Interrelations between Service Quality Attributes , Customer Satisfaction and Customer Loyalty in the Retail Banking Sector in Bangladesh

The main objective of this study is to find the interrelationships between service quality attributes, customer satisfaction and customer loyalty in the retail banking sector in Bangladesh. The study sought to identify the most important attributes in bank settings, which may be used to review characteristics of the banks as experienced by customers. In Bangladesh, no study has yet investigated the above mentioned interrelationship. The purpose of this study is to fill this gap.


Introduction
The main aim of this study is to find the interrelationships between service quality, customer satisfaction and customer loyalty in one frame in the retail banking sector in Bangladesh.
The banking industry in Bangladesh is highly competitive.According to International Finance Corporation -IFC (2009), the financial sector in Bangladesh consists of a central bank (Bangladesh Bank), 4 state-owned banks, 5 Government-owned specialised banks, 30 Domestic Private Commercial Banks, 9 Foreign-Commercial Banks, and 29 Non-Bank Financial Institutions.6562 branches of 49 banks operate in the country (Bangladesh Bank, 2006).According to IMF (2007), Private Commercial Banks command the greatest market share in terms of both industry assets and deposits.In late 2006, the contribution of financial sector assets to GDP was roughly 69%; out of them, the banking sector accounted for 58%.Data shows that retail banking is one of the most important sectors in the Bangladesh economy.The interrelationships between service quality, customer satisfaction and customer loyalty may provide creative ideas for improving services in order to gain a competitive advantage in the retail banking sector in Bangladesh.
Service quality is one of the critical success factors that influence the competitiveness of an organization.A bank can differentiate itself from competitors by providing high quality service.Service quality is one of the most attractive areas for researchers over the last decade in the retail banking sector (Avkiran, 1994;Stafford, 1996;Johnston, 1997;Angur et al, 1999;Lasser et al, 2000;Bahia and Nantel, 2000;Sureshchandar et al, 2002;Gounaris et al, 2003;Choudhury, 2008).However, this study will again investigate the factors that enable banks to attract and maintain their customers.In Bangladesh, customers in the banking sector are in a strong bargaining position due to the significant growth of banks.Therefore, banks have to provide service carefully because of the availability of banks.Banks have to improve the service level continuously.There is no guarantee that what is excellent service today is also applicable for tomorrow.To survive in the competitive banking industry, banks have to develop new strategies which will satisfy their customers.That is why service marketing and bank marketing are important areas in the marketing literature.
Customer satisfaction is one of the important outcomes of marketing activity (Oliver, 1980;Surprenant and Churchill, 1982;Spreng et al, 1996;Mick and Fournier, 1999).In the competitive banking industry, customer satisfaction is considered as the essence of success.According to Hofstede (2001), most of the Asian cultures (like India, Pakistan) are collectivist [People in the collective cultures discriminate in-groups (relatives, institutions and organizations) and out-groups].In this case, word of mouth (WOM) advertisements are important for the banks.File and Prince (1992) argued that the customers who are satisfied tell others about their experiences and this increases WOM advertising.In this way, banks can increase customers.Prabhakaran (2003) mentioned that the customer is the king.High customer satisfaction is important in maintaining a loyal customer base.To link the service quality, customer satisfaction and customer loyalty is important.Kumar et al (2009) stated that high quality of service will result in high customer satisfaction and increases customer loyalty.Heskett et al (1997) argued that profit and growth are stimulated primarily by customer loyalty and loyalty is a direct result of customer satisfaction.Parasuraman et al (1988) and Naeem and Saif (2009) found that customer satisfaction is the outcome of service quality.Caruana (2000) developed a mediational model that links the service quality and service loyalty via customer satisfaction and applied this model in the retail banks in Malta.The results appear to prove the links between service quality, customer satisfaction and customer loyalty.In Bangladesh, however, no study has yet investigated these links.The purpose of this study is to fill this gap.A review of literature has been conducted to explore these relationships and then a research methodology has been proposed.A positive correlation between each of the constructs has been confirmed by the review of literature and the primary analysis.

Objectives of the study
The objectives of the study are as follows: To identify the critical factors of service quality in the retail banking sector in Bangladesh To identify the critical factors of customer loyalty in the retail banking To identify the interrelationships between service quality, customer satisfaction and customer loyalty in the retail banking sector and to identify the benefits of this relationships.

Scope of the study
The scope of the study is: The study is specific only to retail banking sector in Bangladesh

Theoretical Framework
The study will draw up on a range of theoretical frameworks.Service quality is discussed in relation to the SERVQUAL model (Parasuraman et al., 1988).Links between service quality, customer satisfaction and customer loyalty are examined in relation to the Mediational model by Caruana (2000).Gronroos (2000, p.46) defined service as, "A service is a process consisting of a series of more or less intangible activities that normally, but not necessarily always, take place in interactions between the customer and service employees and/or physical resources or goods and/or systems of the service provider, which are provided as solutions to customer problems".Fogli (2006, p.4) define service quality as "a global judgement or attitude relating to a particular service; the customer's overall impression of the relative inferiority or superiority of the organization and its services.Service quality is a cognitive judgement".Mudie and Pirrie (2006) identified the following features of services:

Intangibility
Intangibility is the main feature of service.Service cannot assure the quality because it cannot be counted, measured, tested, verified and inventoried in advance of sale.Most of the firms find it difficult to understand how customers consider their services and evaluate the quality of their services.

Inseparability (or simultaneous production and consumption)
There is a marked distinction between physical goods and services in terms of the sequence of production and consumption.

Variability (or heterogeneity)
An unavoidable consequences of simultaneous production and consumption is variability in performance of a service.The quality of the service may vary depending on who provides it, as well as when and how it is provided.

Perishability
Services cannot be stored for later sales or use.As services are performances they cannot be stored.If demand far exceeds supply it cannot be met, as in manufacturing, by taking goods from a warehouse.Equally if capacity far exceeds demand, the revenue and/or value of that service is lost.
In the competitive business world, service quality is considered as a competitive factor of the Organizations.Moreover, it is also considered as an essential determinant that allows an Organization to differentiate from other Organization.It helps an Organization to gain sustainable competitive advantage.

Service Quality Model
Parasuraman et al (1985) undertook a Qualitative Research to investigate the concept of Service Quality.They arranged an in-depth interview with the executives and Focus Group interviews with customers to develop a model of Service Quality.They proposed the following Service Quality model according to their research: According to figure 1 [see appendix 2], the lower portion shows the gaps revealed by the executive interviews (the marketer side) and the upper portion of the of the figure shows the gaps revealed by the customer interviews (customer side).According to their findings, the five gaps are as follows: Marketer Side: GAP 1 -Consumers' Expectations -Managements' Perceptions Gap (will impact on the customers evaluations of service quality) GAP 2 -Management perceptions of consumer expectations-Service Quality Specifications (will impact on the service quality from the customers viewpoint) GAP 3 -Service Quality Specifications-Service Delivery (will impact on the service quality from the customer's standpoint) GAP 4 -Service Delivery-External Communications (will impact on the service quality from the customer's standpoint)
In 1988, Parasuraman et al arranged a quantitative Research.They revealed an instrument for measuring consumers' perception of Service Quality, after that it became known as SERVQUAL.They collapsed their dimensions from ten to five.The dimensions were: Tangibles -physical facilities, appearance of personnel and equipment Reliability -ability to perform the promised service dependably and accurately Responsiveness -willingness to help customers and provide prompt service Assurance -Assurance (combination of items designed originally to assess Competence, Courtesy, Credibility, and Security) -ability of the organization's employees to inspire trust and confidence in the organization through their knowledge and courtesy.
Empathy -Empathy (combination of items designed originally to assess Access, Communication, and Understanding the customer) -personalized attention given to customer.
Organizations can use SERVQUAL in various ways.Parasuraman et al (1988) mentioned that SERVQUAL can help the Service and Retailing Organizations in assessing the expectations of customers and Service Quality perceptions.It can focus on the core areas where managers need to take attention and action to improve Service Quality.

Criticism of SERVQUAL
Much criticism emerged against the SERVQUAL.Some of the reviewed criticism of SERVQUAL is as follows: Carman (1990) suggested that the five service quality dimensions are inconsistent in cross sectional analysis.He found that some of the items loaded different components when compared to different service providers.As mentioned earlier, Parasuraman et al (1988) converted Understanding and Access component into Empathy.
Carman did not find it appropriate combinations in his research.Carman also noted that the difference between expectations and perceptions concept is operationally difficult to follow.He suggested that future researchers should analyse the expectation and perception at the individual level.Babakus andBoller (1992) supported Carman's (1990) idea about the dimensions of Service Quality.He found that the Service Quality dimensions are under investigation depending on the type of service.He identified that there are some operational problems in the expectations and perceptions gap analysis.Brown et al (1993) argued that the "difference score" (perception minus expectation) has some operational problems.Therefore, they suggested that a "non-difference score" measure is superior to "difference score" measure.
After the criticism of Brown et al (1993), Parasuraman et al wrote an article in the same year where they proved that non-difference score measure is debatable.Brown et al (1993) mentioned that SERVQUAL mean was 0.82 and non-difference score measure mean was 4.51.Parasuraman et al (1993) argued that 0.82 is the ideal standard of expectations because it implies that the average respondents' perceptions fell short of their expectations.In contrast, the mean of 4.51 draws the opposite conclusion.It raises the validity question of non-difference score measure.
In 1992, Cronin and Taylor criticised Parasuraman et al (1988) conceptualization of service quality.Parasuraman et al (1988) described service quality as ".......similar in many ways to an attitude."So, managers and researchers could get more information if the construct measurement was conformed to an attitude-based conceptualization.Therefore, they suggested nullifying the expectation portion from the SERVQUAL.They argued that only performance dimensions could predict behavioural intensions and they termed it as SERVPERF.Gilmore (2003) summarised the criticism of SERVQUAL is as follows: The gaps model -some researchers mention that there is a little evidence that customers assess service quality in terms of performance and expectation gaps.
Dimensionality -SERVQUAL's five dimensions are not universal.The number of dimensions comprising SERVQUAL is contextualized and there is a high degree of intercorrelation between the five dimensions.
Expectations -some researchers argue that measuring expectations is unnecessary.If they are to be measured, expectations and perceptions should be measured on a single scale.
Item Composition -four or five items cannot capture the variability within each SERVQUAL dimension.
Scale Points -the seven-point likert scale is flawed.The mid-range numbers can only be vaguely related to varying degrees of opinions and many respondents may rate these differently.
Polarity -the reversed polarity of items on the scale causes respondent error.In the SERVQUAL instrument some items are reversed to ensure that respondents do not fall into the habit of marking the same scale point for each question; however this can cause confusion.

Service Quality in Retail Banking
Service quality is important in the retail banking sector.Some of the reviewed literatures are presented below: Bahia and Nantel (2000) conducted a research to develop a valid measurement of perceived service quality in the Retail Banking sector in Canada.They argued that the SERVQUAL approach has not except from critics; therefore, they developed a new measurement for perceived service quality in Retail Banking.They proposed a scale that was called Bank Service Quality (BSQ).It comprises 31 items classified across six dimensions as: effectiveness and assurance, access, price, tangibles, range of services offered and accuracy and reliability.They proved that the dimensions of BSQ are more reliable than the dimensions of SERVQUAL.Glaveli et al. (2006) stated that BSQ is more reliable than SERVQUAL.Stafford (1996) conducted research to identify the core elements of BSQ and identified seven attributes in assessing BSQ.They are as follows: Bank atmosphere -environment of the bank including the attitude of the staffs.
Relationship -it indicates the personal relationship with the bank employees.
Rates and charges -an individual's perception of BSQ is affected by the low cost and high interest rates.
Available and convenient services -it indicates the full range of available services, convenient and easily accessible.
ATMs -it indicates the availability of the automatic teller machines.
Reliability/honesty -it emphasises on the solid bank ratings and reliable, honest staff.
Tellers -enough and accessible tellers.Angur et al (1999) examined the applicability of alternative service quality measure in the Retail Banking industry in India.They conducted their research on the consumers of two major banks in India.They use SERVQUAL model to measure the overall service quality.They found that all the dimensions are not equally important in explaining variance in overall service quality.The result indicated that responsiveness and reliability seem to be the most important dimensions followed by the empathy and tangible dimensions; whereas, assurance appears to be the least important dimension.Finally, they concluded that SERVQUAL is the best measure of service quality in banking industry.
The applicability of the SERVQUAL measure is well established in the retail banking industry.As mentioned earlier, Angur et al (1999) stated that SERVQUAL is the best measure of service quality in the retail banking industry in the developing country.Most of the researchers use the SERVQUAL measure or the modified SERVQUAL measure in the retail banking industry.
From the above discussion it can be concluded that SERVQUAL is still suitable as an assessment tool to measure the service quality perceptions in the retail banking industry, whether it is based on difference score, gap score or performance only.

Customer Satisfaction
Customer Satisfaction is one of the most important outcomes in the marketing literature.It serves to link processes culminating purchase and consumption with post purchase phenomena such as attitude change, repeat purchase, and brand loyalty (Surprenant and Churchill, 1982).This definition is supported by Jamal and Naser (2003) and Mishra (2009).
According to Oliver (1980), the customer satisfaction model explains that when the customers compare their perceptions of actual products/services performance with the expectations, then the feelings of satisfaction have arisen.Any discrepancies between the expectations and the performance create the disconfirmation.
Oliver (1980) identified three types of disconfirmation.They are: Positive disconfirmation occurs when Product/service performance expectations.In this case, the customers are highly satisfied.
Negative disconfirmation occurs when Product/service performance expectations.In this case, the customers are highly dissatisfied.
Zero disconfirmation occurs when Product/service performance = expectations.
Based on the above literature and definition of satisfaction, Giese and Cote (2000) identified the following components of satisfaction: Customer satisfaction is one kind of response (cognitive or emotional) The response emphases on a particular focus (product, consumption experience, expectations etc.) The response occurs at a particular time (after choice, based on accumulated experience, after consumption etc.)

Customer Satisfaction in Retail Banking
In the competitive business market, many firms are focusing on their efforts on maintaining a loyal customer base.Most of the retail banks set their strategies towards increasing satisfaction and loyalty of customers through the quality of service.Devlin (2001) pointed out that "customers perceive very little difference in the services offered by retail banks and any new offering is quickly matched by competitors." Zaim et al (2010) find out that tangibility, reliability and empathy are important factor for customer satisfaction, whereas responsiveness and assurance are important factor, found by Mengi (2009).Kumar et al. (2010) and Lai (2004) found that assurance, empathy and tangibles are the important factor, and on the other hand, Baumann et al. (2007) found that tangibles are not related to customer satisfaction and Ahmed et al. (2010) find out that empathy is negatively related to customer satisfaction.Researchers have identified various determinants of customer satisfaction in the retail banking sector.Arasli et al (2005) pointed out that reliability dimension of SERVQUAL has the highest impact on customer satisfaction in Greek Cypriot banking industry, whereas reliability is not related to customer satisfaction, found by Chaniotakis and Lymperopoulos (2009).According to Levesque and McDougall (1996), competitive interest rate is one of the important determinants of customer satisfaction in retail banking sector.They found that a good "employee-customer" relationship can increase the satisfaction level.They pointed out that problem-recovery is important to maintain the customer satisfaction.However, the results did not confirm that satisfactory problem-recovery can increase satisfaction.At least, it can maintain the satisfaction level.Finally, they concluded that competitiveness and convenience of the banks are the two important determinants of customer satisfaction.On the other hand, Jamal and Naser (2003) found that convenience and competitiveness are not the critical factors for all gender, age and income groups.

Relationship between Service Quality and Customer Satisfaction
In marketing literature, Service Quality and Customer Satisfaction have been conceptualized as a distinct, but closely related constructs.There is a positive relationship between the two constructs (Beerli et al., 2004).The relationship between customer satisfaction and service quality is debatable.Some researchers argued that service quality is the antecedent of customer satisfaction, while others argued the opposite relationship.Parasuraman et al (1988) defined service quality and customer satisfaction as "service quality is a global judgement, or attitude, relating to the superiority of the service, whereas satisfaction is related to a specific transaction".Jamal and Naser (2003) stated that service quality is the antecedent of customer satisfaction.However, they found that there is no important relationship between customer satisfaction and tangible aspects of service environment.This finding is contrasted with previous research by Blodgett and Wakefield (1999), but supported by Parasuraman et al (1991).
Most of the researchers found that service quality is the antecedent of customer satisfaction (Bedi, 2010;Kassim and Abdullah, 2010;Kumar et al., 2010;Naeem and Saif 2009;Balaji, 2009;Lee and Hwan, 2005;Athanassopoulos and Iliakopoulos, 2003;Parasuraman et al 1988).Yee et al (2010) found that service quality has a positive influence on customer satisfaction.On the other hand, Bitner (1990) and Bolton and Drew (1991) pointed out that customer satisfaction is the antecedent of service quality.In 2004, Beerli et al supported this finding.Beerli et al mentioned a possible explanation is that the satisfaction construct supposes an evaluative judgement of the value received by the customer.This finding is contrasted with most of the researchers.

Customer Loyalty
Singh and Sirdeshmukh (2000) suggested the customer loyalty as "the market place currency of the twenty-first century".Ndubisi (2005) and Pfeifer (2005) pointed out that the cost of serving a loyal customer is five or six times less than a new customer.This statement shows the importance of customer loyalty.Walsh et al. (2005) mentioned that it is better to look after the existing customer before acquiring new customers.Gee et al. (2008) stated the advantages of customer loyalty are as: The service cost of a loyal customer is less than new customers They will pay higher costs for a set of products; and For a company, a loyal customer will act as a word-of-mouth marketing agent According to Foss and Stone (2001), customer loyalty relates to what customers think and do (or try to do).Most customer loyalty experts would agree that loyalty is best defined as a state of mind, a set of attitudes, beliefs, desires etc. Loyalty is developed by approaches which reinforce and develop a positive state of mind and the associated behaviours.The exchange of information is one of the keys of loyalty, and provides a critical bridge between state of mind and behaviour.Loyal customers are more likely to give information to the service provider (because they trust the service provider and expect from the service provider to use the information with discretion and to their benefits).Managing loyalty is important because it means not only managing behaviour but also managing a state of mind.Dick and Basu (1994) and Bloemer et al (1998) stated that most research on customer loyalty has focused on brand loyalty; on the other hand, a limited number of researches on customer loyalty have focused on service loyalty.Bloemer et al (1998) argued that the findings in the field of brand loyalty did not generalise to service loyalty for the following reasons: Service loyalty is dependent on the development of interpersonal relationships as opposed to loyalty with tangible products.
In case of services, the influence of perceived risk is greater.
Intangible attributes such as confidence and reliability are the important factors to maintain the customer loyalty in the service context (Dick and Basu, 1994).Dick and Basu (1994) identified the two dimensions of loyalty (relative attitude and repeat patronage behaviour) and four categories of loyalty: loyalty (positive relative attitude, high repeat patronage), latent loyalty (positive relative attitude, but low repeat patronage), spurious loyalty (low relative attitude, high repeat patronage), and no loyalty (low on both dimensions).Salegna and Goodwin (2005) mentioned that if the attitude of a customer is poor within the industry, then an organization which has just better than "poor" could elicit a positive customer "relative attitude" score and high repeat patronage.Garland and Gendall (2004) supported Dick and Basu's loyalty typology in the case of personal retail banking.They mentioned that a loyal customer uses few banks, the highest likelihood of increasing business with the main bank, and the lowest probability of defection from that bank.Regression analysis disclosed that attitude and behaviour were the most important determinants of the number of banks used.Palmer et al (2007) suggested [cited from Reichheld, 2003] that in addition to measuring repurchase manager should also consider the likelihood of a customer recommending an organization's product or service to another customer.The willingness to recommend can be seen as a measure of the level of attitudinal customer loyalty.Levesque and McDougall (1996) pointed out that by increasing loyalty, a retail bank: Decreases its servicing cost (i.e.customers do not open or close their accounts)

Fulfils customers need and gains knowledge of financial affairs
Has an opportunity to cross-sell existing and new products and services.

Interrelationships between Service Quality, Customer satisfaction and Customer Loyalty
Researchers argued that service quality has influence on customer satisfaction and generates customer loyalty (Chang et al., 2009).Some of the reviewed literatures are as follows: Zeithaml et al (2008) developed a conceptual model that correlates Service Quality, Customer Satisfaction and Customer Loyalty in one frame.According to the model, service quality is the outcome of reliability, assurance, responsiveness, empathy and tangibles.Customer satisfaction is influenced by the quality of service & product and price as well as the situational and personal factors.Heskett et al (1997) developed a model which is known as "Service Profit Chain".In this model there is a direct and strong relationships between profits; growth; customer loyalty; customer satisfaction; the value of goods and services delivered to customers; and employee capability, satisfaction, loyalty, and productivity.
The link in the chain is as follows: Profit and growth are stimulated primarily by customer loyalty Loyalty is a direct result of customer satisfaction Satisfaction is largely influenced by the value of services provided to customers Value is created by satisfied, loyal and productive employees Employees' satisfaction, in turn, results primarily from high quality support services and policies that enable employees to deliver results to customers.Heskett et al (1997) suggested that in service settings, the relationships were self-reinforcing.That is, satisfied customers contributed to employee satisfaction, and vice versa.

Interrelationships between service quality, customer satisfaction and customer loyalty in Retail banking sector
Several researchers attempt to find the interrelationships between service quality, customer satisfaction and customer loyalty in the retail banking sector.Researchers argue that service quality and customer satisfaction are the predictors of customer loyalty (Tariq and Moussaoui, 2009;Han et al., 2008;Ehigie, 2006).Veloutsou et al. (2004) found the positive relationship among service quality, customer satisfaction and customer loyalty in the banking sector in Greece.Caruana (2000) proposed a meditational model that links the service quality to the service loyalty via customer satisfaction.This model is supported by Santouridis and Trivellas, 2010;Cheng et al., 2008;Bei and Chiao, 2006;Lewis and Soureli, 2006;Butcher et al., 2001. Caruana (2000) mentioned that service quality, customer satisfaction and service loyalty are related to each other.The proposed model is in appendix 3.
He applied this model to the customers of retail banking in Malta.The results proof the above mentioned relationship.The result confirms that SQ is an important input of CS and depicts 53% of its variance.He stated that management should mainly focus on CS, of which SQ is an important antecedent.Finally, he concluded that a better understanding of this model can help managers ensure better targeting of limited marketing resources.Johnston (1997) conducted a research to identify the critical determinants of SQ and to examine the effects of SQ on CS.He found that responsiveness (one of the dimensions of SQ) is one of the important factors of CS, as previously supported by Avkiran (1994), Beery et al (1985) and Bitner et al (1990).
From the above discussion it can be concluded that there is a positive relationships between each of the study constructs.Most of the researchers pointed out that there is a positive relation between the service quality attributes and customer satisfaction and customer satisfaction is positively related to customer loyalty.Caruana (2000) found that service quality is positively related to loyalty mediated by customer satisfaction.This study attempts to test this relationship in the retail banking sector in Bangladesh.

Research Methodology
This section explains the methodology of research to develop the answers to the research questions.At first the research questions are addressed and then the hypotheses have developed.Next a sampling frame and data collection procedure has been discussed.Finally this section tests the validity and reliability analysis of the data.

Research questions
This study addresses the following questions after analysing the literature review: What are the specific service quality attributes that influence customer satisfaction in the retail banking sector in Bangladesh?What are the specific predictors that affect the customer loyalty in the retail banking sector in Bangladesh?How the proposed model contribute in the retail banking sector in Bangladesh?

The Research Model
A research model is developed to find out the interrelationships between service quality, customer satisfaction and customer loyalty in the retail banking sector in Bangladesh.The research model is as follows: Insert figure 1 here Figure 1 demonstrates that the hypothesized relationships between the constructs i.e. service quality, customer satisfaction and customer loyalty.The hypothesized relationships are as follows:
As mentioned earlier in the literature review that in marketing literature, service quality and customer satisfaction have been conceptualized as a distinct, but closely related constructs.There is a positive relationship between the two constructs (Suh and Pedersen, 2010;Saunders and Petzer, 2010;Saha, and Theingi, 2009;Rod et al., 2009;Al-hawari, 2008;Hsu and Hsu, 2008;Eakuru and Mat, 2008;Beerli et al., 2004).The development of hypothesis can be as follows: H1aa: There is a positive correlation between tangibles and customer satisfaction in the retail banking sector in Bangladesh H1a0: There is no correlation between tangibles and customer satisfaction in the retail banking sector in Bangladesh H1ba: There is a positive correlation between reliability and customer satisfaction in the retail banking sector in Bangladesh H2b0: There is no correlation between reliability and customer satisfaction in the retail banking sector in Bangladesh H1ca: There is a positive correlation between responsiveness and customer satisfaction in the retail banking sector in Bangladesh H1c0: There is no correlation between responsiveness and customer satisfaction in the retail banking sector in Bangladesh H1da: There is a positive correlation between assurance and customer satisfaction in the retail banking sector in Bangladesh H1d0: There is no correlation between assurance and customer satisfaction in the retail banking sector in Bangladesh H1ea: There is a positive correlation between empathy and customer satisfaction in the retail banking sector in Bangladesh H1e0: There is no correlation between empathy and customer satisfaction in the retail banking sector in Bangladesh

Customer Satisfaction -Customer Loyalty
There is a strong positive correlation between customer satisfaction and customer loyalty (Donio et al., 2006;Story and Hess, 2006;Cheng et al., 2008).Most of the researchers found that customer satisfaction is the predictor of customer loyalty (Faullant et al., 2008;Leverin and Liljander, 2006;Terblanche, 2006).Pont and McQuilken (2005) found that Customer satisfaction and customer loyalty are related to each other, furthermore, they pointed out that Satisfied customers are not always loyal customers.Al-Wugayan and Pleshko (2010) and Pleshko (2009) pointed out that there is no relationship between customer satisfaction and loyalty.Al-Wugayan and Pleshko (2010) mentions that their findings contrast with most of the researchers.They mention that their findings is only applicable for banks as related to Mutual funds.Based on most of the researchers finding, the development of hypothesis (Customer satisfaction -customer loyalty) can be as follows: H2aa: There is a positive correlation between customer satisfaction and customer loyalty in the retail banking sector in Bangladesh H2a0: There is no correlation between customer satisfaction and customer loyalty in the retail banking sector in Bangladesh

Selecting Samples
The main objective of this research is on the interrelationships among service quality, customer satisfaction and customer loyalty in the retail banking sector in Bangladesh.Therefore, the sample for this study was selected from the retail bank customers in Bangladesh.Anderson et al (1998) suggested that the sample size of 100-200 is adequate for the structural equation modelling.

Data collection procedures
Data were gathered from the retail banking customers in Bangladesh.A set of questionnaire distributed to retail bank customers in Bangladesh.Five post graduate students (interviewers) administered the survey.They administered the survey in Dhaka, Comilla and Feni during two weeks period in July and August, 2010, between 25th July to 7th August, 2010.The first part of the questionnaire consists of the general information of the respondent.Service quality attributes were used in the second part, which is the independent variable of this research.The third part of the questionnaire explains the customer satisfaction and this is the independent/dependent variable of this research.The final part consists of customer loyalty and this is the dependent variable of this research.The interviewers explained each part of the questionnaire to the respondents.Finally, all the questionnaires were posted to the UK.

Operationalization of variables
The variables are operationalized in this research as follows: The Service Quality Attributes consist of five components: Tangibility, Reliability, Responsiveness, Assurance and Empathy.These components are based on SERVQUAL model.This study uses multiple item scale items adapted from Mengi (2009), Caruana (2002) and Cronin and Taylor (1992).
Customer satisfaction is measured using a single scale item.The single scale item adapted from Jamal and Naser, 2002;Mittal andKamakura, 2001 andCronin andTaylor, 1992.The construct of customer loyalty is based on Caruana (2002).(Note: all the parts of the questionnaire use 7-point Likert scales, ranging from "1 -Very dissatisfied" to "7 -Very satisfied")

Validity and Reliability issues
3.6.1 Validity Hair et al. (2007, p.8) defined the validity as "the degree to which a measure accurately represents what it is supposed to".Validity is concerned with how well the concept is defined by the measure(s).Fujun et al. (2007) mentioned [cited from Nunnally andBernstein, 1994] about three types of validity: content validity, Predictive validity, and Construct validity.Duggirala et al. (2008) defined the content validity as the assessment of the correspondence between the individual items and concept.Malhotra (2010) defines the content validity as face validity.This study addresses content validity through the review of literature and adapting instruments used in previous research.

Reliability
Extents to which a variable or set of variables is consistent in what it is intended to measure (Hair et al., 2007).It differs from validity in that it relates not to what should be measured, but instead to how it is measured.Several measures have been used to establish the reliability of the instrument, for example, split-halves, test-retest, equivalent forms and internal consistency method.The current study uses multiple items in all constructs except customer satisfaction.So the internal consistency method should be applied in the current study.Hair et al. (2007) mentioned that the rationale for internal consistency is that the individual items or indicators of the scale should all be measuring the same construct and thus be highly intercorrelated.The validity and reliability test of the current study is in appendix 4 Pallant (2001) pointed out that in terms of reliability the most important figure is the Alpha value.This is Cronbach's Alpha, which, in this case is 0.905 in table 1a.Pallant (2001) also stated that if any of the values in the Cronbach's Alpha If Item Deleted column (table 1b) are higher than the final alpha value then the researcher may want to consider removing this item from the scale.In this study, all the values are less than the final Alpha value; therefore, all the values are reliable.Finally, it can be concluded that all the service quality attributes are internally consistent because all the values exceeded Nunnally and Bernstein (1994) recommended criterion of 0.70 for scale reliability.
The customer satisfaction measurement scale is adapted from Cronin and Taylor (1992).A seven-point Likert scale ranging from "1 -very unsatisfactory" to "7 -very satisfactory" has been used to measure the customer satisfaction.The internal consistency method will not be applied due to single item instrument.
The measurement scale for the customer loyalty is adapted from Caruana (2002).A seven-point Likert scale described at either end by "extremely likely" and "not at all likely" is used to measure the customer loyalty.Reliability statistics table shows that the Alpha value is 0.905.The column (Cronbach's Alpha if Item Deleted) reveals that the Cronbach's Alpha of Customer Loyalty is 0.895 which is less than the original alpha value, therefore the Alpha value of Customer Loyalty is reliable.It has been observed that the Alpha value of customer loyalty also exceeded Nunnally and Bernstein (1994) recommended criterion of 0.70 for scale reliability.

Analysis and Presentation of Findings
At first, this section explains the general sample description and descriptive statistics of each study constructs with the graphical presentation.Then, a hypothesis test has been conducted to find out the relationships between each of the variables of this study.Finally, this section concludes with the summary of findings related to the hypotheses concerning service quality attributes, customer satisfaction and customer loyalty and the summary of open questions.
Based on the 100 sample bank customers, the percentage of male and female respondents are 77 and 23 respectively, which shows the male dominancy of bank customers.In the whole sample, 53% of respondents fell in the age range of 21-30, and 32% fell in the range of 31-40.In terms of qualification, the respondents are almost equal and that is, Undergraduate (31%), Graduate (33%), and Post Graduate (35%).63% of respondents are service holder and 43% of respondents earn more than 20000 Taka per month.

Descriptive Statistics for each study constructs [Note: See the descriptive statistics table on appendix 5]
Based on the 100 samples, Tangibility ranges from 2 to 7 with the Mean of 5.64 and the Standard Deviation of 0.769.Reliability ranges from 3 to 7 and the Mean and Standard Deviation is 5.57 and 0.82 respectively.For Responsiveness, Mean and Standard Deviation is 5.31 and 1.03 respectively with the minimum and maximum value ranges from 2 to 7. Asurance ranges from 3.67 to 7 and the Mean and Stanard Deviation is 5.65 and 0.73respectively.Empathy ranges from 3 to 7 and the Mean is 5.49 and the Standard Deviation is 0.86.The minimum and maximum value for customer satisfaction is 3 to 7 and the Mean and Standard Deviation is 5.64 and 0.90 respectively.Customer loyalty ranges from 2 to 7 and the Mean and Standard Deviation is 5.44 and 1.02 respectively.It has been observed in the table 3 that almost all the Mean are similar.High Standard Deviation means that the data are wide spread, which means that customers give variety of opinion and the low standard deviation means that customers express close opinion.

Pearson Correlation
A correlation coefficient is a very useful way to summarise the relationship between two variables with a single number that falls between -1 and +1 (Welkowitz et al., 2006).Morgan et al. (2004) stated that: -1.0 (a perfect negative correlation) 0.0 (no correlation) +1.0 (a perfect positive correlation) [Note see the correlation table in appendix 6].Pallant (2001) suggested the following interpretation of r-value [cited from Cohen, 1988].This value will indicate the strength of the relationship between two variables.r = .10to .29 or r = -.10 to -.29 Small r = .30to .49or r = -.30 to -.49 Medium r = .50to 1.0 or r = -.50 to -1.0 Large The Pearson correlation analysis obtained for the three intervals scaled variables in the table above.The sample size (N) is 100 and the significant level is 0.01 (p 0.01).
Recalling the first research question: "What are the specific service quality attributes that influence customer satisfaction in the retail banking sector in Bangladesh?" the first hypothesis will examine the correlation between the service quality attributes and customer satisfaction in the retail banking sector in Bangladesh.The hypotheses are as follows:

Tangibles
H1aa: There is a positive correlation between tangibles and customer satisfaction in the retail banking sector in Bangladesh H1a0: There is no correlation between tangibles and customer satisfaction in the retail banking sector in Bangladesh In the table, it can be seen that the correlation (r) of tangibles is 0.491 and the significant level is 0.01 (p .01).The table shows that the p-value is 0.000, which is less than 0.01.We therefore reject the null hypothesis, and concluded that there is a medium positive (r = .491)relationship between tangibles and customer satisfaction in the retail banking sector in Bangladesh.

Reliability
H1ba: There is a positive correlation between reliability and customer satisfaction in the retail banking sector in Bangladesh H1b0: There is no correlation between reliability and customer satisfaction in the retail banking sector in Bangladesh The table shows that the correlation (r) is 0.488 for reliability and the p-value is 0.000, which is less than the significant level (0.01).Therefore, the null hypothesis is rejected and concluded that reliability and customer satisfaction is positively (medium) related in the retail banking sector in Bangladesh.

Responsiveness
H1ca: There is a positive correlation between responsiveness and customer satisfaction in the retail banking sector in Bangladesh H1c0: There is no correlation between responsiveness and customer satisfaction in the retail banking sector in Bangladesh It can be observed in the table that the correlation (r) of responsiveness is 0.493 and the p-value is 0.000, which is less than 0.01.Therefore, the null hypothesis is rejected and it can be concluded that responsiveness is positively (medium) related to customer satisfaction in the retail banking sector in Bangladesh.

Assurance
H1da: There is a positive correlation between assurance and customer satisfaction in the retail banking sector in Bangladesh H1d0: There is no correlation between assurance and customer satisfaction in the retail banking sector in Bangladesh The table shows that there is a large positive correlation between assurance and customer satisfaction in the retail banking sector in Bangladesh where p 0.01 (p=0.000) and r=0.526.So, the null hypothesis is rejected.

Empathy
H1ea: There is a positive correlation between empathy and customer satisfaction in the retail banking sector in Bangladesh H1e0: There is no correlation between empathy and customer satisfaction in the retail banking sector in Bangladesh According to the table, the strongest predictor of customer satisfaction is empathy.The result indicates that the correlation (r) of empathy is 0.545 and the p-value is 0.000, which is less than 0.01.Therefore, the null hypothesis is rejected and concluded that there is a large positive correlation between empathy and customer satisfaction in the retail banking sector in Bangladesh.

Hypothesis # 2
The second hypothesis examined the correlation between customer satisfaction and customer loyalty in the retail banking sector in Bangladesh.Most of the researchers found that customer satisfaction is the predictor of customer loyalty [cited in literature review].This hypothesis recalled the second research question: What are the specific predictors that affect the customer loyalty in the retail banking sector in Bangladesh?The hypothesis is as follows: H2aa: There is a positive correlation between customer satisfaction and customer loyalty in the retail banking sector in Bangladesh H2a0: There is no correlation between customer satisfaction and customer loyalty in the retail banking sector in Bangladesh Referring to the table, it can be observed that the correlation (r) of customer satisfaction is 0.673 and p-value is 0.000.Therefore, the null hypothesis is rejected and it can be concluded that customer satisfaction is positively (large) related to customer loyalty in the retail banking sector in Bangladesh.

Conclusion
The main objective of the current study is to find the interrelationships between service quality, customer satisfaction and customer loyalty in the retail banking sector in Bangladesh.The study sought to identify the most important attributes in bank settings, which may be used to review characteristics of the banks as experienced by customers.
Based on hypothesis H1aa and H1a0, a medium positive correlation was found between tangibility and customer satisfaction.It means that as banks' tangibility increase, customer satisfaction also increases.According to hypothesis H1ba and H1b0, reliability and customer satisfaction has medium positive correlation.This positive relation means that as bank employees' performance and reliability increase, customer satisfaction also increases.Hypothesis H1ca and H1c0 demonstrate the medium positive correlation between responsiveness and customer satisfaction.It means that as the willingness and ability of bank employees' increase, customer satisfaction also increases.Hypothesis H1da and H1d0 shows the large positive correlation between assurance and customer satisfaction.This means that as the assurance from bank employees' increases, customer satisfaction also increases.According to hypothesis H1ea and H1e0, empathy and customer satisfaction has large positive correlation.It means that as the customer and employee interactions increase, customer satisfaction also increases.Hypothesis H2aa and H2a0 demonstrate that there is a large positive correlation between customer satisfaction and customer loyalty.That means that as the customers are satisfied, there is a high chance for loyalty.

Applications and Discussion of Findings
The applicability of the SERVQUAL measure is well established in the retail banking industry.Angur et al. (1999) stated that SERVQUAL instruments are the best measure of service quality perceptions in the retail banking industry in developing country.The current study also uses the SERVQUAL instruments to measure the service quality perceptions in the retail banking sector in Bangladesh.This study uses the SERVQUAL model to determine the relative importance of each of the service quality attributes which influence customer overall quality perceptions.
The hypotheses test confirms that all the service quality attributes are positively correlated with customer satisfaction.Empathy shows the highest positive correlation with customer satisfaction and Assurance demonstrates the second highest positive correlation with customer satisfaction.Tangibility reveals the least positive correlation with customer satisfaction.A large positive relationship between customer satisfaction and customer loyalty was demonstrated in the hypotheses test.The discussion of the findings is as follows:

Empathy
Service customers often have expectations with regard to the extent to which the service provider appears to understand and be concerned about their individual needs and wants.The more the service provider can see things from the customer's point of view, the better.The core concept of empathy is to understand the needs of customers and provide individual attention.Employee and customer interactions are reflected through the empathy dimensions.Data shows that the bank customers in Bangladesh want a high degree of interaction with the bank staff and they also expect personalized service from the bank staff.The bank customers are also looking for front line staff who are capable of understanding their specific needs.The current study confirms that empathy has a large positive correlation with customer satisfaction.Ladhari (2009) also found that empathy is the strongest predictor of customer satisfaction.

Assurance
Assurance is the degree of trust and confidence that the customer feels that the service provider is competent to supply the service.Mostly, this stems from the degree of confidence that the customer has in the service provider's staff.The customer will not be satisfied if he/she does not feel assured about the competence of the service provider.Assurance shows a positive correlation with customer satisfaction in the current study.The possible explanation of this finding is that the bank can instill feelings of confidence in its customers and the banks handle their customers in a professional and competent way.As mentioned earlier in the literature review, Kumar et al. (2010), andLai (2004) also pointed out that assurance is one of the important factors for customer satisfaction.

Responsiveness
Responsiveness refers to the willingness and ability of the service provider to meet and adapt to customers' needs.This study shows that responsiveness and customer satisfaction is positively correlated.The bank customers in Bangladesh prefer a friendly bank, which is willing to help in their banking operations.Willingness to help customers is likely to have an important and positive effect on customer satisfaction in the retail banking sector in Bangladesh.Mengi (2009) also found that responsiveness is positively related to customer satisfaction.

Reliability
Reliability is the extent to which the service is delivered to the standards expected and promised.In essence, it represents the customer getting what they feel they have paid for.According to this study, there is a positive relationship between reliability and customer satisfaction in the retail banking sector in Bangladesh.Data shows that customers are satisfied with the services provided by the bank as promised and handling (speed) of solving the problem.Customers are confident that banks will fulfil the promised terms and conditions which will not go against their (customers) interests.Zim et al. (2010) pointed out that reliability is one of the important factors of customer satisfaction.

Tangibility
The hypotheses test of this study confirms that tangibility and customer satisfaction are positively correlated with each other.The possible explanation of this finding is that the bank customers often look to any tangible indications which may be used as indicators of the service quality.The customers can assess the premises of the banks; or perhaps the appearance of the bank's staff.Lai (2004) also pointed out that tangibility is positively related to customer satisfaction.

Customer Satisfaction and Customer Loyalty
Bowen and Chen (2001) pointed out that a small increase of customer satisfaction leads customer loyalty dramatically.The current study finds that customer satisfaction has a large positive correlation with customer loyalty in the retail banking sector in Bangladesh.Kandampully and Suhartanto (2000) supported this relationship.The strong positive correlation of customer satisfaction and customer loyalty means that the bank customers will recommend the bank to other people.As a consequence, the banks can be assured of a loyal and stable customer base, thereby reducing the cost.Cohen et al. (2007) found that a loyal customer takes less of the company's time during transactions and are less sensitive to price changes.Recalling the Research Question No 3 "How the proposed model contribute in the retail banking sector in Bangladesh?" Reichheld (2006) stated that companies with the highest customer loyalty typically grew revenues at more than twice the rate of their competitors.Therefore the current research model will contribute in the retail banking sector in Bangladesh by using a loyal customer base.

Managerial Implications
The current study has shown the interrelationships among service quality, customer satisfaction and customer loyalty in the retail banking sector in Bangladesh.This study confirms the positive relationship between all the service quality attributes and customer satisfaction.This study also suggests that SERVQUAL is a suitable instrument for measuring the bank service quality in the Bangladeshi context.Therefore, bank managers can use this instrument to assess the bank service quality in Bangladesh.Moreover, because all the dimensions of service quality attributes are positively correlated with customer satisfaction, Bangladeshi bank managers should emphasise all the service quality dimensions in maintaining and improving the service quality that they provide.
Empathy shows the highest positive correlation with customer satisfaction in the current study.The core concept of empathy is employee-customer interactions.Therefore, Bangladeshi bank managers would be well advised to emphasise the employee training programmes so that they can offer personalized service.The main aim should be to develop a long-term relationship with the customers.The current study demonstrates that there is a large positive correlation between customer satisfaction and customer loyalty.That means that if the customers are satisfied then they will become loyal.Jones and Sasser (1995) pointed out that there is a huge difference between merely satisfied and completely satisfied customers.Therefore bank managers should pay attention on the complete customer satisfaction.

Limitations
The current study presented exhibit limitations that should be considered.The limitations are as follows: The sample size of the current study is 100.This sample size meets the minimum requirement.The researcher may use a bigger sample size to find out more about service quality, customer satisfaction and customer loyalty in the retail banking settings in Bangladesh.Other variables may influence customer decision making process in the retail banking industry, such as price, convenience, interest rates etc.This study covers the customers of Government and Private banks in Bangladesh.The Foreign banks that have a role in the retail banking industry in Bangladesh were excluded from this study.

Recommendations for Future Research
Several issues, associated with the limitations inherent in this study, require further research considerations.
Future research should concentrate on a bigger sample size and all types of retail banks [including foreign banks] in Bangladesh.In addition, more decision making variables should also be considered in the research model.

General Information
Note: This information will exclusively be used for research purpose and in no case will be disclosed to anybody.Are you satisfied with the bank service of sending timely bank statement? 1 2 3 4 5 6 7 14.

Assurance
Are you satisfied with the security of the bank? 1 2 3 4 5 6 7 15.
Are you satisfied with the employee's eagerness of instilling confidence to you?
Fujun et al. (2007) pointed out [cited from Nunnally and Bernstein, 1994] that the Cronbach alpha with acceptable cut off point 0.70 demonstrates that all attributes are internally consistent.

Figure 1 .
Figure 1.Research Model Note: All the paths are hypothesized to be positive.

Table 1 .
Summary of the Null Hypotheses Test

Table 3 [Note: SPSS -version. 17 has been used to calculate the above Table]
**. Correlation is significant at the 0.01 level (1-tailed).