Mediating Role of Customer Satisfaction between Service Quality and Customer Loyalty with Non-Interest Bank in Nigeria

The world has become a global village as competition among banks is now intense. Globalization is fuelling increased competition in Nigerian banking industry and as a result, banks are challenged to adopt customer centred strategies that would enable them to gain customer loyalty, maximize profitability and survive the competition. There is no business that survives without customers’ patronage and Islamic banks are not an exception. Customers are key stakeholders to the survival of any business, thus, satisfying the customer through providing high quality services is core to Islamic banking business. The aim of the study is to examine the mediating role of customer satisfaction in investigating the relationship between service quality and customer loyalty in non-interest bank i.e. Jaiz Bank of Kano state metropolis, Nigeria. The variance based (SEM) technique is employed in analyzing the hypothesized model of the study. Service quality positively and significantly influences customer satisfaction and customer loyalty, whereas, customer satisfaction does not significantly mediate the relationship between service quality and customer loyalty among customers of Jaiz bank in Kano state metropolis, Nigeria. The study recommends for providing high quality service in order to achieved maximum loyalty from the customers. It was further recommended on the need for reviving of the existing regulatory bodies or institutions that will ensure proper implementation of good service quality delivery among non-interest banks in Nigeria.


Background
Islamic banking emerges following an attempt by King Abdul Aziz University in Makkah to organized first international conference on Islamic economics and subsequently establishment of the first commercial Islamic bank named Dubai Islamic Bank in the UAE and also the establishment of Islamic Development Bank in Jeddah, Saudi Arabia in year 1973. Establishment of Islamic banks followed suit in Egypt, Sudan, Kuwait, Bahrain and Nigeria (Faosiy, Ahmed, & Adekunle 2016). The history of full-pledge Islamic banking in Nigeria began in January 2012 after the approvement granted by Central Bank of Nigeria to Jaiz Bank to operate as a regional interest-free bank in Nigeria. As a result, Jaiz bank became the first and the only full-fledged Islamic bank operating in the country (Husna & Fuaad, 2017).
Banking has become part and parcel of our everyday life as individuals and business organizations cannot do without engaging in banking activities. While the majority of individuals cannot live comfortably without having a personal bank account for safe keeping of money and withdrawals when the need arises, all enterprises rely on banks to carry out their business activities including money transfers, payment for purchases and salaries and safe keeping of company's assets. Banks play an important role in the financial life of individuals and businesses as their essential activities facilitates the process of production, exchange and distribution of wealth. The basic function upon which other functions depends of banks to individuals and business organizations is the acceptance of deposit and advancing of loans. Banks are the custodians and distribution of liquid capital, which is the lifeblood of our personal needs, commercial and industrial activities as well as the economic well-being of the nation by, ensuring economic and social stability and sustainable growth and development of the economy.
As interest free banking services is offered to better maximize customers satisfaction compared to conventional banking which is based on charging of interest, there is a need to examine the level of customer satisfaction with Islamic banking services especially as interest free banking is touted to be a better alternative to conventional practice of charging interest on capital/loan. More so, in order to be competitive Islamic banks, need to assess the level of customers' satisfaction with their products because without satisfied customers they can neither survive nor beat the competition. Jayasundara, Ngulube and Minishi-Majanja (2016) states that every competitive firm has to be interested in assessing the satisfaction level of its customers because customer loyalty ensures the continuity of a firm, loyal customers are advocates and more tolerant to the firm. Loyal customer and cordial customer relationships are elements for survival and favourable competition in the global world of intense rivalry in business. Pleasing customer satisfaction is important to Islamic banks approach to penetrate the market and gain high market share.
There is no business that survives without customers' patronage and Islamic banks are not an exception. Customers are key stakeholders to the survival of any business, thus, satisfying the customer through high quality service offering is core to the business of Islamic banks. Enhancing Islamic service quality and maximizing the level of customer satisfaction with Islamic banking services is therefore needful for Islamic banks to attract new customers, retain existing ones and survive in the face of intense competition especially as a new entrant with limited market share and restricted customer base relative to conventional banks resulting from different combinations of expectations and performance outcome (Schiffman, Kanuk and Hansen 2012). The SERVQUAL is a model of service quality which was developed by Parasuraman, Valarie, Zeithaml and Len Berry between (1985;1988). Furthermore, five major service quality dimensions are identified i.e. as reliability, assurance tangibles, empathy and responsiveness (Zeithaml, Berry and Parasuraman, 1996;Parasuraman, Berry & Zeithaml, 1991). Customer satisfaction is determined by a number of factors including quality, perceived value, customer support service and price, ease of use (user friendliness), speed and security. Information on which factors are the determinants of customer satisfaction is important to service organizations like banks since customer satisfaction is what the banks depends on to win customer loyalty and cause the customer to repeat purchase of the service. Increased customer satisfaction often leads to a lower turnover of the company's present customers. As customer satisfaction rises, so does customer loyalty and intention to repurchase. In addition to being loyal, satisfied customers are always less price-sensitive and more willing to pay a higher price than other customers. Particularly, the research conducted by Kotler and Armstrong (2010) underlined service quality, service charges and perceived value as the determinants of customer satisfaction in the banking industry.
The world has become a global village as competition among banks has become intense. Globalization is fuelling increased competition in Nigeria banking industry and as a result banks are challenged to adopt customer centred strategies that would enable them to gain customer loyalty, maximize profitability and survive the competition. Conventional banking services even though interest based have gained worldwide acceptance and the majority of customers cutting across all religious spheres are already acclimatized with these products unlike Islamic banking services that are relatively new and restricted in principle to adherents of Islam. Non-interest banks are in a situation of competing with already established conventional banking some of whom have been in existence for some decade in the Nigeria banking environment with well experienced staffs, spread of many branches as well as wide range of customers. therefore, to excel in this competitive market, Islamic banks needs to formulate and effectively implement customer satisfaction driven strategies to attract new customers, retain existing ones and propel existing customers to buy more or patronage more Islamic financial services.
The prospects of Islamic banking in Nigeria are beyond doubt, however, Islamic banking is a new method of banking in Nigeria when compare to conventional banking that is there for some decades, therefore, there is urgent need for Islamic banks to revisit their strategies by creating awareness and to teach Nigerian citizens about how it operates (Unegbu & Onuoha, 2013). Today customers are more significant than ever before and consumers' freedom of choice has no bound. Customers taste for high quality of products or service they are offered is very high and increasing every day by day. Delighting customers and achieving high customer satisfaction scores especially as customers taste changes daily is ever more difficult. And even after Islamic bank satisfy their customer with improved financial services, significant number of them could still leave the bank and start doing business with rival banks. Given that they have freedom of choice and varieties of services amongst which to choose from in the market place they switch from one product to another without notice and change brand loyalty at will. Since the taste, fashion and re-purchase behaviour of customers are constantly changing and unpredictable there is a need therefore to examine the level of customers' satisfaction with Islamic banking services for the purpose of understanding the preferences of customers in taking decision to develop a customer satisfaction programme and improve on the efficiency and product range of Islamic financial services.

Objective
The aim of the study is to examine the role of customer satisfaction as a mediator in the relationship between service quality and customer loyalty in Jaiz Bank plc of Kano state metropolis. Specifically, the study attempts: (i) to determine the influence of service quality dimensions on customer satisfaction in Jaiz Bank plc of Kano state metropolis. (ii) to determine the influence of customer satisfaction on customer loyalty in Jaiz Bank plc of Kano state metropolis. (iii) to determine the mediating effect of customer satisfaction in the relationship between service quality and customer loyalty in Jaiz Bank plc of Kano state metropolis. This paper comprises of five sections. Introduction is presented in first section, both empirical and theoretical literatures were explained in second section. The third section explain methodology of the study. data presentation and analysis were discussed in fourth section. And finally, the fifth section contained conclusion, recommendation and frontier for further studies.

Background Theory
This research chooses Balanced Scorecard (BSC) to be the theoretical framework, BSC theory was developed by Robson (2002). An attempt was made to come up with new methods of performance for companies and enterprises which is later called BSC theory, this effort was as result of the fact that financial measures of performance were insufficient for the modern business enterprise. The BSC theory emerged out of an argument that company success is beyond looking at just the financial statement, because financial statement reveals only the past performance but do not really explain the current situation as well as where the business is likely to be in the nearest future. BSC provides a strategy that is consistence and in reliable manner which can implemented in public sector as well as profit and nonprofit oriented enterprises. Success stories emerged and hence, BSC secure recognition as best suit any types of organization. The sole aim of BSC theory to identify the role play by other factors besides financial statement toward adding value to an organization and directly or indirectly influence the organization to succeed. The study adopted BSC theory because it is a theory that explain the influence non-financial factor I.e., service quality on customer loyal which will subsequently increase the performance of noninterest bank.

Previous Studies
In order to determine a literature gap, there is the need to provide comprehensive review of related empirical literatures, this study was able to review and thematically analysed some previous studies among which are; Slack and Singh (2020) explore the effect of service quality on satisfaction and loyalty of customer and also the mediating role of customer satisfaction between the dependent and independent variable. An exploratory research approach was adopted in the study. In total, 480 customers of supermarket were issued questionnaire. Descriptive statistics were employed to determine the level of service quality, and inferential statistics to test the research hypotheses with the help of paired sample t-test. The findings indicate that service quality provides by supermarkets is perceived as being unsatisfactory, whereas, service quality significantly affects customer satisfaction and loyalty and customer satisfaction partially mediates the relationship between service quality and customer loyalty which subsequently lead to lower customer loyalty. This study provides a hint as to where supermarkets should concentrate when marketing their product and services and may help in knowing the strategies to adopt in order to retain and attract more long-term loyal customers. This research could be useful to government, policy makers and other related regulating agencies in making right decisions and policies. However, a more robust technique of analysis should further be use test the relationship between the latent variables of study.
Teeroovengadum, (2020) aims to reveal gender-based difference with regard to the effect relationship between service quality dimensions and customer satisfaction in the of banking sector. Primary data is collected by using the survey design strategy. The questionnaires were directly administered to bank customers in Mauritius. As result of unavailability of sampling frame because of confidentiality reasons, a nonprobability sampling technique, i.e., the purposive sampling technique was employed. The respondents were asked to respond to the survey but conditionally if they had a bank account and had regularly used bank services during the previous two years. Data is collected from 282 customers of banks. Because of the prediction-oriented aim of the present study, the partial least squares-structural equation modeling technique is used to test a set of relevant hypotheses. The results provide mixed findings and the moderating effect of gender is found to be statistically significant for one service quality dimension but not for the other two. The permutation and partial least squares-multigroup analysis tests both support the hypothesis that gender moderates the relationship between environment quality and customer satisfaction. Future studies could consider the moderating role of categorical variable such as age group, education level and work experience among others.
Surahman, Yasa, and Wahyuni, (2020) were able to assess the effect of service quality on customer loyalty and customer satisfaction as a mediator between the relationship of service quality with customer loyalty. Data was collected from 100 respondents who were tourist that visited the Badung Regency Village. The data were analyzed using structural equation modeling with partial least square approach. The results indicate an insignificant direct effect of service quality on customer loyalty. Significant direct influence on service quality with customer satisfaction, and customer satisfaction with customer loyalty. Furthermore, the results show an indirect effect between service quality and customer loyalty, which is mediated by customer satisfaction. The findings empirically indicate that service quality has a negative effect on customer satisfaction. And customer satisfaction has a full mediating effect, customer satisfaction is important in explaining the quality of service that subsequently drive customer loyalty. Based on the results it can be concluded that this research offers empirical evidence of the relationship of service quality, customer satisfaction and customer loyalty. This research makes a theoretical contribution by clarifying the high level of service quality influence in increasing customer loyalty through customer satisfaction.
Khatoon, Zhengliang and Hussain (2020) examines the influence of electronic (E)-banking service quality dimensions on customer purchasing intentions and also the mediating role of customer satisfaction. 235 sample of questionnaires were distributed among employees and customers working in different banks of Qatar. Both correlation analysis and regression analysis were employed to test the study hypothesis in order to achieved the objective of the study. The results of the study revealed that all the service quality dimension (i.e., responsiveness, reliability, communication, privacy, efficiency, and security) have a significant effect on customers purchasing intentions. Customer purchasing intentions significantly increased with the increase of E-banking service quality. A partial mediating role of customer satisfaction was found between E-banking service quality and customer purchase intentions. However, a more robust technique of analysis (i.e., structural equation modelled) should be further employed to test the relationship between the endogenous, exogenous and moderating variables. Yadav and Rai (2019) examine the mediating effect of customer satisfaction between service quality and customer loyalty with banking industry in India, the study respondents are customers that own a saving or current account with any of four commercial banks namely, SBI Bank. PNB Bank, HDFC Bank and ICICI Banks and also frequently using the account for transaction purpose for more than three years. The study targeted 400 respondents using Snowball sampling technique and questionnaire was distributed equally across the four banks. Simple linear regression was used in the study and the findings revealed a strong influence of service quality dimension on customer loyalty in the banking sector. The research model or equation predictive effect was found to be moderate. The study also finds that customer satisfaction mediates the relationship between service quality and customer loyalty at almost 83.03%. However, more robust technique of analysis can be further employed to properly and simultaneously modelled the direct and indirect causal relationship between the variable of the study i.e., structural equation modelling.
Ozkan, Suer, Keser and Kocakoc (2019) examine the relationship between the customer satisfaction, service quality (i.e., perceived value of services) on customer loyalty, and also the mediating effect of corporate image and corporate reputation between the independent and dependent variable in the Turkish banking sector, the study adopted a survey designed and SEM analysis is conducted in order to study the relationships (i.e., effect) between variables of the study. The findings of the study indicate that corporate image and corporate reputation can be used as indicators that measure performance of a bank. The results also signify that customers perceive quality and satisfaction influence loyalty through perceived value, image and reputation. Meanwhile, this study employed customer satisfaction as independent variable and there is need to further test the mediating role of customer satisfaction and also more studies are needed in other to generalize the results.
Yaqub, Halim and Shehzad (2019) were able to assess the effect of four independent variables namely price fairness, justice to service, perceived service quality and relational bonds on customer loyalty as the dependent variable and also to investigate how customer satisfaction mediates the relationship between them. The study distributed 539 questionnaires among customers that are prepaid subscribers based through a proportionate stratified random sampling in four major capital cities of Pakistan. Data collected was analysed by using SPSS version 23 and the Smart PLS Structure Equation Modelling (PLS-SEM). Findings of the study reveals that perceived service quality and customer satisfaction significantly influence customer loyalty. Also, customer satisfaction actually mediated the relationship between the predictors and explained variable. Hence, there is need to replicate similar study in different case study (i.e., beyond Pakistan) for proper generalization.
Myo, Khalifa and Aye (2019) Empirically examine the relationship between the following independent variables namely service quality (SQ), customer satisfaction (CS) and dependent variable customer loyalty (CL) with Novotel Yangon Max hotel services. The study adopted social exchange theory, the study also assesses the mediating role play by customer satisfaction between Service quality as independent variable and customer loyalty as the dependent variable. The samples are 400 copies of questionnaire were distributed among the customers using simple random sampling. The study adopted SEM to test the study's hypotheses. Findings of the study reveals that SQ have direct impact on CL and also CS mediate the relationship between SQ and CL. however, the paper needs to further find out the extent of the mediating role of CL (i.e., partial or full mediation) and similar study can be conducted by selecting different case study for making general conclusion.  Figure 1 shows that all the three variables are latent variables which are all represented with oval shape, whereas, customer satisfaction is the dependent variable, service quality and customer satisfaction are independent and mediating variable respectively.

III. Methodology
The aim of the research is to examine the role of customer satisfaction as a mediator in the relationship between service quality and customer loyalty with Jaiz Bank plc Kano state, Nigeria.

Data
Survey exploratory research design was employed in this study. The target population of this research are customers of Jaiz bank PLC of Kano State metropolis that span across five branches in the state; namely Hotoro branch 4512 customers; Zoo Road branch 9786 customers; Kabuga branch 14,319 customers; Tafawa Balewa branch 33,606 customers and Bello Road branch 29,866 (Jaiz bank PLC head office, 2019). Service quality delivery is usually aimed at satisfying clients; hence, customers are in good position to answer questions pertaining the extent of service quality delivery of their respective bank. The total population is 92,089 customers and with the aid of the Yamane (1967) formula the study arrived at 400 sample size. Primary data is employed using self-administered questionnaire, the study adopt five likert scales Hence, the five-point Likert-scale that is adopted for this study has the following ratings: strongly agree = 5, agree = 4, Neither agree nor Disagree= 3, disagree = 2, strongly disagree = 1, a convenient sampling technique is employed in the study due to inaccessibility of sampling frame and the distribution of questionnaire is proportionate across the bank's branches.  (2019) and Computed by author.
The Table 1 reveal that sample size was distributed proportionately, there are 4,512 customers in Hotoro branch the sample size is equal to 20, while Zoo Road Jaiz bank branch has 9,786 customers and the proportionate sample size is 42, Kabuga branch has 14,319 customers and sample is 62, Tafawa Balewa has 33,606 customers with proportionate sample size of 146 and Bello Road branch has 29,866 customers with proportionate sample size of 130, this make the population of the study to be 92,089 customers of Jaiz bank in kano metropolis as at 2019.

Model Development
This study empirically tests the research model that contain three variables in which service quality dimensions is the independent variable, and customer loyalty is the dependent variable and lastly customer satisfaction is the mediating variable. All measures used in this research are adapted from existing scales (see. Parasuraman et al., 1994;Zeithaml et al., 1996;Brady et al., 2002;Olorunniwo and Hsu, 2006;Yadav and Rai, 2019;Ozkan, et al., 2019;Yaqub, et al., 2019;Myo, et al., 2019;Slack and Singh, 2020;Teeroovengadum, 2020;Khatoon, et al., 2020;Zhengliang and Hussain, 2020).
Customer loyalty: customer loyalty is defined as a serious commitment to repurchase a particular product/service and also to recommend the product/service to others Oliver (1999). The concept of customer loyalty is more related to attitude and behaviour which exhibited from the mind, from these two concepts a third perspective of customer loyalty a raised through combining both attitudinal and behavioural concepts of loyalty. This integrated approach takes account of both concepts so as to come up with concept of customer loyalty. According to Zeithaml et al., (1996) behavioral aspect of customer loyalty represents actual repeat purchase of products or services from the same company as well as recommending the company to others. Displaying such commitment to the company by showing a resistance to switch to another competitor (Zeithaml et al., 1996).
Service quality: this signifies the difference between expectation and perceived service received by customer and also this entails the extent of service delivery or quality (Gronroos, 1984;Parasuraman et al., 1985). a lot of attempt were made by different previous researchers to come up with different measures of service quality dimensions, meanwhile, an Exploratory research was performed also in an effort to refined with subsequent scale in 1988, this effort yields a result in which original ten dimensions of service quality were merge into five dimensions: reliability, responsiveness, tangibles, assurance and empathy (Parasuraman et al, 1988) Customer satisfaction: Customer satisfaction was defined as the customers' evaluation of product or service in terms of whether that service has met their needs or expectations (Zeithaml and Parasuraman 1996). Satisfaction is the consumer's response or judgment that a product or service rightly feature itself by providing a pleasurable level of consumption (Oliver, 1999). According to Kotler (2008) customer satisfaction is the feeling of happiness or unhappiness as a result of comparing the perceived performance of services or products with the expected performance. customer satisfaction is also defined as the cognitive judgment of a customer resulting out of an interaction between the customer's personality and a company's marketing practices in the perspective of the expectation the customer had with the product and perception of the benefits received (Rai and Srivastava, 2014).
1 and 2 = are parameter estimates of the regression = error term which signifies other factors that could influence dependent variable that are not captured in the model.

Method
Technique for Data analysis for this study was variance based Structural Equation Modelling (SEM) using the partial least square estimator (See. Reinartz, Haenlein and Henseler, 2009;Kock, and Lynn, 2012;Roldán, andSánchez-Franco, 2012 andHair, Hult, Ringle and, the data will be analyse using Statistical Package for Social sciences (SPSS version 22) and Smart-PLs version 3.

Descriptive Analysis
The survey questionnaire required the respondents to answer four demographic questions reflecting their gender, age group, monthly income and occupation status. Out of four hundred respondents (400) that received the questionnaire only three hundred and seven (307) respondents filled and properly return the questionnaire. This represented 76.8% feedback which is high enough to produce valid and reliable results in the study. This section summarises the frequency distribution of respondents on different demographic items as shown in Tables 2.

Descriptive Analysis of Construct of the Study
Descriptive statistics of variable of the study were also statistically analysed. Specifically, three constructs were analysed in order to determine their minimum, maximum, mean and standard deviation values. Table 3 provides a summary of constructs descriptive statistics.  (2019) and Computed by author Table 3 shows that the mean and standard deviation for customer loyalty were 3.77 and 0.6079, respectively. This suggests that respondents are little bit moderate when it comes to issue of loyalty. Table 3 also indicates that the mean for customer satisfaction was 3.89, with a standard deviation of 0.7861; suggesting that the respondents' satisfaction is little bit above moderate. Lastly, the results show a moderate score for service quality performance (Mean = 3.80, Standard deviation = 0.7715).

Pre-Estimation Test
Preliminary analysis is used to address the normality of data and existence of multicollinearity in the model. This is the process of examining the data before further analysis i.e. inferential statistics can be conducted. This process will provide assurance that the data to be examined are of good quality for further analysis. Table 4 shows the normality test for the constructs used in the study, in which the skewness and kurtosis of the data support normality distribution of the variable as both values fall within the range of -1 to +1 for the skewness and -2 to +2 for the kurtosis (Ghasemi & Zahediasl, 2012). Values falling outside this range indicate a non-normal distribution of data. Based on the Table 4, it can be concluded that the data is normally distributed.

Multicollinearity Diagnostic
The Variance Inflation Factor (VIF) identifies the similarities between independent variables and the strength of that similarity. Variance inflation factor between 1-5 and tolerance value of greater than 0.2 suggest that there is a moderate similarity but it is not severe enough to warrant corrective measures. While VIFs greater than 5 and tolerance value of less than 0.2 represent critical levels of multicollinearity where the coefficients are poorly estimated and the p-values are questionable.

PLS-SEM Path Model Results
Structural equation model (PLS-SEM) is a multivariate technique that is use to analyzed the interrelationship between many constructs i.e. exogenous, endogenous, moderating and mediating variable, it is also a more robust statistical tools use to analyzed latent variables with multiple indicators simultaneously. PLS-SEM is divided into two part namely, measurement/outer model and structural/inner model, the later analysed the inner part a model while the former measure the outer part of a model as presented in Figure 2 Hair et al., 2012;Henseler et al., 2009).

Assessment of Measurement Model
In order to assess outer model (measurement or factor model) in Smart-PlS, test such as individual item reliability, internal consistencies of the items, convergent and discriminant validity respectively need to be conducted .

Individual Item Reliability
This refers to the factor loading of items or indicators of particular latent construct, it is examined by analyzing the outer loadings of each construct's measure . Meanwhile, the threshold for retaining indicator with loadings between 0.40 and 0.70, it was discovered that out of 32 indicators, none were deleted because they presented loadings above the yard stick of 0.40.

Internal Consistency
This refers to the convergence of indicators of a particular latent construct in order to measure the same latent variable (Bijttebier, Delva, Vanoost, Bobbaers, Lauwers and Vertommen, 2000). Therefore, composite reliability measures inter-indicator consistency, hence, ensures the existence of correlations among sub-indicator of a latent construct. The study employed three approaches of assessing internal consistency, namely, Cronbach's alpha (Cronbach, 1951), rho_A and composite reliability (Fornell &Larcker, 1981 andDijkstra and. Hence, the present study adopted both approaches i.e. Cronbach's alpha, rho_A and Composite reliability in ascertaining internal consistency level and to clear doubt in the reliability of the study. Based on Bagozzi and Yi (1998) as well as Dijkstra and Henseler (2015) rule of thumb that the reliability most be at least 0.70. thus, the Cronbach's alpha, rho_A and composite reliability of the study were found to be above 0.70, which signifies internal consistency reliability of the measures used in this study.

. Convergent Validity
Convergent validity entails the extent to which a particular indicator represents the intended latent variable and also correlate with other indicators of the same latent variable (Hair, Money, Samouel, and Page, 2007). Hence, this current study adopted the use of AVE, this can be achieved when the AVE is greater than 0.5 (Fornell & Larcker, 1981). As indicated in Table 6, the values of the average variances extracted range between 0.539 and 0.801, suggesting acceptable values.

Discriminant Validity
Discriminant validity is also another form of validity test of measurement model which is aimed at examining the difference that exists between latent variable in a study (Barclay et al., 1995). The assumption is that despite aiming at explaining the same thing, latent variables are expected to be dissimilar, hence, discriminant validity refers to the extent to which a particular latent variable is different from other latent variables (Duarte & Raposo, 2010). Therefore, discriminant validity is the correlations among the latent variable and should be compared with the square root of the average variances extracted, the square root of the average variances extracted (values in bold face) were all greater than the correlations among latent variable as shown in Table 7, this signifies adequate discriminant validity (Fornell & Larcker, 1981).

Assessment of Significance of the inner Model
After examining the outer model, the current study examined the inner model. The current study also applied the standard bootstrapping procedure with a number of 500 bootstrap samples and 307 cases to assess significance of the path coefficients the inner model which is concerned about R 2 , coefficient, P-value, predictive relevance (Q 2 ) and effect size (F 2 ) which will be explained by the structural model with help of bootstrapping as shown in Figure 3.

Direct Relationship
The Figure 3 above provides the graphical display of the regression coefficient (β) and T-values of the hypothesis in this study, this current study uses a significance level of 5% (T-value= 1.96) two-tailed, Table 8 provides standardized path coefficient (β), T-values and confidence intervals as suggested by Hair et al., (2010).  (2019) and Computed by author using Smart-PLS Hypothesis one was stated in alternative form as service quality positively influence customer satisfaction. The result revealed a significant effect between service quality and customer satisfaction at 0.05 significance level with T-value of 1.96 two-tailed (β = 0.546, T-value=10.330 & P-value= 0.000), hence we accept the alternative hypothesis. This finding corroborates with the finding of Khatoon, et al., (2020), Yadav and Rai, (2019), Ozkan, et al., (2019), Yaqub, et al., (2019 and Myo, et al., (2019) who found positive and significant relationship between service quality and customer satisfaction. Hypothesis two was stated in alternative form as service quality positively and directly influence customer loyalty. The result revealed a significant effect between service quality and customer loyalty at 0.05 significance level with T-value of 1.96 two-tailed (β = 0.672, T-value=15.285 & P-value= 0.000), hence we accept the alternative hypothesis. This finding conforms with the finding of Khatoon, et al., (2020), Yadav and Rai, (2019), Ozkan, et al., (2019), Yaqub, et al., (2019 and Myo, et al., (2019) who found positive and significant relationship between service quality and customer satisfaction.
Hypothesis three also was stated in alternative form as customer satisfaction positively influence customer loyalty. The result revealed a significant effect between customer satisfaction and customer loyalty at 0.05 significance level with T-value of 1.96 two-tailed (β = 0.199, T-value=4.524 & P-value= 0.000), hence we accept the alternative hypothesis. This finding conforms with the finding of Khatoon, et al., (2020), Yadav and Rai, (2019), Ozkan, et al., (2019), Yaqub, et al., (2019 and Myo, et al., (2019) who found positive and significant relationship between service quality and customer satisfaction.

Indirect Relationship
According to Albers (2010), the indirect effect is the summation of both direct and indirect effects that exists between two particular variables. The basic idea of testing a mediating effect is to establish if the independent variable influence on the mediating variable can be extended to the dependent variable (Ramayah, Lee and In, 2011). Thus, the mediation of the present study was to ascertain the indirect effect of the exogenous variable (service quality) on the endogenous variable (customer loyalty) through a mediating variable (customer satisfaction). the present study adopted the most robust method of bootstrapping or re-sampling mediation technique. The bootstrapping (re-sampling) technique has the advantage of analyzing both main and indirect complex model effect simultaneously as the case in this study. Thus, making PLS-SEM as the most widely used method in mediation studies (Chin, 1998;Hair et al., 2011;Hayes & Preacher, 2010). However, PLS-SEM version 3.0 provides the result concurrently with that of the direct relationship as shown in Table 9.  Table 9 the relationship between service quality and customer loyalty was established through the mediation of customer satisfaction and hence the indirect effect is significant (β = 0.109, T-value=3.791 & P-value= 0.000).

Variance Accounted For
Furthermore, the study evaluates the variance accounted for (VAF) as recommended by . The VAF is aimed at evaluating the extent effect of the mediating variable. Despite Hayes (2013) submission that the effect of the mediating variable (full or partial mediation) should not be a source of worry so long as the indirect (effect) relationship is significant. However, VAF is a process that explains the proportional effect of indirect effect in relation to the total effect. Accordingly, in line with Ramayah et al., (2011) and  the extent of mediation is ascertained using VAF formula below: Where a = is coefficient value between independent construct and mediating construct, b= is coefficient value between mediating construct and dependent construct, c= is coefficient value between independent construct and dependent construct. Therefore, Variance Accounted For The results indicate that 13.9% of the total effect has been explained by the mediating latent variable (customer satisfaction) on the relationship between service quality and customer loyalty. Accordingly, it can be deduced that there is no mediating effect of customer satisfaction. This is in line with Hair et al. (2013) classification, that VAF value that is less than 20% indicates absence of mediation, while 20% to less than 80% indicates partial mediation and 80% and above indicates full mediation.

Assessment of Variance Explained in the Endogenous Latent Variables
The next is to determine the R-squared value which is an important segment in assessing the validity of the structural model. The R-squared is also referred to as coefficient determinant (Hair et al., 2012). The R-squared value clarifies the variance that exists in explaining endogenous variable as a result of one or two exogenous variables (Hair et al., 2010). Although the acceptable level of R2 value depends on the research context (Hair et al., 2010), an R-squared value of 0.10 is minimum acceptable level.  (2019) and Computed by author using Smart-PLS In predicting the variation in mediating and dependent variable in the model the findings revealed an R-square of (0.637= 63.7percent) and (0.299= 29.9%) respectively and this signifies the variance in customer satisfaction and customer loyalty. According to (Hair, et al., 2012) the acceptable thresh hold value of accepting R-squared is 0.19, 0.50 and 0.75 described as weak, moderate and substantial. Therefore, in case of variance in customer loyalty and customer satisfaction the R-squared are moderate and weak respectively.

Assessment of Effect Size (F 2 )
Having assessed and confirmed the postulated hypotheses of the study, the next criteria for the evolution of the structural model is the effect size (F 2 ) (Hair et al., 2013). Effect size indicates the relative effect of a particular exogenous latent variable on endogenous latent variable(s) by means of changes in the R-squared (Chin, 1998). It is calculated as the increase in Rsquared of the latent variable to which the path is connected, relative to the latent variable's proportion of unexplained variance (Chin, 1998). Cohen (1988) describes F 2 values of 0.02, 0.15 and 0.35 as having weak, moderate, strong effects respectively. Table 11 shows the respective effect sizes of the latent variables of the structural model. Table 11, it can be concluded that service quality as exogenous variables possess strong effect on explaining both customer loyalty and customer satisfaction, whereas, customer satisfaction possess weak effect in explaining customer loyalty.   (2019) and Computed by author using Smart-PLS Subsequent to determining the effect size or variation of the R-square, Hair et al. (2013) recommended that predictive relevance of the model should be ascertained using stone-Geisser's (Q²) (Geisser, 1974;Stone, 1974). The process is only befitting for endogenous reflective constructs to ascertain its predictive relevance in the model. If the Predictive relevance (Q2) value is greater than zero, then the model has predictive relevance (Hair et al., 2013). Table 12 presents the results of the cross-validated redundancy Q2 test.

From the
The Table 12 above revealed that cross-validation redundancy result for direct and indirect relationship in the model is Q²=0.422 and Q 2 =0.157 respectively, hence, all the models i.e. direct and indirect Q 2 is above zero and this signifies predictive relevance of the model.

Model Fitness.
The recent development in PLS-SEM has come with some positive change, especially as a response to the constant critique of the need for a more rigorous analysis, especially for variance-based SEM (Henseler, 2017). One of these culminate development is the bootstrap-based tests of the overall model fit which is one of the key challenges in SEM.
In line with the previous studies and the recent update on the SmartPLS 3.2.6 software, the present studies used Standardized Root Mean Square Residual (SRMR), Root mean square residual covariance (RMS theta) and Exact Model fit test using d ULS to examine the model fit of the present study. The Good of fit (GOF) of the present study as presented in Table 13 above, has indicated that the mode has achieved the threshold to establish that the model and data fit into the study. Specifically, an SRMR value of 0.060 is found to be Ok as compared to the threshold of ≤ 0.08. In the same vein, the dULS values of the study is not significant with values of 2.032. Similarly, the RMS theta value of the study is 0.127 which is within the threshold values of ≤0.12-0.14. this signifies or indicate an overall model fit as suggest by (Henseler et al., 2015., Djikstra and.

Conclusions
The primary aim of this study is to examine the role of customer satisfaction as a mediator in the relationship between service quality and customer loyalty in Jaiz Bank plc Kano state metropolis. Specifically, the study attempts: (i) to determine the influence of service quality dimensions on customer satisfaction with Jaiz Bank plc of Kano metropolis. (ii) to determine the influence of customer satisfaction on customer loyalty in Jaiz Bank plc of Kano metropolis. (iii) to determine the mediating effect of customer satisfaction in the relationship between service quality and customer loyalty in Jaiz Bank plc in Kano state metropolis. To solve this, the survey data is analyzed, and hence, the study drawn the following conclusions.
Service quality positively and significantly influence customer satisfaction and customer loyalty, whereas, customer satisfaction does not significantly mediate the relationship between service quality and customer loyalty among customers of Jaiz bank in Kano state metropolis, Nigeria.

Recommendations
Based on the outcome or empirical findings of the study the following recommendations are made. Service quality is a pivotal factor to be consider by non-interest banks, hence providing good service delivery is necessary so as to maximum loyalty from the customers, the study also recommends the reviving of the existing regulatory bodies or institutions that will ensure proper implementation of service quality dimension i.e. (reliability, responsiveness, assurance, tangibility and empathy) in non-interest banks i.e.(Jaiz bank) in Kano state metropolis, Nigeria.

Recommendation for Further Research
There is need to replicate this study in non-interest banks across all the states in Nigeria in order to compare and contrast the findings so as to achieved proper generalisation.