Understanding the intention to revisit a destination by expanding the theory of planned behaviour (TPB)

Purpose – The purpose of this study is to unearth the factors that influence tourists’ revisit intention. The proposed model of the study is grounded on using the theory of planned behaviour (TPB) and extending it with additional variables, i.e. satisfaction, destination image, perceived risk, service quality and perceived value. Design/methodology/approach – This study adopted a cross-sectional approach to collect data. The data were collected by conducting a field survey questionnaire on 330 respondents and were analysed using partial least squares version 3.2.9. Findings – The results show that perceived behavioural control, perceived value, destination image and satisfaction significantly affect visitors’ revisit intention. The influence of perceived value, perceived service quality and destination image on satisfaction is also confirmed. On the other hand, satisfaction is found to be a significant mediator between perceived service quality, destination image and perceived value. © Ghazanfar Ali Abbasi, Janani Kumaravelu, Yen-Nee Goh and Karpal Singh Dara Singh. Published in Spanish Journal of Marketing – ESIC. Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and noncommercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence maybe seen at http://creativecommons.org/licences/by/4.0/legalcode Authors would like to thank Professor Carlos Flavian and the anonymous reviewers for their useful and valuable suggestions. SJME 25,2


Introduction
Tourism is known as a significant industry of growth among all countries and a foremost foundation of wealth formation, livelihoods and income (Merli et al., 2019;Li et al., 2018). Over the span of years, the tourism industry has seen an exponential growth and diversification, and it has developed as one of the fastest emerging economic areas in the world (Alam and Paramati, 2016). Contemporary tourism is related to development and includes rising number of new destinations. These subtleties have enabled the tourism industry to materialize as a key driver in socioeconomic advancement (Beerli-Palacio and Martín-Santana, 2018;WorldBank, 2017).
There is no doubt that, in the context of travel and tourism-related literature, a great number of scholarly attentions has been dedicated in determining the tourists' revisit intention, Theory of planned behaviour especially in the past ten years Loi et al., 2017;Sohn et al., 2016). Revisit businesses to arrange and assign their resources to the specific elements that drives tourists' revisit intention. TPB is a psychological theory that elucidates the psychological phenomena of human behavioural intention (Ajzen, 1991).
It is evident that the TPB model has been widely applied in several tourism-related studies to examine the tourist perceptions towards a destination (Kamrul Hasan et al., 2020;Soliman, 2019;Meng and Cui, 2020;Meng and Choi, 2019). According to Bianchi et al. (2017), limited research has extended the TPB model to predict the holistic behaviour of tourists in selecting a travel destination. Furthermore, no study has yet used an extended TPB model, consisting of the studied constructs (i.e. destination image, perceived service quality, perceived value and satisfaction) in tourism. Overall, to provide theoretical developments and practical strategies, the present study aims to develop a framework that explains tourists' revisit intentions by extending TPB.
Consequently, the current study aims to: develop a model that provides a more comprehensive understanding of tourist revisit intention by extending the original TPB framework; examine the influence of attitude, subjective norms, perceived behavioural control, perceived value, destination image, perceives risk and perceived service quality on travellers' revisit intention; explore the relationships between perceived service quality, perceived value, destination image and satisfaction; and investigate the mediating role of satisfaction between perceived value, perceives service quality, destination image and tourist revisit intention, see Figure 1.
2. Literature review 2.1 Theory of planned behaviour (TPB) TPB is a psychological theory that elucidates the psychological phenomena of human behavioural intention (Ajzen, 1991). In this theory, both volitional control and non-volitional control are well-thought-out to explain an individual's behaviour (Soliman, 2019;Zailani et al., 2016). According to this theory, individual's intention or repeat intentions are the driver of human Theory of planned behaviour behaviour (Abbasi et al., 2020a(Abbasi et al., , 2020b. Revisit intention is defined as an individual's readiness or willingness to make a repeat visit the same destination (Tosun et al., 2015).
The TPB model has been widely used in the literature to explain consumer behaviour in very diverse contexts such as the adoption of smartcards or artificial intelligence to make financial investments (Belanche et al., 2014(Belanche et al., , 2019, the adoption of self-service technologies (Lien et al., 2019) or the mobile payment (Flavian et al., 2020). In a similar vein, numerous studies have also taken this model as a reference exclusively in the context of tourism (Kamrul Hasan et al., 2020;Meng and Cui, 2020;Meng and Choi, 2019;Soliman, 2019;Han et al., 2019). Meng and Cui (2020) conducted a study with aim to investigate tourist revisit intention to home-based accommodations are formed by inserting additional construct into the TPB model and empirically established that the extended TPB model has better predictive power than original TPB model. Similarly, TPB was again extended in a quest to understand consumers' intention to visit green hotels in China and established that the extended TPB model is more robust in comparison to original TPB model. Thereby, this research also extends the original TPB model by inserting perceived value, perceived risk, satisfaction, service quality and destination image. These additional constructs are added to the proposed model after a thorough literature review. Several scholars have recommended to add variables in TPB original model to enhance its predictive power (Meng and Cui, 2020;Abbasi et al., 2020a;Meng and Choi, 2019;Soliman, 2019;Russell et al., 2017;Mondéjar-Jiménez et al., 2016), as it fails to capture enough of the variance in intended behaviour (Zailani et al., 2016;Hsieh et al., 2016).

Hypothesis development 3.1 Attitudes and revisit intention
Drawing on the TPB, attitude is defined as individual's favourable or unfavourable feelings associated towards the consequences of a behaviour (Bianchi et al., 2017). Attitude is measured as a function of salient beliefs, which may be shaped by secondary information, an inferential process or by observation (Abbasi et al., 2020a;Meng and Choi, 2019). Based on these beliefs, people perceive positive or negative attitudes towards the consequences of a behaviour. Successively, attitudes affect the intentions to perform the behaviour. In the context of travel and tourism, numerous studies find a significant positive relationship between attitudes towards visiting a destination and intentions to visit or vacation destination (Bianchi et al., 2017;Hasan et al., 2017;Hasan et al., 2019). While short-and longterm travellers' beliefs about the repercussions of visiting a destination (e.g. financial, time) are likely to fluctuate, there is no reason to suppose that their attitudes based on those beliefs, whether favourable or unfavourable, will differentially impact intentions to visit or revisit the destination. Thus, we formulate the following hypothesis: H1. Attitude positively affects revisit intention.

Subjective norms and revisit intention
Based on the TPB model, subjective norms are another factor that influences behavioural intention. Subjective norm refers to an individual's perception of social pressure from people who are important to him/her to engage or not in a behaviour (Belanche et al., 2019;Bianchi et al., 2017). Subjective norms are a function of an individual's perception of how referent others (e.g. colleagues, family and friends) interpret the behaviour and the inspiration to observe with these referent's expectations and beliefs (Ajzen and Fishbein, 1970). This implies that individuals will feel communal pressure to establish a behaviour if they believe that important referents will approve the behaviour. The role of subjective norms as a driver of behavioural intentions is well recognized in the marketing and tourism literature (Hasan et al., 2020;Quintal et al., 2015). In a travel and tourism context, prior published literature empirically proves that vacation picks are swayed by travellers' views about whether closed one (family, friend or colleague) will approve of their choice to visitor does not visit a specific vacation destination. There is no sign or motive to imagine that distance between a tourist's destination and tourist's country of origin will inspire subjective norms, their development or their influence on intentions contrarily. This leads to the following hypothesis: H2. Subjective norms positively affect revisit intention.

Perceived behavioural control and revisit intention
Perceived behavioural control is also considered as an antecedent of intention and behaviour (Ajzen, 1991), and it is defined as the perception of the ease or difficulty in performing a behaviour (Song et al., 2015). That is, perceived behavioural control denotes an individual's belief that one is proficient of accomplishing the behaviour. Predominantly, perceived behavioural control contemplates individuals' perceptions of how well they can manage aspects that may enable or constrain an explicit behaviour. In this regard, Cheng et al. (2005) averred that the perceptions of exertion towards a behaviour will negatively affect the intention to perform that behaviour. Perceived behavioural control is a function of an individual's perception regarding the presence or absence of opportunities or resources to achieve a specific behaviour and the perceived level of importance of such opportunities and resources for performing such behaviour (Ajzen and Madden, 1986). Extant literature on destination choice expresses a noteworthy and constructive effect of perceived behavioural control on individuals' behavioural intentions (Bianchi et al., 2017). Factors such as abilities, time and resources are significant in predicting intentions to perform a behaviour. Based on the above discussion, the following hypothesis is formulated: H3. Perceived behavioural control positively affects revisit intention.

Perceived value and revisit intention
Perceived value is the consumer's overall assessment of monetary and non-monetary consideration for a product or service based on perceptions of what is received and what is given (Iniesta-Bonillo et al., 2016). Perceived value is also considered "the consumer's overall assessment of the utility of a product based on perceptions of what is received and what is given" (Zeithaml, 1988). Customer perceived value has been established as a strong predictor of behavioural intention in many consumer behaviour and tourist revisit intention in tourism studies (Ranjbarian and Pool, 2015;Chang et al., 2014). As argued by Chang et al. (2014), perceived value played an important role in the consumption process. Likewise, Allameh et al. (2015), in their study, they concluded that perceived value has a strong influence on tourist revisit intention to sports tourism destination. Allameh (2015) who conducted a study among tourists who travelled to Iran as a sport tourism destination exposed that perceived value has a strong influence on tourist revisit intention to sports tourism destination. This implies that when tourists perceive a higher value to a destination visited, they are likely to develop an intention to visit again. Thus, based on the above discussion, it leads to the following hypothesis: H4. Perceived value positively affects revisit intention.

Perceived risk and revisit intention
Risk is the likelihood of injury or loss. Perceived risk is defined as a potential tourists' perception about the possible uncertain negative outcomes from travel (Tseng and Wang, 2016). Perceived risks upsurge congruently as the extent of any doubt upsurges. International travel involves many uncertainties, such as hygiene, transportation, cultural struggles, security and language. A tourist develops perceived risks involving these uncertainties before commencing the trip. Extant literature has established that when tourist perceive risk relating to any tourist destination; consequently, it is going to negatively affect their intention to revisit (Allameh et al., 2015;Chen et al., 2017;Hasan et al., 2017). Therefore, this study implies that if the tourists perceive any risk associated with the tourist intending destination, it is highly likely the tourist will have negative connotation about it. Hence, based on the above arguments, this study also hypothesizes that: H5. Perceived risk negatively affects revisit intention.

Destination image with satisfaction and revisit intention
Destination image is defined as an individual's overall perception or the total set of impressions of a place (Jalilvand et al., 2012). Destination image has been one of the most investigated topics in the marketing scholarship in tourism studies (Soliman, 2019;Song et al., 2017). Studies carried among golf tourists in Hainan Area, China by Song et al. (2017) found that destination image has a significant impact on tourist revisit intention to a tourism destination. The result from a study among tourist in Iran shows that destination image positively influences tourist revisit intention to Iran as a sport tourism destination (Allameh et al., 2015). Kim et al. (2015) examined the positive relationship between destination image and tourist revisit intention to Crete among high-and also low-spending tourists. Likewise, Sharma and Nayak (2018) empirically established that a positive overall image has a noteworthy influence on the intention of tourists to revisit and recommend the visited destination to others. This implies that if the destination image of a tourist's spot is affected due to any reason, it is going to impact tourist's decision to revisit that place. Thus, based on the above discussion, the following hypothesis is formulated: H6. Destination image positively affects revisit intention.
Moreover, several studies have examined the impact of destination image towards tourists' satisfaction (Hasan et al., 2019;Toudert and Bringas-R abago, 2016;Ramseook-Munhurrun et al., 2015). A study carried out in Macau on tourist shuttle service revealed that destination image predicts intention to revisit through destination satisfaction (Loi et al., 2017). The results of study carried out on international tourists visiting the UAE confirmed the importance of destination image in predicting tourist satisfaction in a destination (Albaity and Melhem, 2017). Therefore, the following hypothesis is postulated: H7. Destination image positively affects satisfaction.
3.7 Perceived service quality with revisit intention, satisfaction and perceived value Perceived service quality is defined as the customer's evaluation of the overall excellence or superiority of the service, amenities and employees (Liu and Lee, 2016). Past studies have also empirically recognized that perceived service quality is an important antecedent of tourist revisit intention (Kim and Thapa, 2018;Adam et al., 2015). Allameh's (2015) study on tourists who travelled to Iran as a sport tourism destination revealed that perceived service quality influenced tourist revisit intention to a sports tourism destination. Another survey conducted among four-star hotels guests revealed that guest perceived service quality affects their revisit intention to hotels (Worsfold et al., 2016). This implies that if the tourist perceives the service of quality higher and superior, it is likely to increase their chances to revisit the same destination spot. Thus, based on the above discussion, the following hypothesis is formulated: H8. Perceived service quality positively affects revisit intention.
Similarly, the relation between perceived service quality and satisfaction has already been developed in prior studies (Ji et al., 2017;Zaibaf et al., 2013). A study conducted by Ji et al. (2017) reveals that perceived service quality positively affects Chinese casino visitor's satisfaction. Zaibaf et al. (2013) supported that perceived service quality is a strong predictor of satisfaction. Multidimensional model is used as a background to synthesize the impact of perceived service quality on hot spring customer satisfaction (Wu et al., 2015). Thus, researchers in this imply that tourists' satisfaction will be influenced by the quality of service if provided to tourists. If tourists perceives the quality of service higher, it will augment their level of satisfaction. Thus, based on the preceding discussion, the following hypothesis is articulated: H9. Perceived service quality positively affects satisfaction.
Also, Raza et al. (2012) found the positive relationship with perceived service quality and perceive value in the study conducted among luxury hotel customers of Pakistan. A study carried out by Allameh et al. (2015), in Mazandaran as a tourism destination in Iran, clearly shows the positive relationship between perceived service quality and perceived value. Similar results were attained by An et al. (2019) who empirically established that element of service quality has a positive and significant effect on tourists perceived value and their satisfaction. Thus, this implies that when tourists visit a destination and receives a top service quality, resultantly, tourists will start perceiving higher value to the destination visited. Therefore, based on the above arguments, the following hypotheses are formulated: H10. Perceived service quality positively affects perceived value.

Satisfaction and revisit intention
Satisfaction refers as an emotional response that follows from cognitive responses to an experience (Smith, 2020;Monferrer et al., 2019;Chong, 2016). Research in the tourism context has already established the association between satisfaction and revisit intention and empirically recognized that satisfaction lays a significant impact on creating a positive revisit intention by tourist (Hasan et al., 2019;Breiby and Slåtten, 2018). An et al. (2019) investigated tourits revist intention for Airbnb guests and established empirically that tourist satisfaction positivly stimulates their intention to revisit the destination. This implies that when tourists are satisfied with a destination, it is highly likely that they will intend to revisit the same destination. Resultantly, the following hypothesis is formulated: H11. Satisfaction positively affects revisit intention.

Perceived value and satisfaction
Prior studies have also established that perceived value can be a contributing factor to customer satisfaction García-Fern andez et al., 2018;Hussain et al., 2015). Waheed and Hassan (2016), in a study on Maldives guest houses, empirically established that perceived value has a strong impact on tourist satisfaction. Similar results were established by a study by Rosoolimanesh et al. (2016) during their investigation about tourist perceived value and satisfaction in a community-based homestay in the Lenggong Valley world heritage site and empirically demonstrated that there is positive relationship between perceived value and satisfaction. Having said this, it implies that tourists value perception is significant for their satisfaction. If tourists perceive higher value to a destination, it is expected to influence their satisfaction level. On the other hand, if tourists perceive inferior value to a destination, it might lead to reduce their satisfaction level associated with the destination. Thus, the following hypothesis is suggested:

Theory of planned behaviour
H12. Perceived value positively affects satisfaction.

The mediation effect of satisfaction
The indirect effect of perceived service quality to revisit intention via satisfaction has been investigated by previous studies. Rajaguru (2016) demonstrated that the influence of perceived service quality on customer satisfaction substantially increases the revisit intention in the case of full-service as compared to low-cost carriers. Likewise, a study by Wu (2014) also established the mediating role of satisfaction between perceived service quality and customer revisit intention.
Moreover, Albaity and Melhem (2017), on tourists visiting the UAE, show that tourist satisfaction positively mediates the relationship between destination image with revisit intention. Similarly, during an investigation carried on Macau's tourist shuttle service, it was empirically concluded that destination image forecasts tourist intention to revisit via destination satisfaction (Loi et al., 2017). Kim et al. (2015) investigated and supported the indirect effect of destination image on revisit intention thorough satisfaction in his study among high-and low-spending tourists in Crete, Greece.
Besides, customer satisfaction as a mediator in the relationship between perceived value and revisit intention is examined by earlier researchers and a positive relationship has been established between them (Kim et al., 2013;Cheng and Lu, 2013). Therefore, the following hypotheses are formulated: H13. Satisfaction mediates the relationship between perceived service quality and revisit intention.
H14. Satisfaction mediates the relationship between destination image and revisit intention.
H15. Satisfaction mediates the relationship between perceived value and revisit intention.

Study context
Malaysia is one of the most popular travel destinations in the Southeast Asian Nations region (Amin et al., 2013). According to Tourism Malaysia (2017), an increasing number of tourists are coming to Malaysia, and over two million tourists visited Malaysia as of July 2017. The tourism industry in Malaysia is getting important in terms of earning foreign exchange, creating employment opportunities and economic development. Besides that, the tourism and hospitality industry in Malaysia is unique in terms of the abundance of natural resources, cultures and local traditions (Kaur et al., 2016). Thus, cultural tourism, eco-tourism and natural tourism are popular in Malaysia compared to other countries. Given the uniqueness of Malaysia, understanding the tourist revisit intention is critical both to the tourism industry and for the economic prosperity of the country. To contrast the proposed model, the case of Penang is taken as a reference. Penang is a located in the northern part of Peninsular Malaysia. The state's population is about 1.5 million, including a mix of religions, ethnicities and cultures (Ghaderi et al., 2012). Penang is a lively State with its capital, George Town with the rare distinction of being one of the UNESCO World Heritage Site (Omar et al., 2017). Moreover, tourism industry is the second main income contributor in Penang state, after manufacturing sector (Leng, 2015). Penang state is the most visited city in Malaysia, with an estimated six million hotel guests reported in 2012 (The Malaysian Insights, 2017). In 2013, the overall arrivals in Penang reached a record 4.70 million (Tourism Malaysia, 2015). Penang has the many places for tourist's attraction, but Penang Hill stands out. Penang Hill is listed as top ten picks of "must-dos" when visiting the Penang by

Measurements
To quantify the constructs included in the empirical analysis, an exhaustive review of the literature was carried out to select the most appropriate scales to measure the different concepts. These scales underwent a process of adaptation to adapt to the particularities of our research. Revisit intention five items were adapted from Tosun et al. (2015). Similarly, for attitude, subjective norm and perceived behavioural control, three, four and three items were adapted from Bianchi et al. (2017), respectively. Whereas, perceived value and perceived risk and satisfaction were measured by four items, each adapted from Iniesta-Bonillo et al. (2016), Tseng and Wang (2016) and Le Chi (2016), respectively. Destination image was measured by using five items from Jalilvand et al. (2012). Lastly, perceived service quality was measured by using three items from Liu and Lee (2016). All the constructs were measured on a five-point Likert scale ranging from (1) strongly disagree to (5) strongly disagree.

Data collection
The unit of analysis of this study is an individual tourist who has visited Penang Hill. This study used a self-administered questionnaire, which was distributed to tourists at the exit location near the Penang Hill station, and respondents were requested to participate voluntarily in the survey. Approval was obtained from Penang Hill corporation management to distribute questionnaires to visitors visiting Penang Hill. The questionnaires were distributed to visitors at Upper Station waiting area where visitors were waiting to board the train down to Lower Station after their visit. Upper station was chosen strategically to collect the data for two main reasons. Firstly, Upper Station waiting areas have several food-related shops and stalls. Visitors after their tiring trip to Penang Hill mostly enjoy their meals and refresh before they start proceeding to Lower Station. Secondly, as this research is focusing on tourist revisit intention, it was deemed important to filter respondents who have actually paid a visit to Penang Hill earlier. Respondents were approached conveniently and were asked to fill the survey once they have fulfilled the filtering criteria by answering two filtering questions: Have you answered this questionnaire before; have you visited Penang Hill before? Convenience sampling was used to collect the data, as it is not possible to have a sampling frame for all the tourists. Only those who actually visited Penang Hill Theory of planned behaviour were asked to fill up the questionnaire. The questionnaires were handled personally to respondents and retrieved on-site after filled. Around 400 questionnaires were distributed and 338 were returned to the researchers, with the response rate of 84.5%. All the returned questionnaires were carefully examined, and 330 were found fit for data analysis. Furthermore, in survey studies, the issue of common method bias (CMB) may affect the reliability and validity of the studied constructs along with its effect on hypothesized relations in the proposed model. In this regard, to cater the CMB, the survey was designed to make sure that it can be catered procedurally as per the guidelines of Podsakoff et al. (2003) by taking several procedural steps. Firstly, the authors assured the respondents on the privacy and confidentiality of their responses; second, well-established scale designs were used to avoid any vagueness; third, we pre-tested the questionnaire and made necessary corrections before conducting any further analyses, to confirm and validate the content and face validity. Pre-testing was done by two academic experts. Only minor issues pertaining to language and grammatical mistakes were highlighted. After making the minor amendments suggested by the panel of two academic experts, the survey questionnaire was disseminated for data collection.
Descriptive analysis found that majority of the respondents are Malaysians (i.e. 70%) compared to non-Malaysians (i.e. 30%). Of the respondents, 54% were male. In this study, the age of the respondents was found between 31 and 40 years old (36.7%). Besides that, 87 respondents, constituting to 27%, had monthly income between MYR2,000 and MYR3,000. Table 1 provides the sociodemographic profile of the respondents and the descriptive analysis of demographic variables in detail.

Results
Concerning the statistical analysis, the proposed research model is tested using the partial least squares structural equation modelling (PLS-SEM) approach. PLS-SEM is considered useful in comparison to covariance-based-SEM (CB-SEM) because of several reasons, such as it can test complex models (Henseler et al., 2015;Chin et al., 2003), small sample size, does not involve normality and able to work without distributional assumptions (Hair et al., 2014) and is predominantly very useful for testing the moderating effect between the constructs (Chin et al., 2003). Furthermore, after reviewing the work of Rönkkö and Evermann (2013), it was suggested that PLS-SEM is more efficient than CB-SEM, particularly in the quest of findings the true model (Hair et al., 2014). Analysis via PLS-SEM consists of two components: the outer model (also known as measurement model) and the inner model (also known as structural model) (Hamidi and Chavoshi, 2018;Ramayah et al., 2018). The outer model measures the reliability and validity, whereas the inner model tests the strength of the relationship between the constructs (Ramayah et al., 2018;Hair et al., 2014).

Measurement model evaluation
To assess the measurement model of reflective constructs, the current study examined their internal reliability, convergent validity and discriminant validity criteria (Hair et al., 2019). According the results, the outer loadings of all items, Cronbach's a, the composite reliability (CR) and the average variance extracted (AVE) of all constructs were above 0.7 and 0.5, respectively, see Table 2. Thus, the proposed measurement model's construct reliability and convergent validity is established as per the suggestions of Hair et al. (2019). To establish the discriminant validity, two criteria were applied in this study, i.e. heterotrait-monotrait (HTMT) ratio (Henseler et al., 2015) and Fornell and Larcker (1981). The HTMT values were revealed to be less than 0.85 (Table 3), thus confirming the discriminant validity of all given variables (Kline, 2015). Moreover, the findings in Appendix indicate that the square roots of the AVEs for all the constructs are greater than the inter-construct correlations, thus Theory of planned behaviour meeting the requirements for the Fornell and Larcker (1981) technique for confirming discriminant validity. In total, the measurement model demonstrated the adequate convergent validity and discriminant validity.

Structural model evaluation
After examining the validity of the measurement model, the current study tested the hypotheses using the structural model. The proposed model demonstrated that 66.0% (R 2 value = 0.660) of the variance of revisit intention was explained, which was above the minimum threshold R 2 value of 25% (Hair et al., 2016). Moreover, the mediating variables of perceived value and satisfaction's R 2 values were 0.451 and 0.533, respectively. In addition to R 2 values, the current study also calculated the predictive relevance (Q 2 ) values by using by running blindfolding procedure. Values more than zero suggest that the model has predictive relevance (Chin, 2010). The current study's results display satisfactory predictive relevance for all endogenous constructs, i.e. revisit intention (0.460), satisfaction (0.397) and perceived value (0.319); all Q 2 values are above zero. Furthermore, the study also examined the effect size (f 2 ), which is used to observe whether a specific exogenous construct has a substantive impact on an endogenous variable. Based upon Cohen's (1988) recommendation, the results of the research show that the f 2 for the supported hypotheses was acceptable (Table 4).
One of the current study objectives was to test the mediating relationship of satisfaction between perceived service quality, destination image, perceived value and revisit intention. In this regard, Hayes (2009) and Preacher and Hayes (2008) estimation of the mediation effect for SEM with the bootstrap test was applied. In regard to mediation analysis, perceived service quality proved to have a significant indirect impact on visitor's revisit intention via satisfaction, and H13 was supported (b = 0.093, p < 0.05). H14 hypothesized destination image has an indirect effect on visitor's revisit intention significantly mediated by satisfaction; thus, H14 was supported in this analysis (b = 0.136, p < 0.05). Finally, H15 (b = 0.175, p < 0.05) suggested that perceived value has an indirect impact on revisit intention thorough satisfaction. As a matter of fact, perceived service quality, destination image, perceived value plays an important role in providing visitor's satisfaction and lead to revisit intention.

Discussion
Grounded on the TPB, this study investigated the tourist revisit intention by extending the TPB original model with the inclusion of perceived service quality, perceived value, destination image, satisfaction and perceived risk. The findings of the study revealed that, among all the constructs used in this study, perceived value and satisfaction had the strongest effect on satisfaction and tourist revisit intention, respectively. Therefore, theoretically, researchers have effectively extended the original TPB in line with Ajzen's (1991) criteria for improving the theory. From the standpoint of tourist overall behaviour, the current research is a progression of broadening and deepening the existing social theory in the tourist behaviour literature (Perugini and Bagozzi, 2001). Specifically, the study contributes imperative understanding into tourists' revisit intention by outspreading the existing TPB framework by adding perceived service quality, perceived risk, destination image and perceived value. To the best of our knowledge, few studies have empirically extended the TPB by adding studied variables. The results demonstrated that all proposed hypotheses relating to direct relationship were supported, except the relationships between attitude, subjective norm, perceived risk and perceived service quality were found to have an insignificant effect on tourists' revisit intention. In addition, the current study also investigated the mediating role of satisfaction between perceive service quality, perceived value, destination image and revisit intention, and the results also exhibited that all Notes: *p < 0.10; **p < 0.05; ***p < 0.01, a b negative sign Theory of planned behaviour hypotheses in this regard were accepted. At the same time, the results of this study disclose complex associations among these constructs.
The result has shown that attitude has no relationship with tourists' revisit intention to Penang Hill. In earlier research, attitude was found to be an important factor in predicting tourist revisit intention (Meng and Choi, 2019;Soliman, 2019;Quintal et al., 2015;Hasan et al., 2017). This implies that not all the tourists posit a positive attitude towards revisiting Penang Hill. Moreover, Sparks (2007) called it as push-and pull-related beliefs that are associated to develop a positive attitude and future intention. This implies that in the quest to inculcate a revisit intention among tourists, destination managers must focus of other factors rather than developing a positive attitude. Moreover, Surprisingly, subjective norm has no relationship with tourists' revisit intention to Penang Hill. The result of this study is not in line with past studies such as by Meng and Cui (2020) and Quintal et al. (2015). Earlier research in the tourism context has shown inconclusive results, e.g. several researchers, just like our current research, have already empirically established that subjective norm does not have a significant influence on tourist behaviour (Shen et al., 2009;Sparks, 2007). The result of our study may be explained by the fact that a large group of respondents are adults between 21 and 40 years old, in which we believed that they are wise enough to make their own decision to visit a tourist destination and not easily get influenced by others. However, researchers have highlighted that the role of subjective norm cannot be disregarded whilst investigating loyalty-related behaviour of tourist, such as, in this case, revisit intention (Han et al., 2019). Therefore, this insignificant relationship can further be strengthened by provide benefits such as delivering appealing atmosphere, highly interactive atmosphere and exhibiting authentic cultural appearances (Meng and Cui, 2020).
Among all the three original TPB constructs, only perceived behavioural control positively affects the revisit intention to Penang Hill. The result of this hypothesis is in line with previous studies (Lee et al., 2014;Cuong and Jian, 2014). In this study, more than 60% of the respondent earning was above RM3,001. They can afford to spend on the tickets resultantly, confirming their revisit intention. Perceived behaviour control is regarded as non-volitional determinant of revisit intention in this study; the results implies that tourists did not perceive difficulty, i.e. financial difficulty, to revisit Penang Hill.
In addition, it was also found in the results of the study that perceived value positively influences tourists' revisit intention to Penang Hill. The results of this study were in line with the earlier studies by Allameh et al. (2015) and Chang et al. (2014). Descriptive analysis of the study reveals that 70% of the respondents are local tourists who have visited Penang Hill. This group of respondents are able to enjoy promotional rate applicable to Malaysian citizen using MyKad. Perceived value is always in the eye of the beholder; overall assessment based on what they feel or believe is worth at that particular time and place compared with what is given. In this case, money, effort, time are used to measure the perceived value among Penang Hill visitors. Considering the amount spent on tickets, the perceived value is strong among them. Apart from monetary value, the respondents perceived the benefits they received from visiting Penang Hill in both monetary and non-monetary terms is more than the sacrifice.
Moreover, the results of the study also exhibited an insignificant relationship between perceived risk and tourists' revisit intention to Penang Hill. This study results were inconsistent with the result of the past studies by Chen et al. (2017) and Hasan et al. (2017). The findings imply that tourists' revisit intention is not influenced by the risk factor associated with the destination. Perception of risk may make more sense to tourists when they are going to a destination for a first time, because this study focuses on the revisit intention of tourists; thus, perception of risk will not have any substantial effect on tourist revisit intention. Furthermore, around 73.1% of tourists are between the age of 18 and 40. According to the generational cohort theory, every cohort portrays a certain behaviour owing to their experiences driven by the related events and technological developments that emerged over the specific era (Duh and Struwig, 2015). Therefore, tourists representing these age cohorts surely would have done a proper researcher planning before revisiting Penang hill, subsequently reducing their risk perception.
This study also further supports the idea of destination image influence revisit intention. Similar results were achieved by Song et al. (2017) and Allameh et al. (2015). Penang Hill, as one of the popular tourist destinations, has been covered in major magazines and online websites, which helps to portray a positive image to encourage revisit intention among the local and international tourists. The result implies that destination image plays a pivotal factor for tourists' destinations; thus, the better the image of the destination, the more people will be attracted to visit/revisit. Furthermore, the result of the current study also established destination image has a positive impact of tourist satisfaction. Prior studies have also empirically proven that destination image is central and play diverse roles in the process of decision-making, as all decision-making factors such as time, money and family are relying on the image of each destination to satisfy the decision makers, subsequently influencing both first-time visit and revisit intention (Allameh et al., 2015;Chen and Tsai, 2007). Thus, the result of the current implies that a positive destination image is critical to tourists, which resultantly has a profound impact on tourist's satisfaction and their revisit intention. From an industry point of view, the integration of destination image with the prior behaviour of tourist from the lens of the TPB can provide a pivotal insight to policymakers and tourism operators and encourages them shed focus on creating a positive image of destination.
Furthermore, the study also found an insignificant relationship between perceived service quality and tourists' revisit intention. However, perceived service quality was found to have significant indirect effect on tourist revisit intention through satisfaction (H13). The results imply that tourists' positive service quality perceptions increase their satisfaction to visiting a destination, but it is not guaranteed to lead to tourist revisit intention as long as the tourists develop satisfaction with their tourist destination. The result of this study is line with Kamrul Hasan et al. (2020) and contradicts with Kim and Thapa (2018) and Adam et al. (2015) studies. Furthermore, this inconsistency in the result can then be attributed to diversity in demographic profile of this study. Kotler (2016) narrated that every customers views service quality from the lens of variety of values such as reliability, assurance, empathy and tangibility. According to Ganesan-Lim et al. (2008), local tourists, as compared to foreign tourists, are more adaptable to the lower quality of service. Furthermore, the result of the study also reveals that perceived service quality positively influences a tourist's satisfaction and perceive value. These results are also in line with previous studies (Rasheed and Abadi, 2014;Wu et al., 2015). The result implies that if tourists' expectations are fulfilled and they perceive the high-quality service, their perception about the destination will also enhances. Thus, it can be stated that experience of using the service significantly impacts their perceived value and tourist's satisfaction.
In overall, they are satisfied with Penang Hill whereby satisfaction significantly influences their revisit intention to Penang Hill. This finding agrees with Hasan et al. (2019) and Breiby and Slåtten (2018) finding, which showed that satisfaction had a direct influence on intention to revisit. This result implies that if tourists are satisfied with their first visit to a destination, resultantly, tourists are more likely to consider revisiting it. Similarly, the study also reveals that perceived value also positively influences a tourist's satisfaction. Similar results were achieved by García-Fern andez et al. (2018) and Waheed and Hassan (2016) and Allameh et al. (2015). The findings highlight and explain how perceived service quality and service convenience could be influencing tourist satisfaction. In particular, high levels of quality yields high value and high satisfaction.
Lastly, the outcome of this study has established the positive and significant mediating effect of tourist satisfaction on perceived service quality, destination image, perceived value and revisit intention. The results of this study are in line with previous studies (Wu et al., 2015;Nazri and Zainal, 2013;Albaity and Melhem, 2017). The results of the study have proven that visitors' level of satisfaction and dissatisfaction with perceived service quality, destination image, perceived value and tourist's satisfaction is influencing the revisit intention to Penang Hill. In this study, about 69% of the visitors have visited Penang Hill more than one time; it is proven that they are satisfied with their earlier visit to Penang Hill and decided to revisit the destination again.

Theoretical implications
The results of this study provide theoretical and practical contributions to the service literature and tourism industry managers. Theoretically, it contributes in several ways. Firstly, the current study used the TPB to study the factors that drive tourists' satisfaction and their revisit intention to Penang Hill particularly. To the best of the researcher's knowledge, this study is the first of its nature that has extended the TPB original model to explains the factors behind tourists' revisit intention. Pervious literature on tourism in Malaysia has mostly focused on tourist visit intention (Mohaidin et al., 2017), but no study has investigated the factors that behaved the tourists' revisit intention. Secondly, this study used perceived service quality, perceived risk, perceived value, destination image and satisfaction to extend the TPB and enhance the tourism-related body of knowledge. The proposed model of this empirical research included the variables of destination image, perceived service quality, perceived value and tourists' satisfaction from revisiting a tourism destination, i.e. Penang Hill, resultantly could elucidate a great proportion of their revisit intention and showed the compatibility of the model. The inclusion of these variables extends the current tourism literature, as no other study has provided a holistic model by inclusion all studied factors to reveal their effect on tourist revisit intention. Thirdly, this study has some interesting results, especially pertaining to the antecedents of the TPB, e.g. insignificant relationships of attitude and subjective norm. These results contradict earlier studies, at the same time, enhance the extant body of knowledge, especially in the context of tourism in Asia-Pacific countries. Malaysia is an emerging Asia-Pacific country, which is culturally, expressively different to developed countries where earlier studies have taken place. Therefore, the application of the TPB in the Asia-Pacific context sheds new light to the path model of the TPB and open the vistas of opportunities for future researchers to investigate how these elements can be explored for future studies.
Lastly, this studied focused on the tourism industry in an emerging country like Malaysia; therefore, this study provides and broadens the extant body of knowledge pertaining to the tourism industry in a contextual perspective. Most of the earlier studies on tourism draw their sample from developed countries. Owing to the cultural differences between countries can limit the extent to which findings from one country may not be applied to other countries that may have different preferences and expectations that are deeply rooted in their shared cultural or societal values (Torres et al., 2014).

Practical implications
Practically, this research is meaningful for both local and global tourist destinations managers who seek to reach and influence tourist revisit intention. To enhance the competitive advantage, tourist destinations should develop their strategies on achieving a greater tourists' satisfaction by providing greater value and quality service, to increase their intention to revisit. For instance, a tourist's satisfaction can be attained by delivering service that is beyond the tourist's expectation, providing greater value to assure that both local and international tourists that they had made the right decision. Especially, in the context of this study's findings, which indicate that the tourist attitude, subjective norm, element of perceived risk and service quality are found insignificant in this study, this implies that there are other factors that draw demands to entice tourist revisit intention, which both academicians and practitioners need to be extremely cognizant of such factors.
Furthermore, the positive effect of destination image on satisfaction and tourist revisit intention highlights the importance of tourist destination spot. Tourists' travel experience could be adversely affected by a single incident, thereby arising dissatisfaction, and subsequently lowering their revisit intention. Managers of tourist destinations and the service providers, therefore, should safeguard that frontline staff uphold high levels of service to incoming tourists. As such, service providers could place larger weight on providing training that encourages employees to deliver quality service to customers and promote a positive image of their area. Furthermore, destination image is contingent on several practical aspects, e.g. tourist amenities, safety and cleanliness. Therefore, a favourable image inspires tourists to make return visits, but unhappy and unsatisfied tourists may not revisit a destination, even though they perceive destination image positively, especially when the practical aspects are mediocre to those elsewhere. Therefore, to enhance the destination image, both local businesses and the government can help to create a positive destination image from tourists' perspectives. There is no doubt, blunders and mistakes are unavoidable in tourism activities, but the crucial point in service activities of tourism is to rectify tourists' difficulties to fascinate their attention and consequently retain them. As the Asian market gaining increasing importance in the global economy, this study offers much needed insights for marketers to target tourists to visit again. It is also noteworthy for destinations marketers to notice that the results of this study also imply that tourist perception of service quality does not affect tourist revisit intention, unless there are satisfied. Once tourists are satisfied, they will start inculcating intention to revisit. Therefore, Penang Hill Corporation needs to give more attention to elements that can stimulate tourist satisfaction, which ultimately affects tourist perception of service quality, enhances the image of tourist destination image, boots perceived value attached to the destinations and creates more urge to revisit.
Comprehending and developing the knowledge driven from this study will enable the tourism industry within the country to strategize to achieve the competitive advantage in this lucrative industry. Moreover, the tourist demographic profile such as nationality is an important source of reference for hill resort operator. For example, local tourists who are Malaysian citizens entitled for a promotional rate to visit Penang Hill with their MyKad. Considering the amount of money spent to visit Penang Hill, the perceived value of Malaysians to revisit is high, as confirmed from this study. Hence, it is recommended that other hill resort operators could offer local tourists a promotional discount to encourage future revisit intention.

Limitations and future research
Although the study met its objectives, it has some limitations as well. First of all, questionnaires were prepared in only one language (English). This study was carried out among local and foreign visitors, and some local respondents may have faced difficulty in understanding the questions in English. So, only certain group of local visitors were able to understand and complete the questionnaire. This shortcoming of the study caused discrepancy in respondent's demographic profile, which is likely to be an important element of revisit Theory of planned behaviour intention. Besides that, the proposed extended theoretical model in this research did not include all potential variables applicable to tourist revisit intention. Its excluded variables such as past experience and word of mouth in examining tourist revisit intention. Future researchers are also encouraged to add context-related variables to increase the predictive power of the original TPB model. Furthermore, this study did not perform the multigroup analysis to exhibit the possible differentiating effects between people from the same country and foreign visitors, which can be considered for future studies. Also, as this study was crosssectional in nature, future researchers may use longitudinal approach to collect data to comprehend the intricacies involved in understating tourist attitude and their subsequent behaviour. Furthermore, this study applied convenience sampling method for data collection; future researchers may try the application of the extended TPB model with a different sampling technique. Lastly, future studies should replicate the studied model with a larger and more geographically diverse sample.

Corresponding author
Yen-Nee Goh can be contacted at: yngoh@usm.my For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: permissions@emeraldinsight.com