designed to verify the context of home-based accommodations and to examine the relationships among the proposed constructs under said context—particularly, to examine the eﬀects of co-creation experiences (experiencescape, perceived value, memorability, co-creation behavior) on the decision-making process. The speciﬁc objectives of the current study are as follows:

formation process of stimulating experiencescape and its eﬀect on memorability through perceived value. Thus, perceived value was believed to be a mediator bridging experiencescape and memorability during this process. Although these co-creation experience constructs illustrate important factors, providing us with an understanding of tourists' intention formation, few empirical eﬀorts have been made to examine individuals' decision-making processes by combining these constructs in the context of home-based accommodation. Therefore, we utilized the TPB by extending the co-creation experience and its related constructs, in order to reﬂect the characteristics of home-based accommodation. As a framework of general human behavior, TPB oﬀers a holistic perspective for predicting various human behaviors (Ajzen, 1991). Tourism scholars argue that TPB could be extended by including the essential constructs for a speciﬁc hospitality or tourism context; Doing this would provide a helpful process for better predicting individuals' behavioral intentions as well as an opportunity to expand existing theories (Ajzen, 1991; Lam & Hsu, 2004; Meng & Choi, 2018). Empirical studies have also supported the eﬃcacy of TPB and its extended models in a variety of human behaviors, including various tourism and hospitality contexts (Japutra, Loureiro, Molinillo, & Ekinci, 2019). For instance, Goh, Ritchie, and Wang (2017) inserted a new ecological paradigm (NEP) of pro-environmental values into the TPB. Additionally, Meng and Choi (2018) extended the TPB into a theme restaurant by adding the constructs of servicescape and authentic perception. More recently, Japutra et al. (2019) examined the relationships between the travelers' mindset and the TPB in Portugal. As discussed above, using the well-stablished TPB framework with its strong predictive ability by adding the co-creation experience as essential factors reﬂecting the characteristics of home-based accommodations (i.e., experiencescape, perceived value, and memorability) would provide a suitable and holistic perspective to gain insight into tourists' revisit intention to home-based accommodations. In TPB, ‘attitude towards a behavior’ refers to ‘the degree to which a person has a favorable or unfavorable evaluation or appraisal of the behavior in question’; ‘subjective norms’ refers to ‘the perceived social pressure to perform or not to perform the behavior’; and ‘perceived behavior control’ refers to ‘the perceived ease or diﬃculty of performing the behavior’ (Ajzen, 1991, p.188). The outcome of these three predictors, ‘behavioral intention’, was described as ‘an indication of how hard people are willing to try, of how much an eﬀort they are planning to exert, in order to perform the behavior’ (Ajzen, 1991, p.188). As such, co-creation experience-related constructs are expected to be such speciﬁc factors explaining revisit intentions in the cases of home-based accommodations. Speciﬁcally, studies revealed that the experiencescape helps form a favorable attitude towards revisiting home-based accommodations (Meng & Choi, 2018; Reimer & Kuehn, 2005), and memorability would positively inﬂuence the intention to revisit home-based accommodations (Coudounaris & Sthapit, 2017; Marschall, 2012). Therefore, attitude and memorability serve as mediators linking co-creation experience-related constructs (i.e., experiencescape/perceived value) to TPB-related constructs (i.e., attitude, and revisit intention). In addition, from the customers' perspective, a co-creation experience could be strengthened by co-creation behavior by the customers themselves (Campos, Mendes, Valle, & Scott, 2018). When customers participate and interact more, they naturally have more opportunity to connect to the elements of experiencescape. Thus, as the beneﬁts of experiencescape accumulate, customers are likely to have higher perceived value and a more memorable experience, which further forms a favorable attitude (Buonincontri et al., 2017; Campos et al., 2017; Zatori, Smith, & Puczko, 2018). As a result, through experiencescape and co-creation behavior, a memorable experience is created, a favorable attitude is formed, and a revisit intention is developed (Priporas, Stylos, Vedanthachari, & Santiwatana, 2017). Thus, co-creation behavior by the customers could be considered a moderator between external factors (i.e., experiencescape) and internal factors (i.e., perceived

1) develop a model that provides a more comprehensive understanding of home-based accommodation by inserting co-creation experience into the TPB framework; 2) compare TPB and extended TPB to explore the superiority of the proposed model; 3) verify the relative importance of the constructs within the proposed model in determining revisit intention; 4) investigate the mediating role of study variables in the customer decision-making process; 5) examine the moderating role of co-creation behavior between the external factor (experiencescape) and its outcomes (perceived value, memorability, and attitude). 2. Literature review 2.1. Co-creation experiences in home-based accommodations Home-based accommodation refers to “types of accommodation where visitors or guests pay to stay in private homes, where interaction takes place with a host and/or family usually living upon the premises and with whom public space is, to a degree shared” (Lynch, 2005, p.534). Although the forms of home-based accommodation have been studied under diﬀerent terms (commercial home, specialist accommodation, Airbnb, homestays, small-business accommodation, host-family accommodations, guest houses), the essential aspects of sharing space and interpersonal interaction are largely similar (Cheng & Jin, 2019; Lynch, 2005; Sood, Lynch, & Anastasiadou, 2017; Wang, Hung, & Li, 2018; Ye, Xiao, & Zhou, 2019). Customers usually have an experience of co-creation with an experiencescape in home-based accommodations, due to the high degree of interaction. Co-creation allows the customer to do certain things, engage in activities for self-development, explore external surroundings, and connect to the people (Eraqi, 2011). At the same time, value is also created through social activities (Rihova, Buhalis, Moital, & Gouthro, 2013), which in turn require active participation and interaction with the service organization or other customers, generating memorable experiences (Prebensen et al., 2013; Reichenberger, 2017). As established by previous literature, the process of generating co-creation experiences includes four key constructs: 1) experiencescape, 2) perceived value, 3) memorability, and 4) co-creation behavior. More speciﬁcally, experiencescape as an external factor positively inﬂuences memorability, and perceived value is a mediator that links experiencescape and memorability (Campos et al., 2017; Pizam & Tasci, 2019; Prebensen & Xie, 2017). Experiencescape originates from the term ‘servicescape’ under consumer-centrism (Edvardsson, Enquist, & Johnston, 2010). ‘Experiencescape’ refers to sensory, functional, social, natural, and cultural stimuli in a service encounter, embedded in a culture of hospitality (Bitner, 1992; Pizam & Tasci, 2019). In the formation of co-creation experiences, experiencescape is the background and stimuli that integrate physical aspects of the environment, social actors, participants, 2

Tourism Management Perspectives 33 (2020) 100581

B. Meng and M. Cui

experiences. Meanwhile, memorability can be directly inﬂuenced by experiencescape (Mathis et al., 2016). Previous literature has indicated that experiences, events, or objects are considered memorable when individuals perceive them as distinctive or salient (Bless, Strack, & Walther, 2001). In home-based accommodations, a distinctive and salient experiencescape that involves interaction among its unique culture, employees, and design (Lynch, 2005) would contribute to memorability. Thus, we posit the following hypotheses:

organizational dynamics, and even features of service delivery (Campos et al., 2017). Perceived value is the customers' overall evaluation of a product or service based on perceptions of what is received and what is given (Ramseook-Munhurrun, Seebaluck, & Naidoo, 2015; Zeithaml, 1988). In home-based accommodations, such value can be created by the participative process, in which people and the organization generate their use of resources together (organizational resources such as experiencescape, and personal resources such as time, eﬀort, and knowledge) (Gronroos & Gummerus, 2014; Ind & Coates, 2013; Prebensen & Xie, 2017). Memorability is usually seen in tourism literature as subjective; it refers to a personal ability to maintain and recall information (Saket, Endert, & Stasko, 2016) or to the subjective feeling that one will remember something in the future (Zimmerman & Kelley, 2010). The last construct, co-creation behavior, refers to customers' physical and/or mental participation in activities and events, including interacting with other participants in the experience environment (Campos et al., 2017; Prebensen & Xie, 2017).

H1. : Experiencescape has a positive inﬂuence on perceived value. H2. : Perceived value has a positive inﬂuence on memorability. H3. : Experiencescape has a positive inﬂuence on memorability. 2.3.2. Experiencescape and attitude Attitude is informed by an evaluation of behavior performance (Ajzen, 1991). In home-based accommodations, behavior performance and services directly contribute to customers' experiences (Dong & Siu, 2013). Since customers' positive perceptions of an experiencescape lead to a high evaluation of customer experience (Dong & Siu, 2013), experiencescape is likely to form a favorable attitude. Empirical studies have supported the notion that experience positively inﬂuences attitude (Dong & Siu, 2013; Reimer & Kuehn, 2005). For instance, Dong and Siu (2013) revealed that the physical and personal dimensions of an experiencescape signiﬁcantly inﬂuence evaluation of a theme park. Reimer and Kuehn (2005) also found that experiencescape is a critical in evaluating service experience. More recently, Meng and Choi (2018) demonstrated that experiencescape signiﬁcantly inﬂuences attitude towards revisiting a theme restaurant. Thus, we posit the following hypothesis:

2.2. The extended theory of planned behavior (TPB) TPB originally derived from the theory of reasoned action (TRA) (Ajzen, 1985). TRA only considered volitional aspects, such as attitude (behavioral beliefs) and subjective norms (subjective normative beliefs); TPB expanded to include non-volitional aspects, such as perceived behavioral control, into the decision-making process (Ajzen, 1991; Lam & Hsu, 2004). Since humans' intentions and behaviors cannot be totally voluntarily controlled in most situations, the integration of non-volitional factors is believed to signiﬁcantly improve the theory's power to predict individuals' intentions (Han & Kim, 2010; Oh & Hsu, 2001; Perugini & Bagozzi, 2001). According to TPB, individuals' behavioral intentions are assumed to be determined by three signiﬁcant predictors: attitude towards the behavior, subjective norms, and perceived behavioral control (Ajzen, 1991). Although TPB has proved a useful framework in predicting human behavior, a number of scholars still assert that the theory needs to be further extended, either by inserting new important variables or by modifying causal relationships based on speciﬁc situations (e.g., Ajzen, 1991; Han, 2015; Meng & Choi, 2018; Perugini & Bagozzi, 2001). Thus, rooted in the TPB framework, our theoretical model maintains that attitudes towards revisit behavior, subjective norms, perceived behavioral control, and memorability are direct predictors of revisit intentions to home-based accommodations. In addition, our model maintains that experiencescape is an external factor that inﬂuences perceived value, memorability, and attitude; perceived value is regarded as an outcome variable of experiencescape, and this value antecedes memorability. The rationale in the current study is that, in the context of home-based accommodations, integrating the co-creation experience can allow a better prediction of the total variance in decisions and can broaden the conceptual understanding of intention formation.

H4. : Experiencescape has a positive inﬂuence on attitude. 2.3.3. Memorability and revisit intention Revisit intention is the likelihood of a customer repeating an activity or revisiting a facility (Baker & Crompton, 2000). Previous literature has indicated that a memorable experience signiﬁcantly inﬂuences revisit intention (Coudounaris & Sthapit, 2017; Marschall, 2012). For example, Coudounaris and Sthapit (2017) show that memorability positively inﬂuences visitors' revisit intentions to the zoo and museum. Zhang, Wu, and Buhalis (2018) also demonstrated that memorable experiences signiﬁcantly inﬂuence revisit intention. When individuals perceive a tourism experience as memorable, they recall experiential elements (excitement, enjoyment, amusement, pleasure, meaning, etc.) (Kim, Ritchie, & McCormick, 2012). Studies have illustrated that these feelings are an important aspect in an individual's evaluation of an experience (Hirschman & Holbrook, 1982). In this way, a favorable evaluation of an experience would increase the likelihood to revisit (Dunman & Mattila, 2005). Thus, we posit the following hypothesis: H5. : Memorability has a positive inﬂuence on revisit intention.

2.3. Hypotheses 2.3.1. The experiencescape, perceived value, and memorability When co-creating customers positively assess experiencescape resources provided by home-based accommodations, they have more cocreation experiences; these experiences create more perceived value, which further enhance memorability (Campos et al., 2018). Therefore, perceived value is a mediator that links experiencescape and memorable experience. Perceived value is formed by perceived beneﬁts (Lovelock, 1996); in home-based accommodations, an experiencescape formulated by the organization beneﬁts the customers. In this sense, high evaluation of experiencescape would contribute to a higher degree of perceived value. In addition, studies have indicated that, with the added value of co-creation of an experience, customers will have a more positive experience (Mathis et al., 2016; Prebensen & Xie, 2017). Thus, perceived value would increase memorability in customers'

2.3.4. Attitude, subjective norms, perceived behavior control, and revisit intention Numerous studies within the TPB framework have supported the causal relationships connecting attitude, subjective norms, and perceived behavior control with revisit intention (e.g., Han, 2015; Hsu & Huang, 2012; Lam & Hsu, 2004). For instance, in the context of hospitality, Meng and Choi (2018) identiﬁed that TPB predictors signiﬁcantly inﬂuence revisit intentions to theme restaurants. Han, Hsu, and Sheu (2010) also indicated that all three precursors of intention were signiﬁcant and positive in a green hotel context. Thus, we posit the following hypotheses: H6. : Attitudes have a positive inﬂuence on revisit intention. H7. : Subjective norms have a positive inﬂuence on revisit intention. 3

Tourism Management Perspectives 33 (2020) 100581

B. Meng and M. Cui

H8. : Perceived behavior control has a positive inﬂuence on revisit intention.

creation behavior. For instance, the physical environment with cultural themes is a stimulus for interaction since tourists believe they may share similar interests; The layout and design in home-based accommodations with open space and grouped chairs would also encourage conversation; Also, the sociable and friendly hosts create a comfortable atmosphere evoking interaction. Second, the tourists' interactive behaviors in home-based accommodations are encouraged by the design of interactive activities. For example, the ‘See Pingyao Again’ live show encourages tourists to walk around diﬀerent scenes, thus strengthening their interactive experience; The ‘Beef Exhibition and Making’ teaches the traditional techniques of preparing beef which leads tourists to participate in co-creation activities such as cooking beef dishes. Finally, ‘Lacquer Work Making’ highlights the traditional method of panting that enables tourists to work together create a piece of lacquer work. As such, these interactive activities require tourists to be deeply immersed, actively participating, and interacting with various environments. As discussed above, with the interaction between experiencescape and tourists' behavior, the co-creation experiences are achieved through the value created by the activities in home-based accommodations in Pingyao. Therefore, Pingyao is an ideal survey site for investigating cocreation experiences in home-based accommodations. In order to get a representative sample, the survey was carried out on both weekdays and weekends by eight well-trained surveyors in the four main entrance/exit gates as well as the main streets of Pingyao. Only the respondents who had stayed in home-based accommodations for at least one night were targeted as potential respondents. A description of the home-based accommodation was ﬁrst presented to the respondents. Then, a screening question was asked. Individuals who had experienced overnight accommodation at Pingyao and who had participated in activities and interaction were chosen as respondents. In order to encourage completeness, when the respondents ﬁnished the questionnaires, they were rewarded with a small gift (a souvenir pen). The surveyors explained questions to respondents when they did not understand them. In total, the surveyors distributed 600 questionnaires to home-based accommodation customers in the four main entrances/ exist gates in Pingyao, and 487 questionnaires were collected (yielding a response rate of 81.1%). After surveys with missing data (n = 22), outliers (n = 23, z-score > 1.96), and multivariate normality (n = 29, according to Mahalanobis distance value) were excluded, 413 respondents were retained for the ﬁnal analysis.

2.3.5. The moderating eﬀect of co-creation behavior Since co-creation (active participation) increases the number of encounters between partners and friends, other customers, and frontline staﬀ, individuals who engage in more co-creation behaviors make the most of experiencescape resources (Choo & Petrick, 2014). In this sense, co-creation behavior could strengthen the eﬀect of experiencescape, further drawing out outcomes such as perceived value, memorability, and favorable attitude. Previous studies have demonstrated that co-creation behavior has a moderating eﬀect between experiencescape and other outcomes. Customers' co-creation behaviors are thought to generate the perception/contacts of experiencescape elements (Buonincontri et al., 2017). Therefore, when individuals perform more co-creation behaviors, they are more likely to obtain the beneﬁts from experiencescape, thus creating higher value and memorability and subsequently forming a favorable attitude. In one empirical study, Oh, Fiore, and Jeoung (2007) applied Pine and Gilmore's (1998) framework of experience realms to a bed-and-breakfast accommodation and illustrated that customers engaging in co-creation behaviors had more experiences. Moreover, a recent study by Campos et al. (2017) in a marine wildlife park indicated that customers' cocreation behaviors made them feel closer to the animals. In this sense, co-creation behavior as a situational factor could strengthen how experiencescape inﬂuences its outcomes. Thus, we posit the following hypotheses: H9a. : Co-creation behavior has a signiﬁcant moderating eﬀect on the relationship between experiencescape and perceived value. H9b. : Co-creation behavior has a signiﬁcant moderating eﬀect on the relationship between experiencescape and memorability. H9c. : Co-creation behavior has a signiﬁcant moderating eﬀect on the relationship between experiencescape and attitude. Therefore, we propose a research model that integrates the original variables in TPB (attitude, subjective norms, perceived behavioral control) with co-creation experience constructs (experiencescape, perceived value, memorability, and co-creation behavior). Our proposed research model is presented in Fig. 1.

3.2. Measures for study variables 3. Method The validity of the instruments used in our study were demonstrated in previous studies on tourism and marketing. In order to reﬂect the tourism and hospitality context, experts from academia and industry were invited to reﬁne the initial questionnaires. The ﬁnal version of the questionnaire included one screening question (‘Did you stay overnight at a home-based accommodation this trip?’), pictures showing the home-based accommodation with a brief description, and questions on constructs of co-creation experience, constructs in TPB, and some demographic information. Speciﬁcally, seven items were used to measure experiencescape (e.g., ‘the atmosphere is appealing to my senses’; ‘the design and layout are functional’) (Pizam & Tasci, 2019). Three items were used to evaluate perceived value (e.g., ‘the home-based accommodation experience here oﬀered a good value for the price’) (Bolton & Drew, 1991; Ramseook-Munhurrun et al., 2015). Four items were used to evaluate memorability (e.g., ‘I will have wonderful memories about my stay in the home-based accommodation’) (Zatori et al., 2018). For the constructs within TPB, ﬁve items were used to evaluate attitude (e.g., ‘I think a revisit to the home-based accommodation would be nice’) (Ajzen, 1991). Three items were used for subjective norms (e.g., ‘Most people who are important to me think I should revisit home-based accommodations’) (Ajzen, 1991). Three items were used to evaluate perceived behavior control (e.g., ‘Whether or not I revisit a home-based accommodation is completely up to me’) (Ajzen, 1991). Three items

3.1. Data collection procedure The survey was carried out by using a nonprobability convenience sampling technique. The reason for choosing such a survey method is that this approach has been widely used in tourists' behaviors when the entire population is too large and impossible to use independent random sampling representing the entire population (e.g., Meng & Choi, 2016). Data collection lasted one month, from November 24–December 23, 2018, at Pingyao Ancient City. Pingyao is a typical tourism destination with various small and private home-based accommodations. As a tourism destination that is on the UNESCO World Heritage List, it oﬀers preserved ancient-style architectures (almost entirely from the Ming and Qing dynasties) and traditional life style (Wikipedia, 2019). Home-based accommodations in Pingyao Ancient City have provided excellent conditions for generating co-creation experiences. Co-creation needs two parties (i.e., a service provider and a customer) who interact with each other to create values (Ind & Coates, 2013; Prebensen & Xie, 2017). In Pingyao, the two parties function by means of an experiencescape of home-based accommodations and customer interaction behaviors. First, the traditional life style, the ancient-style architecture, and the home-based accommodation hosts in Pingyao would play a role as an experiencescape triggering tourists' co4

Tourism Management Perspectives 33 (2020) 100581

B. Meng and M. Cui

Perceived Value

The process of co-creation experience generation

H9a H2

H1 H3

Experiencescape

Memorability

H9b H9c

H4 H5 The original TPB

Attitude H6

H7

Subjective Norms

Revisit Intention

H8 Perceived Behavioral Control

Dotted arrow: Hypothesized moderating effect of co-creation behavior (H9a-c)

Fig. 1. Proposed research model.

were used to evaluate revisit intention (e.g., ‘I am planning to revisit home-based accommodations in the near future’) (Ajzen, 1991). Finally, three items were used to assess co-creation behavior (e.g., ‘In the accommodation experience, I directly interacted with the environment during my stay’) (Campos et al., 2017). All the measurement items were answered on a 7-point Likert scale (See Table 2).

Table 1 Demographic characteristics.

Gender Male Female Age Under 20 21–30 31–40 > 41 Marital status Married Unmarried Education level Less than high school degree High school degree College graduate (or currently enrolled in college) University graduate (or currently enrolled in university) Graduate degree Family monthly income Under 3000 Chinese Yuan 3001–6000 Chinese Yuan 6001–9000 Chinese Yuan 9001–12,000 Chinese Yuan 12,001–15,000 Chinese Yuan 15,001–20,000 Chinese Yuan Over 20,001 Chinese Yuan

3.3. Data analysis SPSS 18 and AMOS 22 were used to analyze the data. A two-step approach was adopted, based on Anderson and Gerbing's (1988) suggestion. In the ﬁrst step, a measurement model was evaluated using Conﬁrmatory Factor Analysis (CFA) and the maximum likelihood estimation method. Then, Structural Equation Modeling (SEM) was used to examine the causal paths in our proposed theoretical model. Modeling comparisons were also performed to identify the superiority of extended TPB compared to original TPB. Finally, the moderating eﬀects of co-creation behavior were examined. Using χ2 diﬀerences with two degrees of freedom (Baron & Kenny, 1986), we compared two models of low-level and high-level interaction behavior.

3.4. Demographic proﬁle Of the 413 respondents, 42.9% of the tourists were male and 57.1% were female. In addition, the majority of the respondents were ages 21–30 years (50.1%), followed by younger than 20 years (36.3%), ages 31–40 years (9.2%), and older than 40 years (4.4%). Moreover, while most of the respondents were single (82.3%), 17.7% of them were married. In terms of family monthly income, the most reported income group was 3001–6000 Chinese Yuan (30.8%), followed by lower than 3000 Chinese Yuan (26.9%) 6001–9000 Chinese Yuan (19.9%), and higher than 9001 Chinese Yuan (14.8%). In terms of education, the vast majority of the respondents held a bachelor's degree or higher (78.4%) (Table 1).

N = 413

%

177 236

42.9 57.1

150 207 38 18

36.3 50.1 9.2 4.4

73 340

17.7 82.3

14 40 60 253 46

3.4 9.7 14.5 61.3 11.1

111 127 82 41 18 16 18

26.9 30.8 19.9 9.9 4.4 3.9 4.4

4. Results 4.1. Measurement model evaluation First, the skewness value was adopted to examine the normality of the data. Since the Zskewness value in the study variables did not exceed ± 2.58 or ± 1.96, the distribution was normal (Hair, Anderson, Tatham, & Black, 2010). Additionally, the correlations between constructs within the proposed model were lower than the cut-oﬀ of 0.75 5

Tourism Management Perspectives 33 (2020) 100581

B. Meng and M. Cui

Table 2 Measures, loadings, and reliability. Measures

Cronbach's α

Standardized loading

Experiencescape (Ex.) The atmosphere is appealing to my senses. The design and layout is functional. The level of crowd is comfortable. The employees are friendly. The customers are sociable. The environment reﬂects nature. The organizational culture shows hospitality to all stakeholders. Perceived value (PV) The home-based accommodation oﬀered good value for the price. The overall value of staying in home-based accommodation was high. The experience home-based accommodation here was worth the money. Memorability (Me.) I will have wonderful memories about my stay in this home-based accommodation. I will remember many positive things about home-based accommodation. I will not forget my experience in home-based accommodation. The overall experience of home-based accommodation will be an everlasting memory. Attitude (Att.) I think revisit to home-based accommodation is good. I think revisit to home-based accommodation is wise. I think revisit to home-based accommodation is pleasant. I think revisit to home-based accommodation is beneﬁcial. I think revisit to home-based accommodation in is attractive. Subjective norms (SN) Most people who are important to me think I should revisit home-based accommodations. Most people who are important to me would want me to revisit home-based accommodation. People whose opinions I value would prefer me to revisit home-based accommodations. Perceived behavioral control (PBC) Whether or not I revisit home-based accommodations is completely up to me. I am conﬁdent that if I want, I can go and revisit home-based accommodations. I have enough resources, time and opportunities to revisit home-based accommodations. Revisit intention (RI) I am planning to revisit home-based accommodations in the near future. I am conﬁdent that if I want, I can go and revisit home-based accommodations. I have enough resources, time and opportunities to revisit home-based accommodations. Co-creation behavior (CB) In the experience of home-based accommodation, I directly interacted with the ambient during my stay. In the experience of home-based accommodations, I had great play time with other people. In the experience of home-based accommodations, I have been physically active.

0.836 0.76 0.77 0.70 0.84 0.72 0.71 0.72 0.908 0.88 0.90 0.86 0.911 0.88 0.90 0.80 0.90 0.966 0.80 0.93 0.94 0.93 0.91 0.925 0.90 0.93 0.91 0.812 0.74 0.82 0.83 0.901 0.81 0.91 0.87 0.874 0.81 0.87 0.82

Measures, loadings, and reliability. Note1. Goodness-of-ﬁt statistics for the measurement model: χ2 = 891.416, df = 406, p < .001, χ2/df = 2.196, RMSEA = 0.054, CFI = 0.95, IFI = 0.95, TLI = 0.94. Note 2. All standardized factor loadings were signiﬁcant (p < .001). Note 3. Measures for variables were evaluated with a seven-point scale from “Strongly disagree” (1) to “Strongly agree” (7). Table 3 .Measurement model results and correlations. Ex. Ex. PV Me. Att. SN PBC RI CB Mean S. D. C. R.

PV

Me

Att.

SN

PBC

RI

CB

0.69 0.70(0.49) 0.63(0.40) 0.61(0.37) 0.66(0.44) 0.58(0.34) 0.64(0.41) 4.989 1.110 0.87

0.63 0.55(0.30) 0.61(0.37) 0.77(0.60) 0.67(0.45) 0.60(0.36) 5.164 1.072 0.87

0.74 0.69(0.48) 0.59(0.35) 0.58(0.36) 0.46(0.21) 5.061 1.333 0.93

0.73 0.69(0.48) 0.68(0.46) 0.50(0.25) 4.680 1.134 0.89

0.58 0.72(0.52) 0.58(0.34) 5.022 1.072 0.80

0.63 0.47(0.22) 4.730 1.215 0.83

0.60 4.908 1.117 0.82

a

0.57 0.63b(0.39) 0.59(0.35) 0.45(0.20) 0.48(0.23) 0.52(0.27) 0.42(0.18) 0.59(0.35) 5.17 0.897 0.88

c

Note1. Ex. = Experiencescape, PV = Perceived value; Me. = Memorability; Att. = Attitude; SN = Subjective norm; PBC = Perceived behavior control; RI = revisit intention; CB = co-creation behavior. Note2. C.R. = composite reliability. a AVE (Average variance extracted) values are bolded and along the diagonal. b Correlations between variables are below the diagonal. c Squared correlations between variables are within parentheses.

6

Tourism Management Perspectives 33 (2020) 100581

B. Meng and M. Cui

(between 0.42 and 0.72). Therefore, no high multicollinearity existed (Green, 1978). Next, reliability and validity were evaluated by the measurement model, using a conﬁrmatory factor analysis (CFA) (See Table 2). The results from the measurement model indicate that the model closely ﬁts the data (χ2 = 891.416, df = 406, χ2/df = 2.196, p < .001, RESEA = 0.054, CFI = 0.95, IFI = 0.95, TLI = 0.94). All items were loaded to their associated latent construct signiﬁcance (p < .001). Cronbach's alpha values of each construct ranged from 0.83 to 0.96. Thus, the reliability of each construct was demonstrated to be high, since they exceeded the suggested cut-oﬀ point of 0.7 (Nunnally & Bernstein, 1994). As seen in Table 3, the composite reliability values were also assessed in order to evaluate the multi-item scales. The values ranged from 0.82 to 0.93, which exceeded the minimum requirement of 0.60 (Bagozzi & Yi, 1988). Therefore, the multi-item measures used in this study are internally consistent and reliable. In addition, all average variance extracted (AVE) values ranged between 0.57 and 0.74, exceeding the suggested cut-oﬀ of 0.50 (Hair et al., 2010). Therefore, convergent validity of the measures was also established. Finally, discriminant validity was evaluated by comparing the AVE values and the squared correlations. Since all AVE values were higher than the squared correlations (Fornell & Larcker, 1981), discriminant validity was also established. The details of the measurement model results are shown in Tables 2 and 3.

Table 5 Result of the structural equation modeling.

The chi-square tests of TPB and extended TPB (ETPB) indicate a signiﬁcant diﬀerence between the two models (Δχ2 (270) = 724.765, p < .05). The R2 for behavioral intention in ETPB improved from 0.60 to 0.65 with the insertion of co-creation experience constructs (experiencescape, perceived value, and memorability). Therefore, the extended TPB was demonstrated to be superior to the original TPB in explaining tourists' revisit intentions. Table 4 presents detailed information of the comparison of the two models.

As seen in Table 5 and Fig. 2, a structural equation modeling (SEM) was carried out to determine the ﬁt of the proposed research model. The study results reveal that the model closely ﬁts the data (χ2 = 913.715, df = 336, p < .01, χ2/df = 2.71, CFI = 0.94, IFI = 0.94; TLI = 0.93; RMSEA = 0.065). Our proposed research model explained 65% of the total variance in revisit intentions. The hypothesized relationships among the constructs related to co-creation experience (Hypotheses 1–3) were examined ﬁrst. The study results indicate that experiencescape had a positive inﬂuence on the evaluation of perceived value (βEx.→PV = 0.84, t = 11.10, p < .001), which in turn contributed to the establishment of memorability (βPA→Me. = 0.46, t = 6.84, p < .001). In addition, our study results indicate that experiencescape also exerted a signiﬁcant direct inﬂuence on memorability (βEx.→Me = 0.47, t = 3.39, p < .001). Therefore, H1, H2, and H3 were supported: perceived value was a partial mediator between

Fit indices χ2

188.95

913.715

df χ2/df RMSEA CFI IFI NFI R2(revisit intention)

66 2.86 0.067 0.97 0.97 0.96 0.60

336 2.71 0.065 0.94 0.94 0.90 0.65

T-values

Hypotheses 1 Hypotheses 2 Hypotheses 3 Hypotheses 4 Hypotheses 5 Hypotheses 6 Hypotheses 7 Hypotheses 8 Variance explained R2 (PV) = 0.495 R2 (Me.) = 0.579 R2 (Att.) = 0.303 R2 (RI) = 0.658

Ex. → PV PV → Me. Ex. → Me. Ex. → Att. Me. → RI Att. → RI SN → RI PBC → RI Total eﬀect on RI βEx-RI = 0.19⁎ βPV-RI = 0.06⁎ βMe.-RI = 0.13⁎ βAtt.-RI = 0.09⁎⁎ βSN-RI = 0.16⁎⁎ βPBC-RI = 0.74⁎⁎⁎

0.84 0.46 0.47 0.79 0.13 0.09 0.16 0.74 Indirect eﬀect on RI βEX.-PV/Me./Att.-RI = 0.07⁎ βPV-Me.-RI = 0.02⁎

11.10⁎⁎⁎ 6.84⁎⁎⁎ 5.51⁎⁎⁎ 9.27⁎⁎⁎ 2.50⁎ 2.14⁎ 2.57⁎⁎ 6.70⁎⁎⁎

4.4. Measurement-invariance model Based on the results of k-means cluster analysis, the sample was divided into groups of low (n = 206) and high (n = 207) co-creation behavior. Then, a non-restricted model was generated for the split group (χ2 = 1572.05, df = 812, p < .05). This non-restricted model was compared to a full-metric invariance model in order to conﬁrm any diﬀerence between the measurement structures (See Table 6). Results of the chi-square diﬀerences revealed that the diﬀerence was insigniﬁcant (Δχ2 (23) = 30.79, p > .05). Therefore, the full-metric invariance was supported.

Table 4 Results of the structural-model comparisons. Proposed model(extended TPB)

Coeﬃcients

experiencescape and memorability. Moreover, the associations between the co-creation constructs (experiencescape and memorability) and the TPB constructs (attitude and revisit intention) were demonstrated to be signiﬁcant (βEx.→Att. = 0.79, t = 9.27, p < .001; βMe.→RI = 0.13, t = 2.50, p < .05). Therefore, H4 and H5 were supported. Furthermore, the relationships among the TPB constructs were examined. Study results indicate that all three predictors of revisit intention signiﬁcantly inﬂuenced revisit intention (βAtt.→RI = 0.09, t = 2.14, p < .05; βSN→RI = 0.16, t = 2.57, p < .01; βPBC→RI = 0.74, t = 6.70, p < .001). Therefore, H6, H7, and H8 were supported. Finally, the total and indirect inﬂuences of certain factors on revisit intention were also examined (See Table 5). Speciﬁcally, the strongest factor inﬂuencing revisit intention was perceived behavior control (βPBC→RI = 0.74, p < .001), followed by experiencescape (βEx..→RI = 0.19, p < .05), subjective norms (βSN.→RI = 0.16, p < .05), memorability (βMe.→RI = 0.13, p < .05), attitude (βAtt.→RI = 0.09, p < .05), and perceived value (βPV→RI = 0.06, p < .05).

4.3. Structural model evaluation

TPB

Paths

Note 1. Ex. = Experiencescape, PV = Perceived value; Me. = Memorability; Att. = Attitude; SN = Subjective norm; PBC=Perceived behavior control; RI = Revisit intention. Note 2. Goodness-of-ﬁt statistics for the measurement model: χ2 = 913.715, df = 336, p < .001, χ2/df = 2.719, RMSEA = 0.065, CFI = 0.94, IFI = 0.94, TLI = 0.93. ⁎ p < 0.05. ⁎⁎ p < 0.01. ⁎⁎⁎ p < 0.001.

4.2. Modeling comparison

Goodness-of-ﬁt statistics & R square

Hypotheses

4.5. Structural-invariance model evaluation The hypothesized moderating eﬀect of co-creation behavior was examined with a series of modeling tests for metric invariance. As seen in Fig. 3 and Table 6, the baseline model (χ2 = 1455.31, df = 672, p < .05, CFI = 0.90, IFI = 0.90 RMSEA = 0.053) satisfactorily ﬁt the data. Then, the baseline model was compared to nested models with a chi-square test. The results from the structural invariance test reveal that the paths from experiencescape to perceived value diﬀered 7

Tourism Management Perspectives 33 (2020) 100581

B. Meng and M. Cui

Perceived Value

H1: .84 Experiencescape

***

2

R = .49 H2: .46

H3: .47

***

The process of co-creation experience generation

***

Memorability 2

R = .57 H4: .79

***

H5: .13

*

Attitude

The original TPB 2

R = .30 Subjective Norms

H6: .09 H7: .16

**

p