Influence of University Service Quality on Student Satisfaction and Loyalty in Bangladesh: With the Mediating Role of University Reputation and External Prestige

Evaluating the service quality of higher educational institutions in Bangladesh has become crucial to ensure continuous improvement and strategy development for tertiary education sector. The present study has aimed to investigate the influence of higher education service quality dimensions on student satisfaction and student loyalty. Moreover, the study also tests the mediating impact of university reputation and external prestige between university service quality and satisfaction and loyalty. The research work has shown the performance of selected public universities in higher education service quality dimensions. Data has been collected through a structured questionnaire from a sample of 390 graduate and undergraduate students of two public universities of Bangladesh. The study reflects that the performance on service quality dimensions of selected public universities is not satisfactory. The results of the study have shown that Teaching, Support services, Library and lab facilities and internationalization have significant influence on student satisfaction except Administrative service and Hostel facilities. Teaching, Administrative services, Library and lab facilities and Internationalization have significant influence on student Loyalty. The mediating analysis has shown that student satisfaction partially mediates between service quality and student loyalty. The results of the study have also indicated that university reputation and external prestige partially mediate between service quality and student satisfaction and loyalty. The previous studies which measure service quality of Bangladeshi higher educational institutions have applied SERVQUAL model which is not absolutely suitable for an educational institution. However, this study has applied higher education service quality model which has been uniquely developed to measure service quality of higher educational institutions.


Introduction
Service quality in the educational sector is considered by various researchers because of its importance and outcomes. Quality in higher education is a complex and multi-faceted concept. (Harvey and Green, 1993). No single definition or theory can capture the essence of 'quality' in higher education sector. In this modern era, multiple groups of stakeholders are well concerned regarding the service quality in higher education sector (Marzo-Navarro et al., 2005). The outcomes of higher education sector (e.g. skilled human resource, ethical citizen, research works etc) make major contribution for economic and social development. Notably international and national rankings are being conducted regularly that are reflecting the condition of education quality in higher education sector in various countries. Hence, Universities are in a continuous race to ensure quality education for its students (Carvalho&Mota, 2010,DeShields, Kara, &Kaynak, 2005 Tertiary institutions had to be concerned regarding how their students feel about their educational experience (Bemowski, 1991). Moreover, Tertiary educators are being called to account for the quality of education that they provide (Abdullah,2006). Customer-orientation in higher education is a generally accepted principle and performance indicators should address student experience (Owlia and Aspinwall, 1997). In recent years universities in developing nations have followed the universities of developed world in adopting quality assurance to enhance the quality of their teaching, research and direct community service programs. The adoption and successful implementation of such quality assurance programs will be beneficial for the developing countries (Lim, 1999). The higher education service is considered the most effective tool for development in less developed countries. However, there is limited theoretical and empirical work on quality of higher education in the developing countries. In addition, it is difficult to get statistical data on quality of higher education from such countries due to bureaucratic and privacy issues (Maringe and Mourad, 2012).
Numerous studies have been conducted to investigate the influence of service quality dimensions on student satisfaction and loyalty in higher educational institutions all over the world from various country contexts. ( However, these studies have not made adequate representation of the real scenario because those studies have serious limitations in various terms. For instance, those studies have adapted service quality model from traditional business practices whereas separate service quality model which is more accurate than traditional service quality model and just for higher educational institutions have been utilized in other studies. The existing studies have also been suffered from limitations in research method contents. Moreover, only satisfaction has been investigated as a dependent variable in those works (Rashid and Rahman, 2017 ;Asaduzzaman et al,2013; Rouf et al, 2016). Moreover, important variables regarding higher educational institutions and service industry have not been covered which are emphasized in other scholarly works such as university reputation (Bontis and Booker, 2015) and external prestige (Sung and Yung,2008). Therefore, the existing works are not sufficient on this issue from Bangladeshi context and insufficient to provide directions to academicians and policymakers because the higher educational institutions in Bangladesh are facing enormous issues related to quality in education. At present, the shifting nature and requirement of higher education services and a raise in competitive intensity necessitates higher performance levels in universities of Bangladesh.
To take appropriate measures about the quality improvement in universities, the policy makers must have to know regarding the existing service quality and the influencing factors that may affect student satisfaction and loyalty in higher educational institutions in Bangladesh. Hence, the current study aims to fulfil this objective by answering the following research questions: How public universities in Bangladesh are performing on higher educational service quality dimensions? Does higher educational service quality dimensions influence student satisfaction and loyalty? How university reputation and external prestige mediate between the overall service quality and student loyalty in public universities of Bangladesh?

Service Quality in Education
O'Neill and Palmer (2004) define service quality in higher education as "the difference between what a student expects to receive and his/her perceptions of actual delivery". According to Hill (1995) the evaluation of service quality in the case of a student could range from encounters with office staff, to encounters with tutors, lecturers, the head of departments, etc. Oldfield and Baron (2000) stated that students have three main criteria that need to be satisfied with services. These has been labeled as Requisite encounters which essentially enable students to fulfill their study obligations; Acceptable encounters which students acknowledge as being desirable but not essential during their course of study and Functional, an encounter of a practical or utilitarian nature.

Service Quality, Student Satisfaction and Student Loyalty
Martha and Priyono (2018) showed a significant positive influence of service quality towards student satisfaction and significant positive influence of student satisfaction towards student loyalty in 13 universities and colleges in Riau Province . Nevertheless, there is no significant and positive influence on the relationship between service quality and student loyalty. Fares et al (2010) conducted a study to test the impact of service quality, student satisfaction and university reputation on student loyalty of the students of International Islamic University Malaysia (IIUM). The study showed a significant positive influence of independent variables on dependent variable. Athiyaman (1997) showed that perceived quality is a consequence of consumer satisfaction. The study also showed the effect of perceived quality on communication behaviour. On both the analysesthat is, using the two different pre-enrolment attitude and disconfirmation measuresthe effect of perceived quality on behavioural intention is greater than that of satisfaction. The results also suggest that pre-enrolment attitude has little or no direct effect on post-enrolment attitude. An important implication of this finding for higher education is that all service encounters should be managed to enhance consumer satisfaction. This in turn would enhance perceived quality.
Mendez et al (2009) hasexamined the relative importance of relationship pathways among key factors affecting student loyalty in the following order: perceived service quality, satisfaction, trust, and commitment. The findings reveal that perceived service quality and student satisfaction do not translate directly into student loyalty, but, rather, indirectly through the mediation of trust and commitment.
Annamdevalu and Bellakmonda (2014) has conducted a study in three oldest state universities in the state of Andhra Pradesh in India to find the relationships between service quality, student satisfaction and student loyalty in higher education sector using structural equation modelling. This study tested the proposed research model and proved the mediator role of student satisfaction between service quality and student loyalty. Service quality has been found to be an important input to student satisfaction.

Service Quality in higher education of Bangladesh
Academicians, education thinkers, education researchers, education policy planners, and other stakeholders are in common consensus that the quality of higher education in Bangladesh has been deteriorating steadily, in some areas quite alarmingly, over the last two decades (Salahuddin & Aminuzzaman, 2011). Osman (2017) has conducted a study to explore the relationship between the dimensions of service quality model (SERVQUAL) and student satisfaction. Factor analysis, multiple regression, ttest, and Analysis of Variance (ANOVA) were employed to analyse data. A sample size of 119 was gathered from four private universities in Dhaka and respondents were students. Finally, 117 were found suitable for analysis. The study reveals that responsiveness, assurance, empathy, and student quality have significant influence on student satisfaction. Among these, assurance illustrated the strongest influence on student satisfaction followed by empathy and student quality. Asaduzzaman et al (2013) showed that the constructs of service quality dimensions are positively correlated with private university student's satisfaction in Bangladesh. Rouf et al (2016) found that the majority of students are satisfied with the facilities provided by universities in Bangladesh. However, the study has been conducted on 160 students of four public universities. The results of the study indicated that all the five dimensions of service quality were correlated with student satisfaction. Rashid and Rahman (2017) showed that there is a significant service gap in all the selected service dimensions such as learning, teaching, recognition, assessment system, internet and library facilities, campus life, and non-academic administration in a science and technology university of Bangladesh. The results also mirror that little more than half (54.1%) students are satisfied with the service quality of the selected university and almost similar proportion (52.1 %) of the students have future interest in studying in the same university. Based on the above-mentioned literature review we can propose the following hypotheses: H1: Higher education service quality dimensions (Teaching, Administrative service, Support services, Hostel facilities, Internationalization , Library and lab facilities) have significant influence on student satisfaction of public universities of Bangladesh.
H2: Higher education service quality dimensions (Teaching, Administrative service, Support services, Hostel facilities, Internationalization, Library and lab facilities) have significant influence on student Loyalty of public universities of Bangladesh.
H3: Student satisfaction mediates the relationship between service quality and student loyalty in public universities of Bangladesh.

University Service quality, University Reputation, Student Satisfaction and Student Loyalty
The link between satisfaction and reputation has received minimal attention and it was found that satisfaction leads to reputation (Anderson and Sullivan, 1993). Parahoo et al (2013) has showed that that the drivers of student satisfaction in the Gulf regiontend to be different from those determined in western contexts, with university reputation playing a key role. Booker and Serenko (2015) showed that organizational reputation mediates the relationship between satisfaction and loyalty. Jiewonto (2012) has conducted a case study that aimed to identify the influence of SERVQUAL to word of mouth influence mediated by student satisfaction and university image. The results indicated that SERVQUAL had a positive impact to the student satisfaction and university image, and then it impacted to the positive WOM Intention. Therefore, based on the discussion and aims of the research we can propose the following hypotheses: H4: University Reputation has a mediating role between university service quality and student satisfaction in public universities of Bangladesh.
H5: University Reputation has a mediating role between university service quality and student Loyalty in public universities of Bangladesh.

University Service quality, External Image, Student Satisfaction and Student Loyalty
An individual's beliefs about how outsiders view the organization is referred to as construed external image (Dutton et al., 1994) or perceived external prestige (PEP; Smidts et al., 2001). Sung and Yung (2008) stated that it is possible that members of the same organization have different perceptions of its external prestige. Among several attempts to assess the quality of programs in higher education, rankings and ratings generated by third parties or agencies are often used to attract prospective students. Carmeli (2005) stated that when assessing an organization's prestige, people who are external consider different aspects of the organization. On the other hand, insiders develop their own view about the organization and its values, and evaluate how outsiders view their organization. Andreasen (1997) showed that Corporate image is positively correlated with perceived quality, customer satisfaction, and customer loyalty. Sung and Yung (2008) proved that perceived external prestige has the strongest impact on students' positive attitudes.
Therefore, based on the discussion and aims of the research we can propose the following hypotheses: H6: External Image has a mediating role between university service quality and student satisfaction in public universities of Bangladesh.
H7: External Image has a mediating role between university service quality and student Loyalty in public universities of Bangladesh.
On the basis of the above-mentioned literature review and proposed hypothesis we have formulated the conceptual framework of the study as below: The conceptual framework shows that the dimensions of higher education service quality model have influence on student satisfaction and student loyalty. The figure shows that university reputation and external prestige have mediating influence between university service quality and student satisfaction. The figure also shows that university reputation and external prestige have mediating influence between university service quality and student loyalty.

Methodology
The research philosophy of this study is based on positivism. The study has developed hypotheses from previous studies and then test those hypotheses through empirical testing. This study is descriptive and quantitative in nature aiming to investigate the impact of university service quality dimensions on student satisfaction and student loyalty with the Mediating Role of External prestige and University Reputation.

Selection of Sample and Data Collection
The study is empirical and based on the primary data. The data has been collected from students of selected public universities of Bangladesh. The population of the study consists of the students of two public universities from the southern part of Bangladesh. The data collection has been carried out on a sample of 500 respondents, through the distribution of structured questionnaires in November-December, 2019. The respondents of the study have been chosen through simple random sampling. All of the respondents who have participated in the study are students of tworecognized public university in Bangladesh (Patuakhali Science and Technology University and Barisal University). A group of BBA students has been given proper instructions and sent as interviewers. The questionnaire has been compiled in such a way that extensive information can be collected on the demographic characteristics of the respondents. The questionnaire contains items regarding students' views on performance on higher education service quality dimensions in their universities, their perception on university reputation and external image, student satisfaction and loyalty toward their university. Around 500 questions have been distributed in the selected public universities of Bangladesh and 390 filled up respondents completed the questionnaire. Hence, the response rate is 78% which is quite satisfactory.

Method of Data Analysis
The statistical package SPSS (version 23.0) has been used for data analysis. The reliability and validity of the questionnaire items have been checked through Cronbach's alpha values, composite reliability, and average variance extraction values. The values of reliability and validity testing cross the cut off values to justify the measurement items.
Structural Equation model has been used to investigate the impact of impact of university service quality dimensions on student satisfaction and student loyalty with the Mediating Role of External prestige and University Reputation. The study investigates the direct and indirect impact of independent and mediating variables on dependent variables. Hence, SEM is an appropriate tool to investigate such relationships.

Scale and Measurement
Higher education service quality performance has been considered multidimensional construct consisting of five elements: Teaching, Administrative service, support service, hostel facilities and social and environmentally responsible dimensions. Each of the components have been measured reliably with multi-item scales. The researchers measure higher education service quality dimensions, student satisfaction, loyalty, university reputation and external prestige through five-point Likert scale (1 = strongly disagree" and 5 = Strongly agree. The researchers will adapt questionnaire items from Allemdevula and Bellamkonda (2016) to measure the higher education service quality performance, student satisfaction and loyalty. The measurement items used by Allemdevula and Bellamkonda (2016) can be applied in Bangladeshi context because the factors adopted by this study are consistent and common for any higher educational institution's service quality performance. Moreover, the study has been conducted on India which is a neighboring country of Bangladesh. To measure university reputation and external image the researchers will adapt questionnaire items from Sung and Yang (1997).

Results and Discussions Demographic Characteristics of the Respondents
Among the 390 respondents more than 70 % are male and rest are female. More than 80 percent of the respondents are undergraduate level students from different semesters. As more students enrolled in undergraduate levels in higher educational institutions in Bangladesh ; hence the inclusion of more undergraduate level students is appropriate here. More than 95% respondents of the study are unmarried which is quite reflective of the original scenario.

Reliability and Validity Analysis
To verify the discriminant validity and convergent validity of the variables, confirmatory factor analyses (CFA) were conducted. The measurement models showed a good fit between the model and the data. Although the chi square was significant, the values of CFI, AGFI, RMR and GFI are within satisfactory level which ensures good fit of the model. The significance of the t-values for each path coefficient of the CFA model was evaluated to ensure convergent validity (Fornell and Lacker, 1981); the construct's estimated average variance extracted (AVE) was above .50 and the composite reliability (CR) was above .70 confirming the convergent validity. According to Fornell and Lacker (1981), for a construct to demonstrate discriminant validity, all of the construct AVE estimates should be larger than the corresponding squared inter construct correlation estimates. All of the constructs passed this requirement and thus convergent validity was confirmed.  Table 1 shows the mean value of university service quality dimensions from the perspectives of public universities of Bangladesh. As far as these descriptive statistics is concerned, the performance of the universities on "Teaching" dimension is at the highest level (with a mean value of 3.46 on a 5point Likert scale) in comparison to other quality dimensions. The other five dimensions which are Administrative service, Support services, Library and Lab facilities, Hostel facilities and Internationalization have mean value below 3.00 (neutral value). The result indicates that the students consider that service quality on these dimensions is not up to the level.  Table 2 shows the summarized results of H1. The hypothesized relationships of Teaching, support services, Library and lab facilities, and internationalization are significant with 5% significance level (P value < 0.05). However, Administrative services and Hostel facilities have insignificant influence on student satisfaction level. Therefore, Hypotheses 1 is not fully supported but partially supported.  Table 3 shows the summarized results of H2. The results depict that Teaching, Administrative services, Library and lab facilities, and internationalization have significant positive influence on student loyalty level with 5% significance level (P value less than 0.05). However, support services and Hostel facilities have insignificant influence on student loyalty level(P value 0.605 and 0.095 respectively). Therefore, Hypotheses 2 is not fully supported but partially supported. Baron and Kenny approach was followed to test the mediating effects of the variables in the study to test H3, H4, H5, H6 and H7. Table 4 shows that student satisfaction has a partial mediating effect between university service quality and student loyalty. The direct impact of service quality on loyalty has coefficient value of 3.26 (P value = 0.00) but the coefficient value has reduced with the inclusion of mediatingvariable student satisfaction (coefficient value 2.22, P value=0.00). Hence, H3 has been partially accepted. Moreover, the table also shows that university reputation and external prestige partially mediates the impact of universityservice quality on student satisfaction and student loyalty. The coefficient value of direct impact of service quality to student satisfaction (4.04, P value = 0.00) has reduced (3.348, P value 0.00) when tested with university reputation (3.348, P value=0.00). The inclusion of university reputation has also reduced the coefficient value from 3.26 (P value 0.00) to 1.86 (p value 0.00) when tested with the impact of service quality to student loyalty. The mediating impact of external prestige has also showed the partial mediating impact on the influence of service quality to satisfaction and service quality to loyalty relationship. Therefore, H4, H5, H6 and H7 are all partially accepted.

Findings and Managerial Implications
The research aims to investigate the performance of public universities service quality through the application of higher education service quality model.
Moreover, the study aims to show the influence of higher education service quality dimensions on student satisfaction and loyalty. The research shows the mediating influence of student satisfaction between university service quality and student loyalty. The study also shows the mediating influence of external image and university reputation between the influence of university service quality on student loyalty and student satisfaction. The result of the study is consistent with several prior studies (Asaduzzaman et al ,2013;Rashid and Rahman ,2017;Osman, 2017;Annamdevalu and Bellakmonda ,2014;Mendez et al ,2009;Serenko ,2015;Jiewonto ,2012;Yung, 2008). However, the findings of the study are somewhat inconsistent with some other prior studies such as Rouf et al (2016).
The result of the study shows that the performance on service quality is below satisfactory below. The performance on significant dimensions of higher education service quality which are administrative service, support services, library and lab facilities, Internationalization and hostel facilities are below neutral value from the perspective of students. The performance on 'Teaching' dimension is highest in comparison to other dimensions but not at a satisfactorily level. Hence, there are lots of areas where administration and policymakers must focus to improve the performance level. This is worth mentioning that without sufficient amount of financial allocation it will be tough for university administrations to perform well in performance dimensions especially in library and lab facilities, hostel facilities and support services areas. The allocation of budget from the government for public universities are the main source of financial resource to work in such areas. However, the allocated budget for the public universities is not up to the level since there are many public universities in Bangladesh and all the universities do not get the sufficient amount of budget. Hence, universities must find alternative ways of funding to develop their performance level. Many foreign universities have strong alumni and corporate network through which a huge amount of funding has been managed. In addition to that those universities have their own earning sources and they employ students as part time students in many of their official works which practice is totally absent in Bangladesh. In public universities in Bangladesh many people are employed and lots of money has been spent there; hence, such type of cost cutting measures can assist a lot the university administrations.
The study demonstrates that Teaching, support services, Library and Lab facilities, and Internationalization have significant positive impact on student satisfaction. Therefore, authorities and academic policymakers must take appropriate steps to improve in those performance dimensions. Nevertheless, a surprising finding of the study is that administrative service and hostel facilities have no significant impact on student satisfaction. A probable reason is in Bangladesh, students of public university have very little contact with administrative service employees. Every public university in Bangladesh offers accommodation to the students with a little charge and it has become a common trend. Hence, a possible explanation for insignificant effect of hostel facilities on satisfaction is this factor has been considered a 'hygiene' factor by the students and the absence of this factor with proper performance can cause serious dissatisfaction on the students.
The study depicts that Teaching, Administrative service, Library and lab facilities, and Internationalization have significant impact on student loyalty. An important fact is 'Administrative service' has been proven as a significant factor to enhance customer loyalty although this factor has not shown significant impact on student satisfaction. This finding implies an important direction that to recommend other students and to spread positive word of mouth university policymakers should emphasize on an efficient 'Administrative Service'.
The study shows that university reputation and external prestige mediates the relationship between university service quality and student satisfaction. However, the perception of the students regarding their university reputation and external image is not up to the mark. Hence, there are lots of scopes to do positive contributions on these two factors. At present, several international rankings exist to evaluate the quality of universities which are situated all over the world. However, the position of the universities in Bangladesh in those rankings are not satisfactory. University administrations must take appropriate measures to secure a good position in those rankings. The public relation departments of the public universities in Bangladesh are not so excelled to enhance and protect the reputation and image of the universities to the outsiders. The personnel who are responsible in those positions have insufficient expertise due to the lack of training and relevant academic background. Hence, appropriate measures should be taken here to ensure that the personnel who will be responsible to work for the image of the university are expert enough. Another issue is the internet existence of the public universities is quite poor. Now a day's people search the websites of a particular university to gain information regarding teacher's profile, rankings, research projects run but a particular university etc. But most of the public universities are not concerned regarding this issue and don't update significant information which can't demonstrate the actual status of that particular university. The universities in North America and Europe maintain official Facebook and twitter pages where stakeholders can contact and get required information.

Limitations and Future Research Directions
This research is not without limitations that may create opportunities for future work. Since our findings cannot be generalized in other settings, therefore future studies could test this model with larger random samples or samples from stakeholders. Future studies could replicate the model in other higher educational institutions. While this research primarily used a quantitative approach to achieve the study objectives, incorporating a qualitative or a meta analysis approach into the proposed framework, may help future studies to develop a more comprehensive theoretical framework. The inclusion of other important factors or more in-depth study on a single service quality dimension of the university can provide useful insight. The service quality model can be adapted and implemented for primary and secondary education sector of Bangladesh to get a holistic view on the quality of overall education sector .