The link between talent management , organizational commitment and turnover intention : A moderated mediation model

Article history: Received: December 14, 2020 Received in revised format: January 29 2021 Accepted: March 9, 2021 Available online: March 9, 2021 It has been suggested that talent management (TM) has a direct and significant relationship with a number of employee outcomes. This is while the number of studies examining the process of TM leading to these outcomes are limited. Therefore, the main purpose of this study is to present a new model for analyzing processes of TM and its linkage with several employee and organizational outcomes that are organizational commitment and turnover intentions. Additionally, the present paper involves a mediating factor (P-O fit) alongside a moderating variable that is, Organizational Culture. A sample of 510 employees were selected from different banks located in Amman, Jordan. Mediation and moderation models were tested through structural equation modeling (SEM). The findings, being in consensus with previous studies, showed that TM has a linkage with both TI and Organizational Commitment. Mediating effect of P-O fit was shown with both aforementioned variables. In addition, the moderation effect of organizational culture on the relationship between TM and TI was found. The study contributes to the literature of the topic by providing a fit model to explain the linkage of TM and a number of organizational and employee outcomes. Bank managers can benefit by being aware and implying TM practices within their firms to further develop company advancements and attending to their employees based on new HRM trends. © 2021 by the authors; licensee Growing Science, Canada


Introduction
Employees and their attitudes and behaviors towards their organization is a highly anticipated factor among both organizations and scholars (Mensah et al., 2016). The interest has taken a rapid growth towards talent management as a concept from the 90s (Thunnissen et al., 2013). This is due to the fact that TM is crucial for organizations, especially in the modern highly competitive market and growth of globalization. Therefore, it is needed for a company to not only attract talent, but to manage and retain them on a sustainable basis for the development of organizational achievements on a global scale. Previous studies have examined different measures of TM and its correlated variables. These extends to job satisfaction, organizational commitment (particularly affective commitment), motivation, work-related encouragement, retention, organizational citizenship behavior, and person-organizational fit as well as turnover intentions alongside organizational culture (i.e. Collings & Mellahi, 2009;Bethke-Langenegger et al., 2011;Bjorkman et al., 2013;Gelens et al., 2015;Mensah et al., 2016).Therefore, the current study presents a model that TM is linked to Organizational Commitment and Turnover Intentions through mediating role of Person-Organization fit, and moderating role of Organizational Culture. Findings of this study are in consensus with previous extant literature of the topic. Contributions of this paper are first to the literature as the existence of the aforementioned variables within a model has not been examined prior to this research. Secondly, this paper extends the borders of the literature by conducting this research in the Banking Sector of Amman, located in Jordan. Structural Equation Modeling (SEM) has been conducted on the gathered data to measure the fitness of the parameters consisting of the hypothesized model of this paper.

Talent Management
The concept of employment has been a subject of change for organizations due to various factors such as, economic pressures, and demographic restrictions and increased mobility for human capital. This has created a shift towards knowledge-based economies (Guthridge, Komm, & Lawson, 2008). Sustaining business through proper adjustment of employees and positioning of people is a key for every organizations' success. This becomes more vivid and crucial as globalization grows rapidly and thus, creates a need for talent and its proper management (Boudreau & Ramstad, 2005). Talent Management (TM) practices are directed towards a specific goal or outcome within an organization. This can be retention of the employees with right talents for their jobs, or to increase the total performance of the firm. Such goals and objectives can be achieved through management by sufficiently performing a set of functions such as, critical TM competency (Saville & Holdsworth, 2000;2001). Talent Management can be defined as means of allocating the right talent in accordance to its position, development of talent and its flexibility, consideration and meeting the needs of various psychological contracts, conduction of proper strategies, and risk management (Sparrow, Hird, & Balain, 2011). TM has been facing an increase of interest from scholars and researchers in the past decade (Meyers & van Woerkom, 2014;Nijs, Gallardo-Gallardo, Dries, & Sels, 2014;Tatoglu, Glaister, & Demirbag, 2016). However, the majority of research has not been empirical, and the subject at hand requires further extensive research and studies to create a better understanding of TM as well as its relationship and/or effects on or from other organizational or individual variables. This can create a better comprehension upon Human Resource Management (HRM) practices and organizational performance, particularly in regard to talent management and its influences.

Talent Management and Model Variables
As a practice of HRM, TM has been not only studied, but proven to be influential on a number of organizational outcomes. This section of our paper shows the most relevant findings and linkages that have been tested by recent studies regarding the matter at hand. Several variables have been examined in relation to TM and its practices within a firm that are namely, job satisfaction, commitment (Affective), motivation, retention, and citizenship behavior (Collings & Mellahi, 2009;Bethke-Langenegger et al., 2011;Bjorkman et al., 2013;Chami-Malaeb & Garavan, 2013;Gelens et al., 2013Gelens et al., , 2015Luna-Arocas and Morley, 2015;Mensah et al., 2016). This is while other influential factors that can act as vital elements in the relationship between TM and organizational outcomes (and particularly employees' outcomes) have been neglected and therefore, the literature requires further research to be conducted on this matter (e.g. Dries, 2013;Gallardo-Gallardo & Thunnissen, 2016;Thunnissen, 2016). Reviewing the literature of TM, it can be seen that local culture (Organizational Culture -OC) is amongst the factors that have influence on TM strategy, specifically within multinational organizations. Through practices of TM, identification, coding, codification, and retention of talented staff is emphasized alongside its development as well as creation of an atmosphere, where organizational culture (OC) is in accordance to mentioned development (Bartlett & Ghoshal, 1992;Meng, et al., 2016). Although organizational culture has been a topic of interest since the 1980s, a consensus does not exist upon its definition (Linnenluecke & Griffiths, 2010). Collective behavioral rules, norms and standards within the organization and externally (Trice & Beyer, 1984;Meng et al., 2016); patterns of interactions among employees, and with clients as well as stakeholders (Meng et al., 2016); shared values and beliefs within the organization and its acceptance; and new staff and empowering them through indoctrination. Organizational Culture can be called a coping mechanism of employees in regard to what was mentioned previously (See Schein, 1983;Hofstede, 2000;Kanungo, 2001). The work of behavioral factors and its practices is contradictory to focusing on assumptions (Hofstede et al., 1984;Alavi et al., 2005Alavi et al., -2006. Empirical work conducted by Hofstede et al. (1984) has shown that perceived shared values and daily routines within the organization, form and shape the culture of organization in its core (including subunits within the firm). Hofstede's work offered a five-dimensional approach towards organizational culture that are namely, result-oriented; tightly controlled; job-oriented; closed system; and professional-oriented subunits (Chang et al., 2010). The current paper follows the previously mentioned 5-dimensional perspective of Hofstede et al. (1984). Another factor related to TM can be considered Person-Organization Fit (P-O fit). This factor can be defined as compatibility level that exists between an employee and his/her skills with the firm that they work in (Kristof, 1996). TM practices specifically address this fit factor as they emphasize on management of talent, its development, culture and retention (Mensah & Bawole, 2018). The effects of P-O fit have been examined by several scholars in regard to its relationship with various employee outcomes (i.e., Chami-Malaeb & Garavan, 2013;Gelens et al., 2015;Luna-Arocas and Morley, 2015). The mediating effect of P-O fit has not been focused and it requires more attention. However, this study follows the work of Mensah and Bawole (2018), in which this mediating effect has been tested in a similar context of Ghanaian Banking System. In their study, P-O fit was found to have a partial mediation effect on employees' job satisfaction. It is deemed that higher fit for the firm leads to lower levels of turnover intentions that is in relation with TM practices as they focus on tending to talents and their needs. The theory of Ability (A), Motivation (M), and Opportunity (O), (AMO) creates a bedrock for investigating the linkages from TM to employees and their outcomes (Appelbaum et al., 2000). Based on this theory, performance of an employee is defined based on their function and participation in the three elements of AMO. Referring to this theory, those employees with possession of skills that are required (A) can perform well. Additionally, with higher levels of interest towards their work (M) and with provision of expressing opportunities within the workplace (O), they can increase their performance level that is beneficial for the organization. TM practices on their nature, focus on motivating and providing incentives for talented employees, who possess desired skill set of the organization, and by allocating talents to right positions within a firm, TM practices affect an employees' performance level (Collings & Mellahi, 2009;Mensah & Bawole, 2018).
Processes and procedures that are used to clearly identify positions of vitality within a firm are the main emphasis of TM practices in a general term. Creation of a pool for talented prospects with high potential in regard to those positions on a constant basis as well as their management is the focus of TM practices. Additionally, through HR practices, talents are ensured to have increased their commitment towards their organization (Lewis & Heckman, 2006;Collings & Mellahi, 2009;Gelens et al., 2013). Therefore, it can be said that the implication of TM practices can further develop employees' commitment towards the firm. Organizational Commitment can be described as an emotional state of belief in organizational objectives and that an employee tends to remain in the firm (Hackett et al., 2001;Allen & Meyer, 1990). In the proposed model of organizational commitment by Allen and Meyer (1990), three distinctive categories were stated for commitment that are, Affective (emotional attachment towards the firm with sense of involvement and identification); Continuance (associated costs of having left a company); and Normative (the sense of obligation towards the company). Similarly, TM has been linked to other variables that are involved in the current study. Turnover intentions (TI) can be described as a tendency towards leaving an organization (Bordia & Blau, 1998). Organizational Culture has been linked to TI as the culture within an organization is a determinant of individual actions and behaviors (Schein, 1983;Hofstede, 2000;Kanungo, 2001;Krumbholz & Maiden 2001). In a study conducted by Bordia and Blau (1998) in the context of India, it was shown that organizational culture has a significant linkage with several employee attitudes and behaviors, including turnover intentions. As a consensus, this was mentioned by other scholars, that organizational culture is considered as an asset, from which performance of employees as well as organization alongside their work attitudes can be greatly impacted. Depending on individualized characteristics, organizational culture may appear desirable or unattractive for a prospect employee (Emerson, 2013). This is related to their shared values and morals and therefore can be interpreted as a predictor of turnover intention (Mensah et al., 2015).

Research Approach
The current research follows one of the main areas of interest for scholars since the 90s that is, Talent Management. After extensive review of the literature, it was noted that talent management has been found to have a positive impact on various attributes and attitudes of employees within the workplace (Allen, et al., 2003;Oehley & Theron, 2010;Magolego, Barkhuizen, & Lesenyheo, 2013;Mpofu & Barkhuizen, 2013;Barkhuizen, 2014;Barkhuizen, Diseko, & Schutte, 2015). This leads to the conception of the fact that talent management is a determinant factor within the workplace and has to be considered as vital. Furthermore, it has been mentioned in the literature of the subject that this concept required more studies and thorough understanding (e.g. Dries, 2013;Gallardo-Gallardo & Thunnissen, 2016;Thunnissen, 2016).Hence, the following hypothesis has emerged based on the existing literature: • When management support is scarce, especially towards talents, it has been noted that it is linked strongly to intentions towards turnover for the employees (Barkhuizen, 2014). This shows that the existence of management support is of necessity. Hence, (H1) There is a significant relationship between talent management and Turnover Intentions (TM&TI). • Having a supervisor with higher levels of availability, has shown to be a determinant of improved commitment towards the organization. This is while in some studies (e.g. academic members) different characteristics of organizational commitment have shown more significance than the others Mohamed & El Shaer, 2013). Therefore, (H2) there is a significant relationship between talent management and organizational commitment. • Talent Management is known to be amongst vital factors, which can link a talent towards an organization that suits him/her. Talent management practices encourage and emphasize on delivering knowledge of values and objectives to the employees through proper means of communication and therefore can be an element that increases the level of Person-Organization Fit (P-O Fit) (Collings & Mellahi, 2009;Poocharoen & Lee, 2013). Thereby. (H3) There is a significant linkage between talent management and Person-organization fit. • Based on H3, and following the existing evidence in the literature, a consistency can be seeing that fit at work can be linked with increased job satisfaction, performance, organizational citizenship, organizational and occupational commitment, and reduced turnover (Bretz & Judge, 1994;Kristof-Brown et al., 2005). Thus, (H4) Person-Organization fit is significantly related to Organizational Commitment. This suggests that P-O Fit plays a mediating role in the aforementioned relationship between TM and Org. Commitment. • In the light of the above, and as previously mentioned, Person-Organization fit has been found to have a significant impact on various work attributes. This includes turnover intentions and the fact that by being fit in a workplace, an employee will have less tendency towards leaving their job (Kristof-Brown et al., 2005). Therefore, (H5) it is expected to see that there is a significant relationship between P-O Fit and Employees' quit intentions. • Management pays a great deal of attention to its organizational culture, that can be described as a set of values, beliefs, and attitudes among members of the firm (Darmawan, 2013 as cited in Hakim, 2015). Within the literature, there has been inconsistency among the findings. While some researchers have found positive relationships between the two variables, others have found negative effects (Yeun & Han, 2015;Kessler, 2014;Dwivedi et al., 2013;Kee, 2010). Therefore, it is possible that situation, and the context, where the research is conducted can have different outcomes.
In addition, Organizational culture has been found to have effects on employee turnover (Lok & Crawford, 2001). Therefore, (H6) Organizational Culture has a moderating effect on the relationship between TM and OC, as well as having a moderating effect on the relationship that exists between TM and TI.

Questionnaire and its Measures
The measures of this study were chosen based on the most relevant and available references at hand. The table below presents the variables and references used for this particular research: All the dimensions selected for this research were paraphrased to fit the purposes of this study. In addition, a 5-item Likert scale was designed for all the questions that were included. Each characteristic of each variable was chosen reflectively to provide better fit for the parameters as well as a sufficient Degree of Freedom. Demographic variables were included. However, no direct questions regarding beliefs or income were asked for decreasing response bias. In addition, the survey was designed anonymously and participation was fully elective. Participants were given information in regard to the research as well as its purposes and the confidentiality of the collected data.

Sampling and Data Collection
The data for this research was gathered through a period of 3 weeks During April 2019. Respondents were selected from a number of banks located in Jordan. Banking sector of Jordan consists of 24 distinctive banks. However, a number of these banks do not possess the elements of talent management. After conducting conversations with managerial levels of various banks as well as observations, 5 banks were chosen to be involved with the data collection. These banks did possess the practices of talent management. Managers were informed about the means of research and necessary permissions were given. Employees were addressed during their breaks and based on their availability. Each respondent was given information and was accompanied throughout the survey. This allowed the researcher to provide more information and create a bedrock for respondents to provide more honest answers. As total number of employees in the bank sector of Jordan reaches above 20,000, a number of respondents above 370 would provide interval error of 0.05 which is a 95% confidence level, considering the number of employees in 5 selected banks. However, a total number of 510 respondents were selected for final analysis. This is after excluding biased responses and the respondents of pilot-test, which was undertaken prior to the main data collection. A convenient sampling method was undertaken due to availability of staff.

Theoretical Model
Below is the theoretical model of the current paper:

Analysis
For analyzing the proposed theoretical model of this study, Structural Equation Modeling (SEM) was used. This process was undertaken via SPSS 22 and AMOS 20. SEM is noted to be a fit approach towards models, in which mediating variables-or moderating-exist (Hair et al., 2010). This analysis was done in three steps; testing the general path of the relationships within the model; examination of mediation effect; and assessment of the moderating variable. Demographic variables were selected as control variables and their effects were not included in the process of hypothesis testing. Three levels of analysis, which were conducted on the data were, Exploratory Factor Analysis (EFA), Confirmatory Factor Analysis (CFA), and the test of relationships among variables through SEM.

Descriptive Statistics
From 510 participants, a number of 280 were "male" and 230 were "Female". The table below presents means, standard deviations, and the correlation that exists within the construct of our model. The square root of average variance extracted is shown in the table below, which is to assess the discriminant validity of the construct. This diagonal element must be greater than other extractions in the column and row that correspond to the diagonal element (Fornell & Larcker, 1981). The results of our data meet the requirements previously mentioned.

Reliability Measurement
To examine the properties of variables of this study, an initial EFA was conducted (extraction and Varimax rotation). This was followed by reliability analysis to test the internal consistency of the data at hand. The reliability was calculated through Cronbach's alpha, which is presented below, and the findings seem to be satisfactory (Hair et al., 2010).  Table below shows Factor Analysis reports conducted on the data for Talent Management. A four factor framework for Talent Management was drawn. Items, which did not contribute to coefficient alpha substantially or with less than 0.5 correlation, were excluded.

Hypothesis Testing
For testing the hypotheses of this research, SEM was used within AMOS 20. General path, mediation effect, and then moderation effect were three steps, from which the data was tested to whether accept or reject our hypotheses. Demographic variables, as previously mentioned were controlled in terms of their effects and therefore, were not included. Confirmatory Factor Analysis (CFA) was conducted to test the hypothesized model of this study by item covariance matrix (maximum likelihood estimation). After being assessed through CFA, the hypothesized model shows the best fit for the model as a measurement. The results show a fit of χ2 = 198.08, df = 118, CFI = 0.94, TLI = 0.95, RMSEA = 0.06, PCLOSE = 0.08, that is against the alternative model that had no direct link between TM and TI. These are the data from the hypothesized 5-factor model (See Fig. 1). The mediation model of P-O Fit, shows a good fit of χ2 = 59.1, df = 41, CFI = 0.95, TLI = 0.94, RMSEA = 0.06. This shows that the mediation has been observed from P-O Fit on the relationship that exists between TM and TI. This is in consensus with previous studies (e.g. Glaister et al., 2018). The findings show that in overall sense, there is a negative relationship between TM and TI, which supports H1. In addition, TM and Org. Commitments were found to have a significantly positive relationship, which supports H2. Moreover, the mediation effect of Person-Organization Fit was found to be in full degree. This finding supports both H3 and H4. Based on the findings of Table 5, it can be seen that H5 is also supported due to significance of linkage between P.O Fit and TI. Following the work of Hair et al. (2010), the moderation effect that was hypothesized in the model was tested. This was shown to be of significance. In other words, the moderating effect of Organizational Culture on the relationship between TM and TI was found to be significant (Non-std. estimate being 0.75* (*p< .01). The analysis of moderation effect was conducted in AMOS 20. The moderation effect of Organizational Culture was found to have a strengthening effect on the linkage between Talent Management Practices and Turnover Intentions. Hence, H6 of the study is also supported. Fig. 2 presents both mediation and moderation effects.

Discussion
Main objectives of this research were to examine the mediating effect of Person-Organization Fit in the relationship between Talent Management and Turnover Intentions as well as Organizational Commitment. The findings of this study were in consensus with previous researches on the topic, showing that Person-Organization Fit does mediate the relationship (Mensah et al., 2016). However, one of the contributions of the current paper is that the mediating effect of Person-Organization Fit has not been tested against Turnover Intentions in the context of Talent Management. According to Thunnissen (2016), Talent management requires further research in terms of empirical studies. Hence, our understanding on the depth of reasons that make Talent management such an important role is dim. The findings of this paper showed that TM has direct and indirect, and significant relationship with both TI and Organizational Commitment. In addition, the mediating effect of P-O Fit was shown in the analysis section of this research. This implies that both variables (TM and P-O Fit) are determinant of employees' outcome within organization, and in particular within the banking sector of Jordan.
Second fold of this study was to assess whether Organizational Culture has a moderating effect on the relationship between TM and TI. This was also found to be of significance within the model fit of this study. Although different aspects of organizational culture have been examined through the literature, this particular aspect has not been tested. Henceforth, this can be considered as another contribution of this paper. Organizational culture moderates the aforementioned relationship as without the presence of Organizational Culture, the strength of TM's effects on TI are considerably lower. Thus, the existence of Organizational Culture and its characteristics, can greatly affect employees' turnover intentions, when it is appropriately fitted within TM and its practices. It can be understood that without having basic implications of TM practices within an organization, other variables involved in this model would not be significant. This is due to the fact that TM cannot exist in isolation and it requires a holistic approach within an organization. Particularly for this research, banks and their management must imply TM practices in their firms. This implication should be involved with every aspect of the organization and therefore, create a bedrock for organizational culture, P-O Fit, and other elements that play a vital role in employees' outcomes to act and coexist with the foundation of TM inside the bank. This enables the culture of organization to be enhanced, fitness of recruits to be considered, and satisfaction, commitment, and lower turnover intentions to be achieved for the firm. Flexibility within TM practices (Collings & Mellahi, 2013) suggests that by properly implicating these practices, the firm can be more adaptable to changes, which in turn will enable the employees to perform better and engage in behaviors that are desired by the organization. Each of the dimensions of TM showed significance in their existence as parameters in the hypothesized Organizational Culture

Turnover Intentions
Org. Commitment 0.44* 0.33* 0.60** 0.47* model of this study. Then, it can be deduced that TM does provide the bedrock for organizational advances as well as enabling other influential variables for employee outcome to become active. This shows that managers must base their decision-making and planning on modern HRM practices such as TM and include this system in processes of development, selection, retention, recruitment, training, and performance. Awareness of managers on implications and outcomes of HRM practices in firms can have an extreme effect on the development planning processes, which in turn will increase social capital, broaden company objectives, enhancement of strategies, and therefore elevate company outcomes and achievements.

Limitations and Future Research Recommendations
This study was limited despite its contributions in terms of presented model and establishment of mediating effect of P-O Fit on TM and TI as well as TM and OC alongside moderation effect of Organizational Culture on TM and TI linkage. This study does not examine talent management and its dimensions from a pre-establishment perspective. Therefore, interpretations, understanding, and conceptualization of this phenomenon is not addressed. Therefore, future researchers can undertake such an approach to further analyze TM. In addition, this research was addressed towards employees and staff of the banking sector in Jordan. A managerial approach can be undertaken. Comparative research can be conducted in different areas or various sectors within Jordan. Due to the lack of sufficient information for the Middle East region, neighboring countries can be investigated to understand differences. Similarly, other factors and variables can be added to the model to further test different influential elements in this context. Cultural aspect of this model can be emphasized in further research to shine upon the depth of cultural impact within organizations.