Impact of Organizational Citizenship Behavior on Perceived Service Quality in Hotel Industry

Though the prior studies have examined the service quality and organizational citizenship behavior in hotel industry, limited understanding exists on how the service quality could be enhanced by effective organizational citizenship behavior (OCB) of customer-contact employees of unclassified hotels. This study concerns about how OCB of customer-contact employees would impact the perceived quality of the customers. Accordingly, it is endeavored to empirically examine the impact OCB of customer contact employees on the service quality in hotel industry. A total of 70 unclassified hotels were selected randomly from the Southern Province, Sri Lanka. The sample consisted of 215 customer-contact employees and 375 customers of these hotels. Structured questionnaires were administered for data collection. Smart-PLS 3 package was used for running the analysis of SEM. The results indicate that the impact of service oriented organizational citizenship behavior on service quality is statistically significant (β = 0.444, p < 0.01) and it is important to note that service-oriented loyalty on service quality appear to be strong. A conceptual and empirical link between OCB and perceived service quality (SQ) of customers is examined. The critical roles of the customer contact employees in ensuring high level of customer services in hotel industry is established.


Introduction
The hotel industry is expanding and becoming one of the competitive sectors in Sri Lankan service industry (Hamsanandini and Park, 2018). In addition, hotel industry has to face challenges for finding ways to retain their customers and attract new customers (Ambepitiya & Dharmasiri, 2017). The Sri Lankan hotel industry is currently confronted with many challenges imposed by the turbulent and rapidly changing external environment that it operates. Among these challenges, increasing competition among rivalry firms, increasing trend of the arrival of budget travelers, and global economic downturn are very notable in the global context (Semasinghe, 2016). In the Sri Lankan context, the advent of new technology, new communication, and entering new competitors in the sector are identified as key challenges emerged in the service industry (Jayawardhana, Silva and Athauda, 2013). The competition has forced hotels to find ways and means to retain current customers and attract new customers (Tsaur and Lin, 2004). As customer expectations and customers' subsequent demands are often subjected to the change, retaining current customers and attracting new customers becoming difficult (Khan, Hussain and Yaqoob, 2013).
The quality of the service has become a critical factor in service organizations to dealing with competitive market (Dominic et al., 2010;Al-Ababneh, 2017). One of the strategies of service organizations that can be used to obtain competitive advantages is to improve the quality of the services that are provided to the customers (Wong & Shoal, 2003;Dominic et al., 2010;Desfitrina, Zulfadhli and Widarti, 2019). Increasing competition in the service industry and changing customer expectations of service cause the service organizations to provide quality service to remain in the competitive market (Kandampully, 1998;Wu and Ko, 2013;Wijethunga, 2016). Service organizations concern about quality of the service to win the customer loyalty and improving organizational performance (Ueltschy et al., 2004).

RESEARCH ARTICLE
The quality of service depends on how employees deal with customers. Therefore, promotion of service quality depends on the way of managing behavior of employees Schneider et al., 1993;Morrison, 1996). Mmutle and Shonhe (2017) emphasized that if employees have a positive attitude, caring and making sure that every customer gets needed attention and the customers feel that they are valued, it therefore would mean that their expectations are met and lead to improve quality of the service provided. Behavioral pattern of employees in the organization can be identified in two ways as "extra-role" service behavior (organizational citizenship behavior) and "role-prescribed" service behavior (Bettencourt and Brown, 1997;Vandaele and Gemmel, 2006;Tang and Tang, 2012;Christensen and Whiting 2018). Extra role behavior refers to the discretionary behavior of customer contact employees in serving customers that extend beyond formal duties appearing in the job descriptions Zhu, 2013). It is extremely important to give extra attention to the customers. An exceptional behavior during the service encounter may enhance quality of the service (Bettencourt & Brown, 1997;Bitner et al., 1990). Customer contact employees should understand expectation of customers and choose how to respond to the expectation of customers. Service organizations also want to select customer contact employees based on behavioral characteristics of the job applicants. That would benefit the organizations to achieve organizational objectives. Extra role behavior leads to delivering customer services according to organizational requirements resulting in improving customer perception of service quality (Bienstock et al., 2003). Drawing from the social exchange perspective, employees' perception of organizational support would produce a sense of obligation to care about the organization and would help to achieve organizational objectives (Eisenberger et al., 1986). If employees perceive that organization provides more support, they are likely produce extra role behaviors . Extra role behavioral responses of customer contact employees can affect customers' judgment of service quality (Bitner, 1990;. However, many researchers have highlighted that limited number of empirical research on extra role behavior and performance relationship (service quality) (Podsakoff & MacKenzie, 1997;Vandaele and Gemmel, 2006). The large and evergrowing research literature has identified a number of factors that may potentially impact on service quality. Therefore, it is very difficult to decide which factors most significantly influence service quality (Ueno, 2010). This study concerns about how extra role service behavior of customer contact employees would affect the perceived quality of the customers.

Research Problem
The tourism sector is one of the largest and rapidly growing service sector industries in Sri Lanka (Kandewatta, Fernando and Silva, 2018). In addition, tourism has become one of the sources of income generation for the economy of Sri Lanka. According to the Central Bank Annual Report (2018) indicated that Tourist arrivals grew by 10.3 per cent during 2018, while earnings from tourism grew by 11.6 per cent to US dollars 4,381 million in 2018 in comparison to US dollars 3,925 million in 2017. The tourism sector continues to perform well and was able to retain its rank in the third level as one of the main sources of foreign exchange earner of the national economy (SLTDA, Annual Statistical Report, 2016). However, due to the dynamic nature of tourism, many changes are taking place in the industry. The changing the political and economic stability after ending 30 years long war has flashed a green light for the tourism industry to grow very faster than earlier (Jayawardhana, Silva & Athauda, 2013). The government of Sri Lanka has paid attention to make new avenues such as destination planning, tourism gateway towns and online awareness campaigns open for the development of tourism sector (Sri Lanka Tourism strategic plan 2017-2020). Accordingly, the tourism industry gained the highest priority in the post war development strategy (Semasingha, 2016). The hotel sector contributes only around 2 percent to the country's Gross Domestic Product (Central Bank of Sri Lanka, 2011). The increased number of hotels Journal of the University of Ruhuna, Sri Lanka 7(1), 2019 has strengthened the competition. Customers now have more choices of hotels offering a wide range of facilities at competitive price. Customers always would expect to gain better service quality for the money that they spend in the hotel (Kandampully & Suhartanto, 2000). Customer satisfaction and customer retention are crucial for the service organization, which required the provision of high quality service (Bowen and Ford, 2002). In the case of hotels, customer contact employees are considered as a unique asset since they are involving in service delivery process from arrival of customer and until when customer check-out the hotel (Bienstock et al., 2003). The customer determines the quality of the service, the way customer contact employees interact with customers in service encounter (Ma & Qu, 2011). There are different types of behaviors that are importance for service excellence. These behaviors are beyond specified by formal job requirements. These behaviors are refereed as organizational citizenship behavior (extra-role behaviors). Over the past decade, a great deal has been written on OCB. However, these studies have ignored the relationship between OCB and organizational outcomes such as service quality (Morrison, 1996). Many researchers have highlighted that limited number of empirical research on OCB and service quality relationship (performance) (Podsakoff & MacKenzie, 1997;Bettencourt, Gwinner and Meuter, 2001;Vandaele and Gemmel, 2006). According to the Quarterly Report of the Sri Lanka Labor Force Survey (2018) indicated during third quarter of 2018, the total number of employed persons in Sri Lanka is estimated at about 8.0 million. Of which, about 46.0 percent engaged in Service sector, 28.7 percent in Industries sector and 25.3 percent in Agriculture sector. Therefore, it is important to study OCB in the service sector in dealing with customers effectively. This indicates importance of studying OCB in service sector in Sri Lankan context. This study concerns about how OCB of customer contact employees would impact the perceived quality of the customers. It is logically argued that customers perceived the quality of service quality based on behavior of employees in the hotels. Therefore, this study endeavors to examine the impact of OCB on the perceived service quality in the unclassified hotels.

Organizational Citizenship Behavior
The concept of organizational citizenship behavior (OCB) was initiated by Barnard (1938), when he was studying system approach to analyze the organization system. He had identified two types of systems in the organizations: formal and informal. The formal system includes rules, regulations and procedure of the organization where individuals develop relationship to achieve organizational goals. The informal systems make foundation of the concept of OCB. According to Organ et al, (2006), the informal system is about the contributions by individuals that go beyond the content of contractual obligations, obedience to legitimate authority or calculated striving for remuneration as mediated by the formal organization. The concept of OCB was discussed in organizational literature in the 80's (Bateman and Organ, 1983, Smith et al., 1983and Hua & Yu, 2010 and has been widely recognized in belief, that behaviors of employees enhance organizational effectiveness (Podsakoff & Mackenzie, 1997). The OCB is about individual behavior that is discretionary, not directly or explicitly recognized by the formal reward system, and that in the aggregate promotes the effective functioning of the organization ).
Many researchers have given different views on the dimensionality of OCB (Williams and Anderson, 1991, Graham 1991, Van Dyne et al., 1994. According to Bolino et al., (2002) OCB is employees' tendency towards motivation to go beyond formal job necessities in order to help each other, to align individual interest with organizational interest, and to have real concern towards general activities and the mission of the organization". This definition indicates two types of characteristics: the OCB cannot be strengthened directly and expected extraordinary behavior from employees to achieve organization success (Tang and Tang, 2012;Christensen and Whiting 2018). OCB have a direct influence on the relationship between customer-contact employees and in turn, customer perception of service quality. This behavior indirectly influences on inter-organizational variables such as service environment, teamwork, and communications. In fact, this will improve the productivity of the employees through which quality Journal of the University of Ruhuna, Sri Lanka 7(1), 2019 of the service would improve (Castro et al, 2004). According to Mackenzie, Podsakoff & Ahearne (1998), OCB includes discretionary behaviors on the part of an employee that directly promote the effective functioning of an organization, independent of an employee's objective productivity. Many researchers identified OCB concept as an extra-role behavior (Schnake, 1991;Van Dyne, Cummings andMcLan-Parks, 1995, Bachrach andJex, 2000;Tang and Tang, 2012;Christensen and Whiting 2018). Van Dyne et al., (1994) highlight that OCBs are deriving behaviors that developed relationship with goals of the organization and its employees, and it provides facilities to achieving organizational goals. Accordingly, an important element of OCB can be identified; it is optional type of behavior of employees, it is not directly related to rewards or punishment, and it is important for effective performance of the organization.
Bettencourt, et al., (2001) have developed a typology of service-oriented OCB (SO-OCB). The SO-OCB consisted of three dimensions: loyalty, service delivery, and participation. Bettencourt, et al., (2001) argued that, this specific form SO--OCB is more appropriate for service industry since customercontact employees play special role in dealing with customers in the hotel environment. The SO-OCB can enhance a situation where improved service quality delivered by the service organizations, develop sound customer relationship, higher customer satisfaction and higher customer retention (Tang & Tang, 2012). This present study considers OCB (Bettencourt, et al., (2001) and its impact on service quality of unclassified hotels.

Perceived Service Quality
Service quality represents a customers' assessment of the overall level of service offered by an organization . This assessment is based upon the perceptions of formulated during service encounter (Bitner et al., 1990, Johnston, 1995. Service encounter denotes interaction between customer and employees (Bettencourt & Brown, 1997;Winsted, 2000). Service quality has been variously defined based on meeting needs and requirement, and how well the service delivery matches customers' expectations. Perceived service quality is a global consumer judgment or attitude relating to the service and results from comparisons by consumer of expectations of service with their perception of actual service performance (Berry, Zeithamal, & Parasuraman, 1985;Gronroos, 1984). According to Bitner & Hubbert (1994) service quality is a customer's overall impression of the relative inferiority and superiority of the organization and its service.
The service quality is a complex concept as it is a sum of activities that representing customer's perception about a service performance (Rossiter, 2002). Perceived quality refers to consumer's judgment about the performance of product or service (Zeithmal, 1987). Service quality is highly subjective and cannot be objectively measured due to its abstraction and elusiveness (Holbrook & Corfman, 1995;Karatepe, Yavas, & Babkus, 2005). The interactions between customer contact employees and customers are important in customer value creation process (Bell & Menguc, 2002). Customer contact employees are central in determining service quality in the hotel industry (Zeithamal, 1987) Behaviors of service employees are important in deciding perceived service quality of the customers (Farrell, Souchon & Durden, 2001).

Organizational Citizenship Behavior (OCB) and Perceived Service Quality
The researchers interested in discussing organizational citizenship behavior in the view of the believing that these behaviors enhance organizational effectiveness (Podsakoff & Mackenzie, 1997). As such, some OCB studies have focused on its antecedents (Organ et al., 2006) while some studies have focused on outcomes of OCB (Podsakkoff, et al., 2009). Kats & Kahn (1966) also highlighted OCB is necessary for effective organization performance. Researchers have claimed that if employees exert more discretionary effort and engage in OCB that favorably influence customer's perception of service quality (Morrison, 1997;Berry, 1999). Customers have become an important factor when employee performance is defined (Bowen & Waldman, 1999). The quality of the interaction between employee and customer is critical in determining customer satisfaction. Borman & Journal of the University of Ruhuna, Sri Lanka 7(1), 2019 Motowidlo (1993) observed that OCB is appropriate for service organizations since it has special requirements in dealing with customers. Considering the nature of influence of OCB on service organizational performance, Bettencourt et al., (2001) has developed a typology of service-oriented OCBs for service setting. The behavior of front-line employees plays an important role in shaping customer's perception of service quality (Liao & Chung, 2004).
The hospitality scholars also emphasized that organizations should have employees with good attitude, who can provide excellent customer service (Kim McCahon, & Miller 2003;Marsh, 1994). Organizational citizenship behavior is important in context of service organization and the five distinct dimensions of OCB: conscientiousness, altruism, civic virtue, sportsmanship, and courtesy capture many of the discretionary behaviors on service quality (Morrison, 1997;Bowen & Lawler, 1992). Koys (2001) has found that OCB influenced customer satisfaction generation of revenues and quality of service.
The present study considered the service-oriented OCB typology developed by Bettencourt et al., (2001). Accordingly, a positive relationship between serviceoriented OCB and service quality is expected i.e. first, OCB can have an immediate effect on customer perceptions then it leads to the perceived service quality and second, OCB may have effect on service quality through internal factors of the organization such as work environment, team cohesiveness and service climate. Bienstock, Moranville & Smith (2003) also highlight three reasons on why OCB would affect customer perceived service quality: First, OCB is a non-specific behavior that is serviceprovider may go beyond the standard guidelines for employee-customer interaction; second, OCB is an independent individual initiative that would enhance interaction with customer; third, OCB would help smooth function of service delivery process which would influence perceived quality of customers. All these depict that customer-contact employees' OCB is significantly linked to service quality (Bell & Mengus, 2002). Sun, Aryee & Law (2007) also reveal that service-oriented OCB is strongly related to service quality strategy. The conceptual framework of this study (see Figure 1) illustrates the impact of OCB on perceived service quality of customers. According to conceptual framework, the dependent variable is perceived service quality and independent variable is OCB. Accordingly, the study hypotheses that there is a positive relationship between OCB and customers' perceived service quality in unclassified hotel industry in Sri Lanka.

Methodology Study Design and Methods
This study focuses on the impact of organizational citizenship behaviors of employees on the perceived service quality of customers of the hotel industry. The study was conducted in the Southern Province, Sri Lanka. Unclassified hotels operating in Sothern Province, Sri Lanka were considered. Unclassified hotels comprised small hotels and guest houses, are important accommodation providers offering limited services and products as compared to the classified hotels and unclassified hotels will continue to form and important part of the hotel industry in Sri Lanka since it represented 43.52% of the accommodation capacity (SLTDA, Annual Statistical Report, 2019). The Southern Province, Sri Lanka was selected by considering its vital economic, social and cultural aspects which have led to be one of the dominant tourist destinations in the Sri Lanka. Further, it has a vast potential resource base for the development of tourism which has diverse attraction to appeal the interest of tourists (Fernando & Sharif 2013). The Ruhuna Tourism Bureau database (2017/2018) was utilized to select 150 unclassified hotels randomly, located in Southern Province Sri Lanka. A total of 100 hotels out of 150 agreed to participate in the survey. The basic criterion of selecting these hotels was, at least five employees should work in the hotel. The respondents of this study were customers and customer-contact employees of these hotels. The customer-contact employees are identified as employees who make frequent face-to-face contact with customers. The service quality of the hotels depends heavily on the effectiveness of customercontact employees who deal with customers (Morrison, 1996;. Previous studies of organizational citizenship behavior-perceived service quality of customer have been used individual (employees), customers, or units/organizations as a unit of analysis (Gracia, Cifre & Grau, 2010;Parasuraman, Zeithamal & Berry, 1988b). Thus, the unit of analysis of the present study is the respondents (i.e. customer contact employees and customers).

Sample and Data Collection
The present study required data pertaining to the OCB of customer contact-employees and perceived service quality of the customers of the unclassified hotels in the Southern Province Sri Lanka. Customer-contact employees who made frequent face-to-face contact with customers were invited to participate in this study. Two questionnaires were administered for data collection: one questionnaire for customer-contact employees and the other one for the customers of the hotels. The questionnaire developed for employees consisted of two sections: 'Section A' focused on employee demographic aspect and 'Section B' focused on employee OCB. The questionnaire developed for the customers also consisted of two sections: 'Section A' focused on customer demographic aspect and 'Section B' focused on the perceived service quality. The questionnaire for customer-contact employees was used for assessing OCB of employees while the questionnaire of customers was used for measuring perceived service quality of customers.
Direct access to the customer-contact employees and the customers of the hotels was not permitted, due to a policy decision imposed by the hotels. Thus, the questionnaire packets were handed over to each selected hotel by the researchers. The hotel Management was asked to collect data at least from five customers and three customers-contact employees. Each packet contained 12 questionnaires (five questionnaires for employees and seven questionnaires for customers). A covering letter was attached to each questionnaire explaining the objective of the study and the important of the study. A total of 100 questionnaire packets was distributed to 100 agreed hotels, contained 1200 questionnaires (500 questionnaires for employees and 700 questionnaires for customers). The administrative officers and human resource managers of the hotels agreed to get questionnaires completed in two-week time. Only 40 hotels had completed a total of 335 questionnaires (125 questionnaires from employees and 210 questionnaires from customers). The hotel managers of the rest of the hotels requested for another two weeks to complete the questionnaires. However, only 30 hotels had completed a total of 250 questionnaires in the second round of the data collection (90 questionnaires from employees and 165 from customers). Table 1 summarizes the sample composition of the study. A total of 210 usable questionnaires from customer-contact employees and 350 usable questionnaires from customers were used for data analysis. The questionnaires of customercontact employees and questionnaires of customers were coded and matched employee's questionnaires with questionnaires of customers. Prior to the data Journal of the University of Ruhuna, Sri Lanka 7(1), 2019 analysis, the questionnaires from customers, collected by the hotel management were averaged and then paired with the questionnaire from the customercontact employees.

Study Variables and Questionnaire Design
As per the above discussion, this study defines OCB as intentional activities/unintentional undertaken by the employees on their own initiative to contribute to organizational success while helping other employees. This indicates that un-intentional activities are not included in their job description as well; the employees do not receive any remuneration for performing these activities. The OCB can positively affect successful service delivery and, in fact, OCB enhances customer perception of service quality. The quality of delivering services depends on the behavior of customer-contact employees (Yoon & Suh, 2003;Hartline & Ferrell, 1996). The service quality in a hotel depends on absolutely on the ability of an operations of employees and how far employees to deal graciously with guests in all situation (Jones and Cotiis, 1986). The employees must have characteristics of innovation and perform spontaneously (Kats, 1964) and this type of behavior is beyond employees' job description (Wener, 2000).
The construct of organizational citizenship behavior is a higher order construct of 20 measurement items with three lower order constructs (i.e. SO-loyalty, SOparticipation, and SO-service delivery) (Bettencourt, et al., 2001;and Nasurdin et al., 2014). And the construct of perceived service quality is one of the higher order constructs with five lower orders constructs (i.e. Tangible, reliability, responsiveness, assurance and empathy), measured using 35 measurement items Farouk & Chris, 1991;Olorunniwo et al., 2006). The questionnaires of the present study were developed based on these constructs.
This study used Five-Point Likert scales with '1' for strongly dis agree to '5' for strongly agree. Table 2 shows the variables used in the present study. A total of 17 measurement items were deleted from OCB construct (two measurement items from Loyalty, five measurement items from -participation and 10 measurement items from SO-service delivery), as the outer loadings of these measurement items were below the threshold value of 0.7 . Accordingly, only Loyalty dimension was retained for the study. The outer loading of 22 measurement items of perceived service quality was below the threshold value of 0.7. As such, 22 measurement items were deleted from perceived service quality construct (seven measurement items from Tangibles, two measurement items from Reliability, four measurement items from Responsiveness, five measurement items from assurance and four measurement items from Empathy).

Results and Discussion
Structural Equation Modeling (SEM) was used for data analysis. Structural equation modeling techniques are capable of two most common type of analysis: confirmatory factor analysis and the estimation of a series of structural equations (Hair et al., 1998). There are several distinct approaches to SEM: the first approach widely applied covariance-based structural equation modeling (CE-SEM). The second approach partial least squares (PLS). The third approach is a component-based SEM known as generalized structural analysis (GSCA). Another way to perform SEM is called nonlinear universal structural relational modeling (NEUSREL) which is widely used in academia (Wong, 2013). These methods are used in different scenarios; CB-SEM has been widely used in the field of social sciences during past several decades (Peres, Moreira and Mesquita, 2019). CB-SEM is applicable, when sample size is large, the data is normally distributed, and the model is correctly specified (Hair, Ringle & Sarstedt, 2011). PLS-SEM is selected in this study as the most appropriate method of analysis for three reasons: first, the present study deals with a model that is not yet tested in the present research context; second, the model is not specified and third, the sample size of the study is small. In Journal of the University of Ruhuna, Sri Lanka 7(1), 2019 addition, PLS-SEM technique as it enables to incorporate unobservable (latent variables) variables measured by indicators (Chin, 1998). Smart-PLS 3 package was used for running the analysis of SEM. Accordingly, steps wise items which do not meet the required criteria were removed for validating the model. PLS Algorithm provides the figures (outer model loading) for the items which are the indicators of latent variables. Items which have outer model loading below 0.7 were deleted from the model. Two step procedures were used in analyzing the collected data, as suggested by Anderson & Gerbing (1988). First, outer model (measurement model) was examined to measure the reliability and validity of the constructs. Next, inner model (structural model) to measure the strength and direction of the relationships among the constructs.
The study constructs can be specified as reflective or formative. Reflective measures dictate all indicator items are caused by the same construct, and indicators associated with a particular construct should be highly correlated with each other . As discussed in the operationalization of variables, dependent and independent variables in the present study were specified as reflective measurements based Journal of the University of Ruhuna, Sri Lanka 7(1), 2019 on classical test theory, which claims that measures represent the effects of an underlying construct . Reflective model should be assessed with respect to reliability and validity (Hair et al., 2011;Henseler et al., 2009). Composite reliability (CR), indicator loadings, average variances extracted (AVE) and Cross-loadings were used to assess the reflective model. Fornell-larcker criterion was used to examine the discriminant validity. Table 03 shows the reliability for each indicator using standardized loadings and its p-values, average variance extracted (AVE) and the composite reliability (CR) of the latent construct. The standardized outer loadings of each items range from 0.771 to 0.941, exceeding the minimum required acceptable value (Chin, 1998 andHair et al., 2010). The standardized outer loadings exceed the value of 0.7 or higher (Hair et al 2010).

Assessment of Reliability and Validity
The internal consistency reliability (ICR) of multiitem scales were measured by Composite reliability (CR). All the constructs met the recommended value of 0.6 (Henseler et al., 2009), suggesting that the scales were reliable. The lowest reported CR value was 0.852. The CR 0.933 of perceived service quality (QS) and 0.852 of organizational citizenship behavior (OCB) exceed the recommended threshold value of 0.7 (Chin, 1998). This shows that all the constructs have high level of internal consistency reliability.
The present study examines two types of validity: convergent and discriminant. PLS-SEM uses AVE as a criterion to assess convergent validity of the latent constructs. An AVE value of at least 0.5 designates sufficient convergent validity. As shown in Table 03, the AVE for OCB is 0.658 and for perceived service quality is 0.520. Discriminant validity refers to the extent to which a construct is truly distinct from other constructs (Hair et al., 2010). The preset study tested the discriminant validity of the latent variables using the Fornell-Larcker criterion. This suggests that a latent variable shares more variance with its assigned indicators than with any other latent variable (Table 4). Journal of the University of Ruhuna, Sri Lanka 7(1), 2019  Table 04 shows that the values in the diagonal (the square root of each construct's AVEs), are greater than the off diagonal elements in their corresponding row and column (correlation with other constructs). Table 5 shows cross-loading for each indicator. The indicators' outer loading on the associated construct is greater than all of its loading on other construct . The cross loading and the Fornell-Larcker (1981) criterion indicate the constructs' discriminant validity. Moreover, Table 6 shows the Heterotrait-Monotrait (HTMT) Ratio results, ranging from 0.398 to 0.870 which is below the threshold value of 0.9 (Gold et al., 2001), indicating that discriminant validity of the study constructs is assured. Furthermore, HTMT inference bootstrap confidence interval (CI) in Table 6 indicates that HTMT values are significantly different from one concluding that the discriminant validity has been established among the constructs.

The Structural Model
The reliable and measurement model permit the evaluation of the structural model. The structural model represents the detailed picture of the results for testing the hypothesis of the study. Tolerance value of 0.20 or lower and a VIF value of 5 and higher respectively indicate a potential collinearity problem (Hair, Ringle & Sarstedt, 2011). This study uses one predictor variable (OCB) hence, Therefore, collinearity is not an issue for the estimation of the PLS path model.

The Model Predictive Accuracy
The Present study used the coefficient of determination (R 2 ) to measure the models' predictive accuracy. The R 2 represents the amount of variance in the endogenous constructs explained by all of the exogenous constructs. According to Chin (1998), the R 2 value range from 0 to1; higher value indicates higher level of predictive accuracy. R 2 value of 0.26, 0.13 or 0.02 for endogenous latent variables are Journal of the University of Ruhuna, Sri Lanka 7(1), 2019 considered as substantial, moderate or weak respectively (Cohen, 1992). The R 2 of 0.197 of the present study shows OCB explains 19.7% of the variance of the endogenous construct of SQ which can be treated as moderate predictive validity.

The Model Predictive Relevance
Predictive relevance of the model was assessed using the stoner-Geisser (Q 2 ) criterion. The Q 2 criterion suggests that the model must be able to provide a prediction of the endogenous latent variables' indicators. A Q 2 value larger than zero indicates that the exogenous constructs have predictive relevance for the endogenous constructs under consideration (Hair, et al., 2013). This study used a blindfolding approach using an omission distance of 6 to calculate the Q 2 value. A Q 2 value of 0.088 of the present study suggests that the model has predictive relevance for SQ.

Hypothesis Testing
The hypothesis of the study was tested using the significance of the individual path coefficient (beta). The bootstrapping procedure was adopted with 5000 re-samples to test the significance of the regression coefficients. The results of table 8 indicate that the impact of OCB on SQ is statistically significant (β = 0.444, p < 0.01). Therefore, the hypothesis of the study is accepted  (2018) The PLS model estimation (Figure 1)

Conclusion
The endeavor of the present study was to examine the impact of organizational citizenship behavior on perceived service quality of hotel industry in Sri Lanka. Respondents of this study were customercontact employees who are working in the unclassified hotels and customers who being served in these hotels. This study investigated the conceptual and empirical link between SO-OCB and perceived service quality of customers. This study contributes to understand critical roles of the customer contact employees in ensuring high level of customer service. It is important to note that SO-loyalty on service quality appear to be strong.
The results obtained from testing hypothesis in this study indicated positive and direct impact of SO-OCB on SQ. This finding confirms the results of previous studies (Bell and Menguc, 2002;Babakus, et al., 1996;Harris et al 2006). Particularly, a favorable organizational climate, positive organizational behavior of employees, employees' mental energy and employee's 'sprit', increase motivation for learning for behavioral change.
Bell and Menguc (2002); Liao, & Chuang (2004) emphasized that there are reasons for positive relationship between OCB and service quality; first, OCB can have an immediate effect customer perception as they are experienced in the actual employee-interactions. Second, OCB can have positive effects on service quality through their impact on internal to the organization. According to Husin, Chelladurai & Musa, (2012), OCB was very highly correlated with perceived service quality. This study highlights the importance of service employees' behavior in customers' formulations of service quality perception. The behavioral construct offers precise guidance for organizational practice that can enhance service quality. Management can concentrate on hiring Journal of the University of Ruhuna, Sri Lanka 7(1), 2019 employees who display appropriate behavioral characteristics and design training program to enhance employees' particular behavior towards OCB. Future research can focus on examining the impact of managerial inputs (empowerment of service employees and leadership) upon the quality of frontline employee's performance. Managers should pay particular attention to the tangible aspect of employee performance in creating favorable perception of overall service quality as customers seek tangibles cues to describe the service encounter. As (Bitner, 1990;Gronroos, 1984) indicated that, customer-contact employees must deliver both technical quality and functional quality in the hotels service. However, much research is needed to examine relationship between OCB and customer satisfaction, employee turnover / retention customerretention and customer complaints (Podsakoff & Mackenzie, 1997). Findings of our study highlight several implications for hotel industry. If organization can provide customer-relationship trainings to the employee that would help to willingly appear to assist customers.
Work behaviors of employee can be categorized as OCB only when they are completely voluntary and culminate in beneficial outcomes for both the individual employees and the organization as a whole. Through the review of existing literature, it was found that there are thee construct to measure service oriented organizational citizenship behavior OCB of employees in the service sector i.e. loyalty, participation, and service delivery. Loyalty is more critical dimension linked to voluntary behaviors of employees. Many pervious researchers used this concept as mediating variable in their studies. Thus, the present study enabled to develop a model considering OCB as an independent variable and open a new way of understanding of direct relationship to the service quality of the organizations. Thus, the study was conducted among frontline employees in one service sector. Future researchers can consider other categories of employees who serve in the hotel industry. This may delimit generalization of findings to other service sectors. The present study also used single cross-sectional data. Data was collected at one point in time. Due to the cross-sectional design used, the causality could not be tested. Thus, future studies may use longitudinal designs to examine the casual relationships among the study variables.