Submitted to the Institute of Graduate Studies and Research in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Tourism Management Do eReferral, eWOM, Familiarity, and Cultural Distance Predict Enrollment Intention among Educational Tourists? Application of Artificial Intelligence Technique Akile Oday Eastern Mediterranean University September 2021 Gazimağusa, North Cyprus Approval of the Institute of Graduate Studies and Research Prof. Dr. Ali Hakan Ulusoy Director Prof. Dr. Hasan Kılıç Dean, Faculty of Tourism Prof. Dr. Mustafa İlkan Co-Supervisor Prof. Dr. Ali Öztüren Supervisor I certify that this thesis satisfies all the requirements as a thesis for the degree of Doctor of Philosophy in Tourism Management. We certify that we have read this thesis and that in our opinion it is fully adequate in scope and quality as a thesis for the degree of Doctor of Philosophy in Tourism Management. Examining Committee 1. Prof. Dr. Serhat Harman 2. Prof. Dr. Mustafa İlkan 3. Prof. Dr. Anıl Kemal Kaya 4. Prof. Dr. Kutay Oktay 5. Prof. Dr. Ali Öztüren 6. Assoc. Prof. Dr. Mine Haktanır 7. Asst. Prof. Dr. İlkay Yorgancı Maloney iii ABSTRACT The extant literature has demonstrated the benefits of electronic word-of-mouth (eWOM), electronic referral (eReferral), familiarity, and cultural distance on behavioral outcomes separately. Research efforts have overlooked their collective effects from educational tourism perspective. This dissertation fecundates the concept of eWOM, eReferral, familiarity, and cultural distance with social network theory to explore their influence on enrollment intention. Cross-sectional data garnered from educational tourists based on a judgmental sampling technique were subjected to linear modeling and artificial neural network modeling in training and testing phases. Empirical analysis based on a single-sourced data of n=931 educational tourists confirmed the influence of eReferral, eWOM, familiarity, and cultural distance on enrollment intentions symmetrically (linear modeling) and asymmetrically (artificial neural network). The artificial neural network technique exerted higher predictive relevance and validity. This dissertation provides meaningful theoretical, practical, and methodological insights into the collective and contributive effects of eReferral, eWOM, familiarity, and cultural distance on ed-tourist enrollment intentions. Practically, implications for university administrators and marketers are prescribed. Methodologically, the research provides incremental insights from orthodox (i.e., linear) and contemporary analytical (i.e., artificial neural network) techniques, which are relevant to the wider management and tourism literature. The results suggest that eReferral, eWOM, familiarity and cultural distance can predict intention to enroll in both symmetrically (linear modelling) and asymmetric ally ( Artificial Neural Network ) manner. The asymmetric modeling possesses greater predictive validity and relevance. This study contributes theoretically and methodologically to the management literature iv by validating the proposed relationships and deploying contemporary method such as Artificial Neural Network. Keywords: Familiarity, eReferral, eWOM, Online reviews, Cultural distance, Enrollment, Educational tourism v ÖZ MeYcut literatr, elektronik ağızGan ağıza iletiúim e:2M , elektronik y|nlenGirme e5eIerral , aúinalık Ye kltrel mesaIenin GaYranıúsal sonuçlar zerinGeki IayGalarını ayrı ayrı g|stermiútir $raútırma çaEaları, eğitim turizmi perspektiIinGen kolektiI etkilerini g|zGen kaçırmıútır %u tez, kayıt niyeti zerinGeki etkilerini araútırmak için e:2M, e5eIerral, aúinalık Ye kltrel mesaIe kaYramlarını sosyal ağ teorisi ile Eesler (ğitim turistlerinGen yargısal |rnekleme tekniğine Gayalı olarak elGe eGilen kesitsel Yeriler, eğitim Ye test aúamalarınGa Goğrusal moGelleme Ye yapay sinir ağı moGellemesine taEi tutulmuútur n  eğitim turistinin tek kaynaklı Yerilerine Gayanan ampirik analiz, e5eIerral, e:2M, aúinalık Ye kltrel mesaIenin kayıt niyetleri zerinGeki etkisini simetrik Goğrusal moGelleme Ye asimetrik yapay sinir ağı olarak GoğrulaGı Yapay sinir ağı tekniği, Gaha yksek |ng|rc alaka Ye geçerlilik uygulaGı %u tez, e5eIerral, e:2M, aúinalık Ye kltrel mesaIenin eG-turist kayıt niyetleri zerinGeki kolektiI Ye katkıGa Eulunan etkilerine Gair anlamlı teorik, pratik Ye metoGoloMik içg|rler sağlar 3ratik olarak, niYersite y|neticileri Ye pazarlamacılar için çıkarımlar |ng|rlmútr MetoGoloMik olarak, araútırma, Gaha geniú y|netim Ye turizm literatr ile ilgili olan ortoGoks yani Goğrusal Ye çağGaú analitik yani yapay sinir ağı tekniklerinGen artan içg|rler sağlar 6onuçlar, e5eIerral, e:2M, aúinalık Ye kltrel mesaIenin hem simetrik Goğrusal moGelleme hem Ge asimetrik Yapay 6inir $ğı úekilGe kayGolma niyetini tahmin eGeEileceğini g|stermekteGir $simetrik moGelleme, Gaha Iazla tahmin geçerliliğine Ye alaka Gzeyine sahiptir %u çalıúma, |nerilen iliúkileri Goğrulayarak Ye Yapay 6inir $ğı giEi çağGaú Eir y|ntemi kullanarak y|netim literatrne teorik Ye metoGoloMik olarak katkıGa EulunmaktaGır vi Anahtar Kelimeler: $úinalık, eReferral, eWOM, deYrimiçi incelemeler, Kltrel mesaIe, Kayıt, (ğitim turizmi vii DEDICATION To My Father viii ACKNOWLEDGEMENT I am extremely grateful to my supervisors. The completion of this study could not been possiEle Zithout the continueG support anG e[pertise oI 3roI 'r $li gztren, 3roI 'r MustaIa İlkan anG Assoc. Prof. Dr. Abubakar Mohammed. With Many thanks to my beloved staff of the School of Computing and Technology. I could not have completed this dissertation without the support of my friends who provided stimulating discussions as well as happy distractions to rest my mind outside of my research. Last but not the least, I would like to thank m y family, without you none of this would indeed be possible. I am extremely grateful to my family for their love, prayers, caring and sacrifices for educating and preparing me for my future. Finally, my thanks go to all the people who have supported me to complete the thesis work directly or indirectly. ix 1 TABLE OF CONTENTS ABSTRACT ................................................................................................................ iii g= ................................................................................................................................ v DEDICATION ........................................................................................................... vii ACKNOWLEDGEMENT ........................................................................................ viii LIST OF TABLES ..................................................................................................... xii LIST OF FIGURES .................................................................................................. xiii 1 INTRODUCTION .................................................................................................... 1 1.1 Problem Statement .............................................................................................. 7 1.2 Research Purpose And Rationale Study .............................................................. 9 1.2.1 Purpose of the Dissertation ......................................................................... 9 1.2.2 Contribution and Significance of the Dissertation ..................................... 9 1.3 Outline of the Dissertation ................................................................................ 10 2 LITERATURE REVIEW ....................................................................................... 11 2.1 Educational Tourism ........................................................................................ 11 2.2 Social Network ................................................................................................. 17 2.3 eReferral ........................................................................................................... 20 2.4 eWOM .............................................................................................................. 21 2.5 Familiarity ........................................................................................................ 28 2.6 Cultural Distance .............................................................................................. 31 2.7 Theoretical Framework .................................................................................... 34 2.7.1 Social Network T heory ............................................................................ 34 2.8 Hypotheses Development ................................................................................. 38 2.8 .1 eReferral, eWOM and Enrollment Intentions .......................................... 39 x 2.8 .2 Familiarity and Enrollment Intentions ..................................................... 40 2.8 .3 Cultural Distance and Enrollment Intentions .......................................... 41 3 RESEARCH METHODOLOGY ............................................................................ 44 3.1 Research Philosophy ........................................................................................ 44 3.1.1 Strategy ( Survey) ..................................................................................... 46 3.1.2 Research Philosophy ( Positivism) ........................................................... 46 3.1.3 Methodological Choice ( Q uantitative Method) ....................................... 48 3.1.4 Time Horizon ( Cross Sectional) .............................................................. 49 3.2 Study Context ................................................................................................... 49 3.3 Data Collection ................................................................................................. 51 3.3.1 Pilot Study ............................................................................................... 51 3.3.2 Sample and Procedure ............................................................................. 51 3.4 Study Measures ................................................................................................ 57 3.5 Analytical Approaches ..................................................................................... 58 4 RESULTS OF THE STUDY .................................................................................. 60 4.1 Demographic B reakdown ................................................................................. 60 4.2 Results of Coefficients Between V ariables ...................................................... 62 4.3 Results of Neural N etwork ............................................................................... 67 5 DISCUSSION AND CONCLUSION .................................................................... 68 5.1 Discussion ........................................................................................................ 68 5.2 Theoretical Implications ................................................................................... 73 5.3 Practical Implications ....................................................................................... 75 5.4 Limitations and R ecommendations for Future Research ................................. 77 REFERENCES .......................................................................................................... 79 APPENDI CES ......................................................................................................... 108 xi Appendix A: The List of Items and Descriptive Statistics .................................... 109 Appendix B: The Analytical Steps of Flowch art .................................................. 111 Appendix C: R Co des ............................................................................................ 112 xii LIST OF TABLES Table 1: Demographic variables ................................................................................ 61 Table 2: Linear modeling coefficients for enrollment intentions............................... 62 Table 3: Training and testing networks ...................................................................... 67 xiii LIST OF FIGURES Figure 1: Online users influenced by reading social media 2018, by country ............. 3 Figure 2: Social media platforms used by marketers worldwide 2021 ........................ 6 Figure 3: Ritchie's educational tourism segmentation model .................................... 15 Figure 4: A social network is a web of relations ........................................................ 18 Figure 5: Traditional social network .......................................................................... 19 Figure 6: Attitudes towards online shopping in Turkey 2020 ................................... 31 Figure 7: 2pinion leaGers¶ netZork Zith a high degree of centrality, proximity centrality, and betweenness centrality........................................................................ 37 Figure 8 : Research Model .......................................................................................... 43 Figure 9 : Methodology Scheme ................................................................................. 45 Figure 10: Sampling Process Steps ............................................................................ 52 Figure 11: Sampling Strategies .................................................................................. 54 Figure 12: Relative importance of the predictor variables ......................................... 63 Figure 13: Artificial neural network modeling .......................................................... 65 Figure 14: Generalized weight distribution for artificial neural network modeling .. 66 1 Chapter 1 1 INTRODUCTION This section discusses the goals and the contribution to theory and practice. Additionally, research limitations and shortcomings are addressed, as is the study's premise. Furthermore, a compilation of other chapters is included. The development of both the education and tourism industries has resulted in increased awareness of both sectors from an economic and social viewpoint in recent decades. As a result, tourism companies must use digital platforms and Web 2.0 marketing techniques to remain competitive ( Cristobal-Fransi et al., 2017). The enjoyment of sharing and exchanging information about locations and other aspects of holidays is a vital component of the travel experience (Munar et al., 2013). Tourism managers have been inspired to develop many of their core functions online as a result of the widespread use of media and sharing platforms, as well as the rapid advancement of ICTs: advertising, networking, product and service delivery, destination image, brand, and identity tracking, knowledge and information formation, customer interaction, and expanding communications networks are just a few of the serv ices that are available to consumers (Gon et al., 2016). Education and healthcare are two examples of services that are increasingly being provided online. In both, e-service quality, or the level of service that consumers receive through online channels, is essential (Blut et al., 2015). 2 The implied commitment to one's self to travel to a place to enjoy the attractions or activities is referred to as the intention to visit. Visit intention is critical for marketing practitioners because it demonstrates visitors' involvement and engagement to the actual location. Kim et al. (2009) showed that knowledge about the destination influences visits intention significantly. Furthermore, the subjective knowledge obtained via involvement in tourist activities leads to better decisions about visitation intentions (Sharifpour et al., 2014). A previous study on the factors that influence visit intentions (Loureiro and Sarmento, 2018) concluded that people's emotional attachment to the destination impacts their choice to visit such a location. As of March 2018, this statistic shows the percentage of online users worldwide whose online buying behavior is influenced by social media according to country. Respondents said that reading reviews and comments on social media affects their online shopping habits (Figure 1 ). 3 Figure 1: Online users influenced by reading social media 2018, by country ( Source: https://www.statista. com/statistics/297006/internet -users-expert-opinions- before-purchase/ ) 4 Tourists on the demand side of today's experience economy spend a significant amount of time in virtual worlds searching for information before making purchasing choices. On the supply side, the ongoing advancement of ICT technology has placed destination marketing organizations (DMOs) in a highly competitive environment where more innovative tactics are required to distinguish destinations. Current brand management views emphasize the significance of delivering a good brand experience to customers when they interact with brands (e.g., via physical shops, advertising campaigns, or companies' websites or social media platforms) for this reason ( Jimenez-Barreto et al., 2019). The advent of social media has transformed visitors from passive recipients of services given by conventional advertising and government tourism sources into practical information producers. According to Sigala et al. (2012), today's travelers are producing a notable quantity of data, which is widely distributed on the Internet. To create and access new content, they work closely with their colleagues and interact with them (Boyd and Ellison 2008). Online social networks (OSNs) are usually cited as the most significant technical development in recent years, with a beneficial impact on the tourism sector (Bilgihan et al., 2013). OSNs include well -known sites such as Facebook and Twitter. It allows passengers to exchange their experiences, thoughts and opinions, hotel reviews, vacation ideas, and package offer via the use of OSNs (Bilgihan et al., 2013). The usage of OSNs has a significant effect on the lives of members of Generation Y (those born between 1978 and 1994) as they adopt the essence of interactive digital media. According to Kim et al. (2018), consumers like reading user reviews, which have been shown to influence their purchasing decisions; trust in online review forums, however, remains questionable. 5 Tourists are increasingly dependent on internet platforms to acquire necessary travel information, according to Bronner and de Hoog (2011). Aside from obtaining information, visitors may record and share their experiences, which serve as a resource for future travels. Platforms such as Instagram, Twitter, and Facebook accommodate and harbor several information generation activities through interactions, adverts, products, brands, and news. These features have transformed these sites into relevant marketing platforms (Leung et al., 2017). Such activities on such pla tforms can empower institutions to attract and retain international students in the context of educational tourism (ed -tour) ( Harazneh et al., 2018) . Tourists are more inclined to utilize social media to learn about a location or organization (Peruta and S hields, 2018; Xiong et al., 2018). According to Ritchie (2003 ) and McGladdery and Lubbe (2017), socio -psychological forces such as experience, familiarity, and culture are important factors for tourism product consumption. Familiarity reflects knowledge on how to acquire information about a business entity or institution. Facebook was the most widely utilized social media network among marketers worldwide as of January 2021. According to a global poll, 93 percent of social media marketers said that they us ed the platform to promote their brand, with another 78 percent using Instagram. When asked which social media network was most crucial for their business in early 2021, the great majority of global marketers chose Facebook. Advertisers and marketers have realized Facebook's potential as a marketing tool to reach new and existing customers as a result of its global reach, and have come up with new and unique ways to interact with their target audiences as a result (Figure 2 ) . 6 Figure 2: Social media platfo rms used by marketers worldwide 2021 ( Source: https://www.statista.com/statistics/259379/social -media-platforms-used-by- marketers-worldwide/ ) Familiarity with destinations, products, websites, and social media platforms appears to exert a significant influence on decision-making processes )laYi n et al,  -en- Hwa Hu et al., 2017). In addition, culture reflects individuals' values, which exert s a meaningful impact on behavioral outcomes (Farzin and Fattahi, 2018). Past research shoZeG that ³the Gesire to see anG e[perience a GiIIerent culture is a motiYator Ior leisure tourists, Zhich inIluences Gestination preIerences anG choices´ Kozak, , p. 4). However, cultural differences can also demotivate visit intentions (Qian et al., 2018). In essence, cultural distance functions as both a motivator and a demotivator depending on the tourism context. Interestingly, research has rarely examined how cultural distance and familiarity regulate visitation and destination choices in the context of edu-tour. 7 Existing empirical work has identified electronic referral (eReferral; Al -Htibat and Garanti, 2019), electronic word -of-mouth (eWOM; Ladhari and Mic haud, 2015), familiarity (Mittendorf, 2018), and cultural distance (Dang and Nandakumar, 2017; McGladdery and Lubbe, 2017a) as determinants of purchase decisions, tourist engagement, and visit intentions in other contexts. The lack of empirical evidence in the edu-tour context warrants further investigations. This study aims to contribute to the literature in various ways. The first is to shed light on the eReferral (i.e., based on strong ties), eWOM (i.e., based on weak ties), cultural distance, and famili arity concepts by examining their effects on intention to enroll using social network theory. Second, conceptualizing eReferral, eWOM, familiarity, and cultural distance can create an alternative and better medium to reach educational tourists (ed u-tourists). Third, this study aims to provide answers regarding the relative importance of the aforementioned antecedents in predicting edu-tourists¶ enrollment Gecisions )ourth, this study utilizes an artificial intelligence technique to predict behaviors (e.g., enrollment decisions) in the ed u-tour context. 1 .1 Problem Statement Online reviews such as eWOM and eReferral are becoming an increasingly important part of the digital marketplace and functions as facilitator for reputation. Scholars have reached a consensus that reviews assist firms to get noticed, increase sales, boost search engine results, convince clients to patronize a brand, firm, or business (Akile et al., 2021; Hu, & Yang, 2021; Jia, 2020). However, research in the field of educational tourism is scarce and hard to come by. Fan, Qiu, Jenkins and Lau (2020) also posited that cultural distance can motivate travel and interaction intentions between locals and tourists but acknowledged that perceived image of the destination may alter such nexus. Since online reviews have been shown to shape brand image, destination image 8 and trust (Abubakar & Ilkan, 2016; Abubakar et al., 2016). This research theorizes that online reviews can alter tourist ed-tourists destination image and trust which further shape perceived cultural distance. On the other hand, Mariani and Matarazzo (2020) argued that cultural distance between service provider and the tourist can influence online review ratings. The extant empirical findings highlight the presence of a bidirectional association, it is therefore imperative to analyze their collective effects on an outcome variable such as travel intention or intention to enroll in the context of educational tourism. With regard to familiarity, tourists with lower levels of familiarity depend on external information about a destination, primarily online reviews sources such as eWOM and eReferral. According to Chi, Huang and Nguyen (2020), high levels of familiarity create emotional attachment among tourists, rather than a sense of novelty and has been shown to shape intention to travel. Ying et al. (2020) also argued irrespective of context or type of familiarity, the resulting outcomes are always associated with motivation to engage in something. Artificial intelligence and linear modeling approaches were used in the present study to evaluate the probable relationships between eWOM and eReferral marketing strategies and how they influence edu-tourists' decision -making based on the social network model. Culture and cultural distance are seldom examined in educational tourism research due to their rarity. Besides, this research exploring the impact of the strength of ties in online communications, familiarity and cultural distance and their interaction, in order to determine individual decision behaviors. Therefore, this thesis is carried out to provide answers to the following questions: 1. 'oes )aceEook e:2M preGict eGucational tourists¶ enrollment Gecisions" 9 2. 'oes )aceEook e5eIerral preGict eGucational tourists¶ enrollment Gecisions" 3. Does familiarity preGict eGucational tourists¶ enrollment Gecisions" 4. 'oes cultural Gistance preGict eGucational tourists¶ enrollment Gecisions" 5. Are these associations linear or non-linear? Or both? 1 .2 Research Purpose and Rationale of Study 1.2.1 Purpose of the Dissertation This study adopts a network approach to unveil the possible interactions of eWOM and eReferral marketing strategies and how they influence edu-tourists¶ Gecision- making. In line with social network theory, this research investigates the impacts of eReferral, eWOM, familiarity, and cultural distance on behavioral outcomes, particularly in the setting of educational tourism. In this study, linear modelling is device to determine relevant predictors and machine learning for prediction. Combination of linear modeling and ANN has been addressed their deficiencies through allowing the two methods to complement each other. 1.2.2 Contribution and S ignificance of the Dissertation Existing empirical work has identified electronic referral (eReferral; Al -Htibat and Garanti, 2019), electronic word -of-mouth (eWOM; Ladhari and Michaud, 2015), familiarity (Mittendorf, 2018), and cultural distance (Dang and Nandakumar, 2017; McGladdery and Lubbe, 2017a) as determinants of purchase decisions, tourist engagement, and visit intentions in other contexts. The lack of empirical evidence in the edu-tour context warrants further investigations. This research aims to contribute to the knowledge in various ways. The first is to shed light on the eReferral (i.e., based on strong ties), eWO M (i.e., based on weak ties), cultural distance, and familiarity concepts by examining their effects on intention to enroll using social network theory. Second, conceptualizing eReferral, eWOM, familiarity, and cultural distance can 10 create an alternative and better medium to reach educational tourists (ed -tourists). Third, this study aims to provide answers regarding the relative importance of the aforementioned antecedents in predicting edu-tourists¶ enrollment Gecisions )ourth, this study utilizes an artificial intelligence technique to predict behaviors (e.g., enrollment decisions) in the ed u-tour context. 1 .3 Outline of the Dissertation The study is structured as the second section provides a thorough examination of the research factors in terms of theory and what has been researched. The section also briefly describes the work of previous researchers on the possible associations of eWOM, eReferral, familiarity and cultural distance on intention to enroll. In addition, the section illustrates the theoretical and hypothetical interaction of the proposed variables. Section three provides a description of the methodological approach employed in the research, and a brief explanation regarding the type of data analyses, approaches used, and why such methods were used for this study. Section four presents the research results and findings of the current empirical study. The chapter also provides detailed explanations for each hypothesis. The implications of the study for research and practice are discussed in the concluding chapter. Best practices and recommendations for practice are presented as well as methods and caveats of the present study and future research. 11 Chapter 2 2 LITERATURE REVIEW This chapter presents a theoretical background of the proposed research variables, a thorough literature review, and how the variables might interact with each other. This section also presents the research hypotheses, and how each hypothesis was developed. 2 .1 Educational Tourism There are many similarities between Cyprus and other small island states in the Mediterranean Sea, such as its geographical size and political isolation from the mainland. North Cyprus was forced to prioritize the services industry due to the country's political isolation and economic embargoes in virtua lly every field (Arici et al., 2014). The Turkish Republic of Northern Cyprus (TRNC) was formed in 1983 on a split island and is unrecognized by all countries except mainland Turkey. North Cyprus is strategically positioned in the Eastern Mediterranean. Be cause of its political non-recognition, the TRNC has no external trade connections with nations other than Turkey. Hence, international tourism and the growth of the higher education sector are two important sources of foreign exchange for this small island. However, the tourism industry has a hard time luring foreign visitors because of issues, including the absence of direct flights to North Cyprus and expensive transportation expenses. By the 1990s, the demand for higher education in North Cyprus had inc reased significantly, owing mostly to students from Turkey and foreign advertising, particularly in Africa and the Middle East. Since 1982, international students have started traveling to North Cyprus for higher education. Since then, the number of international students has steadily 12 increased. Thus, higher education is currently a major industry in North Cyprus, generating significant foreign exchange and helping to the development of this small and unrecognized island state (Katircioglu, 2010). Famagusta city gets the bulk of educational tourists and is home to Eastern Mediterranean University, North Cyprus's oldest and largest higher education institution (Katircioglu, 2014). In their research on the variables that drove Northern Cyprus students to trave l for educational purposes, Abubakar et al. (2014) argued that the quality of life in the host community is a motivating factor for students to study abroad. In addition, natural beauty, safety, school quality, and employment possibilities are all significant motivators. People become tourists as they voluntarily depart from their present environment to see another. Regardless of how close or far this destination is, these people will usually engage in a variety of activities (Camilleri, 2018). As a result, tourists are visitors, and what they do while visiting another location may be considered tourism. The tourism industry agreed to use the term "visitors" (as opposed to "residents") to describe people who travel to another country. This definition encompasses two types of visitors: Tourists were defined as visitors who stayed in a location for at least 24 hours. Their visit could be classified as leisure if they are traveling for recreation, health, sport, vacation, study, or religious activities (Camiller i, 2018). In both developed and developing countries, tourism is one of the fastest-growing industries in the world. Tourism growth was fueled by leisure growth combined with the rise in authority and the desire to escape and enjoy holidays at home and int ernationally. Tourism definitions differ in terms of whether the term is supply-side (industry) or demand -side (consumer). 13 Educational facilities, such as universities and other institutions of higher learning around the globe, are beginning to offer undergraduate degrees in travel and tourism. This higher education or vocational institution will supplement the already well- established higher national diplomas, postgraduate diplomas, and master's degrees in tourism and hospitality management. Due to the popularity of degree-level tourism programs, the number of courses offered has significantly increased over the last few years. Among the top universities in hospitality and tourism management are the Hong Kong Polytechnic University, the University of Central Florida, and the Eastern Mediterranean University (ARWU 2021). In recent decades, the growth of both education and tourism as industries has resulted in increased recognition of these industries from both an economic and social point of view. According to 5itchie  , ³as countries Eecome more interGepenGent, their success, growth, and economic prosperity will largely depend on the ability of two industries ± education and tourism ± to create the possibilities required to support international transfer anG learning´ As a result of changes in the tourism industry, tourism and education have converged in which education enables mobility and learning to become a vital part of the tourist experience. According to Ritchie (2003), the association between education and tourism needs to be empirically investigated. It is vital to acknowledge that tourism and education are "difficult bedfellows" to understand educational tourism (Pitman et al., 2010). Opponents generally agree that it entails traveling outside with the major or minor aim of studying in a new setting (Richards, 2011; Stoner et al., 2014 ). Furthermore, conceptions of educational tourism differ. The majority of the research makes technical recommendations to address educational tourism regarding the industrial sectors or segments it encompasses or 14 ignores. Educational tourism is rising as a result of cultural tourism disruption, according to Magrizos et al. (2021), and it differs from volunteer tourism, language tourism, and innovative tourism. From an anthropological perspective, Ritchie's (2003) segmentation model of educational tourism examines motivating factors that influence visitors of different ages. Accordingly, the desire to learn attributes to educational tourism. According to Ritchie's app roach, learning can be a major or secondary purpose for travel (2003, p. 14); it can take place in an official capacity (with the help of an expert or guide), or it can take place informally (p. 11). Figure 3 shows his conceptual framework for education tourism's major components . Ritchie's (2003) approach has been widely recognized worldwide as a modeling approach for education tourism during the past decade. However, using a market segment strategy to define educational tourism drives the risk of eliminating industry sectors in which learning is a significant activity. Pitman et al. (2010) suggest a process framework for defining educational tourism to avoid this. Richards (2011) proposes the idea of educational tourism as a transformational experience. They also suggest that educational travel be viewed as an opportunity to learn about international citizenship. 15 Figure 3: Ritchie's educational tourism segmentation model Ritchie's model has some shortcomings. It does, however, help to conceptualize educational tourism as a specialized industry. It is challenging to adopt a motivating dichotomy of 'tourist first' or 'education first.' 7he term µeGucational tourism¶ refers to "purposeful and deliberate experiential learning" (Pitman et al., 2010, p. 221). If the notion of the learning process, which is the learning of new skills and expertise, is accepted, several other tourism industries could be covered by education tourism. The sharing of knowledge and abilities between a visitor and a host is defined as "creative tourism" by Richards (2011), even though he specifically excluded them from his definition (p. 35). 16 Today's educational tourists are more knowledgeable, academically orientated, have higher purchasing power, and are more environmentally and culturally conscious ( Slocum et al., 2019 ). According to Richards (2011), e ducational tourists are seeking new, genuine experiences rather than mass-produced ones and are particularly wary of cultural commercialization (Lyons et al., 2012). An additional criterion for an adequate educational tourism program has been added by Van't Klooster (2014), who agrees with international administrators. It must be distinct from the tourists' everyday experiences for educational travel to be fruitful. Most people think of educational tourism as a phenomenon that occurs at the higher level of education. Formal academic credit-bearing programs for educational tourism are examples of informal self-development excursions. Learners may utilize cognitive, emotional, and behavioral outcomes to assess their progress via experiential learning. Learning about the global environment may be recognized as a principal objective of educational tourism that takes place in a cultural or environmental setting when the tourist is unfamiliar with; or when it has an international travel component. Under these circumstances, it is possible to develop global knowledge (cognitive results), universal mindedness (affective results), and global competence (behavioral results) . The "compassion gap" ca n be traversed by educational tourism. Intercultural or international educational tourism may have global learning as an aim. Authenticity, the commoditization of traditional conventions, and the conservation of cultures are some of the challenges that educational tourism may help address. 17 2 .2 Social Network Traditionally, a social network was defined as a specific set of connections between a defined set of individuals, groups, and business entities (Tichy et al., 1979). The characteristics of these connections could be used to interpret the individuals' social behavior, or shared values, visions, ideas, social contacts, financial or commercial exchanges (Serrat, 2017). A social network is determined by a number of individuals and organizations or other social institutions connected by a series of socially meaningful connections, like friendship, co-working, or the sharing of information 2¶Murchu et al,   6uEseTuent consumer searches anG Euying Gecisions haYe become significantly popular on social media websites (Lee & Koo, 2015). Online information-sharing systems, such as social networking sites, are referred to as "social media" (e.g., Facebook). An ongoing revolution is a social network in which members are identified as the audiences. A social network, according to Figure 4, can be seen as a "web" that surrounds individuals with direct or indirect social relations (illustrating the linkage between two circles) (illustrated by circles). For example, person A is directly connected to person C and may create indirect connections with people D, E, and F via person C. Person B, however, has five direct connections with others that lead to more indirect relationships between its social networks. Interactions between participants that can be direct or indirect promote social integration and allow them to form social networks with their peers. In addition, social networks may differ concerning their variability and size (Garton et al., 1997). It is common for traditional labor teams and rural communities to have 18 small, homogenous networks. Higher online communities have more variety in their members' social traits as well as more complexity in their structure . Residents are the participants in conventional social networks (e.g., family, friends, and relatives, people in face-to-face communities (such as those we encounter at the workplace or during group activities). In most cases, relationships are formed between people who live in the community, and services are provided to help the locals improve their social and economic conditions. As shown in Figure 5, conventional social networking sites are often modest in size, show homogeneity in terms of members and network nature, are constrained by the physical availability of representatives, and are separated from other networks. Figure 4: A social network is a web of relations (Source: Garton et al., 1997 ) 19 Figure 5: Traditional social network (Source: Garton et al., 1997 ) According to Kimball and Rheingold (2000), as technology develops, "social networks evolve as a result of human interactions through time, as well as the technical infrastructure that connects those individuals," evolving into online social networks generated by networked computers. Through the convergence of video conferencing, real-time interaction tools, SMS, and shared online workspaces, formerly in-person connections between persons are now conducted online. These digital environments help to boost inventiveness by allowing people to interact and create socioeconomic powers at higher levels. Social networks are also crucial in commercial activities and economic development. Users may exchange information on the wide variety of items provided, charges or pricing, and service quality using computer-mediated social networks to improve the reputation of a company's website and the services it gives (Hogg & Adamic, 2004). Global Trendyol users, for example, purchase things, debate products, share best practices and interests, and receive support and feedback from others. They also promote items that help them build their reputations in this online community. Through the increased interactions between people and small company 20 units that are loosely connected, this sort of virtual community boosts economic activity. 2 .3 eReferral As the profit-motivating eWOM proliferates, people tend to rely on trusted information sources or eWOMs, commonly known as e-referrals (Al -Htibat et al.,2019). What is missing is the interactive eReferral forming of the involvement of tourists, the intention to visit, and the sharing of e-referrals. eReferral is sent between friends, relatives, and social partners. Interactivity within the framework of the study is to what degree eReferral nature and contents promote cooperation between educational tourists (e.g., sound, videos, photos, emojis, and textual fonts). It is thus necessary to complement conventional electronic referral with interactive eReferral, which contains various videos, audio files, and pictures, emojis, and text fonts. According to the theory of social tie, connected associates become interested in and more likely to imitate each other's operations and decisions. In the area of travel information sharing and recovery, e-Referral takes attention (Jeong and Jang, 2011). In a hospitality context, the positive effect of eWOM, e-referral, and other types of on tourists' decision -making and destination choice has been confirmed (Ladhari & Michaud, 2015). Relevantly, research findings have shown that e -Referrals make up the vast majority of sales volume (Nielsen, 2012). When it comes to electronic referral (eReferral), the two are frequently conflated. eReferral is distinct from the other two ideas because it occurs among people with significant social connections. Affiliate marketing or referral marketing is one of the least developed aspects of internet marketing. Reciprocal referrals and client referrals 21 are the two aspects of eReferral in terms of technicality. Whenever two or more businesses agree to cross-refer clients, this is known as a reciprocal referral (Abubakar et al., 2016). In general, consumers prefer to seek out reliable information through eWOM (electronic word -of-mouth) (eReferral). A referral is "any good or negative remark made by a friend about a product or business, and made accessible to friends, family, coworkers, and others through the Internet" (Abubakar and Ilkan, 2016). Tourist engagement, visit intention, and eReferral sharing behavior are not well documented in hospitality marketing literature ( Al-Htibat, A., & Garanti, Z., 2019 ) . 2 .4 eWOM An influential and useful source of information for customers, electronic word of mouth (e -WOM) continues to dominate online marketing. It has been shown to be the main determining factor in approximately 20 to 50 percent of all buying choices (Bughin, Doogan, & Vetvik, 2010) . As argued by Bulbul, Gross, Shin and Katz (2014) , eWOM is ranked as the top source of trustworthy information and most influential factors by customers in their purchase of brands. One of the most important forms of eWOM in tourism and hospitality industry is online hotel reviews. 81% of the respondents to Boykin's ( 2015) survey said that internet user reviews were a significant source of information when selecting hotels, while 49% said that they only book hotels after reading online feedback prior to their judgment. Online reviews are perceived to be substantially credible which was confirmed by Anderson's (2014) research, where eighty eight percent of the respondents who are travelers said they trust online reviews as much as recommendations from family 22 members or friends. The term "prosumer" evolved as a result of the development of eWOM, which demonstrates how influential consumers have become as a result of their capacity to simply, rapidly, and freely share their experiences with goods or services with a big group of individuals (Grinevich, 2017) . The majority of the material on review websites like Yelp, Google Reviews, TripAdvisor, and social networking sites, namely Instagram, Twitter, and Facebook, originates from user submissions, making them producer-consumers (prosumers) or co -creators. Through sharing of experiences and co-creation of contents, Siuda and Troszynski (2017) believe that for marketers, prosumers are becoming the essential part of brand- imaging process. Social communities and review platforms have given prosumers easy access to discover information as well as easily share information. The same individuals who post reviews of their experiences in the morning, acting as prosumers can at night reaG other¶s reYieZ Ior the purpose oI choosing a hotel, restaurant or some other services, thereby becoming consumers. According to Ladhari and Michaud (2015) , in the hotel industry, purchasing decision is perceived to be high risk because in the decision-making process, the reference group evaluation is important which further increases the impact that prosumers have on sharing eWOM. According to Jalilvand, Esfahani, and Samiei (2010), traditionally, word-of-mouth (WOM) refers to the interchange of communication between people about a specific product or service. WOM communication irrefutably has influence on both consumers¶ Gecision anG purchasing EehaYior a proGuct or serYice can Geter or attract consumers depending on the friendliness of the conversation. For hotel managers, word-of-mouth communication and reviews are not new concepts, however, the rise in the eWOM communication as well as its influence on the purchasing power of 23 consumers have motivated managers to put their focus on review-generating initiatives and reputation management. Typically, WOM communication occurs on a one-on-one conversational basis, however, due to the capability and accessibility of online communities and internet, the concept has been entirely flipped and has made eWOM communication more important. Jalilvand et al. (2010) describe eWOM as any positive or informative review customers submit online about a good or service, allowing easy access to a wide variety of prospective or existing consumers. Marketing practitioners and researchers haYe GiscoYereG that aEout tZenty IiYe percent oI consumers¶ reYieZs are negative or critical, revealing that eWOM can not only be advantageous but disadvantageous too (Sotiriadis & Van Zyl, 2013) . Even though positive word-of-mouth can positively and significantly affect consumers¶ purchasing anG Gecision-making decision, research also has it that negative word-of-mouth reviews can have a greater impact and influence on consumers¶ EehaYior anG attituGe (Cheng & Ho, 2015) which is due to the fact that customers who are dissatisfied are often more aggressive when communicating their experiences, that is they seek to inform more people about their purchases or experiences than those satisfied. According to Breazeale (2009), customers who look to other consumers for information about products and brands are more likely to pay heed to negative comments and reviews than good remarks or reviews. )rom the customary marketing knoZleGge, consumers¶ negatiYe messages anG complaints are important and research has it that when there is complaint from consumers, it indicates that a relationship exist between the consumers and the brand 24 and are concerned about the situation. All forms of communication from customers, whether positive or negative is a means for the organization to receive ideas and feedback on what should be changed or improved in the company¶s operations, products or services. For companies that can handle consumer complaints, Meik, Brock, and Blut (2014) view them as a valuable source of information and knowledge . 'ue to the increase in internet usage, customers¶ Ieelings are easily expressed online; Positive user-generated content, such as social media posts and reviews, might detrimentally affect a brand's image. It has been noted by researchers that there are types of negative customer-generated content especially in the social media context that organizations to professionally manage because of the likelihood of it being shared. Berger and Milkman (2012) argued that it is important for marketers to address the angry or unpleasant experiences of customers because the customers often transmit their feeling of disappointment or anxiety to potential customers. Several researches such as Cantallops and Salvi (2014); Cro tts, Mason and Davis (2009); Di Pietro, Di Virgilio and Pantano (2012) have extensively published and reviewed the importance and significance of eWOM marketing for organizational success. ninety percent of consumers, according to a study by Cheung and Thadani (2012), utilize eWOM, such as blogs and online reviews, to obtain knowledge about new goods; however, King, Racherla and Bush (2014) believed that the relationship will be stronger in the hotel context. There has been several research on eWOM communication in relation to satisfaction (Barreda & Bilgihan, 2013; Loureiro & Kastenholz, 2011) , role of gender (Memarzadeh, Blum, & Adams, 201 5; L. B. Sun & Qu, 2011) as well as analyzing failures in services 6inchez-Garcta Curris-3prez, 2011; Swanson & Hsu, 2009) . 25 There have been limited studies on the variables that drive customers to write eWOM reviews, as reported by Cantallops and Salvi (2 014a). Yoo, Sanders, and Moon (2013) investigated the frequency of eWOM on social media (Facebook) concerning the tone of the message and the personality of the reviewer. Moreover, Chen et al. (2013) investigated consumer participation in eWOM review-writing and the effect on customer loyalty. A research by Yoo and Gretzel (2008) assessing 1,200 respondents using TripAdvisor travel platform was carried out to evaluate factors that motivates consumers to write eWOM reviews. The research was conducted to not just identify the motiYes Ior consumers¶ Zriting online reYieZs Eut also assess these motiYes Zith respect to their demographic differences. Findings from their research revealed that eWOM communication, unlike the traditional WOM does not consist of exchange of conversation but entails reviews posted anonymously without the reviewer expecting a conversation or response in return; a lso discovered was the list of motivators which includes self-enhancement, enjoyment, expressing positive feelings, power, helping the company, venting negative feelings and concern for other consumers (Yoo & Gretzel, 2008) . Consumers' buying choices are undoubtedly influenced by WOM since they tend to consult relatives or friends whose views they trust. Researchers started studying the perception of trust offline and online as soon as online review became common (Urban, Amyx, & Lorenzon, 2009) . Fine, Gironda and Petrescu (2017, p. 6) Tuoting Glen mentioneG that, ³Trust often connotes trustworthiness, integrity, dependability, confidence, and kindness in the conventional [offline] meaning ´ also research shoZs that same characteristics associated with trust offline is associated with that of trust online (Urban et al., 2009) . 26 Several researches discovered that consumers perceive online review as more trustZorthy that organizations¶ aGYertisement aimed at mass marketing their products (Cheung & Thadani, 2012) . According to Pan and Chiou (2011) , since reading of online reYieZ has Eecome an important Iactor inIluencing consumers¶ purchasing decision, especially for purchases that are high-risk like hotel stays, destinations and flights, researches have studied specific clues in online review that can generate consumer trust like negative or positive reviews, review rating and connection to poster. In online reviews, when organizations identify the general level of trust by consumers, it can help them have a better understanding of factors that affects the frequencies in which consumers write reviews; however, due to self -reported behavior issues, trust was eliminated from Fine et al.'s (2017) research. In order to extend the research on eWOM, several scholars have used Hennig-Thurau, Gwinner, Walsh and Gremler's (2004) research as a foundation for that. A research by Rensink (2013) rooted in Web 2.0 was carried out to identify why consumers switch from consumer to prosumer, motivations on their participation in writing negative and/or positive online review and the influence of personality. Drawing on Hennig-Thurau et al.'s (2004) seven motivational factors, Rensink (2013) added some factors that motivate negative and positive tradition WOM communication and came up with other lists which includes helping the company, self- enhancement, Yenting negatiYe Ieelings, social EeneIits, concern Ior other consumers¶, warning other consumers and advice seeking. Rensink ( 2013) constructed hypotheses for the five personal traits: openness, conscientiousness, extraversion, agreeableness, and neuroticism to identify the moderating effect of personality characteristics on 27 eWOM exchange. The motivation for user-generated content were the independent variables, involvement was the dependent variable while personality and positive/negative user -generated content were used as the moderating variables. )inGings reYealeG that µsocial EeneIits¶ motiYation is the only motiYation that is applicable for consumers to be more involved in creating online reviews; also, the study discovered that several differences exist between motivations in creating negative and positive online reviews. Also, the personality traits have an insignificant efIect on consumes¶ motiYation to post online reYieZ Rensink, 2013) . Unlike traditional WOM, eWOM recommendations are typically generated by anonymous users in a text-based format. Because of this anonymity, consumers have difficulty determining the veracity of the material and are thus more circumspect (Chatterjee, 2001). Previous eWOM studies overlooked the degree to which social connections develop inside the domain of social networks and the impact they have on a consumer's choice to purchase a pr oduct or service. Indeed, among the most apparent features of online communications is that they are frequently one-way. The majority of online consumers are classified as "lurkers," who read information and reviews but infrequently or never take part (Kim et al., 2018). The study on the social impacts of eWOM communication is sparse, even though social connections influence consumer behavior and decisions (Granovetter, 1983) . Online word-of-mouth, or eWOM, has become more prevalent since the advent of the internet. In the online environment, eWOM is defined as any good or negative comment made about a product or firm by future, existing, or past consumers open to a large number of people and organizations over the web (Hennig -Thurau et al., 2004, p. 39). Be cause they are founded on group commonalities, the ideas disseminated via 28 the network via eWOM are viewed as remarkably credible by users (Fatma et al., 2020). Reviews of a company's products and services posted on an online platform (e.g., company website , third-party retailer, online community or forum, search engine, or social media platform) by a person who claims to have used or purchased the product or service are examples of eWOM (Filieri et al., 2021). Informal customer conversations regarding a product or company are known as eWOM (Assaker et al., 2021). Electronic word -of-mouth (eWOM) communication is gaining increasing interest and attention in commercial disciplines such as marketing strategies, consumer behavior, economic management, and information systems (Chu et al., 2020) . 2 .5 Familiarity Consumer behavior is significantly influenced by familiarity. Familiarity refers to an individual's direct and indirect understanding of a service or product. It is a term that refers to an individual's evaluat ion of a brand or service as a result of personal experience, advertising, or WOM (Kaya et al., 2019). The familiarity of a website can influence consumers' views and purchasing decisions. Familiarity is a prerequisite of trust because it creates a framework and understanding of the environment and the trusted party within which the expectations of trust can be explicated (D. Gefen, 2000). Most customers need a certain level of familiarity with services to make them feel secure and comfortable. Accordingly, some writers believe that familiarity is gained not only by product usage (internal sources) but also through knowledge received from external sources such as advertising or word-of-mouth (Casalo et al., 2008) . Participants that were familiar with the website gained more knowledge about how to utilize the system, allowing them to traverse the website more easily than non- familiarized participants, according to Veldof and Beavers (2001). Consumers' 29 ultimate conclusions may be influenced by their familiarity with the website, such that individuals change their judgments based on their experience with website usage. Customer familiarity with a specific store affects how they absorb information (Blanco et al., 2010). Personal experience of a website or familiarity of a website refers to a consumer's knowledge of the website, such as knowledge of the website and the relevant procedures, such as the search for information. Familiarity with a website can influence consumer behavior on a familiar website. The client learns what information the website collects, how data is used, or how to control the information and its use; such knowledge is difficult to obtain on an unfamiliar website. According to research, customer familiarity with a website leads to a feeling of intimacy (Lee et al., 2011), and intimacy, which is a matter of privacy (Li, 2014), facilitates the self-disclosure of personal information on the consistent use of the website. Reynolds and Simintiras (2006) observed that because clients are hesitant to invest time, energy, and effort in researching web pages, familiarity may increase website loyalty. People who were born during the advent and spread of the digital and social revolution are referred to as Digital Natives. They are heavy Web 2.0 users, searching, accessing, consuming, purchasing, and producing massive amounts of data, goods, and services. As a result, they constitute a different generation endowed with unique digital inherent skills that have elevated them to the status of 'informal expe rts' in the fields of the Internet, digital culture, media platforms, and ICT (Line et al., 2011). People who are "digital natives" have natural intelligence and abilities in interacting wit h digital portable devices (e.g., tablets and smartphones ). In addition to being frequent users of 30 social media, they are likely familiar with Web 2.0 technologies, participate in online communities, and produce user-generated content (Gon et al., 2016). In their research on website familiarity, Qiu -Ying et al. (2012) h ypothesized that familiarity expands the area of future expectations and enables individuals to develop verified confidence in the anticipated future based on prior interactions. According to Qiu -Ying et al. (2012), consumer familiarity with shopping websi tes, such as Amazon.com, where consumers may inquire and buy books, investigate the transaction process, and the more consumers are familiar with the shopping website, the more likely it creates sound expectations, emotions, and behavior. As a result of the previous study, we define website familiarity as the degree to which consumers are familiar with the design and content of an online shopping website. The data in the chart shows the attitudes towards online shopping according to the results of the Statista Global Consumer Survey conducted in Turkey in 2020. About 72 percent of respondents said online comments were very helpful. Before purchasing a product, many online shoppers looked at what other customers had to say about it (Fig ure 6). 31 Figure 6: Attitudes towards online shopping in Turkey 2020 ( Source: https://www.statista.com/forecasts/1003017/attitudes -towards-online- shopping-in-turkey) As stated by Kaya et al. (2019) the website familiarity significantly and directly affects the evaluation of advertising and indirectly affects the quality of the website. Not only does familiarity refer to product use (internal sources), but also knowledge gained from ads and WOM. The term "website familiarity" relates to one's awareness of how a website works, what it presents, and what it admires. It represents the breadth of experience with the acquired goods, services, and consuming settings of a website. 2 .6 Cultural Distance According to tourism scholars, social interaction and cultural distance are crucial in forming knowledge about travel experiences, intergroup interactions, and tourist attitudes. Intergroup interactions can improve mutual understanding, reduce prejudice and stereotypes, and improve intergroup interactions (Fan et al., 2017). According to 32 Goeldner et al. (2012), the larger the cultural distance, the stronger the resistance. A better investigation of the connections between social interaction and cultural distance may substantially strengthen ties across areas, particularly those that are tense and resentful of one another. Personal ties and an understanding of different cultures may help overcome political divisions through tourism, which may connect people. As a result of these factors, tourism is recognized as one of the most effective means of promoting world peace (Goeldner et al., 2012). Goeldner et al. (2012) define culture as "a complex totality that encompasses knowledge, belief, art, morals, law, tradition, and any other capacities and habits acquired by man as a member of society." It affects how people choose, analyze, interprets, and utilize information. Cultural distance is a term used in tourism research to describe the degree to which the culture of the origin region differs from that of the host location (Goeldner et al., 2012). In addition to cultural differences, attitudes toward leisure (for visitors) and work (for hosts) may generate social obstacles in the interaction between the two groups; for example, effective communication, behavioral patterns, and quality of service norms, among others. According to Cohen (1972), the most significant elements to compare are the variations in cultural qualities between tourists and hosts, in tourism research. Because culture may be interpreted in several ways, different researchers provide diverse views of this topic in their research. According to Cohen (1972), the perceived cultural distance of visitors is connected to various variables, namely cuisine, privacy, cleanliness, social conduct, communication, and cultural values. Culture has a significant impact on how passengers interact throughout their trip planning process, which makes tourism among the most internationally integrated 33 industries in the world today (Qian et al., 2018). The tourism industry must place a greater emphasis on cross-cultural variables to create a positive destination image among Yisitors anG, as a result, attract more oI them Culture is GescriEeG as ³the configuration of learned behavior and results of behavior whose component aspects are shared and distributed by memEers oI a speciIic society´ /inton,   Culture affects every member's conduct and perception of another's behavior. Consumer decisions are influenced by culture. The majority of research indicates that culture is consistently one of the most influential factors in consumer decision-making (Solomon, 2004). People's decision -making processes are impacted by culture (Rokeach, 1973). Consumer behavior is influenced by cultural values (Farzin and Fattahi, 2018). Culture has long been thought to have a significant impact on human behavior, with the presumption that an individual's behavior reflects their cultural value system. Individual preferences and decision-making are thought to be influenced by culture (Farzin and Fattahi, 2018). The cultural dime nsion is when people in a nation prefer to act in an independent or individualistic rather than interdependent or collectivistic manner (Blut et al., 2015). Individualistic cultures, like masculine cultures, are expected to be more agentic, focusing on the positive, seeking risk, and having a functional orientation. Collectivistic cultures, like feminine cultures, are expected to be more communal, with an emphasis on loss prevention and an experiential orientation (He et al., 2008). Cultural distance quantifies the degree to which national cultures differ and are similar to the host culture (Crotts, 2004). According to Qian et al. (2018), the degree of cultural distance between the source and host areas may be quantified. Cultural differences 34 have been proposed as a probable explanation for how consumers and managers in different countries make different decisions (Tahir and Larimo, 2004). Jackson (2001) empirically examined such association on cultural aspects, with individuals from high- level collectivistic societies tending to select culturally different destinations. People's perceived risks were higher when they visited less familiar (or more culturally distant) destinations, according to Lepp and Gibson (2003), due to their ignorance of local languages, signs, and customs (Lepp & Gibson) (2003). 2 .7 Theoretical Framework Interactivity within the framework of the study is to what degree eReferral nature and contents promote cooperation between educational tourists. This dissertation fecundates the concept of eWOM, eReferral, familiarity, and cultural distance with social network theory to explore their influence on enrollment intention. 2.7.1 Social Network Theory The status of social network theory at the moment has been shaped by a diverse range of research traditions. Scott (1998 ) asserts that three fields of research aided in the initial development of the idea. First, there is the tradition of sociometric assessment that is founded on approaches from mathematical graph theory. Second, interpersonal interaction tradition is predicated on the development of cliques within a group of individuals. Finally, there is the tradition of anthropology that measures the structure of community interactions in a particular context. Innovations are spread between individuals or groups within a social system. The network of individuals responsible for originating, transmitting, and accepting innovations may be thought of as a social platform, with network connections consisting of friendship, communications, and social aid. Diffusion is fundamentally 35 a networked process. How quickly innovations are accepted may depend on the structure and features of a social-relationship network (Valente, 1996 ). Until the 1960s, these research traditions remained disjointed and did not form a cohesive theoretical framework. Numerous sociologists have contributed significantly to the social network technique by combining and building on previous academic concepts to grasp formal and informal social connections. The sociometric perspective on social networks, in fact, has been created, concentrating on structural aspects such as the relative position of specific network nodes. Additionally, inspectors enhanced social network approaches during this period by introducing block modeling and multidimensional scaling. The position of a node in a social network is taken into account while modeling blocks. It helps scholars to discover nodes with relative network locations or architecturally identical nodes. On the other hand, the scaling method enables researchers to translate social ties into sociometric distances, mapping them in a social space (Wasserman and Faust, 1994) . The degree of connectivity among a collection of nodes is measured by network cohesion. This metric has proven effective for detecting subgroups or cliques inside a broader social network over a lengthy period (Burt, 1987). Network cohesiveness is a fundamental structural feature in the field of media impact research because it serves as a moderator of interpersonal network influence. Among other aspects, Friedkin's (1993) longitudinal examination revealed an increase in personal impact in more cohesive social networks compared to less cohesive ones. Previously, social network theory provided three network centrality metrics: degree, betweenness, and closeness to distinguish the favorable view that opinion conductors 36 typically own (Freeman, 1979). The number of connections to and from a person in a network is defined by the extent of centrality. Greater centrality gives people more opportunities to learn and spread information, which increases their chances of becoming opinion leaders (see Figure 7, black node). Betweenness centrality is a measure of the frequency with which an individual node is on the shortest route connecting other nodes in the network. Network bridges with high centrality are more likely to connect fragmented clusters. Information may not reach other areas of a network of individuals with high centrality in the betweenness oppose its spread, much as gatekeepers prevent it from reaching other parts of a network. The node in Figure 7 that is light gray fills this critical position. It also estimates the average distance between a single node and the network as a whole. Less distance between the central person and everyone else in the network means that they can send information faster. Because they are good at reaching out to other people in their network, individuals with high closeness centrality are influential. There is a significant degree of proximity among the dark gray nodes in Figure 7. 37 Figure 7: 2pinion leaGers¶ netZork Zith a high Gegree oI centrality, pro[imity centrality, and betweenness centrality ( 6ource $GapteG Irom (Yerett¶s kite, in %ranGes anG HilGenEranG, ) Social network theory relates to social relationships via nodes (i.e., individuals) and ties (i.e., interaction between individuals). $ social netZork represents a ³set oI people, organizations, or other social entities, connected by a set of socially meaningful relationships´ ( Kim, Kandampully & Bilgihan, 2018, pg. 244 ). Individuals use social networks and social media for several reasons, spanning social, economic, psychological, and emotional gains (Granovetter, 1973, 1983). Kim et al. (2018) argued that social networking ideology is premised on the Iact that  ³social networks play a significant role in determining individual attributes and actions (e.g., by exposure to information and ideas), and (2) the network of relationships in which the individuals are embedded is more important in explaining behavior than are the intrinsic attriEutes oI the inGiYiGuals themselYes´ pg   38 Our study is based on social network theory and investigates the impacts of eReferral, eWOM, familiarity, and cultural distance on behavioral outcomes, particularly in the setting of educational tourism (Figure 8) . Luo et al. (2015) used social network analysis to investigate the communication features of travel-related eWOM on SNSs from both the ego and entire network perspectives. The findings reveal that travel- related eWOM communication via SNSs (Social Network Sites) relies on pre -existing social relationships, with links classified as strong, moderate, or weak. Gray et al. (2011) contended that taking a social network approach to understanding how social bookmarking systems assist individuals in crossing structural gaps to obtain further information might help explain why some workers are more inventive than others. Their research indicates that social bookmarking systems may aid in the lubrication of ideas as they travel between social environments. Casanueva et al. (2016) aim to offer Social Network Analysis methodologies used in tourism research that are based on reliable information about their current use using social network analysis. It then suggests future advancements in the same field. According to Chang (2021), social networks (intelligence, friendship, and counsel) influence three factors of tourist attitudes development (cognition, affection, and action tendency) . 2 .8 Hypotheses Development In light of the above discussions, four hypotheses are developed. H1: eReferral has a positive and significant effect on enrollment intentions. H2: eWOM has a positive and significant effect on enrollment intentions. H3: Familiarity has a positive and significant effect on enrollment intentions. H4: Cultural distance has a positive and significant effect on enrollment intentions. 39 2.8.1 eReferral, eWOM and Enrollment Intentions Online reviews are and will continue to be an essential source of information for tourists (Litvin et al., 2017; Wong et al., 2020). The concept of Referral is not well understood in the literature because scholars often interchangeably mix it with eWOM. The abstraction of each concept differs on two factors (i) strength of social ties and (ii) the onymous nature of eReferral (Abubakar et al., 2016). Strong social ties ensure a shared sense of social identity, which facilitates group-wise decisions. Information from eReferral is transmitted and shared among individuals (i.e., friends, family members, close associates) who share common social ties (Abubakar et al., 2016) instead of eWOM. Standing et al. (2016) found that strong ties were more efficient than weak ties, and friends' recommendations impact a person's purchase intentions. The internet is not the single information source utilized to select a location; in other words, inGiYiGuals Zithin one¶s social circle contriEute anG somehoZ YaliGate one's findings, for example. De la Hoz-Correa anG Muxoz-Leiva (2018) stated that ³Opinions and suggestions from key figures of reference such as friends, family members, colleagues, and tour operators affect the process of making choices towards medical tourism locations.´ pg  . Individuals share information because they want to help and guide their friends, acquaintances, and family members to select the right products, destinations, and services (Bilgihan et al., 2016). It has been demonstrated that eWOM has a positive impact on visitors' intent to travel (Abubakar and Ilkan, 2016; Filieri, 2015). Despite the ample evidence on the relationship between eWOM and consumer decisions, little is known in the edu-tour context. Additionally, the literature lacks knowledge on the simultaneous effects of eWOM and eReferral on potential consumers¶ intentions, anG ther are calls for additional studies in other contexts (Ladhari and Michaud, 2015; Erkan and Evans, 2016). Close associates, 40 family members, and friends are influential in decision making due to the shared strong social ties. In light of the above discussions, the following hypotheses are proposed: H1: eReferral has a positive and significant effect on enrollment intentions. H2: eWOM has a positive and significant effect on enrollment intentions. 2.8.2 Familiarity and Enrollment Intentions )amiliarity ³is an unGerstanding, often based on previous interactions, experiences, anG learning oI Zhat, Zhy, Zhere anG Zhen others Go Zhat they Go´ /uhmann,  in Gefen, 2000, p. 727). According to Mittendorf (2018), familiarity "means allowing for relatively safe future predictions by assuming asymmetric relationships between a system and its surroundings, so reducing risk exposure" (p. 379). It has a notable inIluence on consumers¶ Gecision-making processes, such as their intention to continue using a site (Jen -Hwa Hu et al., 2017), their platform engagement (Chen et al., 2011), destination image and travel intentions (Chen and Lin, 2012), and their online shopping behavior; familiarity also reduces disorientation (Suki and Suki, 2013). Tourists exert efforts (i.e., informa tion search or experience) to gain familiarity (De la Hoz-Correa anG Muxoz-Leiva, 2018), and familiarity with websites and social media platforms can facilitate the decision-making process )laYi n et al,   3rospectiYe tourists not only build on their experiences to gain familiarity, but also spend time and energy on searching for and acquiring useful information and sources. For example, familiarity with educational tourism promotional site or the focal university websites, social media outlets, and other online information sources would facilitate potential tourists to search, retrieve information, get in touch with the institution, like, comment, or share social media posts and probably develop enrollment intention. In the context of ed-tour, this study theorizes that familiarity with websites and social media pages 41 might increase stuGents¶ enrollment intentions ,n light oI the aEoYe Giscussions, the following hypothesis is proposed: H3: Familiarity has a positive and significant effect on enrollment intentions. 2.8.3 Cultural Distance and Enrollment Intentions Individuals' attitudes and behaviors reflect their exiting cultural values, norms and beliefs )arzin anG )attahi,   3erceiYeG cultural Gistance is the ³e[tent to Zhich people from one culture perceive people from other cultures to be different from them in terms oI their ethnicity, nationality, language, Yalues anG customs´ 6harma anG Wu, 2015, p. 3). In line of this, the current study theorizes that when perceived cultural distance is high, the level of interpersonal communication might be discouraged, which further limits interaction among individuals and related outcomes. On the other hand, the current study theorizes that a low perceived cultural distance would encourage interaction and its potential outcomes. Culture can influence consumers' purchase decisions or visit intentions (Dang et al., 2017) and acculturation orientations (Liu et al., 2018). Investigations unveil that cultural distance has a negative impact on destination choice (Y ang et al., 2016; Yang et al., 2018). This study argues that in the context of ed-tour, cultural distance is a motivating factor, as prospective students are more likely to enroll in a university at a destination with a different culture. Ed-tourists are novelty seekers (i.e., they aim to experience a new and different culture or learn a new language) as opposed to traditional tourists (Abubakar et al., 2014; Harazneh et al., 2018; McGladdery and Lubbe, 2017a). Calls for research on how culture influences visit intention have been issued in past work (Chen and Law, 2016; McGladdery and Lubbe, 2017b; Martin et al., 2017). In response to these calls and in light of the above discussions, the following hypothesis is proposed: 42 H4: Cultural distance has a positive and significant effect on enrollment intentions. Figure 8 : Research Model (Source: authors) eREFERRAL eWOM FAMILIARITY CULTURAL DISTANCE INTENTION TO ENROLL H1 H2 H3 H4 44 Chapter 3 3 RESEARCH METHODOLOGY As it is indicated in the title, this chapter includes the research methodology of the dissertation. In more details, in this part the author outlines the research strategy, the methodological approach and conceptual model employed in this study, philosophy of research and discusses the sampling method, plan, size, location of the study, data collection instruments, procedural and statistical analyses. 3 .1 Research Philosophy The Methodology Scheme in Figure 9 provides a clear and structured approach to ensure that you can identify each of the choices you make when selecting my research design for my thesis. The development of a research design starts with the location in a particular research paradigm of my proposed work. However, it should be noted that some data collection and analysis methods are not determined by certain paradigms. The capacity to recognize and justify the interloc king decisions as a research design is vital (Saunders et al., 20 07). 45 Figure 9 : Methodology Scheme ( Saunders et al., 2007) 46 3.1.1 Strategy ( S urvey) A popular strategy among hospitality and tourism researchers is the quantitative survey search approach. This research strategy is synonymous and closely related to deductive approach (Altinay et al., 2015). It is a process whereby a researcher selects a sample of informants from the general population. Consequently, a standardized questionnaire or structured survey is administered, and respondents or participants are expected to provide answers according to their perceptions or experiences. 3.1.2 Research Philosophy ( Positivism) Positivism advocates a more objective view of reality by relying on concrete facts from surveys. Positivism has always been connected with scientific study (Altinay et al., 2015). Positivists advocate quantitative techniques with high reliability and representativeness, such as social surveys, structured questionnaires, and government statistics. Sociologists in positivist research look for connections, or 'correlations,' between two or more variables. Positivism's core premise is that there is an external world that may be investigated and known. Positivists believe that nature is fundamentally organized and regular and that an objective reality exists irrespective of human observation a nd is waiting to be discovered. Much scientific work under the positivist paradigm is devoted to elucidating the fundamental causes of natural events (Persoon , 2010). Positivism is frequently offered as an inadequate starting point from which other and improved viewpoints may be explored that are more suited to the search issues and possibilities at hand. Specifically, scholars use positivism to advocate for a more nuanced understanding of science. This dissertation takes a deductive research approach. The researcher examines previous work, reads existing theories, and then evaluates ideas derived from those theories. 47 Whereas empiricists maintain that the only legitimate method of learning about the world is through observation, experiment, and experience, rationalism proponents believe that reason is the fundamental source of knowledge (Mukherji and Albon, 2018). In the opinion of rationalists, it is possible to obtain knowledge of a subject without having directly witnessed the phenomena in question through the acts of thinking and reasoning. Mukherji and Albon (2018) distinguish between two forms of reasoning: deductive and inductive reasoning .  As a result of prior information that is known to be accurate, deductive reasoning allows one to draw inferences about a subject. Consider the following statement (premise): 'Newborn infants cannot communicate.' One might accept that this statement (premise) is accurate: The fact that 'baby John can pronounce DaDa' suggests that he is not a newborn. It only works if the first premise is true.  As previously stated, inductive reasoning is the process through which one draws conclusions about something based on the likelihood that a claim is accurate in light of the events that have occurred earlier. Consider the following scenario: you've noticed that every cat you've seen has a furry coat, and you've come to the conclusion that all cats have furry coats. It is always possible that an exception to the rule will arise while using inductive reasoning, therefore disproving the hypothesis. However, this is extremely unlikely. The philosophy of positivism has its beginnings in the field of physical science that conducts research methodically and scientifically. Mukherji and Albon (2018) claim that positivism considers the cosmos as being ruled by immutable, universal laws and argues that knowledge of these universal principles can account for everything that 48 occurs around us. To grasp these universal laws, we must first observe and record events and phenomena in a systematic manner and then derive the underlying principle that 'caused' the event. The tale of Sir Isaac Newton and the apple is an illustration of this mechanism in action. Isaac Newton is said to have seen an apple fall straight to the ground while strolling through an apple tree. He began to develop the gravity theory after becoming curious about how high above the earth the force of gravity affected. The observable event in this scenario was a falling apple, and the underlying universal rule was gravity (Mukherji and Albon, 2018) . 3.1.3 Methodological Choice (Quantitative Method) A quantitative study was conducted to meet the dissertation's objectives. The main feature of quantitative research is that it provides a comprehensive description and analysis of a research topic without limiting the scope of the study to the nature of participants' responses (Collis and Hussey, 2003). We chose an appropriate method for data analysis. These approaches are broadly classified as deductive, and they are typically used to analyze quantitative data. The positivist worldview lends itself to the use of quantitative approaches due to its empirical, methodical approach to study. Investigators that employ a quantitative research methodology frequently (but not always) focus on the confirmatory stages of the research cycle, namely a hypothesis development and the collection of the numerical data to test it. In contrast to the qualitative approach, which focuses on expressing experiences, emphasizing the importance, and delving into the heart of an issue, the quantitative technique aims to quantify, measure, or find the size of phenomena. 49 3.1.4 Time Horizon ( Cross S ectional) According to Saunders et al. (2007), empirical research time horizons is a prerequisite for successful implementation of any given research methodology. Cross-sectional and longitudinal are the two popular research time horizons, in which cross sectional studies are associated with limited time span and/or pre -set time is created for data acquisition e.g., collecting information about the phenomenon of interest at a given time without interval. Cross-sectional study can be thought of as providing a snapshot of a phenomenon's behavior a nd characteristics at a specific time (Saunders et al., 2007). Longitudinal studies are repeated over an extended period, where researchers collect data within an interval, for instance, 1 week or month or more. For the present study, data will be collected from individuals studying at the university and at a time, thus this study can be classified as a cross-sectional study. 3 .2 Study Context Ed-tour is an increasingly popular trend in the tourism industry. It involves deliberate and explicit learning experiences that lead to the acquisition of knowledge and skills (Pitman et al., 2010; Ritchie et al., 2003). The phenomenon occurs in two ways: µtourism Iirst¶ or µeGucation Iirst¶ 5itchie,   7he Iormer is generally Ior inIormal education and learning purposes with touristic experiences as complementary byproducts and services, for example, excursions to amusement parks or zoos, or camping. The latter is generally for formal education and learning purposes, such as professional certification, language, exchange, undergraduate, and graduate programs. Touristic experiences are often secondary to educational activities (Ritchie, 2003). Ed-tourists "are individuals or groups who travel to and stay in places outside their usual environment for more than 24 hours and not more than one year for study, 50 business, leisure, and other activities" (Abubakar et al., 2014, p. 58). According to McGladdery and Lubbe (2017b), the achievement of intercultural competence motivates tourists to participate in edu-tour. This is because multicultural or cultural intelligence contributes to personal and career development. Other motivators and determinants for participation include English language programs, the quality of teaching, the reputation of the destination and its institutions, and the desire to encounter a new culture and experiences (Abubakar et al., 2014; Harazneh et al., 2018). Political, safety, and discrimination concerns can function as demotivators (Harazneh et al., 2018; Nagai and Kashiwagi, 2018). Online reviews from past and present students, social media promotions, opinion leaderships, and influencers are substantial predictors of enrollment (Peruta and Shields, 2018). Students are not experts but are perceived as trustworthy ambassadors in the ed-tour context (G ymez et al,   eWOM is described as "any constructive or negative feedback about services and products provided by current or past customers and made available to other customers over the internet" (Hennig -Thurau et al., 2004, p. 39). eWOMs are review s from different people with limited social ties, Zith the possiEility oI proIit motiYations, Ior e[ample, Iirms¶ sponsoreG posts that impersonate consumers. eReferrals are reviews from people with strong social ties. eWOM is stronger than eReferral because the coverage area of the message transmitted through weak ties is limitless (Granovetter, 1982). Social media sites entail groups and a community of friends who share some level of ties, such as family members, friends, and acquaintances (Abubakar et al. , 2016). This internal structure and network aids stuGents¶ Gecision-making through increased access to knowledge from experienced peers (Granovetter, 1982). 51 3 .3 Data Collection 3.3.1 Pilot Study A pilot test is a vital stage in any research project. It is a small study used to evaluate research procedures, data collection tools, sample recruitment tactics, and other research methodologies in advance of conducting a huger study. A pilot study is a critical step in any research endeavor since it identifies possible issue areas (Hassan et al., 2006) . Pilot studies are a critical stage in the research process. A pilot study was used to determine the viability of a technique that will be utilized in larger-sized research. Pilot studies are small-scale, exploratory studies that are used to examine the feasibility of critical components of a more comprehensive investigation. Pilot studies are frequently used to refer to a quantitative technique for evaluating a research instrument. Conducting pilot research does not ensure that the main study will be successful, but it does improve the probability (Conelly, 2018) . In our research a small-case version of the complete survey was tested with 30 students. The pilot also tested whether the questionnaire was comprehensible and appropriate, and that the questions were well defined, clearly understood and presented in a consistent manner. We observed that there is no issues for changes. Our pilot study has demonstrated that the study protocol is feasible. 3.3.2 Sample and Procedure It is almost impossible for scholars to collect data from all cases and the propensity that all cases will responds or answer the study questions is quite low. This primary due to constraints such as financial, access, timing, scope, and reach. Overtime social 52 science scholars devised a feasible way to collect responses and/or access a social phenomenon using a sample (a sub -set of population) . The population refers to the total group of instances from which the researcher draws a sample. Because researchers lack the time and resources necessary to analyze the whole population, they employ sampling techniques to minimize the number of instances. Figure 10 illustrates the stages that are likely to go through when conducting sampling. Figure 10: Sampling Process Steps ( Taherdoost, 2016) 53 The sampling procedure begins with a well-defined target population. The term "population" is frequently used to refer to the whole people who live in a specific nation as the name suggests, the sample size is a set of real-world examples from which a sample will be taken. The target population must be representative of the population as a whole. For starters, it's vital to understand what sampling includes and why researchers would pick a particular sample. The process of picking a subset from a chosen sample size or the entire population is known as sampling. For example, a sample might be used to establish generalizations about a population or to test pre- existing ideas. Basically , this is dependent on the sampling strategy chosen. Overall, there are two major sampling techniques:  Probability or random sampling  Non - probability or non- random sampling A wide sampling method must be selected before settling on a particular kind of sample approach. The Figure 11 illustrates the various sampling strategies (Taherdoost, 2016) . 54 Figure 11: Sampling Strategies ( Taherdoost, 2016) The term "probability sampling" refers to the fact that every item in the population has an equal chance of being included in the sample. Probability sampling technique are free of classification error, sampling bias, and estimation errors.  Simple random sampling is a random sample that indicates the potency of every instance in a given population having an equal chance of being included in the sample.  Systematic sampling is a technique in which every nth example following a random start is chosen.  Stratified random sampling divides the population into strata (or subgroups) and draws a random sample from each subgroup.  Cluster random sampling is a technique in which the entire population is split into clusters or groups. Then a sample drawn from the clusters or some of the clusters depending on intent or strategy. 55  Multi-stage random sampling is the process of shifting from a large to a focused sample in a step-by-step manner. Non -probability sampling is a valuable sampling approach that may be utilized in quantitative, qualitative, and mixed methods quantitative research. Non -probability sampling is a sampling strategy in which the researchers choose samples based on their subjective assessment rather than random selection as oppose to probability sampling technique (Taherdoost, 2016) .  Quota sampling is a non -random sampling approach in which participants are chosen based on predefined qualities; the entire sample has the same distribution of characteristics as the larger population.  Snowball sampling is a non-random sampling approach that employs a few instances to persuade additional cases to participate in the research, expanding the sample size.  Convenience sampling is the practice of selecting participants based on their availability and ease of access. For instance, surveying every other person that enters a caIp Zhere the researcher is sitting haYing hisher meal  Purposive or judgmental sampling is a method in which specific locations, individuals, or events are chosen purposefully to provide critical information that cannot be acquired by other choices. It occurs when the researcher includes cases or participants in the sample solely based on their belief that they should be included. Sampling bias or sample selection bias leads to errors that appear in research studies when the researchers do not accurately select their participants. A random sample 56 needs to be of sufficient size to reduce the risk of bias in sampling. After establishing the target population, sampling frame, sampling procedure, and sample size, the next step is to gather data. According to Churchill (1996 ), in the MuGgmental sampling techniTue ³the sample items are selected on the basis that they are considered to represent the specified target population" (p. 582). According to the information obtained from the ministry of education, there are 102,944 students (local inclusive) studying in Northern Cyprus. The total number of population size is , anG Ze are comIortaEle Zith a “ margin of error. We can see we will need at least 383 (it was obtained using survey monkey sample size calculator) student to take our survey. The informed consent form and questionnaires were sent to the University Ethics Committee and necessary permissions were obtained prior to data collection. Permission to conduct the survey was issued by the management of the universities. Paper-based survey were distributed by the researcher to the related instructor and were collected within two days by the researcher. Using a judgment sampling technique, 1200 survey packets were dispatched to undergraduate and graduate students studying in Northern Cyprus. The purpose of the research was explained to the participants, and they were told that there were no right or wrong answers and that they could quit at any time. The researchers assured them of information confidentiality. Proximity often leads to consistency in responses to nearby items and variables (Weijters et al. , 2009). To overcome this problem, we placed the construct items on separate pages to make them appear unrelated. This procedure was done to evade the potential threats of social desirability and the common method bias (Podsakoff et al., 2012). At the end of the survey, 972 packets were returned, yielding an 81% response rate. 41 packets had missing 57 information and were therefore discarded from the study, leaving behind 931 valid forms. 3 .4 Study Measures Abubakar et al. (2016) created 4 -items to measure eReferral. "I frequently evaluate institutions recommended by my friends, colleagues, and family," for example. The mean (M) was 3.63, and the standard deviation (SD) was 0.77. Abubakar et al. (2017) used a 6 -item questionnaire to assess electronic word-of-mouth (eWOM). For instance, "I frequently read other students' online reviews to ensure I enroll in a good institution." The mean is 3.63, and the standard deviation is.71. Gefen's (2000) 4 -item was used to assess familiarity. "I am familiar with the processe s of searching for and obtaining information about institutions online," for example. The mean is 3.36, and the standard deviation is.89. Sharma et al. (2015) used a 6-item scale to assess cultural distance. For instance, "customs are very different from m ine." The mean is 3.80, and the standard deviation is.74. Shukla's (2010) 3-item questionnaire was used to assess students' intentions to enroll. For example, "I enroll in this institution rather than any other available institution." The mean is 3.85, and the standard deviation is.74. For all the measures student rated their response options on a 5-point Likert scale spanning from 5 (strongly agree) to 1 (strongly disagree). Demographic data includes gender, age, marital status, enrolled program, class and country. 58 eReferral was measured with 4 items utilized by Abubakar et al. (2016). eWOM was measured with 6 items utilized by Abubakar et al. (2017). Familiarity was measured with 4 items utilized by Gefen (2000). Cultural distance was measured with 6 items utilizeG Ey 6harma et al   6tuGents¶ enrollment intentions Zere measureG Zith 3 items utilized by Shukla (2010). The measures were rated on a 5 -point Likert scale spanning from 5 (strongly agree) to 1 (strongly disagree). The list of items is given in Appendix A. The demographic data include gender, age, marital status, enrolled program, year/class of study, and country of origin. 3 .5 Analytical Approaches This dissertation employs a multi-method approach namely linear and asymmetric techniques. At the first instance, the linear method was deployed to identify the importance of the examined predictors, while the asymmetric a kind of method machine learning method ( Artificial neural network (ANN) ) was deployed for predictive purpose. Researcher argued that ANN i s an integral component of artificial intelligence and machine learning algorithm and has been championed to be smarter compare to traditional canonical correlation, regression and structural equation modeling analyses (Abubakar et al., 2019). ANN s are useful for accurate and precise decision-making and are less restrictive than multivariate assumptions as homoscedasticity, linearity, and normality are not enforced (Abubakar et al., 2019; Sharifi et al., 2019). Despite these features and capabiliti es, due to its black-box operational nature, ANN is not completely exemplary for evaluating causations. Linear modeling has the tendency to oversimplify complexities, which in turn harms decision- making processes $EuEakar,  Gnther & Fritsch, 2010). This thesis amalgamates linear and ANN techniques to address their deficiencies through allowing the two methods to complement each other. For linear and ANN modeling R program 59 version 1.0.136 was deployed, utilizing linear and relaimpo packages and neuralnet package. The analytical steps followed are illustrated using a flowchart in Appendix B and the R codes are reported in Appendix C. 60 Chapter 4 4 RESULTS OF STUDY This section of the thesis provides a thorough summary of the empirical findings from the analysis of the results conducted on data obtained from different universities. More precisely, the chapter includes a detailed analysis of the respondents' demographic data as well as the outcomes of the hypothesis testing. 4 .1 Demographic Breakdown The present study garnered a total of 931 respon ses from the participants. The demographic breakdown of the participants is as follows: male participants make-up 64 percent and the rest comprise of females. Regarding marital status, single respondents are the majo rity summing up to (96 percent) and the re maining 4 percent of the respondents were married. Age wise, about 35 percent of the participants were in the 18 ±20 age range, 62 percent were in the 21±30 age range, and the rest were older than 30 years. 77 percent oI the responGents Zere enrolleG in Eachelor¶s Gegree programs, 12 percent in associate degree programs, and the rest in postgraduate degree programs. Majority of the participants (39 percent) are freshmen (first year students), 34 percent are sophomores ( second year students), 15 percent are junior s ( third year) and the remaining are seniors (fourth year students). Consequently, most of the respondents (61.44 Percent) are from mainland Turkey, 8.38 percent of the respondents are from Nigeria, 3.65 perce nt for the respondents are from Pakistan, 3.54 percent for the respondents are from Zimbabwe, 3.11 percent for the respondents are from Iran and the rest from other countries (See Table 1 for details). 61 Table 1: Demographic variables ( n = 931) Number of s tudents Percentage Gender Male 600 64.4 Female 331 35.6 Age 20 or younger 330 35.4 21±30 574 61.7 31-40 22 2.4 Older than 40 5 0.5 Marital status Single 896 96.2 Married 35 3.8 Education/enrolled program Associate degree 107 11.5 Bach elor¶s Gegree 715 76.8 Higher degree 109 11.7 Year/class of study First year 360 38.7 Second year 316 33.9 Third year 142 15.3 Fourth year 113 12.1 Nationality Turk 572 61.44 Nigerian 78 8.38 Pakistani 34 3.65 Zimbabwean 33 3.54 Iranian 29 3.11 Jordanian 28 3.01 Syrian 24 2.58 Egyptian 15 1.61 Palestinian 13 1.40 Libyan 12 1.29 Yemeni 9 0.97 Iraqi 6 0.64 South African 6 0.64 Kazakhstani 5 0.54 Azerbaijani 4 0.43 Eritrean 4 0.43 Ugandan 4 0.43 British 3 0.32 Congolese 3 0.32 Sudanese 3 0.32 Turkmenistan 3 0.32 62 4 .2 Results of Coefficients between V ariable s The results in Table 2 show that eReferral, eWOM, familiarity, and cultural distance have a symmetric (linear) impact on enrollment decisions. Four methods of relative relevance were employed to determine the influence of predictor factors. The first method depicts the contribution of each variable when included first, which is the squared covariance between the response variable and the focal predictor variable. The last method depicts the contribution of each variable when included last, also called usefulness. Lmg method depicts R2 contribution averaged over orderings among regressors proposed by Chevan and Sutherland (1991) and Lindeman et al. (1980). Pratt method depicts the product of the standardized coefficient and the correlation. It shows that the predictor exerted a positive and significant influence on the response variable, with 32.44% of the variance expl ained by the model (see Figure 12 ). Table 2: Linear modeling coefficients for enrollment intentions Exogeneous variable Estimates SE t ρ (Inte rcept) 1.179 0.141 8.379 *** eReferral 0.062 0.029 2.136 * eWOM 0.099 0.031 3.144 ** Familiarity 0.258 0.024 10.942 *** Cultural distance 0.320 0.029 10.804 * ** Multiple R-squared = .324; Adjusted R -squared = .321 F-statistic = 111.2; Degree of freedom = 926 Note : estimate = unstandardized coefficient; SE, standard error; t = t statistics; *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001 Figure 12: Relative imp ortance of the predictor variables (Source: authors) 64 Intelligence delineates the ability of a system to obtain information and retain it as knowledge to be applied towards adaptive behaviors within an environment or context. Resilient backpropagation with weight backtracking algorithm with the aid of multiple layers was used. As for differentiable error function and activation function, the sum of squared errors and logistic function were employed. The logistic function was used for all layers (i.e., hidd en and output layers). These algorithms provide support for ANN to minimize errors during the learning or training process. In the training stage, random synaptic weights are usually assigned to connect layers to adjust and obtain minimal errors. The lm model yielded a mean square of error (MSE) value of .301. Using (2, 2) multi -layer hidden nodes, ANN yielded an MSE value of .004, which denotes that ANN was better at predicting the performance and accuracy of the research model. Figure 13 shows the synaptic weights of the nodes (input, hidden, and output). The training process took 372 steps for the absolute partial derivatives of the error function to satisfy the condition (<0.001). $lthough $NN has ³superior preGictiYe poZer´ anG Ile[iEle non-linear response values, it has issues with pictorial cluttering that affect interaction weights visualization and other undesired behaviors in terms of interpretation (Beck, 2015). Scholars recommended the use of generalized weight distributions for ease and usefulness in interpreting the effects of a predictor variable on the response variable (Alice, 2015). The dist ribution weight plot in Figure 14 shows that eReferral, eWOM, familiarity, and cultural distance exert an asymmetric effect on intention to enroll as the generalized weights are above zero. Figure 13: Artificial neural network modeling (Source: authors) Figure 14: Generalized weight distribution for artificial neural network modeling (Source: authors) 67 4 .3 Results of N eural N etwork Ten simulated networks were subjected to a cross -validation test that used 75% and 25% of the data for training and testing, respectively, to avoid over -fitting. According to the MSEs obtained from the simulated networks shown in Table 3, the test data predicted the same outcome as the training data. Thus, H1, H2, H3, and H4 received empirical support. Table 3: Training and testing networks # Network Training Testing 1 2 3 4 5 6 7 8 9 10 Mean MSE .004 .004 .009 .009 .009 .003 .004 .009 .009 .004 .007 .011 .010 .004 .003 .005 .011 .010 .002 .004 .009 .007 Note: MSE = mean square of error 68 Chapter 5 5 DISCUSSION AND CONCLUSION 5 .1 Discussion This dissertation cross-fertilizes the literature of online reviews, cultural perceptions, and familiarity in the context of ed-tour. By doing so, this research underline the relevance of eWOM, eReferral, familiarity, and cultural distance on the intention to enroll in the context of ed-tour by using artificial intelligence and linear modeling techniques. The employed techniques confirmed the predictive nature of the exogenous variables on the response variables. In sum, the predictors can facilitate the path toward a sustainable ed-tour marketing and promotion mix. The results showed that all predictors have a significant symmetrical and asymmetrical effect on the intention to enroll. ANN (asymmetric modeling) appears to exert more significant levels of predicting performance and accuracy, in that it predicted enrollment intention better than the linear (symmetric) modeling. The findings for hypotheses 1 and 2 show that both eReferral and eWOM have a positive symmetrical and asymmetrical influence on the intention to enroll. Prior work delineates that eWOM and eReferral have significant impacts on brand image and consumer buying intentions (Abubakar et al., 2016). Previous findings ad vocate that social ties and individuals' networks of friends have a profound influence on intention to use (Chen et al., 2017; Kim et al., 2018). The results of the study contribute to the social network theory (Granovetter, 1983) by bringing to light the vital roles of social networks in ed- 69 tourists¶ enrollment Gecisions 7he tourists, Zho are the actors, attach importance to the information obtained from the people in their electronic social networks, which are defined as ties in determining their decisions. They behave in accordance with these ties/networks (Viren et al., 2015). Furthermore, this research detailed the mechanisms through which eReferral and eWOM influence enrollment decisions by providing a richer understanding of the differences between eWOM (i.e., weak ties) and eReferral (i.e., strong ties). Individuals are affected by the strength of their relationships. The higher the amount of interactions and similarities individuals have with one another, the higher the level of influence. People with strong social ties tend to share similar views (Jen -Hwa Hu et al., 2017). While the shared strong ties between friends, family members, and group members are categorized as eReferral, the dissemination of information between people with weak social ties is commonly realized as eWOM (Abubakar et al., 2016). It appears that the symmetrical influence of eWOM is higher than that of eReferral, whereas the asymmetrical influence of eReferral is higher than that of eWOM. Individuals who share strong ties have a higher number of interactions and similarities, which explains the more substantial asymmetrical impact of eReferral on the intention to enroll and the more substantial symmetrical impact of eWOM on the intention to enroll. An informed individual who shares a strong tie with a potential ed-tourist is more likely to influence enrollment decisions than less informed individuals. However, the level of influence for eWOM seems to be stable due to the lack of 100% credibility and the inability to ascertain and confirm the level of knowledge of the eWOM source. 70 Regarding hypothesis 3, this research unveils that familiarity has a positive symmetrical and asymmetrical influence on intention to enroll. Prior work asserts that familiar travelers are expected to have higher intentions and interest in traveling to a destination (Kerstetter and Cho, 2004). Similarly, familiarity with a website or social networking system influences destination selection (Chung et al., 2015) and the perception of the destination's prese nce on the social network (Di Pietro et al., 2012). Consumers become familiar by obtaining more information from websites and feeling knowledgeable (Choi et al., 2016). Moreover, familiarity with technology may be directly affected by family culture and values, together with usage of and access to technology by family members (Zhang et al., 2017). Familiarity serves as a trust and confidence mechanism. Our finding is imperative as familiarity with a website or social media page appears to have a notable influence on the intention to enroll due to acquired knowledge. In sum, in this research, we find that familiarity has a prominent influence on enrollment intentions. Contrary to the mainstream literature which posits that cultural distance is negatively related to travel intention, the results for hypothesis 4 reveal a positive symmetrical and asymmetrical influence of cultural distance on the intention to enroll. For instance, a study conducted in Hong Kong in the context of leisure tourism suggested that marketers should focus on markets with less cultural distance (Qian et al., 2018). Prior research shows that Western (individualist) tourists are more likely to visit less culturally distant destinations, as opposed to Asian (collectivist ± https://www.hofst ede-insights.com/country -comparison/ ) tourists who are more likely to visit culturally distant destinations for a holiday (Jackson, 2001; Martin et al., 2017). The researchers argued that Western tourists' choices are shaped by the desire to avoid culture shock, for example, in terms of food, lifestyle, and entertainment. 71 Consequently, the lower the perceived cultural distance, the higher the intention to travel (Zhang et al., 2017). These studies imply that individuals are attracted to others whose attitudes and beliefs are like theirs and that similarities presumably provide reassurance for affiliation and recognition for leisure tourism. In line with our assertion, Liu et al. (2018) found an insignificant relationship between perceived cultural distance and destination choice, which means that high cultural distance does not necessarily put culturally distant destinations at a disadvantage. Learning and experiencing a new culture can be perceived as one of the motivators for an ed-tourist (Abubakar et al. , 2014; Richards and Wilson, 2004; Sie et al., 2018). Tourists who perceive that the destinations they visit are close to their own culture are people who buy mostly entertainment and recreational tourism products. Tourists who are motivated to learn and experience tend to go to destinations that are different from their own culture (Falk et al., 2012). International tourists who go to another destination with a learning orientation desire to increase their personal development by acquiring new information, experiences, horizons, and insights, and learning academic or profession-specific skills (Falk et al., 2012; Sie et al., 2018). In particular, the desire to experience a different culture or food or learn a new language are among ed-tour motivations (Hara zneh et al., 2018; McGladdery and Lubbe, 2017a). Novelty seekers are more willing to travel to culturally distant destinations in pursuit of different cultural experiences. The sample in this study consists of students who are mostly from collectivist cultures, which explains their desire to study in a culturally distant destination. Because of the growing significance of electronic marketing, eWOM and eReferral campaigns have become an important element of marketing strategy. Experts have 72 argued that increasing the number of online reviews might help minimize bad comments; others have indicated that responding quickly to consumer complaints can help neutralize negative remarks. Social networks contribute to social and economic progress. Virtual communitie s make use of modern information and communication technologies to create a forum for individuals from various groups to engage and exchange knowledge, experience, and mutual interests. Participants visit nations with a considerable cultural gap. When customers wish to go shopping online, they are more likely to land on recognized and more familiar websites, which will effectively enhance the amount of self-determination satisfaction, accelerate happiness perception, and ignite involvement behavior. Simultaneously, with autonomous participation willingness, consumers may commit more time to the virtual world, sense better levels of fulfillment in demands than in a controlled setting. Thus, consumer participation intentions will improve proportionally. Individuals' emotive judgment influences perceived cultural distance, which is one form of perception. Tourists' attitudes about their hosts improve when they engage with locals, and they begin to see themselves as more like the locals. Second, the acculturation process indicates that the major force shaping individuals' cultural values is interaction. To a greater extent, the more tourists interact with destination hosts, the more psychological and perceptual changes occur; the more they immerse themselves in local culture, the more they see themselves as similar to the locals. As educational tourism is under-researched, there are many possibilities for study and development. Educational tourism can help to promote global peace and equality. Educational tourism is an excellent means of bridging the "compassion gap." Global 73 learning may be included as an aim of any variety of educational tourism that has an intercultural or international component. Educational tourism addresses issues about authenticity, cultural stereotype propagation, and cultural commercialization while simultaneously promoting global tolerance and peace (McGladdery et al., 2017). Practitioners and theorists of international educational tourism might rely on research from the intimately associated field of international education. The opportunity for value creation exists when educational tourism is integrated with other areas of the tourist or industry sectors. Perhaps more than any other type of travel, educational tourism has the most opportunity to address these issues. Global learning components may be included in bulk, if not all, as educational tours for visitors of all ages. UNESCO's Global Monitoring Report on Education for All captures the contradictions of contemporary society: While t echnology advancement increases connectivity and creates new opportunities for interchange, collaboration, and solidarity, high intolerance, political mobilization based on identity and violence are also side effects of this process (UNESCO, 2015). 5 .2 Theoretical Implications This dissertation extends the nascent research in the domain of eReferral, eWOM, familiarity and cultural distance in various dimensions. Considering the strategic relevance of social media networks and platforms, our results add to what we know to produce a more inclusive theory that deals with intention to enroll. In other words, we extend and explain the nature and degree of influence of concepts based on strong and weak ties simultaneously. Given that social network illustrates how edu-tourists are connected and their degree of centrality. The findings of this dissertation delineate and explain how the antecedents of intention to enroll 74 operate cognitively. Some of these underlying mechanisms are network conYergence, GeIineG as ³the e[tent to which people develop a common social circle´ 7seng et al, , p  , anG interGepenGence, GeIineG as ³the Gegree to Zhich memEers in a community rely on each other to make Gecisions´ 7seng et al., 2015, p. 603). Specifically, we show that eRe ferral are messages originating from close acquittances and eWOM mostly comes from individuals without identities or friend friends can foster enrollment intentions among edu-tourist. Existing empirical efforts demonstrate a rudimentary understanding of the concepts of intention to enroll, its antecedents and implications in the context of edu-tourism. We also showcase how familiarity creates an attraction and sense of relatedness to a destination. Theoretically, there are mixed results on cultural distance, does it foster or hinder travel intentions? Our study spiced up the factors and consider cultural distance amid eWOM, eReferral and familiarity with the aim of offsetting the clash of and direction of antecedents. We showcase the theoretical relevance of familiarity and cultural distance in edu-tourism context to be positive. This research proposes that decision behavior is guided by a set of features: eWOM, eReferral, familiarity, and cultural distance. The study sheds theoretical light on the eReferral, eWOM, familiarity, and cultural distance concepts by examining their effects on intention to enroll in the edu-tour context. Second, despite the abundant literature on referral marketing, only a few studies have examined the influence of electronic referrals on enrollment intention. This study enriches our understanding on the simultaneous effects of eWOM and eReferral, which offers theoretical and marketing insights on how universities can make better use of these concepts. Third, the incorporation of cultural distance and familiarity allows for theory building as these concepts were mostly considered as ideal phenomena. We therefore 75 extend the literature by exploring their intersection. Fourth, this study contributes to the ed-tour literature methodologically by revealing the presence of symmetrical and asymmetrical associations between the concepts investigated. 5 .3 Practical Implications Social networks are web shaped social capital that picturize how individuals are connected to others with an illustration on the level of relationship centrality. $ccorGing to -ohnson , p  , Gegree centrality reIers to ³the numEer oI peers with whom a student has face-to-face interactions, and so is a marker of the size of that stuGent¶s microsystem s ´ 'raZing on social network theory our study contributes to edu-tourist attraction managerial understanding. Specifically, we explore antecedents that foster intention to enroll and or study in a foreign country. Differing from conventional studies and approaches that rely on marketing and promotions, we generate insights from social network theory in the context of social media usage and networking based on weak and strong ties concepts. Our findings echoed strong message for practitioners and informed them about the presence of heterogenous actors in the SNS that is based on relational relevance. For instance, individuals with weak ties have varied level of influence and/or persuading power compare to individuals with strong social ties. Thus, the stakeholders in edu-tourism are encouraged to design and take into account the power of eReferral and eWOM. For instance, administrators can capitalize on eReferral on mobile based platforms and apps, and eWOM on web- based or general platforms. Because individuals relate mor e with close associates on mobile based apps and more with distant or unknown messages on web-based apps. Nevertheless, a mixture of both could be used based on connecting devices as mobile is about to become the standard access points for most user. Furthermore, the findings of this dissertations have practical relevance from familiarity and cultural dimensions, 76 we offer critical take home message for managers in edu-tourism industry who strive to attract edu-tourists. Since familiarity and cultural distance are motivating factors, managers could build on this to attract edu-tourists from a distant culture and also target and bombard users with enticing and appealing promotions on their social media webpages. However, it is important to regulate the rate as excessive promotions and marketing campaigns have been shown to diminish its effectiveness (Alwreikat & Rjoub). The research model and propositions stands to help destination marketers and administrators to understand how and what they can build on to establish destination competitiveness. The outcomes of the present study can assist ed-tour managers in creating a tailored and persuasive marketing strategy to increase their market share. Our findings indicate that positive eReferral and eWOM can contribute to uniYersities¶ marketing strategies. Consequently, universities should study effective ways of enhancing positive eReferral and eWOM in the market. The concept of eReferral offers an alternative and better medium to reach ed-tourists using social networks. The empirical evidence indicates that cultural distance and familiarity should be incorporated in marketing and management strategies for ed-tour, given their central role in enrollment decision-making. In order to penetrate the information sharing on social media and promote e:2M anG e5eIerral, Gestinations¶ anG institutions¶ marketing eIIorts should include collaborations with bloggers, vloggers, and influencers. Administrators and managers can identify weak points by encouraging tourists to air their anger or dissatisfaction while they are still in-house (Alrawadieh and Dincer, 2019). This approach could reduce the number of negative online reviews. 77 5 .4 Limitations and Recommendations for Future Research We build on social network theory and theoretical underpinnings to propose the study conceptual model. Although, we hope that the study will spart debate among researchers in edu-tourism domain, we also wish to cautioned readers about the sample coverage as subject and faculty or domain of study was not ca ptured. We believe this can have profound impact of the choice of studying abroad and/or desire to engage in edu-tourism. For instance, medicine and other important areas are sometimes unavailable, scarce, underdeveloped and very competitive in many countries around the world. Thus, the issue of necessity and not necessarily the studied concept may play a dominant role. Future studies are encouraged to diagnose the role of study subject, availability, and competition in home countries of edu -tourists when investigating the tested associations as majority of social sciences concepts are not universals. Furthermore, governmental regulations and agreements with host countries can have profound impact on the choice of studying abroad and faculty of study, which therefore limits the factors studied. A comprehensive study considering all this factors concurrently is a fruitful research avenue, and we are certain valuable insights can be generated for practical use. Applicable to most empirical work, this dissertation has a few limitations that worth stating. First and foremost, the study data were garnered from individuals categorized as ed-tourists who are studying at universities located in Northern Cyprus. Since most of the respondents have Turkish origin, we suspect that the generalizability of the results cannot be inferred. Thus, scholars are therefore encouraged to extend investigation in other destinations with varied cultural settings and makeup. Second, even though classical technique (i.e., linear modeli ng) and contemporary technique 78 (i.e., artificial intelligence techniques), t he cross-sectional and self-reported survey design has the potency to created biased outcomes (i.e., superficial high or low associations among the studied variables) . Upcoming research are therefore encouraged to use larger sample size, use longitudinal and experimental design, multi- sourced and/or secondary data . Third, characteristics such as major, social impact, motivation to study, and reputation of the destination or university were not taken into consideration. Future studies may consider these variables for better comprehension. Finally, the study has limitations on the number of considered antecedents and contextual variables, for instance, destination image, trust and awareness were not examined. These variables have been shown to exert and explained significant degree of variance in intention to enroll. 79 REFERENCE S Abubakar, A. M. (2018). Linking work -family interference, workplace incivility, gender and psychological distress. Journal of Management Development, 37(3), 226-242. https://doi.org/10.1108/JMD -06-2017-0207 Abubakar, A. M., Behravesh, E., Rezapouraghdam, H., & Yildiz, S. B. (2019). Applying artificial intelligence technique to predict knowledge hiding behavior. International Journal of Information Management, 49, 45-57. Abubakar, A. M., Ilkan, M., & Sahin, P. (2016). eWOM, eReferral and gender in the virtual community. Marketing Intelligence & Planning, 34( 5), 692 -710. https://doi.org/10.1108/MIP -05-2015-0090 Abubakar, A. M., Ilkan, M., Al-Tal, R. M., & Eluwole, K. K. (2017). eWOM, revisit intention, destination trust and gender. Journal of Hospitality and Tourism Management, 31, 220-227. Abubakar, A. M., Karadal, H., Bayighomog, S. W., & Merdan, E. (2018). Workplace injuries, safety climate and behaviors: application of an artificial neural network. International journal of occupational safety and ergonomics, 1-11. https://doi.org/10.1080/10803548.2018.14 54635 Abubakar, A. M., Shneikat, B. H. T., & Oday, A. (2014). Motivational factors for educational tourism: A case study in Northern Cyprus. Tourism management perspectives, 11, 58 -62. 80 Al-Htibat, A., & Garanti, Z. (2019). Impact of interactive eReferral on tourists behavioral intentions. Marketing Intelligence & Planning. Alice, M., ( 2021) . Fitting a Neural Network in R; neuralnet package. [online] DataScience+. Available at: [Accessed 22 September 2021]. Alrawadieh, Z., & Dincer, M. Z. (2019). Reputation management in cyberspace: eYiGence Irom -orGan¶s lu[ury hotel market Journal of Hospitality and Tourism Technology, 10 ( 1), 107-120. Altinay, L., Paraskevas, A., & Jang, S. S. (2015). Planning research in hospitality and tourism. Routledge. Anderson, M. and Anderson, M., 2021. 88% Of Consumers Trust Online Reviews As Much As Personal Recommendations. [online] Search Engine Land. Available at: [Accessed 22 September 2021]. $rici, H (, (rtrk, M, 2rcan, 2   $ 6tuGy on (Gucational 7ourism ,mpacts of Foreign Students on The Perception of Local Turkish Students: Evidence From Northern Cyprus . Journal of Gastronomy and Studies, 2(1), 3 -12. Èrnason, 3  , 7houghts on (Gucational Geo 7ourism, 8niYersity oI ,celanG hornaIM|rèur rural research centre, Hiskyli ,slanGs 81 $ssaker, G, 2¶Connor, 3   e:2M platIorms in moGerating the relationships between political and terrorism risk, destination image, and travel intent: the case of Lebanon. Journal of Travel Research, 60(3), 503 -519. Balabanis, G., Reynolds, N., & Simintiras, A. (2006). Bases of e -store loyalty: Perceived switching barriers and satisfaction. Journal of Business Research, 59(2), 214 -224. Barreda, A., & Bilgihan, A. (2013). An analysis of user -generated content for hotel experiences. Journal of Hospitality and Tourism Technology, 4(3), 263 ±280. Beck, M. W. (2018). NeuralNe tTools: visualization and analysis tools for neural networks. Journal of statistical software, 85(11), 1. Berger, J., & Milkman, K. L. (2012). What makes online content viral?. Journal of marketing research, 49 (2), 192 -205. Berger, J., & Schwartz, E. M. (2 011). What drives immediate and ongoing word of mouth?. Journal of marketing research, 48(5), 869 -880. Bilgihan, A., Barreda, A., Okumus, F., & Nusair, K. (2016). Consumer perception of knowledge-sharing in travel-related online social networks. Tourism Management, 52, 287 -296. 82 %ilgihan, $, 2kumus, ), CoEanoglu, C   Generation Y traYelers¶ commitment to online social network websites. Tourism Management, 35, 13- 22. Blanco, C. F., Sarasa, R. G., & Sanclemente, C. O. (2010). Effects of visual and t extual information in online product presentations: looking for the best combination in website design. European Journal of information systems, 19(6), 668 -686. https://doi.org/10.1057/ejis.2010.42 Blut, M., Chowdhry, N., Mittal, V., & Brock, C. (2015). E -service quality: A meta - analytic review. Journal of retailing, 91(4), 679 -700. Boyd, D. M., & Ellison, N. B. (2007). Social network sites: Definition, history, and scholarship. Journal of computer‐mediated Communication, 13(1), 210 -230. Boykin, J. (2015). Start monitoring your online reviews now: No excuse for ignoring online reputation. [online] Internetmarketingninjas.com. Available at: [Accessed 22 September 202 1]. Brandes, U., & Hildenbrand, J. (2014). Smallest graphs with distinct singleton centers. Network Science, 2(3), 416 -418. doi: 10.1017/nws.2014.25MC Breazeale, M. (2009). An assessment of electronic word -of-mouth research. International Journal of Market Research, 51(3), 297 ±318. Briggs, S. R., & Cheek, J. M. (1986). The role of factor analysis in the development 83 and evaluation of personality scales. Journal of personality, 54(1), 106 -148. Bronner, F., & De Hoog, R. (2011). Vacationers and eWOM: who posts , and why, where, and what?. Journal of travel research, 50(1), 15 -26. https://doi.org/10.1177/00472 87509355324 Bughin, J., Doogan, J., & Vetvik, O. J. (2010). A new way to measure word -of-mouth marketing. McKinsey Quarterly, 2, 113±116. Bulbul, C., Gross , N., Shin, S., & Katz, J. (2014). When the path to purchase becomes the path to purpose. Think with Google. Burt, R. S. (1987). Social contagion and innovation: Cohesion versus structural equivalence. American journal of Sociology, 92(6), 1287 -1335. doi: 10.1086/228667 Camilleri, M. A. (2018). Travel marketing, tourism economics and the airline product: An introduction to theory and practice. springer. Cantallops, A. S., & Salvi, F. (2014). New consumer behavior: A review of research on eWOM and hotels. International Journal of Hospitality Management, 36, 41-51. Casaly, /, )laYi n, C, Guinaltu, M   7he role oI perceiYeG usaEility, reputation, satisfaction and consumer familiarity on the website loyalty formation process. Computers in Human behavior, 24(2), 325 -345. 84 https://doi.org/10.1016/j.chb.2007.01.017 CasanueYa, C, Gallego, È, Garcta-6inchez, M 5   6ocial netZork analysis in tourism. Current Issues in Tourism, 19(12), 1190 -1209. Chang, K. C. (2021). The affecting tourism development attitudes based on the social exchange theory and the social network theory. Asia Pacific Journal of Tourism Research, 26(2), 167 -182. Chatterjee, P. (2001). Online Reviews: Do Consumers Use Them ? In M.C. Gilly & J. MyersLevy (Eds.), Proceedings of the ACR 2001, pp. 129 -134. Provo, UT: Association for Consumer Research. Chen, C. C., & Lin, Y. H. (2012). Segmenting mainland Chinese tourists to Taiwan by Gestination Iamiliarity $ Iactorဨcluster approach. International Journal of Tourism Research, 14(4), 339 -352. Chen, C. Y., Chen, T. H., Chen, Y. H., Chen, C. L., & Yu, S. E. (2013). The spatio - temporal distribution of different types of messages and personality traits affecting the eWOM of Facebook. Natural hazards, 65(3), 2077 -2103. Chen, J. V., Lin, C., Yen, D. C., & Linn, K. P. (2011). The interaction effects of familiarity, breadth and media usage on web browsing experience. Computers in Human Behavior, 27(6), 2141 -2152. Chen, K., Luo, P., & Wan g, H. (2017). An influence framework on product word -of- 85 mouth (WoM) measurement. Information & Management, 54(2), 228 -240. Chen, Y. F., & Law, R. (2016). A review of research on electronic word -of-mouth in hospitality and tourism management. International Journal of Hospitality & Tourism Administration, 17 (4), 347 -372. Cheng, Y. -H., & Ho, H. -Y   6ocial inIluence¶s impact on reaGer perceptions oI online reviews. Journal of Business Research, 68 (4), 883 ±887. Cheung, C. M. K., & Thadani, D. R. (2012). The impact of electronic word-of-mouth communication: A literature analysis and integrative model. Decision Support Systems, 54(1), 461 ±470Fine, M. B., Gironda, J., & Petrescu, M. (2017). Prosumer motivations for electronic word-of-mouth communication behaviors. Journal of Hospitality and Tourism Technology, 8(2), 280 -295. https://doi.org/10.1108/JHTT -09 -2016-0048 Chevan, A., & Sutherland, M. (1991). Hierarchical partitioning. The American Statistician, 45(2), 90 -96. Chi, H. K., Huang, K. C., & Nguyen, H. M . (2020). Elements of destination brand equity and destination familiarity regarding travel intention. Journal of Retailing and Consumer Services, 52, 101728. Choi, J., Ok, C., & Choi, S. (2016). Outcomes of destination marketing organization website navigation: The role of telepresence. Journal of Travel & Tourism Marketing, 33(1), 46 -62. 86 Chu, 6 C, Chen, H 7, Gan, C   Consumers¶ engagement Zith corporate social responsibility (CSR) communication in social media: Evidence from China and the United States. Journal of Business Research, 110, 260-271. Chung, N., Lee, H., Lee, S. J., & Koo, C. (2015). The influence of tourism website on tourists' behavior to determine destination selection: A case study of creative economy in Korea. Technological Forecasting and Social Change, 96, 130- 143. Churchill, G. A., Brown, T. J., & Suter, T. A. (1996). Basic marketing research. Orlando, FL: Dryden Press. Cohen, E. (1972). Toward a sociology of international tourism. Social research, 164- 182. Collis, J., & Hus sey, R. (2013). Business research: A practical guide for undergraduate and postgraduate students. Macmillan International Higher Education. Connell, J., & Meyer, D. (2009). Balamory revisited: An evaluation of the screen tourism destination-tourist nexus. Tourism Management, 30(2), 194 -207. Connelly, L. M. (2008). Pilot studies. Medsurg nursing, 17(6), 411. Cristobal-Fransi, E., Daries-5amon, N, Marinp-Roig, E., & Martin -Fuentes, E. (2017). Implementation of Web 2.0 in the snow tourism industry: Analysis o f 87 the online presence and e-commerce of ski resorts. Spanish Journal of Marketing-ESIC, 21(2), 117 -130. Crotts, J. C. (2004). The effect of cultural distance on overseas travel behaviors. Journal of Travel Research, 43(1), 83 -88. Crotts, J. C., Mason, P. R ., & Davis, B. (2009). Measuring guest satisfaction and competitive position in the hospitality and tourism industry: An application of stance-shift analysis to travel blog narratives. Journal of Travel Research, 48(2), 139 -151. Dang, A., & Nandakumar, G. S. (2017). Cultural Values and Their Impact on Electronic Word-of-Mouth (eWOM) Behavior. In Marketing at the Confluence between Entertainment and Analytics (pp. 1281 -1282). Springer, Cham. de la Hoz-Correa, $, Muxoz-Leiva, F. (2019). The role of informa tion sources and image on the intention to visit a medical tourism destination: a cross -cultural analysis. Journal of Travel & Tourism Marketing, 36(2), 204 -219. Di Pietro, L., Di Virgilio, F., & Pantano, E. (2012). Social network for the choice of tourist destination: Attitude and behavioural intention. Journal of Hospitality and Tourism Technology, 3(1), 60 ±76. (rkan, ,, (Yans, C   7he inIluence oI e:2M in social meGia on consumers¶ purchase intentions: An extended approach to information adoption. Computers in human behavior, 61, 47-55. 88 Falk, J. H., Ballantyne, R., Packer, J., & Benckendorff, P. (2012). Travel and learning: A neglected tourism research area. Annals of Tourism Research, 39(2), 908 - 927. Fan, D. X., Qiu, H., Jenkins, C. L., & Lau, C. (2020). Towards a better tourist -host relationship the role oI social contact EetZeen tourists¶ perceiYeG cultural distance and travel attitude. Journal of Sustainable Tourism, 1-25. Fan, D. X., Zhang, H. Q., Jenkins, C. L., & Lin, P. M. (2017). Does tourist±host social contact reduce perceived cultural distance?. Journal of Travel Research, 56 (8), 998 -1010. https://doi.org/10.1177/0047287517696979 Farzin, M., & Fattahi, M. (2018). eWOM through s ocial networking sites and impact on purchase intention and brand image in Iran. Journal of Advances in Management Research, 15( 2), 161-183. https://doi.org/10.1108/JAMR -05- 2017-0062 Fatma, M., Ruiz, A. P., Khan, I., & Rahman, Z. (2020). The effect of CSR engagement on eWOM on social media. International Journal of Organizational Analysis. Ferguson, D., Sairamesh, J., & Feldman, S. (2004). Open frameworks for information cities. Communications of the ACM, 47(2), 45 -49. Filieri, R. (2015). What makes online reviews helpful? A diagnosticity -adoption framework to explain informational and normative influences in e- WOM. Journal of business research, 68(6), 1261 -1270. 89 Filieri, R., Galati, F., & Raguseo, E . (2021). The impact of service attributes and category on eWOM helpfulness: An investigation of extremely negative and positive ratings using latent semantic analytics and regression analysis. Computers in Human Behavior, 114, 106527. )laYi n, C, Guinaltu, M., & Gurrea, R. (2006). The influence of familiarity and usability on loyalty to online journalistic services: The role of user experience. Journal of Retailing and Consumer Services, 13(5), 363 -375. Freeman, L. C. (1978). Centrality in social networks conceptual clarification. Social networks, 1(3), 215 -239. doi: 10.1016/0378 -8733(78)90021 -7 Friedkin, N. E. (1993). Structural bases of interpersonal influence in groups: A longitudinal case study. American Sociological Review, 861 -872. doi: 10.2307/20959 55 Garton, L., Haythornthwaite, C., & Wellman, B. (1997). Studying online social networks. Journal of computer-mediated communication, 3(1), JCMC313. Gefen, D. (2000). E -commerce: the role of familiarity and trust. Omega, 28(6), 725 - 737. Goeldner, C. R., & Ritchie, J. R. B. (2012). Tourism: Principles, practices, philosophies (12th ed.). Wiley. Gymez, M, ,mhoII, %, Mart n-Consuegra, D., Molina, A., & Santos -Vijande, M. L. 90 (2018). Language tourism: The drivers that determine destination choice intention among US students. Tourism management perspectives, 27, 125-135. Gon, M., Pechlaner, H., & Marangon, F. (2016). Destination management organizations 'M2s anG 'igital NatiYes the neglecteG ³inIormal e[pertise´ in web 2.0 implementation and social media presence. Insights from the Italian Friuli Venezia Giulia DMO. Information Technology & Tourism, 16(4), 435 - 455. G|tzl, , , 2ctoEer  7eleCities²Digital Cities Network. In Kyoto Workshop on Digital Cities (pp. 101 -109). Springer, Berlin, Heidelberg. Granovetter, M. S. (1973). The strength of weak ties. American journal of sociology, 78(6), 1360 -1380. https://doi.org/10.1086/225469 Granovetter, M. S. (1982). The strength of weak ties: A network theory revisited, In P. V. Marsden, & N. Lin (Eds.), Social structure and network analysis. Beverly Hills: Sage. Granovetter, M., & Soong, R. (1983). Threshold models of diffusion and collective behavior. Journal of Mathematical sociology, 9(3), 165 -179. Gray, P. H., Parise, S., & Iyer, B. (2011). Innovation impacts of using social bookmarking systems. MIS quarterly, 629 -643. Gnther, ), )ritsch, 6   Neuralnet training oI neural netZorks R J., 2(1), 91 30. Harazneh, I., Al-Tall, R. M., Al-Zyoud, M. F., & Abuba kar, A. M. (2018). Motivational factors for educational tourism: marketing insights. Management & Marketing, 13(1), 796 -811. Hassan, Z. A., Schattner, P., & Mazza, D. (2006). Doing a pilot study: why is it essential?. Malaysian family physician: the official journal of the Academy of Family Physicians of Malaysia, 1(2 -3), 70. Hausemer, P., Rzepecka, J., Dragulin, M., Vitiello, S., Rabuel, L., Nunu, M., ... & Baines, L. (2017). Exploratory Study of consumer issues in peer -to-peer platform markets. He, X., I nman, J. J., & Mittal, V. (2008). Gender jeopardy in financial risk taking. Journal of Marketing Research, 45(4), 414 -424. Hennig-Thurau, T., Gwinner, K. P., Walsh, G., & Gremler, D. D. (2004). Electronic word-of-mouth via consumer-opinion platforms: what motivates consumers to articulate themselves on the internet?. Journal of interactive marketing, 18(1), 38 -52. Hennig-Thurau, T., Malthouse, E. C., Friege, C., Gensler, S., Lobschat, L., Rangaswamy, A., & Skiera, B. (2010). The impact of new media on custo mer relationships. Journal of Service Research, 13 (3), 311 ±330. 92 Hogg, T., & Adamic, L. (2004, May). Enhancing reputation mechanisms via online social networks. In Proceedings of the 5th ACM Conference on Electronic Commerce (pp. 236 -237). Hu, X., & Yang, Y . (2021). What makes online reviews helpful in tourism and hospitality? a bare -bones meta-analysis. Journal of Hospitality Marketing & Management, 30(2), 139 -158. Jackson, M. (2001). Cultural influences on tourist destination choices of 21pacific rim nations. Paper presented at the CAUTHE national research conference, 166± 176. Jalilvand, M. R., Esfahani, S. S., & Samiei, N. (2010). Electronic word -of-mouth: Challenges and opportunities. Procedia Computer Science, 3, 42±46. Jen -Hwa Hu, P., Han-fen, H., & Xia o, F. (2017). Examining the Mediating Roles of Cognitive Load and Performance Outcomes in User Satisfaction with a Website: A Field Quasi -Experiment. MIS Quarterly, 41(3). Jeong, E., & Jang, S. S. (2011). Restaurant experiences triggering positive electro nic word-of-mouth (eWOM) motivations. International Journal of Hospitality Management, 30(2), 356 -366. Jia, S. S. (2020). Motivation and satisfaction of Chinese and US tourists in restaurants: A cross-cultural text mining of online reviews. Tourism Management, 78, 104071. 93 Jimenez -Barreto, J., Sthapit, E., Rubio, N., & Campo, S. (2019). Exploring the dimensions of online destination brand experience: Spanish and North American tourists' perspectives. Tourism Management Perspectives, 31, 348 - 360. Katircioğlu, 6   Higher eGucation anG health care GeYelopment 7he case oI Northern Cyprus. Applied Research in Quality of Life, 9(3), 605 -616. Katircioğlu, 6 7   ,nternational tourism, higher eGucation anG economic growth: The case of North Cyprus. The World Economy, 33(12), 1955 -1972. Kaya, %, %ehraYesh, (, $EuEakar, $ M, Kaya, 2 6, 2r~s, C   7he moderating role of website familiarity in the relationships between e-service quality, e-satisfaction and e-loyalty. Journal of Internet Commerce, 18 (4), 369 -394. Kerstetter, D., & Cho, M. H. (2004). Prior knowledge, credibility and information search. Annals of Tourism research, 31(4), 961 -985. Kim, S., Kandampully, J., & Bilgihan, A. (2018). The influence of eWOM communications: An application of online social network framework. Computers in Human Behavior, 80, 243-254. Kim, S., Lehto, X., & Kandampully, J. (2019). The role of familiarity in consumer destination image formation. Tourism Review. 94 Kim, Y., Park, Y., Lee, Y., & Park, K.   'o Ze alZays aGopt )aceEook IrienGs¶ eWOM postings? The role of social identity and threat. International Journal of Advertising, 37(1), 86 -104. Kimball, L. & Rheingold, H. (2000). How online social networks benefit organizations, Rheingold Associates, Poplar Mill Valley, CA. King, R. A., Racherla, P., & Bush, V. D. (2014). What we know and don't know about online word-of-mouth: A review and synthesis of the literature. Journal of interactive marketing, 28(3), 167 -183. Komninos, N. (2013). Intelligent cities: innovation, knowledge systems and digital spaces. Spon Press, London. Ladhari, R., & Michaud, M. (2015). eWOM effects on hotel booking intentions, attitudes, trust, and website perceptions. International Journal of Hospitality Management, 46, 36±45. Lee, K. T., & Koo, D. M. (2015). Evaluating right versus just evaluating online consumer reviews. Computers in Human Behavior, 45, 316-327. Lee, Y., & Kwon, O. (2011). Intimacy, familiarity and continuance intention: An extended expectation±confirmation model in web-based services. Electronic Commerce Research and Applications, 10(3), 342 -357. Lepp, A., & Gibson, H. (2003). Tourist roles, perceived risk and international 95 tourism. Annals of tourism research, 30(3), 606 -624. Leung, X., Tanford, S., & Ji ang, L. (2017). Is a picture really worth a thousand words? An experiment on hotel Facebook message effectiveness. Journal of Hospitality and Tourism Technology. https://doi.org/10.1108/JHTT -08 -2016- 0039 Li, Y. (2014). The impact of disposition to privacy, website reputation and website familiarity on information privacy concerns. Decision support systems, 57, 343-354. Lindeman, R.H., Merenda, P.F. & Gold, R.Z. (1980). Introduction to Bivariate and Multivariate Analysis. Glenview IL: Scott, Foresman. Line, T., Jain, J., & Lyons, G. (2011). The role of ICTs in everyday mobile lives. Journal of Transport Geography, 19(6), 1490 -1499. Linton, R. (1945). The cultural background of personality. New York: D. Appleton- Century. Litvin, S. W., Goldsmith, R. E., & Pan, B. (2008). Electronic word -of-mouth in hospitality and tourism management. Tourism management, 29(3), 458 -468. /iu, H, /i, ; 5, CirGenas, ' $, Yang, Y   3erceiYeG cultural Gistance and international destination choice: The role of destination familiarity, geographic distance, and cultural motivation. Journal of Destination 96 Marketing & Management, 9, 300-309. Loureiro, S. M. C., & Kastenholz, E. (2011). Corporate reputation, satisfac tion, delight, and loyalty towards rural lodging units in Portugal. International Journal of Hospitality Management, 30(3), 575 ±583. Loureiro, S. M. C., & Sarmento, E. M. (2019). Place attachment and tourist engagement of major visitor attractions in Lisbo n. Tourism and Hospitality Research, 19(3), 368 -381. Luhmann, N. (1979). Trust and Power 2 works. Chichester: Wiley. Luhmann, N. (2018). Trust and power. John Wiley & Sons, Chichester, UK. Luo, Q., & Zhong, D. (2015). Using social network analysis to expl ain communication characteristics of travel-related electronic word-of-mouth on social networking sites. Tourism Management, 46, 274-282. Lyons, K., Hanley, J., Wearing, S., & Neil, J. (2012). Gap year volunteer tourism: Myths of global citizenship?. Annals of tourism research, 39(1), 361 -378. Magrizos, S., Kostopoulos, I., & Powers, L. (2021). Volunteer Tourism as a Transformative Experience: A Mixed Methods Empirical Study. Journal of Travel Research, 60 (4), 878 -895. Mariani, M. M., & Matarazzo, M. (2020) . Does cultural distance affect online review 97 ratings" Measuring international customers¶ satisIaction Zith serYices leveraging digital platforms and big data. Journal of Management and Governance, 1-22. Martin, B. A., Jin, H. S., & Trang, N. V. (2017). Th e entitled tourist: The influence of psychological entitlement and cultural distance on tourist judgments in a hotel context. Journal of Travel & Tourism Marketing, 34(1), 99 -112. McGladdery, C. A., & Lubbe, B. A. (2017). Rethinking educational tourism: proposing a new model and future directions. Tourism Review. McGladdery, C. A., & Lubbe, B. A. (2017). Rethinking educational tourism: proposing a new model and future directions. Tourism Review, 72( 3), 319 -329. https://doi.org/10.1108/TR -03-2017-0055 McGladdery, C. A., & Lubbe, B. A. (2017 a). International educational tourism: Does it foster global learning? A survey of South African high school learners. Tourism Management, 62, 292 -301. Meik, J., Brock, C., & Blut, M. (2014, April). Complaining customers a s innovation contributors: stimulating service innovation through multichannel complaint management. In 2014 Annual SRII Global Conference (pp. 125 -132). IEEE. Memarzadeh, F., Blum, S. C., & Adams, C. (2015). The impact of positive and negative e-comments on Eusiness traYelers¶ intention to purchase a hotel room Journal of Hospitality and Tourism Technology, 6(3), 258 ±270. 98 Mittendorf, C. (2018). Collaborative consumption: the role of familiarity and trust among Millennials. Journal of Consumer Marketing. 35( 4), 377-391. https://doi.org/10.1108/JCM -12-2016-2040. Mukherji, P., & Albon, D. (2018). Research methods in early childhood: An introductory guide. Sage. Munar, A. M., & Jacobsen, J. K. S. (2013). Trust and involvement in tourism social media and web-based travel information sources. Scandinavian Journal of Hospitality and Tourism, 13 (1), 1 -19. Nagai, H., & Kashiwagi, S. (2018). Japanese students on educational tourism: current trends and challenges. In Asian youth travellers (pp. 117 -134). Springer, Singapore. Nielsen (2012). Under the Influence: Consumer Trust In Advertising. Nielsen.com. Retrieved 22 September 2021, from https://www.nielsen.com/ma/en/insights/article/2013/under -the-influence- consumer-trust-in-advertising/. O'Murchu, I., Breslin, J. G. , & Decker, S. (2004, August). Online Social and Business Networking Communities. In ECAI Workshop on Application of Semantic Web Technologies to Web Communities (Vol. 107). Perry Hinton, D., Hinton, P. R., McMurray, I., & Brownlow, C. (2004). SPSS explained. Routledge. London and New York. 99 Persson, J. (2010). Misconceptions of positivism and five unnecessary science theoretic mistakes they bring in their train. International journal of nursing studies, 47(5), 651 -661. Peruta, A., & Shields, A. B. (2018). Marketing your university on social media: A content analysis of Facebook post types and formats. Journal of Marketing for Higher Education, 28(2), 175 -191. Piedmont, R. L. (2014). Encyclopedia of quality of life and well -being research. Michalos ACeditor. Inter-item Correlations. Springer. https://doi.org/10.1007/978 -94 -007-0753-5_1493 . Pitman, T., Broomhall, S., McEwan, J., & Majocha, E. (2010). Adult learning in educational tourism. Australian Journal of Adult Learning, 50(2), 219 -238. Podsakoff, P. M., MacKenzie, S. B., & Podsakoff, N. P. (2012). Sources of method bias in social science research and recommendations on how to control it. Annual review of psychology, 63, 539 -569. Qian, J., Law, R., & Wei, J. (2018). Effect of cultural distance on tourism: A study of pleasure visitors in Hong Kong. Journal of Quality Assurance in Hospitality & Tourism, 19(2), 269 -284. Rensink, J. M. (2013). What motivates people to write online reviews and which role does personality play?: a study providing insights in the influence of seven motivations on the involvement to write positive and negative online reviews 100 and how five personality traits play a role (Master's thesis, University of Twente). Richards, G. W. (2011). Cultural tourism trends in Europe: a context for t he development of Cultural Routes. In Impact of European Cultural Routes on SMEs’ innovation and competitiveness (pp. 21 -39). Council of Europe Publishing. Richards, G., & Wilson, J. (2004). The international student travel market: Travelstyle, motivations, and activities. Tourism Review International, 8(2), 57 -67. Ritchie, B. W. (2003). 1. An Introduction to Educational Tourism. In Managing educational tourism (pp. 1 -24). Channel View Publications. Ritchie, B. W., Carr, N., & Cooper, C. P. (2003). Managing Educational Tourism; Channel View Publications: Clevedon, UK. Ritchie, B.W. (2003), Managing Educational Tourism, Channel View Publications, Clevedon. Rokeach, M. (1973). The nature of human values. Free press. 6inchez-Garcta, ,, Curris-3prez, 5 ). Effects of dissatisfaction in tourist services: The role of anger and regret. Tourism Management, 32(6), 1397 - 1406. 101 Saunders, M., Lewis, P. H. I. L. I. P., & Thornhill, A. D. R. I. A. N. (2007). Research methods. Business Students 4th edition Pearson Education Limited, England. Scott, J. (1988). Social network analysis. Sociology, 22(1), 109 -127. Serrat, O. (2017). Social network analysis. In Knowledge solutions (pp. 39 -43). Springer, Singapore. Sharifi, M., Pool, J. K., Jalilvand, M. R., Tabaeeian, R. A ., & Jooybari, M. G. (2019). Forecasting of advertising effectiveness for renewable energy technologies: A neural network analysis. Technological Forecasting and Social Change, 143, 154-161. Sharifpour, M., Walters, G., Ritchie, B. W., & Winter, C. (2014). Investigating the role of prior knowledge in tourist decision making: A structural equation model of risk perceptions and information search. Journal of Travel Research, 53(3), 307-322. Sharma, P., & Wu, Z. (2015). Consumer ethnocentrism vs. intercultural competence as moderators in intercultural service encounters. Journal of Services Marketing. Shukla, P. (2011). Impact of interpersonal influences, brand origin and brand image on luxury purchase intentions: Measuring interfunctional interactions and a cross-national comparison. Journal of world business, 46(2), 242 -252. https://doi.org/10.1016/j.jwb.2010.11.002 102 Sie, L., Phelan, K. V., & Pegg, S. (2018). The interrelationships between self - determined motivations, memorable experiences and overall satisfaction: A case of older Australian educational tourists. Journal of Hospitality and Tourism Technology, 9(3), 354-379. Sigala, M., Christou, E., & Gretzel, U. (Eds.). (2012). Social media in travel, tourism and hospitality: Theory, practice and cases. Ashgate Publishing, Ltd.. Sim, J. J., Tan, G. W. H., Wong, J. C., Ooi, K. B., & Hew, T. S. (2014). Understanding and predicting the motivators of mobile music acceptance±a multi-stage MRA- artificial neural network approach. Telematics and Informatics, 31(4), 569 - 584. Siuda, P., & Troszynski, M. (2017). Natives and tourists of prosumer capitalism: On the varied pro-prosumer activities of producers exemplified in the Polish pop culture industry. International Journal of Cultural Studies, 20(5), 545 -563. Slocum, S. L., Dimitrov, D. Y., & Webb, K. (2019). The impact of neoliberalism on higher education tourism programs: Meeting the 2030 sustainable development goals with the next generation. Tourism Management Perspectives, 30, 33-42. Solomon, M. (2014). Consumer behavior: Buying, having, and being. Engelwood Cliffs. Sotiriadis, M. D., & Van Zyl, C. (2013). Electronic word -of-mouth and online reviews in tourism services: the use of twitter by tourists. Electronic Commerce 103 Research, 13(1), 103 ±124. Standing, C., Holzweber, M., & Mattsson, J. (2016). Exploring emotional expressions in e-word-of-mouth from online communities. Information Processing & Management, 52(5), 721 -732. Stoner, K. R., Tarrant, M. A., Perry, L., Stoner, L., Wearing, S., & Lyons, K. (2014). Global citizenship as a learning outcome of educational travel. Journal of Teaching in Travel & Tourism, 14(2), 149 -163. Suki, N. M., & Suki, N. M. (2014). Consumer online shopping behavior: The effect of internet marketing environment, product characteristics, familiarity and confidence, and promotional offer. International Journal of Economics and Management Engineering, 7(3), 814 -819. Sun, L. B., & Qu, H. (2011). Is there any gender effect on the relatio nship between service quality and word-of-mouth?. Journal of Travel & Tourism Marketing, 28(2), 210 -224. Swanson, S. R., & Hsu, M. K. (2009). Critical incidents in tourism: Failure, recovery, customer sZitching, anG ZorGဨoIဨmouth EehaYiors Journal of Travel & Tourism Marketing, 26(2), 180 -194. Sweeney, J., Soutar, G., & Mazzarol, T. (2014). Factors enhancing word-of-mouth influence: positive and negative service -related messages. European Journal of Marketing, 48(1/2), 336 ±359. 104 Taherdoost, H. (2016). Sampling methods in research methodology; how to choose a sampling technique for research. How to Choose a Sampling Technique for Research (April 10, 2016). Tahir, R., & Larimo, J. (2004). Understanding the location strategies of the European firms in Asian countries. Journal of American Academy of Business, 5 (1/2), 102-109. The New York Times. (2012). Buy Reviews on Yelp, Get Black Mark (Published 2012). Nytimes.com. Retrieved 22 September 2021, from http://www.nytimes.com/2012/10/18/technology/yelp -tries-to-halt-deceptive- reviews.html?_r=3&StreitfeldStreitfeld. Tichy, N. M., Tushman, M. L., & Fombrun, C. (1979). Social network analysis for organizations. Academy of management review, 4 (4), 507 -519. UNESCO (2015). Rethinking education: Towards a global common good? Paris: UNESCO. http://une sdoc.unesco.org/images/0023/002325/232555e.pdf Urban, G. L., Amyx, C., & Lorenzon, A. (2009). Online trust: state of the art, new frontiers, and research potential. Journal of Interactive Marketing, 23(2), 179 ± 190. Valente, T. W. (1996). Social network thresholds in the diffusion of innovations. Social networks, 18(1), 69 -89. 105 van't Klooster, E. (2014). Travel to learn: The influence of cultural distance on competence development in educational travel (No. EPS -2014-312-MKT). Veldof, J., & Beavers, K. (2001). Going mental: Tackling mental models for the online library tutorial. Research Strategies, 18(1), 3 -20. Viren, P. P., Vogt, C. A., Kline, C., Rummel, A. M., & Tsao, J. (2015). Social network participation and coverage by tourism industry sector. Journal of Destination Marketing & Management, 4(2), 110 -119. Wasserman, S., & Faust, K. (1994). Social network analysis: Methods and applications. Cambridge, UK: Cambridge University Press. Weijters, B., Geuens, M., & Schillewaert, N. (2009). The proximity effect: The role of inter-item distance on reverse-item bias. International Journal of Research in Marketing, 26(1), 2 -12. Wong, E., Rasoolimanesh, S. M., & Sharif, S. P. (2020). Using online travel agent platforms to determine factors influencing hotel guest satisfaction. Journal of Hospitality and Tourism Technology, 11( 3), 425 -445. https://doi.org/10.1108/JHTT -07-2019 -0099 Xiong, L., Alsadoon, A., Maag, A., Prasad, P. W. C., Hoe, L. S., & Elchouem i, A. (2018, August). Rise of social media marketing: A perspective on higher education. In 2018 13th International Conference on Computer Science & Education (ICCSE) (pp. 1 -6). IEEE. 106 Yang, Y., Liu, H., & Li, X. (2016). The influences of national cultural distance on international destination choice of Japanese and Chinese residents. Tourism Tribune, 31(10), 45 -55. Yang, Y., Liu, H., & Li, X. (2019). The world is flatter? Examining the relationship between cultural distance and international tourist flows. Journal of Travel Research, 58(2), 224 -240. Ying, T., Tan, X., Ye, S., Ka, X., & Zhou, Y. (2020). Examining tourist mindfulness in museums: the roles of familiarity and motivation. Asia Pacific Journal of Tourism Research, 25(9), 981 -996. Yoo, C. W., Kim, Y. J., & Sanders, G. L. (2015). The impact of interactivity of electronic word of mouth systems and E-Quality on decision support in the context of the e-marketplace. Information & Management, 52(4), 496 -505. Yoo, C. W., Sanders, G. L., & Moon, J. (2013). Exploring the effect of e-WOM participation on e-Loyalty in e -commerce. Decision Support Systems, 55(3), 669 -678. Yoo, K. H., & Gretzel, U. (2008). What motivates consumers to write online travel reviews?. Information Technology & Tourism, 10 (4), 283 -295. =hang, 7 C, 2mran, % $, CoEanoglu, C   Generation Y¶s positiYe anG negative eWOM: use of social media and mobile technology. International Journal of Contemporary Hospitality Management, 29( 2), 732-761. 107 Zhang, Y., Li, X., & Wu, T. (2019). The impacts of cultural values on bilateral international tourist flows: A panel data gravity model. Current Issues in Tourism, 22(8), 967 -981. Zheng, Q. Y., Yao, T., Qiu, Q., & Cao, H. R. (2012, September). Research on the mechanism of self-determination for online customer participation psychology and behavior. In 2012 International Conference on Management Science & Engineering 19th Annual Conference Proceedings (pp. 53 -58). IEEE. 108 APPENDICES Appendix A : The List of Items and Descriptive Statisti cs Scale Items eReferral - Mean = 3.6 2; Std. Deviation = .77; Alpha = .53. Mean Std. Deviation RFQ1 ± ³, oIten consiGer institutions reIerreG Ey my IrienGs, colleagues anG Iamily ´ 3.66 1.06 RFQ2 ± ³, oIten consiGer institutions reIerreG Ey trusteG Iirms anG social netZork site ´ 3.80 1.05 RFQ3 ± ³:hen , enroll in an institution that is not reIerreG, , Zorry aEout my Gecision ´ 3.49 1.09 RFQ4 ± ³2nline reIerrals increase my conIiGence in enrolling to an institution ´ 3.52 1.03 Electronic word - of- mouth (eWOM) - Mean = 3.63; Std. Deviation = .71; Alpha = .76. WMQ1 - , oIten reaG other stuGents¶ online reYieZs to knoZ Zhich institutions make gooG impressions on others´ 3.72 1.09 WMQ2 ± µ7o make sure , enroll in a gooG institution, , oIten reaG other stuGents¶ online reYieZs ´ 3.84 1.05 WMQ3 ± ³, oIten consult other stuGents¶ online reYieZs to help choose the right institution´ 3.73 1.05 WMQ4 ± ³, IreTuently gather inIormation Irom online reYieZs EeIore , make a Gecision on a certain institution ´ 3.83 1.00 WMQ5 ± ³,I , Gon¶t reaG stuGents¶ online reYieZs, , Zorry aEout my enrollment Gecision´ 3.08 1.16 WMQ6 ± ³6tuGents¶ online reYieZs make me conIiGent in enrolling to an institution´ 3.59 1.02 Familiarity - Mean = 3.36; Std. Deviation = .89; Alpha = .71. FMQ1 ± ³, am Iamiliar Zith searching Ior inIormation online ´ 3.19 1.20 FMQ2 ± ³, am Iamiliar Zith social meGia platIorms ´ 3.38 1.17 FMQ3 ± ³, am Iamiliar Zith the processes oI searching anG getting inIormation aEout institutions online ´ 3.52 1.13 FMQ4 ± ³, am Iamiliar Zith inTuiring aEout the ratings oI institutions online ´ 3.34 1.14 Students’ enrolment intentions - Mean = 3.85; Std. Deviation = .74; Alpha = .50. ENQ1 ± ³, enroll in this institution rather than any other institution aYailaEle ´ 3.48 1.11 ENQ2 ± ³, am Zilling to recommenG others to enroll in this institution ´ 3.91 1.01 ENQ3 ± ³, intenG to enroll to this institution in the Iuture ´ 4.14 .97 Cultural distance - I feel comfortable dealing with people whose Mean = 3.80; Std. Deviation = .74, Alpha = .72. CDQ1 ± ³(thnicity is Yery GiIIerent Irom me ´ 4.13 .94 CDQ2 ± ³Nationality is Yery GiIIerent Irom me ´ 4.04 .95 CDQ3 ± ³/anguage is Yery GiIIerent Irom me´ 3.98 .96 CDQ4 ± ³Customs is Yery GiIIerent Irom me ´ 3.49 .95 CDQ5 ± ³5eligious is Yery GiIIerent Irom me ´ 3.58 .99 CDQ6 ± ³:ay oI liIe is Yery GiIIerent Irom me ´ 3.48 1.05 Notes: Short scales with less than 5 items can have alpha between .50 to .70 (Hinton et al., 2004; Piedmont, 2014). When short scales have persistent low alpha value, cross-checking the mean inter-item correlations with the optimal range of .20 to .40 (Briggs & Cheek, 1986). This study measures have low alpha value due to their nature, and the inter-item correlations were adequate. Besides, this research employed ANN that has superior predictive power(Sim et al., 2014); no mandates for multivariate assumptions such as homoscedasticity, normality, internal consistency due to ANN’s ability to handle noisy data (Abubakar et al., 2019; Sharifi et al., 2019). Thus, the low alpha scores appear not to be a major problem. Appendix B : The Analytical S teps Flowchart 112 Appendix C: R Codes # check for missing values apply(Akiledata ,2,function(x) sum(is.na(x))) # 75% for training and the 25% for testing index <- sample (1:nrow(Akiledata ),round(0.75*nrow(Akiledata ))) train1 <- Akiledata [index,] test1 <- Akiledata [-index,] # appling LM on the model lm.fit <- lm (ENR ~ EREF + EWOM + FAM + CLD, data = Akiledata) # generating results (e.g., estimates and t-values) summary(lm.fit) # testing the model with the test for Akiledata pr.lm <- predict(lm.fit,test1) MSE.lm <- sum((pr.lm - test1$ENR)^2)/nrow(test1) print(paste(MSE.lm)) ##################################################### yukleme library(relaimpo) # Calculate Relative Importance for Each Predictor calc.relimp(lm.fit,type=c("lmg","last","first","pratt"), rela=TRUE) # Bootstrap Measures of Relative Importance (1000 samples) boot <- boot.relimp(lm.fit, b = 1000, type = c("lmg","last", "first", "pratt"), rank = TRUE, diff = TRUE, rela = TRUE) # to print result booteval.relimp(boot) # to plot result plot(booteval.relimp(boot,sort=TRUE)) ####################################### # NEURAL NET FITTİNG ####################################### maxs <- apply(Akiledata , 2, max) mins <- apply(Akiledata , 2, min) scaled <- as.data.frame(scale(Akiledata , center = mins, scale = maxs - mins)) train_ <- scaled[index,] test_ <- scaled[-index,] library(neuralnet) nn <- neuralnet (ENR ~ EREF + EWOM + FAM + CLD, data = train_, hidden = c(2,2), err.fct="sse", linear.output= FALSE) plot(nn) par(mfrow=c(2,2)) gwplot(nn, selected.covariate= "EREF", selected.response = "ENR", min=-5, max=5) gwplot(nn,selected.covariate= "EWOM", selected.response = "ENR", min=-5, max=5) gwplot(nn,selected.covariate= "FAM", selected.response = "ENR", min=-5, max=5) gwplot(nn,selected.covariate= "CLD", selected.response = "ENR", min=-5, max=5) nn$result.matrix columns <- c("EREF", "EWOM", "FAM", "CLD") covariate <- subset(test_ , select = columns) pr.nn <- compute(nn, covariate, rep=1) # Next step pr.nn_ <- pr.nn$net.result*(max(test_$ENR)-min(test_$ENR))+min(test_$ENR) test.r <- (test_$ENR)*(max(test_$ENR)-min(test_$ENR))+min(test_$ENR) # Calculating MSE MSE.nn <- sum((test.r - pr.nn_)^2)/nrow(test_) #Compare the two MSEs print(paste(MSE.lm, MSE.nn)) 113 #############################Cross validation for linear model########################### library(boot) set.seed(200) lm.fit <- glm(ENR ~ EREF + EWOM + FAM + CLD, data = Akiledata ) cv.glm(Akiledata ,lm.fit,K=10)$delta[1] ############################FOR TRAİNİNG###################### set.seed(450) cv.error <- NULL k <- 10 library(plyr) pbar <- create_progress_bar('text') pbar$init(k) for(i in 1:k) { index <- sample(1:nrow(Akiledata ),round(0.75*nrow(Akiledata ))) train.cv <- scaled[index,] test.cv <- scaled[-index,] library(neuralnet) nn <- neuralnet (ENR ~ EREF + EWOM + FAM + CLD, data = train_, hidden = c(2,2), err.fct="sse", linear.output=FALSE) columns <- c("EREF", "EWOM", "FAM", "CLD") covariate <- subset(train.cv, select = columns) pr.nn <- compute(nn, covariate, rep=1) pr.nn <- pr.nn$net.result*(max(train.cv$ENR)-min(train.cv$ENR))+min(train.cv$ENR) train.cv.r <- (train.cv$ENR)*(max(train.cv$ENR)-min(train.cv$ENR))+min(train.cv$ENR) cv.error[i] <- sum((train.cv.r - pr.nn)^2)/nrow(train.cv) print(paste(cv.error[i])) pbar$step() } mean(cv.error) ############################FOR TESTİNG################## set.seed(450) cv.error <- NULL k <- 10 library(plyr) pbar <- create_progress_bar('text') pbar$init(k) for(i in 1:k) { index <- sample(1:nrow(Akiledata ),round(0.75*nrow(Akiledata ))) train.cv <- scaled[index,] test.cv <- scaled[-index,] library(neuralnet) nn <- neuralnet (ENR ~ EREF + EWOM + FAM + CLD, data = test_, hidden = c(2,2), err.fct="sse", linear.output= FALSE) columns <- c("EREF", "EWOM", "FAM", "CLD") covariate <- subset(test.cv, select = columns) pr.nn <- compute(nn, covariate, rep=1) pr.nn <- pr.nn$net.result*(max(test.cv$ENR)-min(test.cv$ENR))+min(test.cv$ENR) test.cv.r <- (test.cv$ENR)*(max(test.cv$ENR)-min(test.cv$ENR))+min(test.cv$ENR) cv.error[i] <- sum((test.cv.r - pr.nn)^2)/nrow(test.cv) print(paste(cv.error[i])) pbar$step() } mean(cv.error)