The moderating effects of organizational commitment on the relationship between employee satisfaction and employee performance in Jordanian Islamic banks Fu'ad

Article history: Received: April 29, 2020 Received in revised format: April 3


Introduction
In the current dynamic global business environment, competition is growing and every organization is exposed to it (Cavusgil & Knight, 2015). Today, almost all organizations in the world, particularly banks, operate in a risky and uncertain competitive environment. To survive in a competitive environment, an organization's employees are the most important and leading factor for its success (Ali & Ibrahim, 2019). Currently, the banking sector is confronting unstable financial times, related to liquidity, inflation, exchange rate, credit, cash flow, etc. (Al-Ettayyem & Zu'bi, 2015). Thus, banks need to build procedures and tools to face these difficulties by enhancing their employee performance (EP) to survive economic hardships (Al-Ettayyem & Zu'bi, 2015). The banking sector in Jordan is one of the most important financial sectors that continuously seeks to create a suitable working environment for employees, in light of the intense competition between banks for their distinctive human capabilities and competencies toward the expansion of local banks (Almajali, AL-majali & Bashabsheh, 2016). However, employees in Jordanian banks are deficient in the performance of services rendered to customers due to lack of experience and inability to meet customer demands efficiently; in addition, the bank management does not have the ability to deal with the problems faced by its employees (Shammout & Haddad, 2014). Yet, for a bank or any organization to achieve success, employees are the root. Human resources management is one of the most important activities related to the management of an organization and is considered a cornerstone of most organizations (Jiang, Hu, Liu, & Lepak, 2017). It aims to strengthen organizational capabilities by enabling organizations to attract as well as to qualify the skills necessary to keep up with current and future challenges and success (Jiang et al., 2017). Besides, if an organization can manage its employees, their commitment can bring about beneficial results, such as increases in productivity, performance, and effectiveness and decreases in absenteeism and turnover, at both organizational and individual levels (Inanc & Ozdilek, 2015). Proper employee management leads to increased Employee Performance (EP), as it is still a problem that is always faced by the management (Mosisa, 2017). As stated by Armstrong (2009), employees' outstanding performance is achieved when they perform their job-related activities or organizational objectives effectively and efficiently. Most organizations that are aware of this fact have focused on enhancing EP by satisfying their employees (Dinc, 2017). Several previous studies have proved that there was a strong relationship between EP and employee satisfaction (ES) (Huang, Rundle-Thiele, & Chen, 2019;Triwahyuni & Ekowati, 2017;Hira & Waqas, 2012). On the other hand, ES and organizational commitment (OC) are also studied together frequently, as some researchers stated that ES influences OC significantly (Aftab, 2012;Abdirahman, 2018;Huang, You, & Tsai, 2012;Fu & Deshpande, 2014), while others studied the relationship between EP and OC (Abdirahman, 2018;Setyaningrum, Setiawan, & Surachman, 2017;Memari, Mahdieh, & Marnani, 2013). However, no previous study has tested OC as a moderator among ES and EP. Thus, this study aimed to analyze the effect of ES on EP and to examine the effect of OC as a moderator in their relationship in Jordanian Islamic banks.

Employee Performance (EP)
EP is essential as it reveals employees' capacity and capability in completing the tasks allocated to them. So, an organization needs to set clear and measurable criteria as a reference to measure EP (Triwahyuni & Ekowati, 2017). According to Kawiana, Dewi, Martini and Suardana (2018) EP may be defined as "the work quality and quantity achieved by an employee in performing their duties in accordance with the responsibilities given to them". Nassazi (2013) expressed that EP is generally a result of the capacity of employees in finishing their tasks to fulfill organizations targets. Kirimi and Maende (2019) stated in a research paper on EP and advancement that employees are the key component in an organization and its failure or success depends essentially on EP.

Employee Satisfaction (ES)
On the other hand, ES is crucial to the development of any organization (Sarker & Ashrafi, 2018). According to Ali (2016), a standout among the most frequently considered concepts in organizational sciences is ES. In addition, ES defined as an effective reaction to a job that results from the comparison of perceived outcomes with those that are desired (Sarker & Ashrafi, 2018). As per Ahamed (2014), ES was otherwise called an employee's feeling of success and achievement that is connected directly to productivity and personal prosperity. Besides, ES explains the way in which an employee's job is satisfactory (Abdirahman, 2018). It is assumed that an employee will feel fulfilled with their work if they are satisfied and enjoy doing their job (Noah & Steve, 2012). Where ES look after there are many material advantages which are reflected on an employee from the work, even the work environment is essential to gain these vital advantages; if the work environment is not effective or satisfactory, it might reduce the spirit of the employee (Adigun, Rotimi, & Hamzat, 2017).

Organizational Commitment (OC)
In this study, OC refers to a sense of belonging of the employees toward the company, thus affecting their ability to perform beyond expectations in achieving the company's goals (Ibama, 2016). OC is an important factor to gain more attention from the bank management because the OC has a significant impact on increasing EP and ES. Accordingly, the banks should consider the importance of ES to obtain EP. Bank executives can instill OC toward the employees, by motivating and rewarding those who perform well and have high loyalty to the bank, to reach the desired corporate goals, and create values and culture. Omar (2016) shows in his study that OC could be defined as "an attitude in the form of an attachment that exists between the individual and the organization, and is reflected in the relative strength of an employee's psychological identification and involvement with the organization". As shown in the studies by Abdirahman (2018), Susanty et al. (2013), and Huang et al., (2012), OC has a strong influence on ES as well as On EP.

Relationship between Employee Satisfaction and Employee Performance
The relationship between ES and EP has drawn the attention of many researchers (Huang et al., 2019;Triwahyuni & Ekowati, 2017;Adigun et al., 2017). ES improves EP and is affected by different factors inside and outside the workplace (Bello, Ogundipe, & Eze, 2017). For example, Huang et al. (2019) concluded that EP is fundamental to ES, which, thus, increases customer satisfaction and loyalty. Also, ES is a significant determinant of EP and may influence organization performance (Chen, Wu, Chang, & Lin, 2015). It is imperative to quantify ES as low levels of ES can expand employees' intent to leave their current employment when opportunities are presented to them (Yao, Qiu, & Wei, 2019). However, this study expands on the totaled previously mentioned outcomes to assume the following: H1: There is a positive relationship between employee satisfaction and employee performance in Jordanian Islamic banks.

Organizational Commitment, Employee Satisfaction, and Employee Performance
This study aims to examine the effects of ES and EP through OC as a moderator. OC is created by the formal and casual interactions and cooperation among individuals and organizations. In administrative organizations that provide services, contact among employees and the association is occurs mainly through interactive procedures at work (Yao et al., 2019). Abou Hashish (2017) describes OC as the general relative strength of identifying and involving individuals in a specific organization by three factors: first, a strong belief in the goals and values of an organization; second, a desire to exert a determined effort for the organization's benefit; and third, a strong willingness to be one of the members of the organization. Many studies conducted to illustrate the relationships between ES, EP, and OC (Kawiana, 2018;Shibeika, 2016;Triwahyuni & Ekowati, 2017;Dinc, 2017) concluded that ES and EP have a positive and significant influence on OC. For more than three decades, OC and ES have generated a lot of interest among researchers (Ali & Ibrahim, 2019). Besides, if OC is managed properly, employee commitment can promote beneficial results, for example, increased performance and productivity, at both the organizational and individual levels (Inanc & Ozdilek, 2015). Various studies have demonstrated that work-related factors are major determinants of ES and OC among employees. Also, there is a direct relationship between these factors and EP (Ali & Ibrahim, 2019). However, the correlation between all three concepts has been studied rarely; also, no previous studies have examined OC as a moderator among the variables of ES and EP. Unlike previous literature, this paper not only examines the impact of ES on EP and interdependency between them but, also illustrates and tests the effect of OC as a moderator of the relationship between ES and EP as shown in Fig. 1. Hence, this study developed on the totally previously mentioned outcomes to assume the following: H2: The positive relationship between employee satisfaction and employee performance will be stronger for higher organizational commitment in Jordanian Islamic banks.

Sample and data collection
This study was conducted using a quantitative research design method with a probability sampling design by using simple random sampling to select the sample size. The target population of this paper was composed of 3841 employees of Islamic banks in the Hashemite Kingdom of Jordan which are Jordan Islamic Bank, Islamic International Arab Bank, Safwa Islamic Bank, and Al-Rajhi Bank. Besides, according to Sekaran (2003), the sample size was 351 respondents from the target population, selected from all bank employee rank, from the frontline to the managerial level. Therefore, after we distributed questionnaires for this research, we utilized the self-administered process to collect the questionnaire responses from the respondents. For the purposes of scientific research and analysis, 400 questionnaires were distributed to the respondents. Several steps were taken by the researchers to improve the response rate for the current research; first, questionnaires with a cover letter that introduces the purpose of this study were personally delivered; second, respondents were informed that all the information they provided in the questionnaires would be highly confidential. In addition, the researchers relied on a few employees in each bank who were an internal point of contact. These employees took care of gathering the questionnaires from different employees and encouraged them to take an interest in the questionnaires. A total of 303 questionnaires were finally collected and usable for analysis, which represented a valid response rate of 75.7%. Babbie (1992, p. 267) mentions that this is a very good rate; "60% response is good and 70% is very good." Thus, the data tested the reliability and validity of the instruments by using SPSS software version 25, to conduct a descriptive analysis of the demographic profile of the respondents; the data were also screened and cleaned to remove outliers. The measurement and structural model of the research framework was analyzed by using SmartPLS 3.2.9.

Measurement
In this study, the questionnaire included all the constructs' measurement items by using the five point Likert-type scale of agreement, ranging from 1 "strongly disagree" to 5 "strongly agree." The instruments involving 23 items related to the three constructs of this model were adapted from past literature. Based on the context of this study the 23 items were divided into 6 items for ES measured, which were adapted from Hassan (2015), and 6 items for OC measured, which were adapted from Abdallah (2014) and Tabouli, Habtoor, and Nashief (2016). In the end, 11 items for EP included 5 items related to efficiency which were adapted from Tabouli et al. (2016) and Fadlullah and Sulaiman (2016), and 6 items for creativity adapted from Tabouli et al. (2016) and Wang and Miao (2015). Table 1 shows the items with their sources. There is an emotional attachment to this bank where I work.

EF1
I feel dedicated, confident, and able to take responsibility to doing my job.

EF2
I enjoy the professional skill and technical knowledge required to carry out the work efficiently. EF3 I do my work according to specific policies and procedures. EF4 I have the ability to provide the best services to the customers. EF5 I have the ability to do the jobs without making mistakes. CR1 I am careful to make changes in the working methods of each period. CR2 I would like to stay away from repeating what others do in solving problems related to work. CR3 I feel bored of repeating the same procedures in doing the work. CR4 I have the ability to put forward ideas and solutions rapidly to face work-related problems. CR5 I have the ability to express my thoughts fluently and freely. CR6 The services I offer come with new ideas to meet the needs of customers.

Results
This study used a quantitative approach and used partial least squares structural equation modeling (PLS-SEM) to analyze the hypotheses and test the moderator and direct effects (Sarstedt et al., 2014). The model was analyzed to asure the reliability and validity of the proposed measurement scales, using a two-stage approach: measurement model and structural model analyses (Hair et al., 2017).

Demographic profile
The demographic profile of the sample is given in Table 2; of 303 questionnaires, 222 respondents (73.3%) were male and the remaining 81 respondents (26.7%) were female. As Table 2 shows, the majority of the respondents were aged between 26 and 35 years, representing 170 (56.1%) of the total respondents. As for marital status, 72.3% of the respondents were married. Furthermore, the employee's educational background section indicated that doctorate holders, master's degree, bachelor's degree, diploma, and certificate holders were 2.3%, 12.9%, 74.6%, 8.9%, and 1.3% respectively. Regarding the position section, frontline employees (37.6%) were the highest respondents, followed by the supervisors (23.1%). Finally, the majority had between 6 and 10 years of work experience, with a percentage of 33.0%.

Measurement model analysis
SmartPLS 3.2.9 software was used to test the model with latent variables and to perform measurement model tests. First, this study found 27 cases called outliers that exceeded the permissible limit and were deleted from the total number of questionnaires. This reduced the sample number from 303 to 276. After that, the measurement model was examined to check the reliability and validity (outer model) of the instruments (Hair, Ringle, & Sarstedt, 2011). Sekaran and Bougie (2010) described reliability as the way an instrument consistently measures the concept it is meant to measure, while validity is the ability of the instrument to measure the concept it is intended to measure.
The PLS measurement model analysis comprised reliability (internal consistency) and validity (convergent and discriminant validity). The measurement model criteria are as follows: all the item loadings should be more than 0.7 or 0.6 (Tan et al., 2017;Hair et al., 2017); the composite reliability (CR) value ought to be 0.7 or greater; and the average variance extracted (AVE) should be at least 0.5 (Hair et al., 2010(Hair et al., , 2017. Meanwhile for validity testing, the convergent validity reflects whether a specific item estimates a latent variable that it is expected to measure (Tan et al., 2017;Urbach & Ahlemann, 2010), while the AVE assesses the measure of change that a build catches from its contrasting markers and the sum because of the estimation mistake (Tan et al., 2017;Fornell & Larcker, 1981;Ringle et al. 2015). Table 3 presents the latest reliable as well as valid measurement model conducted using SmartPLS version 3.2.9.  Table 3 shows the results of the reliability and validity assessments of the measurement model. As per Table 3, 3 items were deleted (ES6, OC4, and CR6) due to low factor loadings and the remaining items were above 0.6. The results also depict that the CR of all constructs exceeded the threshold value of 0.7, ranging from 0.889 to 0.963. This illustrates that the measurement model of this study is reliable (Hair et al., 2017). Additionally, the validity was examined based on its convergent and discriminant validity for the measurement model. The convergent validity of the measures was accepted, with AVE values greater than the recommended level of 0.5, ranging from 0.616 to 0.837 (Hair et al., 2017). Discriminant validity is confirmed when item loading should load more highly on own construct than on other constructs. Hence, researchers assessed discriminant validity by differentiating the square root of the AVE value of each construct with all the other constructs to be larger than the correlations between two factors (Fornell & Larcker, 1981;Barclay et al., 1995). Thus, Table 4 reveals that the correlation scores between each construct with itself is greater than all other constructs. This indicates that the discriminant validity was met by all constructs and it is acceptable. Therefore, it can be concluded that the measurement model is reliable and valid. Based on this, the data' reliability and validity of the instrument were confirmed.

Structural model analysis
The structural model consists of the constructs or latent variables and the paths that connect them with each other (Hair et al., 2017). Figure 2, is the schematic diagram of the structural model, which starts with ES, OC, and EP. The arrows linking the constructs are determined by the direction of the two hypotheses proposed in the paper. The standardized estimate for the structural model of this study shows the direct relationship between ES and EP, as well as the moderating impact of OC on the relationship between ES and EP is presented in Fig. 2. To clarify more, the coefficient of path values ranges from -1 to + 1, which expresses the depth of the relationship between any two constructs (Hair et al., 2017). Ramayah et al. (2016) specified the critical values for significance in 2-tailed tests as follows: p < 10% (1.64), p < 5% (1.96), and p < 1 (2.58); also, as usual, the researchers in marketing utilized the significance level of p < 5%. Hair et al. (2017) stated that the bias-corrected bootstrap confidence intervals (lower limit, upper limit) allow testing whether a path coefficient is significantly different from zero. If the confidence interval for an estimated path coefficient does not include zero, this means there is a significant effect of this path (Hair et al., 2017). Therefore, the researcher set 5000 subsamples with a replacement from the bootstrap cases equal to the original set of data which is 276 and the results of the structural model for this study are shown in Table 5. Note: Significance level at **p < 0.01, *p < 0.05 (two-tailed). LL, lower limit at 5%; UL, upper limit at 95% confidence interval; BC, bias corrected.

Fig. 2. Structural Model Results
As can be observed in Table 5, the path coefficients of ES were found to have a significant relationship with EP (direct effect); the result indicated that (β = 0. 601; t-statistic of 13.149), which supported H1. As for the moderation test results of OC with an interaction term (indirect effect), they were found to be statistically significant (ES→OC→EP, β = -0.176, t-statistic of 2.212), which indicates that H2 is statistically supported. Therefore, when OC is greater, the beneficial correlation between ES and EP would be greater. Also, the indirect effect 95% confidence interval (lower limit = -0.307, upper limit = -0.002) does not straddle a 0 in the middle, which also indicated that H2 is supported.
The effect size of the moderator (F 2 ) was determined for a complete description of the results, as seen in Table 6, by using the effect size calculator in Microsoft Excel or by using F Square in the SmartPLS algorithm results report. Based on the Kenny criterion (2016), the F 2 values of 0.005, 0.01, and 0.025 were found, to be low, medium, and high, respectively. Based on the results in Table 6, the interaction term's F 2 effect size for ES*OC has a value of 0.0482. According to Kenny (2016), this value indicates a medium effect. In addition, the researchers must select the standardized beta coefficients (2-way standardiesd.xls) to the interaction plot. Therefore, Dawson (2014) proposed how an interaction plot could be created via web pages such as http:/www.jeremydawson.co.uk/slopes.htm, providing internet tools for calculations and simple slope plot extractions to follow up on significant interactions, as shown in Fig. 3.

Fig. 3. Interpretation of the Interaction Results of ES×OC
The interaction plots' interpretation seems to be to look at the slope's gradient. As shown in Fig. 3, when compared to the red line (high OC), which has a positive and less steep gradient, the blue line (low OC) has a positive and steeper gradient, implying that when OC seems to be low, the positive relationship is stronger. Hence, hypothesis H2 is supported.

Discussion
In this paper, two hypotheses have been created in which ES was considered as an independent variable, whereas EP was the dependent variable. OC was a moderator between both of them. The first result of the hypothesis illustrates the positive as well as significant relationship between ES and EP. In contrast, the outcomes of the second hypothesis explain that OC had also a significant moderating relationship between ES and EP. This demonstrates that improving the level of ES in Islamic banks in Jordan (Jordan Islamic Bank, Islamic International Arab Bank, Safwa Islamic Bank, and Al-Rajhi Bank) also leads to increasing EP at work. Additionally, the findings are generalized into both the situation and the context in which employees actually feel the value of commitment, which influences their work performance significantly. The relationship between ES and EP actually recommends examining the effect between them. Several research studies have focused on the satisfaction of employees as a single variable that affects EP and their results were positive relationships (Kirimi & Maende, 2019;Huang et al., 2019;Triwahyuni & Ekowati, 2017;Adigun et al., 2017;Chen, et al., 2015). To confirm the results of this study, Triwahyuni and Ekowati (2017) indicated that ES plays a role in achieving better expectations of quality and productivity, and provides the chance to create stable employees, to use human resources more efficiently and to increase effective performance. However, the results of this study are different from Sani's (2013), which revealed that ES does not positively affect EP. Additionally, Sani (2013) has illustrated that there are factors that bring happiness and satisfaction for employees in an organization: among others challenging work, applying a fair system of rewards, stimulating working conditions, and the behavior of colleagues. Crossman and Abou-Zaki (2003) also showed that a negative relationship exists between ES and EP.

Moderator
Low OC

High OC
In this regard, ES play a fundamental role in any organization. When management appreciates their employees, they tend to make them enthusiastic and motivated. As a result, employees will shift their motivating power to works harder, which leads to a significant positive effect on their performance. On the other hand, OC also showed a positive correlation with the relationship between ES and EP in the Jordanian Islamic banks. These results clarify that the positive commitment of employees in Jordanian Islamic banks is beneficial, which may also lead to better results and improved EP. Moreover, when there is a strong alignment between employees' targets and individual values, employees expect greater levels of OC. When the employees understand, acknowledge and promote the goals of the bank, they are much more likely to grant all their loyalty to the bank and be willing to advance their level of performance.

Conclusion
This study sought to identify the relationship between ES and EP in Jordanian Islamic banks, with the moderating role of OC in one model. This study attempts to provide fresh insights into this topic. The two main objectives in this study have been accomplished and achieved through different methods of analyses. There was a clear indication that the independent variable (ES) was positively related to the dependent variable (EP). Overall, this suggests that ES has a significant relationship with EP. Additionally, the moderator, OC has a significant effect on ES and EP.
The current study certainly contributes well to the literature, since it demonstrated that a correlation between ES and EP with the moderating role of OC is yet to be investigated in the context of Jordanian Islamic banks. From the point of view of the management, the findings of this research have several important implications. These implications could be useful not only for the financial sector, such as banks and money exchange shops, but also for organization or social associations and several key stakeholders. At the same time, the success of employees in Islamic banks is connected mainly with the satisfaction of employees. Consequently, the current study believes that it is necessary to indicate how this aspect impacts the desired performance in the Jordan banking sector.
The fact that this study has provided some empirical evidence that verified OC, which is significantly relevant for EP and ES, helps bank managers to know and develop their commitment. Nonetheless, it is an important task for managing the bank to understand and know the impact of OC on positive results for its employees. In all of this, the management of the Islamic banks in Jordan must promote the dedication of their employees, prioritize them, and nurture them. This research is exceptional and unique in Jordan because of its practical application to the Islamic banking sector. This will help the Jordanian Islamic banks identify and understand factors that contribute significantly to the growth of employees and improve the bank's good reputation for consumers, as the employees strive to provide them with excellent services.