Application of social exchange theory on relationship marketing dynamism from higher education service destination loyalty perspective

aNational Orientation Agency, Federal Secretariat Complex, PMB 1470, Ilorin, Kwara State, Nigeria bFaculty of Management Sciences, Prince of Songkla University, P. O. Box 5, Korhong, Hat Yai, Songkhla 90112, Thailand cSchool of Business Innovation and Technoprenuership, Universiti Malaysia Perlis, Malaysia dFaculty of Business, Information Technology and Creative Arts, Toi-OHOMAI Institute of Technology, New Zealand eCollege of Law, Government and International Studies, Universiti Utara Malaysia, Malaysia C H R O N I C L E A B S T R A C T


Introduction
The global decrease in education financing, transformation from manufacturing to knowledge based economy and adoption of consumerist approach to higher education have put pressure on institutions/countries to design various retention/loyalty strategies to survive in the competitive international education market (Wu & Naidoo, 2016;Abubakar & Mokhtar, 2015a;Bowden, 2013;Maringe & Carter, 2007; Jain et al., 2011).Consequently, there is a need for countries involved in international higher education services to understand multifaceted determinants of international students' choices and destination loyalty since students choose country first before institution (Arambewela & Hall, 2009;Cubillo, Sanchez & Cervino, 2006;Ehigie & Taylor, 2009).
The use of marketing theories has plunged into higher education from the business world as part of survival strategy to cope with government disinvestment in education and multiple choice options open to students globally pose some challenges to service providers (Wu & Naidoo, 2016;Moore & Bowden-Everson, 2012;Hemsely-Brown & Oplatka, 2006).These challenges resulting from international student's mobility across the globe left each country to design coping strategies such as robust students' support schemes, reduced tuition fee, financial aids, course waivers, conducive environment and advertising/market approaches not only to attract but importantly to retain students at the destination (Abubakar, 2015;Hashim, Abdullateef & Sarkindaji, 2015;Arambewela & Hall, 2009;Naidoo & Wu, 2014;Biodun et al., 2012).Essentially, the target is to gain and sustain competitive advantage in the market.
Hemsley- Brown and Oplatka (2006) observed that countries that wish to open their boarders to international students must recognize that they need to market both the country and the institutions in a climate of international competition.The key strategy is to first understand variables that trigger prospective student's choice of a country and then higher education institution (Mazzarol & Soutar, 2002).Interestingly, the market for international education is getting more and more competitive with increased number of countries but yet less satisfaction and low level of loyalty (Prahalad & Ramaswamy, 2004;Cubillo et al., 2006;Bowden, 2013).Based on the report of UNESCO (2013), over 4 million students studied outside their home countries and projected to hit over 7 million by year 2025.
Extant literature reveals that destination loyalty has been established to enhance performance through increase in price premium, referral and growth in revenue while it reduces switching cost, lowers marketing cost and attrition rate (McMullan & Gilmore, 2008;Moore & Bowden-Everson, 2012).Surprisingly, destination loyalty is fast declining despite its importance for sustainable competitive advantage in business to customer (B2C) relationship (Gounaries & Stathakopoulos, 2004) including tertiary education services (Naidoo, 2007;Abubakar & Mokhtar, 2015;Gummeson & Gronroos, 2012;Jamaludin et al., 2016).Considering the importance of destination loyalty in competitive business environment (Suhartanto, 2011), exploring the determinants of destination loyalty from country perspective will enhance the performance of international education service industry particularly for emerging destinations (Abubakar, 2015;Ali et al., 2016;Kusumawati et al., 2010;Kusumawati, 2013).Thus, it becomes imperative for destination promoters in emerging destinations to understand the factors influencing the purchase intention of prospective students and the nature of the relationship among those factors (Cubillo et al., 2006).This is because international students can manifest their loyalty to a country in several ways: they may choose to stay and complete their programs, and they may re-enroll for next degree or even both, thus generating higher revenue for the country (Chen, 2016).They may also advocate the country, concerned by playing a powerful role in the decision making of others, thus reducing the country's marketing communication costs and also increase alumni participation (Ali et al., 2016;  Moisescu, 2014; Abubakar, 2015; Carvalho & de Oliveira Mota, 2010).

Challenging Issues for Emerging Destinations
In 2011, global education business was estimated at US$65 billion which represented roughly 3 per cent of world services export (Naidoo & Wu, 2014).This trend reflects both global competition among higher education institutions (HEIs) as well as international mobility with critical implication for attraction and retention of students at both domestic and international levels (Thomas, 2011).These global developments have challenged destination countries to apply marketing approach to education services to synthesize dominant factors influencing students' loyalty to particular destination despite threat from competing alternatives (Abubakar & Mokhtar, 2015b;Mazzarol & Soutar, 2012).
At a global level, USA which is the leading education destination in the world has more than 40 per cent of college entrants leaving Higher Education Institutions (HEIs) before earning a degree and 75 per cent of these students drop out in the first year (Bowden, 2013;Fares et al., 2013).Similarly, in Australia, 50 per cent of all attrition at the tertiary level occurs to students in the first year, while 30 per cent of these students are not satisfied with their stay (Bowden, 2013).National Education Statistics of Higher Education Sector revealed that in 2011, 48% of international students' entrance to Malaysian public universities left before graduation (MOHE, 2014).This alarming global surge and fragile situation resulting from stiff competition made countries to be concerned about loyalty in order to survive in view of students' high switching rate (Chen, 2006).
Research has revealed that 60% of international students harbor migration plans, only using study as a stepping stone to migrate either for employment or further studies (Chen, 2006(Chen, , 2008)).Students' switching rate or non-loyalty at higher institution has been associated with the level of engagement between the students and faculty officers (academic and supporting staffs), feedback processes, universities policies and facilities (Normala & Dileep, 2012).This prompted Jamaludin et al. (2016) to suggest further work on the predictors of destination loyalty.
Despite these shortcomings, more study destination countries continue to emerge at booming education market each year without unique destination characteristics (UNESCO, 2010).These countries provide greater varieties but are less able to differentiate themselves, while students also have more choices of destination but less satisfaction and less loyal to a destination (Prahalad & Ramaswamy, 2015).Like other tourism services, International education is highly price sensitive and competitive with low levels of customer loyalty (Richard & Zhang, 2012).Empirical studies have been conducted on destination loyalty particularly on tourism sites on determinants of patronage, reuse or repurchased (Oliver, 1999;Yoon & Uysal, 2005;Suhartanto, 2011;Chi, 2012;Wu, 2015;Ekinci & Hosany, 2006).Nevertheless, there is a need for more studies particularly in higher education marketing to integrate some of the variables suggested as drivers of destination loyalty.
Marketing researchers agreed that customer loyalty is a valuable asset in highly competitive market, therefore understanding factors affecting students' loyalty is important to destination countries who are seeking ways to maintain a strong customer base in international education supply (Abubakar & Mokhtar, 2015;Gummeson & Gronroos, 2012;Richard & Zhang, 2012;Henning-Thurau, Langer & Hansen, 2001;Ivy, 2001).Ironically, most of previous studies on international education concentrated on recruitment strategies of new students with less research output on loyalty (OECD, 2012;UNESCO, 2013;Hendarman, 2013;Abubakar, 2015).However, Helgesen and Nesset (2007a,b) viewed student's retention to be as important as attracting them to a study destination.

Peculiarities in Higher Education Services
Education services are grouped into primary, secondary, tertiary and adult or non-formal education (Mazzarol & Soutar, 2001).However, the focus of this study is on higher education as service industry particularly operating in highly competitive international education market.Famous researchers in the field of service marketing like Zeithaml et al. (1985) identified four major characteristics of services: intangibility, inseparability, heterogeneity and perishability.All these attributes are applicable to education services (Oldfield & Baron, 2000).For instance, education is meant to be a mental work in the mind rather than the body, people based rather equipment control which makes it intangible.Also the production and consumption of education service is simultaneous, from the lecturer to the student while growing a mutual tie and loyalty relationship between the client and service providers.Thus, university education falls into the service marketing category where service performances are situation specific or transaction specific (Hennig-Thurau et al. 2001;Schoefer & Ennew, 2005).This implies that service performs under different settings and by different individual cannot be treated as identical (Zeithaml & Bitner, 2000).The increasing competition in the international education arena makes the emphasis on quality inevitable if the goal is to attract and retain foreign students in their institution or the country (Lim, et al., 2011).The desire to address the issues of service satisfaction from the perspective of the student's need is being widely discussed by researchers using different determinants (Joseph, Yakhou, & Stone, 2005;Oldfield & Baron, 2000;Baldwin & James, 2000).

Loyalty from Education Perspective
Destination loyalty is synonymous to a process that allows a destination to be identified and differentiated from other alternatives with the construction of messages through which the provider tries to attract students to the destination for consumption (Ng, 2011;Qu et al., 2011).Basically, destination loyalty aims at accentuating the singularity of a student relationship and presents a favorable image of the mentioned provider to a target market and so the marketing communications has become important for destinations to promote their identity and image to target audiences (Chaudhuri & Holbrook, 2001;Roy & Hoque, 2015).
Similar to customer loyalty, students' loyalty is also comprised of attitudinal and behavioral components with emphasizes on mutual benefits as canvassed in social exchange theory (SET) (Abubakar, 2015;Hennig-Thurau et al., 2001).The attitudinal component can be described as tripartite, consisting of cognitive, affective and conative elements, whereas the behavioral component can be perceived as being related to decisions that students make regarding their mobility option which is action taken (Abubakar, 2015;Helgesen & Nesset, 2011).Favorable destination loyalty could encompasses continuous study (i.e.willing to complete study in the same destination/institution), re-enrollment (i.e.coming back to the same university for learning/studying some day in future after graduation or try to pass the entrance examination for admission to post-graduate program in the same university), provision of positive words-ofmouth, and encouraging friends and relatives to apply for admission in to the particular country/university (Abubakar, 2015;Sheu, 2011).

Drivers of Loyalty
Generally, the established loyalty drivers have been service quality, perceived value, and customer satisfaction (Brodie et al., 2009;Cronin et al., 2000;Ramaya & Lee, 2011;Suhartanto, 2011).These constructs have also been considered as the building blocks of customer loyalty in service industry including education (Lovelock & Wirtz, 2007;Cronin et al., 2000).But, Sumaedi et al. (2012) argued that firms that satisfied their customers but fail to create favorable intention among their customers will not make significant progress.Favorable intention is synonymous to conation loyalty (Oliver, 1999).However, due to peculiarities in different service industry, additional constructs will be required to fully explain student's destination loyalty.Thus, perceived image of the country, city and institution where service is consumed were added.Perceived image has an important role in the service industry, and specifically in the context of higher education destination loyalty, where competing services are perceived as virtually identical in term of performance, price, and availability (Schuiling & Kapferer, 2004;Suhartanto, 2011).Alves and Raposo (2010) observed that literature on country/university image as perceived by its students and how this image affects their behavior remains scarce.Thus, perceived Image of country, institution, city and people are significant in building destination loyalty (Suter & Jandi, 2008;Lee, 2004;Cubillo et al., 2006).
Interpersonal relationship provide platform to share ideas and exchange pleasantry such as word of mouth (WOM).WOM accounts for up to 60% of sales to new customer (Reichheld & Sasser, 1990).Therefore, there is a need to test the efficacy of this construct as conceptualized by Cubillo et al. (2006) to determine international students' personal judgment in decision making and loyalty behavior at study destination.Jamaludin et al. (2016) concluded that personal reasons and well-being are context-dependent and stressing the need to further examine the role of personal reasons in relation to destination loyalty, suggested further studies to consider psychological situation of the student.
Therefore, this study investigated and empirically justified synthesized major drivers of destination loyalty behavior of international students using social exchange theory.These factors are: service quality, students' satisfaction, perceived image, perceived value and personal reasons.

Literature Review
The focus of early researchers on destination loyalty was mainly towards tourism locations and associated sites.The choice variables of visiting a tourist site and revisit intention was greatly researched (Quintal & Polczynski, 2010;Chen & Tsai, 2007;Oh, 1999;Yoon & Uysal, 2005;Chen & Gursory, 2001).However, there appeared to be a slight variation on the choice factors and revisit factors.The discrepancy is largely on experience and willingness to always try a new location (Kozak, 2001).While the choice factors were grounded on motivation theories, loyalty factors use relationship marketing theories.The effects of globalization which saw education services being packaged as export commodity shift research paradigm from the direction of determining factors for choosing a study destination to factors that keep students at a particular place despite availability of competing alternatives (Hemsley-Brown & Oplatka, 2006; Li & Bray, 2007).
The research on generally acceptable framework to determine destination loyalty is still in progress by various marketing scholars as more variables are added from various sectors (Suhartanto, 2011;Nesset & Helgesen, 2009).However, empirical studies have established some variables as core determinants of loyalty while proposing integration of others.After critical review of extant literatures, this study identified the three building blocks variables for loyalty and added two new independent variables that are recommended by scholars to determine loyalty.

Service Quality in Higher Education
Service quality research is expanding and applicable to different fields of study including higher education services (Ali et al., 2012;Sultan & Wong, 2013).Critical review of extant literature on service marketing revealed two main conceptualization of service quality: Nordic and American schools of thought (Goroons, 1978(Goroons, , 2000;;Parasuraman et al., 1985Parasuraman et al., , 1988)).
Service quality researchers have continued to expand the scope and measurement of service quality from both schools of thought and even combination of both.For instance, Firdaus (2005Firdaus ( , 2006) ) specifically operationalized dimension for education services called HEdPERF.Sultan and Wong (2013) also conceptualized service quality in higher education PHEd which provides comprehensive dimensionalities of education service sector.The HEdPERF measure and the PHEd measure were conceptualized on the perception-only scale, however recent researches confirmed that service quality not only measures dimensionally but also in hierarchy (Clemes et al., 2010).The findings of these studies showed that the dimensions of service quality measures in the context of higher education depend on study perspectives (Sultan & Wong, 2013;Purgailis & Zaksa, 2012).This assertion was based on the complexity in education services and multi-stakeholders in the industry (Jancey & Burns, 2013).Therefore, defining service quality in higher education from the perspectives of students implies the "difference between what a student expects to receive and his/her perceptions of actual delivery" or the general student's experiences with different aspects of services provided during their student years (Jancey & Burns, 2013;O'Neill & Palmer, 2004).Firdaus (2006) confirmed the effectiveness of performance only measurement of service quality in higher education.Similarly, Firdaus (2005) proposed five dimensions performance only for higher education (HEdPERF).They are: academic aspects, non-academic aspects, program issues, reputation and access.He tested the scale on 409 students from 6 Malaysian public universities and found some aspect of service to be weak except access.The results of the study showed that students perceived 'access' to be the only significant determinant of service quality.Hence, further validation of this scale was proposed by Firdaus (2005).In 2006, Firdaus conducted a comparative study between SERVQUAL and SERPERF instruments and found that performance instrument of HEdPERF yielded a better reliability than SERVQUAL.Also, Brochado (2009) investigated the efficacies of both instruments and concluded that the measurement of service quality by means of the HEdPERF method yielded more reliable estimations, greater criterion and construct validity, better explained variance, and consequently HEdPERF was found to be a better fit than the other two instruments.
In a related development, Chavan et al. (2014) found that there are significant challenges with regard to the use of student expectations as the foundation for assessing students' service quality evaluations.The reason is that students were in the habit of having vague expectations, and limited prior experience of the service from which to shape their expectations.Further, students were likely to form their expectations as they consumed the service, throughout their degree program which often spanned three to four years depending on the duration of their course.The authors concluded that for student's service quality dimension to be fully captured, social benefits and Co-Creation/participation of both relational parties is important.Their result was validated using total of six focus groups discussions with a total of 36 International students and domestic third year students of a famous Australian university (Chavan et al., 2014).This research adopted HEdPERF to assess the service quality of students in Higher Education Institutions, considering it as a complete measurement instrument that is able to capture the authentic determinants of service quality within the higher education sector (Firdaus, 2006).This notion was supported by Sultan and Wong (2013) who concurred that the scale is comprehensive and considered broad range of service attributes in the contest of universities.Firdaus (2005Firdaus ( , 2006) ) HEdPERF dimensions were found to greatly influenced students' loyalty and were tested thus: H1: Service quality has positive and significant relationship with destination loyalty.

Students' Satisfaction in Higher Education
Elliott and Shin ( 2002) described students' satisfaction as "the favorability of a student's subjective evaluation of the various outcomes and experiences associated with education".Factors influencing students' satisfaction is divided into institutional and personal factors (Appleton-Knapp & Krentler, 2006;Elliott & Healy, 2001).The institutional factors are the academic and non-academic activities while personal factors are student's physical and psychological ability (Ali et al., 2016).In education context, students are treated like any other customer because of the fee paying system and marketization of higher education (Hill, 1995;Brookes, 2003).
Previous researches conducted on students' satisfaction in developed countries found that satisfied student will not only continue study but will also re-enroll for next degree and recommend the institution/country to others and speak well of the institution/country (Maringe & Carter, 2007;Nesset & Helgesen, 2009;Wilkins & Huisman, 2011).Arambewela and Hall (2009) investigated the perceptions of post graduate business students from China, India, Indonesia and Thailand undertaking study in Australia on their level of satisfaction with education service.Although the study was conducted in developed education destination, the unit of analysis was students from emerging nations and findings indicated that the importance of service satisfaction factors related to both educational and non-educational services varies among nationality groups.
A meta-analysis of students' satisfaction in higher education institution shows that student's satisfaction is a strong determinant of student's loyalty at both institutional and country level (Arambewela & Hall, 2006;Helgesen & Nesset, 2007b;Nesset & Helgesen, 2009).The previous researches focused on interrelationships between antecedence and consequence of satisfaction and sometimes measure satisfaction as mediation or intervening variable (Helgesen & Nesset, 2007b;Sheu, 2010Sheu, , 2011;;Rojas-Mendez et al., 2009).The multi-direction effect is indicating that student's satisfaction have both direct and indirect effect on loyalty.Dado et al. (2012) believe that previous studies still did not convincingly explain the interrelationship between students satisfaction and students loyalty to alma mater.Thus, they call for further research to determine the strength of the relationship between satisfaction and loyalty.

Image as perceived by Higher Education students
Image covered name, logo, architecture, variety of services, tradition, ideology and impression about people communicating with the public (Nguyen & LeBlanc, 2001;Hankinson, 2004).The image of destination is critical in shaping students' attitude towards destination, particularly behavioral loyalty such as re-enrolment (Sirgy et al., 2010).This is because a repeat visitor is different from first timer who only relied on external information sources, but a repeater is considered to be experienced and will take many factors into consideration or else will prefer to try a new location (Chen & Tsai, 2007).Generally speaking, academics and practitioners agreed on the need to market and brand places to form an image synonymous with the attraction or destination especially in tourism related businesses (Konecnik & Gartner, 2007).Four major reasons have been advocated why nations should manage their images (Stock, 2009): first, to attract visitors; second, to promote the service/product from such country; third, to generate direct foreign investment (DFI); and fourth, to gain human capital influx in form of brain gain (Stock, 2009).
The image of a place (country or institution) is considered a key factor in the student's selection and loyalty decision-making process (Fragos et al., 2015;Tasci & Gartner, 2007;Wu, 2015).International education is facing global competition among established countries and emerging destinations, and in order to succeed places must be distinctive and enhanced their image (Hakala et al, 2013).Places are evaluated not solely in terms of real or imagined attributes but also according to the "uniqueness".At best, nation branding can help a nation improve its Image, but it must be borne in mind that there are many other factors -political, economic and social -that affect the Image.Branding is not the solution if there are fundamental problems in those factors; rather the perception of Image is preferred (Fan, 2006).
The importance of Image in education context was explained by Nyugen and LeBlanc (2001) in their study of image and reputation of higher education institutions in students' retention decision.The researchers maintain that perceived Image of the institution have strong and positive influence on both students' selection of tertiary institution and also have impact on students' decision to stay for advance studies.Similarly, Cubillo et al. (2006) conceptualize image in higher education choice into three dimensions.They are: Image of the country, Image of the city, and Image of the institution.They proposed the integration of different Image perception in to the model for international student's decision making process in higher education choice.The dimensions proposed were validated in this study.
Most previous studies concentrated on institutional Image or corporate Image of the service provider to establish relationship with loyalty (Alves & Raposo, 2010;Weiwei, 2007;Helgesen & Nesset, 2007a;Simões & Soares, 2010).Despite these findings, some studies also have contrary result in the relationship between Image and loyalty.Brown and Mazzarol (2009) established indirect relationship between institution's Image and loyalty through satisfaction.Others find partial effect in their studies between the constructs (Quintal & Polczynski, 2010).However, the focus of these studies is on university providing the services leaving out the macro effect of education sector on the image of the country as a destination.More so, students and other stakeholders in education industry consider country image more when making a destination choice than decision to remain loyal (Helgesen & Nesset, 2007a;Nguyen & LeBlanc, 2001).Although, Hashim et al. (2015) considered Image as perceived by students to determine loyalty, their study was limited to only one university and few variables as determinants of loyalty.Wei and Wonglorsaichon, (2010) developed a model of foreign student loyalty studying in Thailand and found reputation to be most influential driver of loyalty.
Despite the importance of perceived image in the literature, little work has been conducted on perceived image in relation to destination loyalty in higher education (Alves & Raposo, 2010;Brown & Mazzarol, 2009;Binsardi & Ekwulugo, 2003).Extending the model of destination loyalty to include perceived image as one of its determinant in the higher education industry context will inevitably contribute to both theory and practice.In higher education service, the importance of image (institution) on students' loyalty was confirmed by Helgensen and Nesset (2007).Arambewela and Hall (2009) confirmed country image and Cubillo et al (2006) conceptualized city image.Put together, the three dimensions of image play significant role in destination loyalty of international students, thus hypothesis: H3: Perceived image has positive and significant relationship with destination loyalty.

Service Value in Higher Education from Students' Perspective
In higher education consumption, students are not actually buying the degree but the benefit that comes with the certificate such as employment, prestige, status in the society (Binsardi & Ekwulugo, 2003).Therefore, international students are often conscious of both physical and psychological gains when making decision.Interestingly, loyalty is a post consumption decision which makes evaluation critical since other factors are equally considered (Mazzarol & Soutar, 2002; Vauterin et al., 2011).Petruzzellis and Romanazzi, (2010) argued that universities can enhance students loyalty by concentrating on providing high value services.But value just like beauty is in the eye of the beholder.To explore the effect of value on student's loyalty, Hashim et al. (2015) maintained that university value is perceived to be high according to institution's relationship with the public.
The demand for value in education is multi-faceted covering money, time and convenience (Sheu, 2011).But the judgment of value is purely a prerogative of students who are at the liberty to interpret what they receive and juxtapose with what they loss.The literature is rich with a widespread evidence of the links between perceived value and destination loyalty (He & Li, 2012;He & Mukherjee, 2007;Spiteri & Dion, 2004).Sirdeshmukh et al. (2002) found some support for direct associations between perceived value and loyalty; they concluded that more research is needed to evaluate the nature of the link between perceived value and involvement (Harris, and Goode 2004).Therefore, we expect that Perceived Value relates positively to destination loyalty.Hence, the following is hypothesized: H4: Perceived value has positive and significant relationship with destination loyalty.Binsardi and Ekwulugo (2003) argued that students are not only buying degree when they venture into international studies but the benefits that comes with the degree such as employment, social status, prestige, and remuneration.Student's decision to remain at a destination will be subjected to evaluation based on the percentage of previous students that study in same country and their respective placement in the society (Brown & Mazzarol, 2009).Aside image, ease of employment is one of the strength of established educational destination countries like US and UK (Berker & Kolster, 2012;UNESCO, 2009).The assumption is once you hold certificate from certain countries, you become a potential global employee (Bostan et al., 2015).Thus, personal improvement played critical role in international students' loyalty decision (Brown & Mazzarol, 2009;Cubillo et al., 2006).Kotler and Fox (1995) referred to future employers as true customers who are conscious of "buy" highly trained students from pool of varieties in the labor market.Career prospect and higher status ranked most in Bourke (2000) research on study abroad program.

Personal Reasons in Higher education Students' Loyalty
Student's harbor different reasons for their study at chosen destination and this reason(s) constantly guide their future action or inaction (Bostan et al., 2015).If in tone with their aspiration, they will exhibit both behavioral and attitudinal loyalty in form of re-enrolling for their next degree and also recommend same destination to family and friends in accordance with social exchange theory.Social learning theory proposes that behavior is predicted by the expectancy that, if a person behaves in a certain way, that person will be rewarded by the extent that the person values the reward (McLeod & Wainwright, 2009).Most previous studies that evaluate personal reasons considered it in respect of determining student's satisfaction and not loyalty (Ali et al., 2016).However, this study relied on social exchange theory which is more encompassing in dealing with other integrated variables than social learning theory.In a study conducted on personal values, subjective well-being and destination-loyalty intention of international students, personal values that emphasize justice, equity and self-fulfillment matters to students' well-being (Jamaludin et al., 2016).Thus, the need to investigate personal reasons as determinant of destination loyalty will establish the linkage between choice and loyalty as conceptualized by Cubillo et al. (2006).
Most of previous studies on personal reasons, personal value, personal factors, individual characteristics or personality concentrated on destination selection by international students (Cubillo et al., 2006;Chahal & Devi, 2013;Zikmund et al., 2013) leaving issues relating to attitudinal and/or behavior loyalty for further research.It is therefore necessary to determine the influence of personal reason on decision to continue study, re-enroll, recommend or spread word of mouth (Ong et al., 2015;Srikatanyoo & Gnoth, 2002).Customers will choose a particular destination either to express their personality or to show case their status appropriate to the situation (Aaker, 1991).Following this, international students and their parent deeply concern about the social status of the family when making decision about study destination and directly affect their willingness to stay at a particular destination.Thus, the hypothesis: H5: Personal reasons positively and significantly influence destination loyalty.

Research Methodology
Seven Malaysian public universities with large number of international students were carefully selected.Structured questionnaire was administered on international students of these universities based on proportionate stratified simple random sampling according to descriptive research design.After preliminary analysis and data screening, a total of 498 useable samples were analyzed from 760 questionnaires used for this survey.Demographic analysis was carried out to view percentages of respondents in different thematic.Descriptive statistics was also conducted to purify the data.Finally, partial least square structural equation modeling was used to generate measurement model and structural model.

Demographic Profile of Respondents
This section describes the characteristics of those who participated in the study at individual level.The unit of analysis of this study is individual international students in selected Malaysian public universities.The features that are examined in this section include gender, age, marital status, current Program enrolled at the University, Nationality/Continent of origin, name of the University, length of enrolment for current study, status of enrolment in Malaysian University, source of funding for the program.The individual characteristics have been measured on nominal and ordinal scales.
Table 1 shows the summary of respondent's characteristics.Table 1 indicates that the respondents were mostly males, who constituted 74.1 per cent while the remaining 25.9 per cent were females.The male dominance is accounted for by the structure of global international students' population in which males account for 68% of the population (UNISCO, 2013).A more cogent reason is the gender disparity between males and females international students enrolled at Malaysia HEIs.It is reported that males represent 75 per cent of international students in public universities in Malaysia as against 25 per cent for females (MOH, 2014).As for the respondents' age, Table 1 depicts that majority of the respondents (39 per cent) are within the age range of 26-35 years, closely followed by age group 18-25 years, while the lowest are those respondents whose age is above 55 years (representing 0.6 per cent).On marital status, there seems to be no significant difference between those married and singled (each group at almost 50%).Related to age is the respondents' program, the majority of international students in Malaysian public universities are postgraduate students (Master students 34.1 per cent and PhD students 34.7 percent), while undergraduate students are less in population due to restriction on number of their admission (representing 31.1 per cent).The demographic statistics for age and program depicted above is consistent with the goal of the study which is to predict students' loyalty to their study destination, those students in the age group 18-35 years (71.3 per cent) and those on Bachelor's and master degree programs (65.2 per cent) are the more likely to have the intention to go back to the university for higher degree programs in the future.On the Nationality composition of the respondents, majority of them are Africans (42.6 per cent), followed by Asians (27.3 per cent) and Arab gulf (24.3 per cent) which is consistent with the structure of the international students' population in Malaysia.Of the 100 Countries in Malaysia, Africans are quite visible and active at their respective institutions, account for about 12 per cent of the population of international students at public universities in 2013, Asian 35 per cent, Arab, 46 per cent and the other Nationalities account for the remaining 7 per cent (MOH, 2014).
On the status of enrolment, this study captured majority of international students that registered for studies in Malaysia for the first time (64.3 per cent) suggesting that once they are satisfied there is possibility to register for the next degree in the country.Similarly, the students who register between 1-3 years are the majority (61.4 per cent) with great potential to continue if all retention variables are favorable to them.The noticeable threat to international loyalty based on demographic statistics above is the sponsorship.Most of the international students surveyed are self-sponsored students (57.6 per cent).The likely challenges could be finance even if they wish to continue their study or enroll for next degree.However, Malaysia provides reasonable cost of living and lower cost of study for international students relative to other study destinations (EMGS, 2015).

Assessment of the Measurement Model
This study used Partial Lease Square structural equation modeling (PLS-SEM) to estimate its theoretical model using the software application SmartPLS 2.0 (Ringle et al., 2005).PLS-SEM lies on two important multivariate techniques including factor analysis, and multiple regressions (Hair et al. 2010).Basically, there are two main criteria that are used in PLS analysis for the assessment of the measurement model and these include validity and reliability (Ramayah et al., 2011).In an elaborate form, researchers assessed outer model with individual item reliability, construct internal consistency and construct validity as shown in the model below.Based on the criterion given by Hair et al. (2014), all the Cronbach alpha values are more than the acceptable level (0.70) and all the composite reliability values are more than 0.80 which confirms the reliability of data, all the item loadings are 0.60 and above, all AVE values are more than 0.50; hence all these values meet the criteria for convergent validity of data as required in quality criteria (Ringle et al., 2005).
Furthermore, discriminant validity was determined using Chin's (1998) benchmark by comparing the indicator loadings with other reflective indicators in the cross loadings.First, as a rule of thumb for evaluating discriminant validity, Fornell and Larcker (1981) recommended the use of AVE with a score of .50 and above.To realize adequate discriminant validity, Fornell and Larcker (1981) further suggested that the square root of the AVE should be higher than the correlations among latent constructs.Table 3 shows that the square root of AVE is higher among the inter constructs correlations in the particular columns which indicate the discriminant validity of data.The two conditions suggested by Fornell and Lacker (1981) were met: all constructs have AVE above 0.5 and the square root of AVE above the correlation for each of the construct in a particular column.The R 2 value represents the combined effects of the exogenous latent variables on the latent endogenous variable (Hair et al., 2010;Hair et al., 2014).However, the R 2 value of the endogenous variable of the direct relationships model is presented in Table 4 below: As shown in Table 4, the exogenous latent constructs of this study (i.e., service quality, students' satisfaction, perceived image, perceived value, personal reasons) explain 60 percent variance of destination loyalty.Following Chin (1998) recommendation, the R 2 value explained by the exogenous constructs on the endogenous construct in the relationships is very close to substantial effect.It indicates that destination loyalty is 60 percent dependent on the five predictors considered in this study.The remaining 40 percent may be explained by other factors.

Assessment of the Structural Model for Hypotheses Testing
Having discussed the measurement model of this study in the preceding sections, the next step is to consider the structural model otherwise called inner model.This step allows the researcher to inspect the standardized path coefficients and the significance level with bootstrapping in order to test the hypotheses of this study.Specifically, standard bootstrapping procedure was employed using a number of 5000 bootstrap samples for 498 cases to assess the significance of the path coefficients.Bootstrapping is meant to generate the t-value that is used to test the significance of the relationship (Fig 2).
The results of the structural model based on the relationships between the predictors and criterion variables of this study are presented in Table 5.These results are interpreted using the coefficients (Beta) of the path relationship, the standard error (SE), t-value (T Statistics) and P-value.

Discussions on the findings
Hypothesis 1: Service quality has positive and significant relationship with destination loyalty.This hypothesis got strong support as Table 5 depicted, the path coefficient value is 0.516 and the corresponding t statistics is 13.731 (P<0.000) that indicates 1% significance level.From this result, it is accepted that service quality has positive and significant relationship with destination loyalty.This finding is consistent with the findings of previous researchers that used HEdPERF to measure education service quality (Sultan & Wong, 2013;Firdaus, 2005Firdaus, , 2006;;Ali et al., 2016).
Hypothesis 2: Students' satisfaction positively and significantly influences destination loyalty.The present study proves this hypothesis.The path coefficient here is 0.118 with a positive sign and this value is significant at 1% (t value; 3.135; P, <.001) level.So it is accepted that Students' Satisfaction positively and significantly influences destination loyalty.This finding is consistent with the findings of (Brown & Mazzarol 2009;Alves & Raposo, 2007;Arif & Ilyas, 2013;Helgesen & Nesset, 2011;Abubakar, 2015) who confirmed through their empirical studies that satisfaction is an important predictor of loyalty.Hence it is again proved in the context of international higher education that students' satisfaction is a strong determinant of destination loyalty.
Hypothesis 3: Perceived image has positive and significant relationship with destination loyalty.The path coefficient for this variable was 0.036 and the t statistic was 0.943 (P, 0.345).However, the findings revealed that perceived image doesn't have any significant relationship with destination loyalty, which leads to the decision that hypothesis H3a, is not accepted.This finding is inconsistent with the findings of Kandampully and Suhartanto ,2000;Helgensen and Nesset, 2007;Arambewela and Hall, 2009;and Cubillo et al (2006) who found in their studies in different contexts that perceived image is an important predictor of loyalty.
Hypothesis 4: Perceived value has positive and significant relationship with destination loyalty.This hypothesis is supported as the path coefficient got a positive value of 0.161 and this value is also significant at 1% level (P<0.000;t, 3.664).So the perceived value is positively and significantly correlated with destination loyalty in the context of higher education.This finding is consistent with the findings of previous researchers (He & Li, 2012;He & Mukherjee, 2007;Parasuraman & Grewal, 2000;Spiteri and Dion, 2004) who confirmed that perceived value is a strong predictor of loyalty in service sector including higher education.
Hypothesis 5: Personal reasons positively and significantly influence destination loyalty.This hypothesis is also supported as the path coefficient got a positive value of 0.237 with a t value of 6.392 and this value is significant at 1% level (p<0.000).So the Personal reasons are positively and significantly correlated to destination loyalty in the context of higher education.This finding is in line with the notions of previous findings on this construct (Aaker, 1991;Freling & Forbes, 2005;Govers & Schoormans, 2005;Vila-Lopez & Rodriguez-Molina, 2013) who found through their studies that personality attributes and individual reason were important factors for customers while evaluating a particular service.

Assessment of the Effect Size on the Relationships
The effect size specifies the relative effect of a specific exogenous latent variable on the latent endogenous variable based on the changes in the R 2 value as a result of excluding the former (Chin, 1998(Chin, , 2010)).Consequently, the effect size is measured using Cohen's formula (Cohen, 1988;Hair et al., 2014).
f 2 = R 2 included-R 2 excluded 1-R 2 included Based on the above formula, the F2 values of 0.02, 0.15, and 0.35, indicate small, medium, and large effects respectively (Cohen, 1988).Table 6 represents the effects size assessment of the respective exogenous latent variables on the endogenous variable in their direct relationships.As seen from the aforesaid table, all the exogenous latent constructs that significantly affect the endogenous latent variable have a small effect on the endogenous latent variable except service quality which has a large effect size.

Predictive Relevance for the Relationships
This study used cross-validated redundancy criterion to examine the predictive relevance (Q 2 ) of the exogenous latent variables on the reflective endogenous latent variable (Geisser, 1974;Hair et al., 2014;Hair et al., 2013;Ringle et al., 2012;Stone, 1974).Consequently, a model with the Q greater than zero is assumed to have predictive relevance (Henseler et al., 2009;Coelho & Henseler, 2012), and thus the higher the Q the greater the predictive relevance (Duarte & Roposo, 2010).The Q value obtained using the blindfolding procedure is presented in table 7 below.(Chin, 1998;Hair et al., 2014).

Theoretical and Practical Contributions
The conceptual framework of this study was derived from the past empirical evidences and theoretical gaps identified in the literatures.The framework was supported and explained from theoretical underpinnings, the social exchange theory of relationship marketing (Blau, 1960(Blau, , 1964;;Lambe et al., 2001).Expectedly, a new theory emerged that form a major contribution of this study and empirically justified.This theory is code named '2S2P' and it has been recommended for destination loyalty studies particularly at emerging education destinations.'2S' stands for service quality and students' satisfaction, while '2P' stands for perceived value and personal reasons which together determines destination loyalty.The '2S2P' model of this study contributes to the growing body of knowledge in understanding the determinants of destination loyalty for education sector by establishing a theory-based empirical link between loyalty and its drivers.The current study extended SET by investigating a broad range of fundamental RM dynamics of marketing variables otherwise known as the building block of customer loyalty in service marketing (Lovelock & Wirtz, 2007).Similarly, instead of validating the research framework in popular contexts such as banking, retail business, hospitality, manufacturing, or restaurants, the study extend SET'S application by empirically testing the model in the context of HEI -a pivotal industry that has received scant research attention by RM researchers.In addition, prior empirical studies have not paid the desired attention to the validation of the construct of personal reasons integrated into the model of this study perhaps due to the conceptual ambiguity associated with the word personal reasons or in some literatures called personal value (Cubillo et al., 2006;Brown & Mazzarol, 2009) and the difficulty in its practical implementation by service providers (Binsardi & Ekwulugo, 2003;Berker & Kolster, 2012).This study filled this void with empirical justification.Another important contribution was recorded from service quality.The SERVQUAL dimensions was not a stable instrument in education sector, thus this study made important contribution to body of knowledge by revalidating HEdPERF modified by Firdaus (2006) and Sultan and Wong (2013) as reliable dimension for HEIs.
The findings of this study vividly implies that to achieve students' satisfaction and good impression (image and value) which are prerequisites to student's loyalty, the university should focus on identifying individual personal reasons of choosing the institution.This strategy will strengthen ties with students including interpersonal communication with students, rapport, and alumni activities with student's participation in decision making roles which are hallmark of loyalty.Specifically, university management should enhance student-university ties through the development of social interaction based on familiarity, competence, rapport, care and confidence that can spur them to look back to the country even after completion of studies.The managers responsible for the recruitment of international students should be wary and expect different reactions from actual and prospective students of different cultural backgrounds.University students are not likely to respond uniformly to university adverts, appeals and incentives.While acknowledging the fact that it will be difficult for university management to discern the differences in individual student's concerns, it is managerially safer to provide both short term and long term inceptives to students in the quest for students' retention and loyalty.The study findings encourage service providers to reexamine how they can facilitate unique education experiences and the settings.

Limitations and Recommendations
Notwithstanding the many insightful findings of the present study, the results need to be interpreted with consideration to a number of limitations.Firstly, the study adopted a cross-sectional survey instead of using a longitudinal approach of data collection which will cover a long period.Further studies could apply longitudinal and compare the findings.Secondly, data for this study was collected only from Malaysian public universities to the exclusion of other HEIs particularly the private higher education institutions which hitherto was driving Malaysia international education sector for decades.Future studies may wish to collect data across the entire spectrum of the HEIs and compare their results with the present findings.A meaningful future research direction could be an attempt to compare results from the different components of the HEI to see whether variations could exist.Thirdly, many of the scales used in the current study were adapted from prior studies conducted in contexts other than higher education.Thus, future studies in the context of HEI are needed to revalidate these measurement scales.

Conclusion
While several studies abound on the link between relationship marketing dynamics and destination loyalty, the present study, however addressed theoretical gaps by testing the relative impact of personal reasons which was hitherto rarely investigated.Also, different dimensions of image as perceived by the students were tested.However, image dimensions (institution, city and country) proved to be insignificant as influencer of destination loyalty.Though the focus of the study was on emerging destinations, other education exporting countries stands to benefits from the contributions of this study to gain and sustain competitive advantage.Implementation of the findings of this study can help destination countries curb ever increasing students' switching rate at study destination.

Fig. 1 .
Fig. 1. 2S2P Measurement Model Above model shows items reliability, construct reliability and construct validity.Table 2 shows the figures generated from the PLS algorithm for items loading, composite reliability (CR), Cronbatch Alfa and average variance extraction (AVE).

Table 1
Demographic Characteristics of Respondents

Table 6
Assessment of the Effect Size (F-Square)

Table 7
represents the blindfolding result of the cross-validated redundancy (Q 2 ) of the latent endogenous variable of the direct relationships model of this study.As this cross-validated redundancy (Q 2 ) is greater than zero, it clearly indicates the presence of path model predictive relevance