HRM PRACTICES AS DETERMINATES OF THE EMPLOYEE INTENTION TO STAY: THE MODERATING EFFECT OF TRANSFORMATIONAL LEADERSHIP: A CASE OF HIGHER EDUCATION INSTITUTES IN UAE

Purpose of Study: The main objective of the current study is to investigate the impact of HRM practices on the intention to stay among the lecturers working in the higher education institutes of UAE. In addition to that the study has examined the moderating role of transformational leadership in the relationship between HRM practices and intention to stay has examined. This study is important to the management and HR department in the education industry because through this study, higher learning education institutions are informed of the results of HRM practices and transformational leadership style which might have an influence on the lecturers’ intention to stay. The study is among the pioneering studies on the issues related to the HRM practices on the intention to stay. Methodology: So, current study has used SEM-PLS as statistical tool to answer the research questions raised in this study and research objectives envisaged in the current study. Results: The findings of the current study have provided support to with the hypothesized results. The study has argued that the to deal with the current external opportunities and menaces, new knowledge and skills for improving their existing and future performances must be provided for the organizations This study will be helpful for policymakers and researchers in examining the link between HRM practices on the intention to stay the higher education institutes of UAE.


INTRODUCTION
There is not denying human resource is an important asset in an organization because human resource supports the management team to achieve organization's goals and objectives in order to produce their business functions such as managing cash flow, business transactions and dealing with customers to promote their products and services. Without human resource, organizations will not be able to produce their business results and the management will not achieve the organization's vision and mission. Nowadays, job-hopping or resignation is becoming a trend among employees in any organization. Many researchers have studied the topic of employee intention to leave. In these studies, the researchers identified many reasons employees harbour the intention to leave the organization. Basically, employees will leave because they are unhappy or dissatisfied with many factors related to the organization.
The decision of employees to leave an organization is influenced through seven factors. These factors include mismatch job abilities and job requirement, limited coaching and provision of feedback, less growth and development opportunities, workplace lacking the expectations, workload stress, non-recognition, and undervaluation. Other factors include imbalance of work life and mistrust relation with the senior management.
Trend of employees' resign from occupation becomes a problem to the organization resulting in the loss of talented employees, loss of organization's sales and profits, loss of customers' satisfaction and loss of business opportunities. Besides, employees' resignations also put pressures on existing employees because they have to carry extra workload, meanwhile the organization also has to bear high cost to recruit and provide training for new employees.
The mushrooming of higher education institutions in UAE shows that higher education sector is an important sector in UAE because it helps the country's economic development, for it to transform into a high income nation by the year 2020. It also impacts productivity and human capital development in UAE. According to Ministry of Education (2014), in UAE more than 900,000 students pursuing their higher education in 20 public universities, 79 private universities, 7 branch campus of foreign universities, 30 polytechnics, 80 community colleges and 341 private's colleges. Higher education institutions in UAE are involved in teaching as well as conducting research because these institutions also provide professional training to students to prepare them for high-level job requirements and contribute to the development of social economics (Ng'ethe et al., 2012). Even though higher education institutions in UAE are growing and the number of students increase every year, high turnover rate among lecturers still exists. According to Researcher, the intention to leave among private lecturers from UAE higher education institutions was increased from 10% to 16% in 2009 until 2010. While, UAE Employers Federation (2012), recorded the turnover rate in the education sector scored forth highest in turnover rate with 29.28% in 2010 until 2011. Thus, this statistic shows that the turnover rate among lecturers teaching at private higher education institutions in UAE increases at an alarming rate every year (Merriam et al., 2007). UAE is among world fastest growing economies with over 7 million of local and international workforce. According to survey by Widodo (2018) the turnover rate in UAE were expected to raise by twenty three percent by 2018. According to latest statistics, in the salary and employment report of UAE government, the turnover rate in 2012 was 20 percent which in 2018 has reached to 23.16 percent (See figure 1) Other arising issue is the unfair or bias in evaluating performance appraisal. Leaders who badly evaluate their lecturers' performance due to the dissatisfaction they feel towards them. In the performance appraisal or review process, leaders annually record his/ her suggestions of the lecturers' performance on a document delivered by the Human Resource (HR) Department. Normally, the leaders evaluate the lecturers' performance based on the most current events that reflect what the leaders can remember and overlook pass outstanding performance. This outcome is sometimes demotivate the lecturers. One of the most important factors that have influence on intention to stay is the interaction transformational leadership style in the organization. Leaders must know how to implement the performance appraisal process because through the evaluation result and conductive feedback from transformational leadership, it will helps existing lecturers to be promoted and salary increment. Thus, based on this relationship, existing lecturers will look they have a job secure and influence them to continue working in this college for a long period of time.
Therefore, Human Resource (HR) Department must play a pivotal role to manage HRM practices in the organization because a good practice in HRM functions and implement transformational leadership style in the organization will influence lecturers' intention to stay and also help organizations in making the right decisions when formulating future action plans to support the achievement of business outcomes while sustaining an economical or optimal level of resources.
This study is important to the management and HR department in the education industry because through this study, higher learning education institutions are informed of the results of HRM practices and transformational leadership style which might have an influence on the lecturers' intention to stay. Some situations such as career development opportunities, performance appraisal, compensation and intervention transformational leadership in organization may lead to lecturers' intention to stay. There are many factors influencing lecturers to stay in one organisation for example they are given the opportunity to do their best effort to complete their tasks and highly appreciated by their leaders. Some of them feel happy and satisfied when their work is recognized and being trusted to carry out a task. This study is not only beneficial for the education industry. Leaders in other industries such as hospitality, manufacturing and food can also get a clear picture on the conditions and factor that impact the intention of lecturers and instructors to stay in one organisation. Hence, from the viewpoint of a company or institution, they will recognize the effects of HRM practices and transformational leadership style towards the lecturers' intention to stay in the current organization. Consequently, the management will try to increase the intention to stay among lecturers by paying attention to the HRM practices and transformational leadership.

Relationship between Career Development Opportunities and Intention to Stay
Employees, who feel that they have an opportunity to develop their career, will normally enjoy their job and perform well. However, if their position does not come with career development, the chance for them to find other jobs is high. A career without the possibility to advance in the four stages, namely establishment, advancement, maintenance and withdrawal has a major influence on the employee's intention to leave the organization (Super, 1957). Usually, in organizations there are limited positions and not everyone can get promoted. When employees have different career opportunities to develop their career, they will have different solutions. For example, an employee who is comfortable with the position may tolerate this kind of situation and continues to work in the organization. But, an employee who is ambitious may not have the intention to stay for a long time in the organization and may leave this job as soon as possible. Hence, everyone will look for other jobs that are close to his or her career development opportunities and it will affect one's decision to stay or leave the job.
When opportunity is offered to the employees for development of their professional career, they are likely to stay in the organization. Employees are motivation through provision of promotion opportunities in the organization. Employees feel . Good career opportunities should be provided to the employees as it makes them valuable resource for the organization. They will work harder for the organizational performance and engage them in the job related activities. The risk of decision for leaving the job will be reduced. Employees have greater chances of getting promotion, which makes them stay longer in the organization. Provision of career development opportunities makes it an appreciation for the employees.
H1: Career development is in significant relationship with intention to stay

Relationship between Performance Appraisal and Intention to Stay
Performance appraisal clearly has an effect on employees' performance. Through the fair process in performance appraisal, employees can be more productive and committed to perform their tasks. Appraisal process is perceived by employees to be fair when performance evaluation is done based on standards and fair policies (Erdogan, 2002;Subair and Oriogu, 2016;Tambun et al., 2018). The performance appraisal perception of employees is reflected through their behavior at work. The perception of fairness is contributed by an effective performance appraisal system. This makes them to take decision for staying the organization. A fair treatment by employers is taken into consideration by the employees for continuing with their job. Alternatively, an unfair treatment of employees makes it feel negative about the work environment and a decision to quit the job. The working behavior or performance of employees is influenced by the performance appraisal and opportunities for promotion. This also influences their intention to stay in the organization.
H2: Performance appraisal is in significant relationship with intention to stay

Association between Compensation Strategy and Employee's Intention to Stay
Generally, compensation can be defined as the payment provided by the organization to employees in a fixed period. According to Mondy and Noe (2005), compensation is a cumulative financial and non-financial rewards payable by organizations as a return for their employees' services. Hasibuan (2006) mentioned compensation as something that is received by the employees as the reward to pay their services in terms of money, either directly or indirectly. According to scholar, direct financial payments are salaries, wages, incentive bonus and commissions that an organization pays for the work performed and it tends to reflect the value of the work or skills; and indirect financial payment such as insurance to protect the employees during his/ her tenure with the organization.
Past studies have shown the positive relationship between compensation and intention to stay (Gamage and Herath, 2013; Tyagi and Siddiqui, 2017). Employees who are satisfied with the total of compensation are more important and more closely related with overall satisfaction. Thus, it is not surprising if compensation becomes a major reason why employees remain in the organization. According to Tzeng (2002), employees who are not satisfied with the compensation they receive are significantly related to incidence of organizational stress and depression, productivity loss and high intention to leave. Thus, it is very important for organizations to provide equitable compensation systems because it will influence employees' positive attitude towards the organization such as performing well in their tasks and willing to remain in the organization ( H3: Compensation Strategy is in significant relationship with intention to stay

Relation between Employee's Intention to stay and Transformational Leadership
The intention of employees to stay in an organization is positively related to transformational leadership (Avolio et al., 2004). The role of transformational leadership is to empower its followers and motivate them to accomplish tasks. H4: Transformational Leadership is in significant relationship with intention to stay

Transformational Leadership as a Moderating Variable
The focus of transformational leadership is on improving the performance of employees and offers them development opportunities. Transformational leaders consider four characteristics such as charisma, motivation of inspiration, intellectual stimulation, and individual recognition to the employees or followers (Bass and Avolio, 1994). The role of transformational leadership is important to be explored as a mediating variable in the relation of HRM practices and employee's intention to stay affects the decision to stay in the organization. This will reveal the conceptual framework and explain the mechanism of association. The role of transformational leadership on strengthening or weakening the association between HRM practices and intention to stay will be determined through analysis (Sarwar et al., 2014). Variables improving the association between employee's intention to stay and HRM practices have been identified in this research study. Transformational leadership is a crucial variable, which have the ability to influence this relation. It has been found that the level of satisfaction of employees and their performance is positively influenced through transformational leadership. This also affects on their decision to continue with the current job.
The environment is created by transformational leadership, which is conducive to learning. This increases the level of their knowledge, abilities, skills through training and development programs. Along with the skill development through training programs, employees come across opportunities for promotion (Sarwar et al., 2014). The work performance is enhanced through growth and development opportunities, which are provided by the transformational leadership (Sosik et al., 2004). A clear way is provided to the employees by transformational leader. Employees have long journey for career development in the organization, which are willing to continue their job. It is ensured by the transformational leader that the relation between employees and the organization is long term. Job security is guaranteed by the transformational leader even when there are financial issues. Employees are motivation through this economic activity to work more than expectations. Employees are explained by transformational leadership about rewards on the achievement of set target. The intention of employees to stay in the organization is influence through transformational style of leadership, which has an intervening role. It creates a sense of shared vision and tasks, which improve the level of job satisfaction of the employees.
H5 : Transformational leadership moderates the relationship between Career development and intention to stay.
H6: Transformational leadership moderates the relationship between performance Appraisal and intention to stay.
H7: Transformational leadership moderates the relationship between compensation strategy and intention to stay. respectively. For populations of 100,000 people, 1,058 respondents are required for 3% accuracy and 383 respondents are required for a 5% accurate at alpha = 0.05. According to the researcher, the larger the sample size, the higher will be the confidence level, the smaller will be the error variance, the better will be representation of results, the more homogenous will be the sample, and the richer will be the information. Krejcie and Morgan (1970) describe, for a population of 6,327 as in the present study, a sample size of 364 is adequate. According to the table, a sample size of 361-364 should be obtained from a population ranging from 6000-7000 and as the population of the study is 6327, the researcher selected 364 as the sample size.The present study employed a purposive sample method to select the sample because of the difficulties faced in using random sampling method. Employees who participated in soft skill training or those who were sponsored to attend by the universities are selected as respondents to the questionnaire.

METHODOLOGY
In this study, an aggregate of 450 questionnaires were distributed to academic staff in ten public universities in Nigeria within five out of the six geo-political zones in Nigeria with the exemption of North-East geo-political zone of the country due to the prevalence of insecurity in that geo-political zone. 367 copies were returned which represent 81.6 % response rate. However, out of these 367 questionnaires that were returned, a total of 346 were usable for data analysis. The remaining 21 questionnaires were excluded due to large numbers of un-responded items in the questionnaire. This accounted for 76.9% valid response rate which is higher than 30% .

RESULTS
Fornell and Larcker (1982) PLS-SEM was used for the analysis of the data because it is recommended to be used by Cassel et al. (1999) when the model is complex and in this study there are four constructs which are in second order form and 99 indicators. It is also essential because of the items in the study are formative and reflective in nature which other software analysis may not appropriately handled (Hair Jr et al., 2014;Yaron, 2018). Furthermore, PLS-SEM is suitable for this study as it takes into account measurement error and as a matter of fact, it is categorically require by some highly indexed journals.
Using the PLS approach of structural equation modelling to source information regarding the relationship between the variables of the study; the analysis of the model was carried out viz-a viz measurement model and the structural model . PLS was used to assess the measurement model for theory confirmation, to suggest possible relationship among variables and for prediction since PLS assumes that all measured variance can be explained in a study.
Measurement model according to Hair Jr et al. (2014) indicates the association between the latent or unobserved variables and the observed/measured variables (items/indicators/ scales for each construct). In evaluating the measurement model, the confirmatory factor analysis was carried out to measure the construct validity (discriminate validity and convergent validity) and reliability of the items. The average variance extracted (AVE) as well as the composite reliability (CR) for the variables in the study was calculated. According to Nunnally and Bernstein (1994), the CR must be ≥ 0.7, AVE ≥ 0.5 while Nunnally and Bernstein (1994) recommends a value of 0.7 for the Cronbach alpha. The convergent validity are ascertained when the outer model loadings are greater than 1.96 at 0.05 level of significance. Once the measurement model is validated and found reliable, the structural model was also assessed. Indicator reliability can be defined as the proportion of indicator variance that is explained by the latent variable. The value is between 0 and 1. According to Hulland (1999), indicator reliability is assessed by examining the outer loadings of each construct measures. That is, when indicator and latent variable are standardized, the indicator reliability equals the squared indicator loading. Following the rule of thumb that any reflective indicators whose loadings within the PLS model are smaller than 0.4 should be eliminated (Hair Jr et al., 2014). However, no item in the measurement model is deleted as the least and only loading less than 0.7

Figure 2: Measurement Model
The internal consistency reliability is assessed after the unidimensionality of the indicators have been carried out. The Partial Least Square Structural Equation Modelling (PLS-SEM) employs the use of composite reliability (ρc) instead of Cronbach's alpha (α) which estimate the reliability based on the inter-correlations of the observed indicators variables to measure the internal consistency reliability. The prioritization of items in accordance with their individual reliability by PLS-SEM couples with the limitations of Cronbach's alpha (α) such as it assumes equality of all indicators loadings; it is sensitive to the number of indicators on a construct; and it underestimate the internal consistency reliability has made it imperatives for an alternative means of measuring internal consistency reliability which composite reliability (ρc) has readily fill the gaps. This measured the extent to which each indicator of a constructs share a high proportions of variance and converges in comparison to indicators measuring other constructs. Convergent validity tests if whether an item measures the construct it is expected to Measure. The criterion for measuring convergent validity is the AVE proposed by . AVE which is equivalent to the communality of a construct is the sum of square loadings of indicators associated with a construct divided by the number of indicators. Convergent validity is achieved when the AVE value is 0.50 and above which means that, the construct explains more than half of the variance of its indicators on the average. When the value of AVE is below the threshold value of 0.50, convergent validity is not achieved because the construct on the average cannot explain the variance of its indicators due to errors in the items (Urbach and Ahlemann, 2010;Hair Jr et al., 2014). As shown in Table …, there is adequate convergent validity of the measures as their AVE values ranges from 0.610 to 0.814. Which exceed the minimum acceptable level of 0.5 as suggested by Bagozzi and Yi (1988) and thus, indicating adequate convergent validity. Discriminant validity is defined as the extent to which the measures of a construct are distinct from the measures of another constructs by empirical standards. When discriminant validity is established, it means that, a constructs is distinct in its representation of a phenomena in comparison to other constructs in the model. Two methods have been proposed for measuring discriminant validity in a reflective measurement model, namely: (a) Examination of the indicators crossloadings. The indicators loadings for a particular construct should be greater than its loadings (cross-loadings) on the other constructs in the same model under consideration. Where any of the cross-loading is greater than the actual construct loading, then discriminant validity is violated and not achieved for that particular construct; (b)  criterion-a conservative method of assessing discriminant validity examine and compare the square root of AVE of each latent construct with the latent variable correlations of other latent construct. The square root of AVE should be greater than its correlations with other constructs, otherwise Fornell and Larcker (1981) discriminant validity criterion assumed not meet for a reflective measurement models (Hair Jr et al., 2014). After confirmation for the non-existence of collinearity issue, estimating structural model is the next step. The key measures for estimating PLS-SEM's structural model are coefficient of determination (R 2 ), significance of path coefficients, predictive relevance (Q 2 ), and the effect size (f2).
Afterwards, a bootstrapping method was performed which begins with the determination of path model for the direct association among the dependent and independent variables, having no mediators. The path models involve t-values and path coefficients by using bootstrapping method and algorithm (Hair Jr et al., 2013). The next step involves path model estimation through the mediating variable. The main concern is to observe whether the relationship between mediator and independent variables and between dependent variable and a mediator are significant. Assessing mediating effect is necessary but is insufficient. Lastly, in order to analyze the significant role of indirect effect, two significant path coefficients' product is divided by the product's standard error.
A systematic model analysis was done for the structural model, in order to get clear picture of the outcomes for comprehensive testing of the proposed hypotheses. The inner model is evaluated by assessing the direct association among the independent and the dependent variables. PLS-SEM Algorithm was employed to determine the size of path coefficients, whereas, bootstrapping procedure is also employed in SmartPLS 3.0 to check the significance of association. For the number of cases, the original number is used and for the bootstrapping procedure 5000 samples were taken (Henseler et al., 2009;Hair et al., 2011;Hair Jr et al., 2013). The direct association among the dependent and independent variables were analyzed in the first model i.e. H1-H4, and a mediating variable is added into the second model to check the relation among this mediator and independent variables, i.e. H6-H8. In addition, the relation among dependent variable and mediator was also analyzed.  Predictive relevance of the structural model has also been observed for the study using Stone-Geisser criterion. The criterion observed that whether the inner model predict the endogenous indicators of the latent construct (Henseler et al., 2009). Thus, Stone-Geisser's Q2 test was performed through blindfolding procedure to check the predictive relevance of Q2 (Henseler et al., 2009;Hair Jr et al., 2013) and to obtain the measures of cross-validated redundancy for the latent construct.

CONCLUSION
The previous studies were based on finding the direct relation between HRM practices and Intention to Stay. However, this study has explored the relation through intervening force of transformational leadership. It is interesting to reveal the effect of moderator on the relation. In future studies, other variables can be incorporated such as strategic HRM, employee relations, and selection and recruitment policies as exogenous variables. The transformational leadership can be regarded as a mediator and intention to stay as a dependent variable. An organization with greater turnover rate can be selected to evaluate its outcome and their effectiveness for the organization. The results will be presented as solution to the problems. The loyalty of lecturers working in the college s exposed to change. It has been showing by this research study that compensation and performance appraisal is positively associated with the employee intention to stay.
A positive impact is not incurred on lecturers through the intervention of transformational leadership. There is need for evaluation other HRM practices to identify the significant factors. The intention to stay in college is not positively affected by the mediating role of transformational leadership. There is need for the management to explore ways for encouraging the lecturers to stay in college through recognition and appreciation. Review sessions at regular interval play important role in making lecturers recognized of their value. Lecturers are likely to achieve high ranks and work for long tenure. They must be informed about the update policies and procedures. It is important to take an interview before an employee leaves the organization. This could avoid the risk of losing other employees through improving the current policies, environment, or practices. Most significant factors behind an employee leaving the job include lack of recognition, lack of association with the management and colleagues, less opportunities for growth, etc. the turnover rate can be reduced by identifying the possible reasons and taking action against them.