The Holistic Model of Labour Retention: The Impact of Workplace Wellbeing Factors on Employee Retention

Abstract

This paper explores the holistic context of workforce retention. The global labour shortages in developed countries have made employers realise that in a world of scarce resources, employee wellbeing and retention are key factors in competitiveness. The aim of the research is to create a model to identify the key determinants of employee well-being and workforce retention. A quantitative research methodology was applied, using a questionnaire with 58 validated statements, completed online by the research participants (n = 406). The PLS-SEM method was used for data analysis and inner and outer modelling. The measurement model was tested for internal consistency reliability and convergent and discriminant validity. Cronbach’s α and CR values were above the 0.7 threshold for all constructions, indicating high internal consistency of measurements. In our model, there are a total of 36 significant relationships between latent variables. Based on the research results, the effect of organizational commitment on the intention to quit was determined. If the organizational commitment within an organization changes, the intention to resign changes. Thus, critical variables affecting workforce retention (benefits, promotion, communication, nature of work, coworkers, and normative commitment) have been identified, the changing of which will affect organizational commitment. The results support that if employees perceive these factors negatively, their commitment will drastically decrease.

1. Introduction

The success and competitiveness of every organization depend primarily on a talented and committed workforce (Hadi and Ahmed 2018). According to literature sources, committed employees perform their tasks more effectively, can be burdened at a higher level, are more open to challenges and cooperation (Boehm and Lyubomirsky 2008; Zelenski et al. 2008), which also results in positive feedback from partners (Raišienė et al. 2023; Schaufeli and Greenglass 2001). Employers are beginning to recognize that employee wellbeing is a key factor in competitiveness, as employees’ expertise and loyalty have a fundamental impact on organizational performance (Santos and Lousã 2022). Organizations that pay attention to employee wellbeing can gain a competitive advantage in the long run (Binnewies and Wörnlein 2011; Hussain et al. 2022; Wright and Cropanzano 2004). Employee wellbeing is extremely important, as the positive consequences of employee wellbeing include higher productivity, higher levels of organizational commitment, lower intent to leave, and better retention rates (Horváthová et al. 2021; Spector 1997; Suárez-Albanchez et al. 2022).

Today, retaining valuable employees has become more important than ever, as the workforce remains one of the few resources that can give organizations a sustainable competitive advantage (Barney 1991; Bite and Konczos-Szombathelyi 2020; Hall 1993). It follows that managing employee turnover is one of the biggest organizational challenges of our time (Bite et al. 2020; Pfeffer and Sutton 2006), as replacing employees can only be achieved through costly and time-consuming processes. This includes recruitment, selection, onboarding of new employees, and the training and development needed to achieve good performance. Consequently, if an employee with the appropriate expertise and organization-specific knowledge decides to leave the organization, it means that, in addition to the significant costs, the organizational efforts so far have been wasted (Collins and Smith 2006).

Increasing employee engagement and reducing employee turnover cannot be achieved without corporate strategies and measures that target retention. The ability of organizations to retain a workforce is determined by their ability to meet the expectations of employees and to provide working conditions that are adapted to the needs of employees. The retention strategy should therefore include measures to encourage employees to stay longer in the organization by increasing their satisfaction. Thus, an effective retention strategy cannot be developed without knowing employees’ satisfaction with different work-related factors (Kóczy et al. 2022; Pimentel and Pereira 2022).

The study contributes to the literature by filling several research gaps. Previous research has studied the subject from a number of angles, including turnover. Based on Dysvik and Kuvaas (2010), some authors examined the demographic factors of turnover (e.g., Griffeth et al. 2000), the relationship between employees and organizations (e.g., Meyer et al. 2002; Rhoades and Eisenberger 2002), and the effect of workplace stress (e.g., Podsakoff et al. 2007; Raza et al. 2023). However, literature sources have come to different conclusions regarding the factors predicting turnover, which, according to Dysvik and Kuvaas (2010), is mainly due to different cultural characteristics. As far as we know, no similar research has been carried out in Hungary before, so the study reveals the peculiarities of the subject that have not been examined so far.

Research on employee intention to quit has a long history, though the assumption that intention to quit and employee commitment are influenced by different factors has only gained traction recently (Madigan and Kim 2021; Holtom et al. 2008; Lee et al. 2004). According to Rai et al. (2019), the literature on employee retention has mainly focused on the factors associated with employee departures and much less is known about the factors that reinforce employees’ intention to stay. This study aims to contribute to the international literature on the subject by taking a holistic approach to the factors that influence employee retention, such as intention to leave, organizational commitment, and normative commitment.

As a criticism of studies on employee wellbeing, it is important to note that most of them only take a few factors into account when studying the topic. Based on the literature search, few studies have been published that take a holistic approach and aim to examine as many factors of employee wellbeing as possible. In our view, a holistic approach is needed to examine the issues of employee wellbeing and labour retention.

The aim of the study is to create a model that identifies the key determinants of employee wellbeing and workforce retention. The research will attempt to assess the wellbeing factors affecting employee retention that affect both organizational commitment and intention to quit. The research examines the question of which workplace wellbeing factors influence employee retention and how these factors affect organisational commitment and intention to leave.

2. Literature Review

The aim of the literature review was to explore and compare the previously published international results on the topic. The chapter consists of two parts, the first of which presents the dimensions of employee wellbeing, while the other part presents the dimensions of labour retention. The factors affecting employee wellbeing were explained through the satisfaction factors of Spector (1985). The first chapter of the literature review presents the positions of the literature on a total of nine factors, while the topic of workforce retention is presented through the dimensions of organizational commitment, normative commitment, and intention to quit.

2.1. Dimensions of Employee Wellbeing

2.1.1. Nature of Work

The nature of work includes factors such as meaningfulness, interest, complexity, clarity and variety of tasks, cooperation with colleagues, and autonomy in the work. According to Liu and Li (2012), Jackson et al. (2003), and Van Dijk et al. (2012), meaningful, challenging, complex tasks and responsible, independent, and flexible work positively affect the wellbeing of employees, while routine work processes that do not require creativity do not. According to Wrzesniewski et al. (2003), the meaning of work is the perception of employees about their activities at work and their importance, which includes factors such as the variety of work, identification with tasks, and the importance of tasks and autonomy. Suárez-Albanchez et al. (20emphasizeizes that managers should ensure that talented employees can perform their tasks independently and participate in decision-making in their field of expertise. It is also essential that workers are given meaningful and responsible tasks that enable them to use their skills as effectively as possible. In relation to retention, several authors concluded that meaningful tasks at work reduce employees’ intention to quit (Arnoux-Nicolas et al. 2016). In line with this, other authors have identified positive consequences of meaning in work such as job satisfaction, higher levels of engagement, performance, motivation, and openness (Jung and Yoon 2016; Ochoa Pacheco et al. 2023; Karacsony et al. 2022). In addition to the meaning of the work, Tims and Bakker (2010) concluded that an adequate level of workplace requirements also contributes to reducing the intention of employees to quit.

2.1.2. Coworkers

The literature is unified on the role of social relationships in the workplace and the workplace community in labour retention (Glew 2012; Nassar et al. 2022; Pitts et al. 2011). Experts studying the subject agree that good employee relationships contribute to long-term employee engagement, so employers should strive to build a supportive and cohesive workplace community (Alexander et al. 1998; Zayed et al. 2022). In order to retain employees, the issue of team integration should also be an important consideration during selection, as conflicts arising from the different personalities of employees can lead to the deterioration of employee relations (Allen 2006; Shouman et al. 2022; To and Yu 2022; Szentgróti and Tapolczai 2011). If the relationship between direct colleagues is inadequate or there is a conflict, there may be a rivalry in the workplace, which not only negatively affects the quality of work, but may even result in the loss of talented employees (Glew 2012). Jasper (2007) draws attention to the importance of the manager–subordinate relationship. According to his results, a bad relationship with the leader is the second most common reason for termination. According to Allen (2006), managers have an important role to play in maintaining a positive work environment, integrating new employees, and resolving conflicts between employees, so a good manager–subordinate relationship positively affects employee retention.

2.1.3. Supervision

Employees’ views on the organization are heavily influenced by their relationship with their supervisor (Eisenberger et al. 1990; Karacsony et al. 2020). According to Irshad and Afridi (2007), employees who terminate their employment do not actually leave the organization, but their manager. Accordingly, Bryant and Allen (2013) found that the strength of the relationship between employee and manager is one of the most reliable predictors of fluctuation.

Several authors agree that leadership style plays a decisive role in retaining employees. According to Chen and Silverthorne (2005), Romão et al. (2022), Yücel (2021), and Vinh et al. (2022) leadership style can positively affect job satisfaction, which in turn positively affects organizational commitment and job performance. Irshad and Afridi (2007), Joel et al. (2023) and Zayed et al. (2022) found that if superiors support their subordinates, communicate openly, and maintain good relationships with them, employees’ turnover intentions will decrease. Alblooshi et al. (2021) draw attention to the fact that the organization must employ managers who support employees and create a work environment in which employees are happy to work. In his opinion, employee satisfaction can be increased if managers ensure that employees can make better use of their skills. According to Noah (2008), it is extremely important for employees’ commitment that managers trust their opinions. If employees are given the opportunity to participate in decision making, they will feel valued by the organization and that they are an important part of the organization, and this will contribute significantly to increasing their loyalty (Curtis and Wright 2001; Nagyová and Gyurián 2018).

Eisenberger et al. (2002) also draw attention to the role of leadership support. In their opinion, employees who feel supported by their supervisor in their daily work will be committed to the organization. In addition to leadership support, feedback on performance is also extremely important. In order to support them, employees need to be aware of their duties and performance, which requires continuous management feedback. Positive feedback on performance is essential to motivate employees (Curtis and Wright 2001). According to Irshad and Afridi (2007), employees who feel valued are actively involved in achieving the goals of the organisation and are productive and engaged, which reduces absenteeism and fluctuation.

2.1.4. Pay

Researchers’ views on employee satisfaction with pay levels and labour retention are divided. Some authors argue that pay plays a primary role in building engagement, while others argue that it has no direct impact on retention. According to Williams and Dreher (1992), pay plays an important role in attracting and retaining employees. Several authors have stated that workers who are not satisfied with their salaries will leave the organization (Irshad and Afridi 2007; Nawab and Bhatti 2011). According to Trevor et al. (1997) and Williams et al. (2008), wage increases enhance the ability of organizations to retain employees, though, in addition to a wage increase, the transparency of the wage increase process and the effectiveness of communication about the wage increase also affect employee retention. According to Salleh and Memon (2015), and Shtembari et al. (2022), employees who are satisfied with their salaries feel that they are treated fairly and receive adequate compensation for their performance. Arnold (2005) points out that it is the responsibility of managers to develop a fair and equitable wage system, which Heshizer (1994) considers to be critical for labour retention. However, according to Bryant and Allen (2013), wage dissatisfaction has a weak correlation with fluctuation intention. Pfeffer (1998) argues that pay only plays a role in attracting workers and is not in itself sufficient to retain them. Many organizations can be effective in retaining employees without high wages. Highhouse et al. (1999) and Hajli (2014) believe that in the absence of a high wage level, the right working environment and the support of employees and superiors can motivate employees to stay, so an emphasis should be placed on improving the quality of life at work in order to ensure the long-term commitment of employees.

2.1.5. Contingent Rewards

According to Agarwal (1998), performance pay is a consideration for work and performance, which, according to Hytter (2007), and Garg and Rastogi (2006), promotes employee retention. Arnold (2005) believes that reward plays an important role in labour retention since it gives employees the impression that they are valued in the organization. Employees who consider the extent of their reward to be unfair in terms of their effort and performance are likely to leave the organization. This conclusion is related, among other things, to Heshizer’s (1994) findings on wage fairness. However, there are also divergent opinions on the role of contingent rewards in labour retention, according to which reward alone is not an important retention factor (Hayes et al. 2006; Shields and Ward 2001; Shtembari et al. 2022).

2.1.6. Benefits

According to Arnold (2005), the efficiency of organizations’ retention of the workforce can be improved mainly by means of benefits determined on the basis of working time, and flexible benefit packages can also contribute to increasing employee satisfaction. According to Bryant and Allen (2013), the main contribution to managing fluctuation is the benefits that employees can only receive after a longer—though not too long—effort, which motivates them to maintain their employment. Examples of such benefits are share options and pension schemes. According to Bryant and Allen (2013), health-related employee benefits had a negative impact on the termination rate, while Sutton (1985) found that benefits related to different types of insurance and retirement reduced employee turnover. As with pay and contingent rewards, Bryant and Allen (2013) stressed the importance of fair and equitable benefit conditions to reduce turnover. Arnold (2005) pointed out that managers should devote sufficient time to regular communication on wages, salaries, and benefits.

2.1.7. Promotion

Career planning, career development, and the development of suitable career paths are in the interest of both employers and employees. Career development is a planned activity designed to strike a balance between organizational interests and individual career goals in order to reap mutual benefits. Some authors have investigated the relationship between career dissatisfaction and leaving a job. According to Curtis and Wright (2001), dissatisfaction with career opportunities is one of the main causes of fluctuation. Many authors have also identified positive consequences of satisfaction with opportunities for advancement. Pergamit and Veum (1999) found a close and positive relationship between promotion and job satisfaction, while others concluded that career satisfaction has a positive effect on the commitment to an organization (Cardy and Lengnick-Hall 2011; Hiltrop 1999).

According to Curtis and Wright (2001) and Mak and Sockel (2001), in order to keep employees effective, employers should adapt their career development measures to the needs of employees. Gaffney (2005) and Arnold (2005) draw attention to the fact that the development of individual career plans should be carried out in parallel with the business plan, which is the responsibility of managers. According to Prince (2005) and Sánchez-Hernández et al. (2019), employees should be informed in detail about the available career opportunities at the time of recruitment.

2.1.8. Operating Procedures

Organizational operating procedures are the rules that must be followed during a workflow. According to Downes et al. (2002), the introduction of operational procedures to coordinate work processes requires standardization of work activities and the establishment of clear and unambiguous rules that promote efficient operation. Strict operating procedures reduce employees’ creativity, ability to innovate, and effective use of individual skills. According to Valaei and Rezaei (2016), dissatisfaction with operational procedures reduces employee wellbeing and commitment, which requires a review of existing procedures. According to Shalley et al. (2000), instead of overly bureaucratic operating procedures, an innovative organizational culture should be sought, as employees show a higher degree of satisfaction and commitment in an innovative environment.

2.1.9. Communication

Cropanzano and Mitchell (2005) see organizational internal communication as a key process that has a positive impact on productivity and employees’ attitudes towards the organization. Organizational internal communication plays an important role in predicting workplace engagement and retaining employees (Gomes et al. 2023; Gyurián Nagy and Gyurián 2022). Singh (2019) emphasizes that the task of management is to achieve effective organizational communication, which, according to Sinha and Sinha (2012) and Stazyk et al. (2021), promotes the identification of employees with organizational goals and the culture of the organization, and contributes to the creation of an open, friendly, and trusting working environment. Arnold (2005) draws attention to the need for managers to develop an effective multidirectional communication system in order to inform employees about the most important organizational policies. In his view, particular attention should be paid to organizational communication on pay, job security, promotion, grievance handling, health and safety regulations, and idea suggestion schemes.

2.2. Dimensions of Employee Retention

2.2.1. Normative Commitment

Normative commitment is when an employee feels a moral obligation to be a member of an organization (McDonald and Makin 2000). Normative commitment can be formed on the one hand as a result of the individual socialization process, previous experiences, and cultural factors that determine the level of employee commitment. Employees with high normative commitment show a higher level of commitment to a particular organization. The normative commitment of employees can also be triggered by organizational efforts (such as training, promotion, recognition, etc.), which provoke a sense of reciprocity from employees. Normatively engaged workers, therefore, stay in the workplace because they feel obliged to do so for some reason, with a number of known positive consequences. From an organizational and employee perspective, positive consequences include a negative correlation between normative commitment and intention to quit, as well as a positive correlation between workplace presence, employee performance, health, and wellbeing. (Bayode and Duarte 2022; Meyer and Allen 1991).

2.2.2. Organizational Commitment

According to Mowday et al. (1979), organizational commitment is the emotional attachment of employees to the organization and the workplace. Organizational commitment shows how much effort an individual is willing to make for an organization, how much he or she accepts its values and goals, and how strongly he or she insists on maintaining organizational membership (Martins et al. 2023). Steers (1977), Mascarenhas et al. (2022), and Mottaz (1988) found that organizational commitment is related to job satisfaction, which is influenced by a number of work-related factors. Schaufeli and Greenglass (2001) believe that employee organizational commitment is an important competitive factor. Employers are interested in increasing the organizational commitment of their employees, since they are passionately committed to their workplaces, show a high level of initiative, are constantly looking for new challenges, and are committed to high-quality work. The high level of commitment of employees results in more efficient work and higher performance, which positively affects both customer satisfaction and the perception of the organization (Lulewicz-Sas et al. 2022).

2.2.3. Turnover Intention

The intention to quit is a voluntary decision of the employee to leave the employment relationship (Arendt and Grabowski 2022; Zeffane 1994). According to Price (2001), employees’ decisions about their intention to quit may be influenced by external and internal factors. External factors include but are not limited to, environmental factors such as better labour market opportunities or influence from human relationships (e.g., family, friends). Internal factors of intention to quit include factors in the workplace that determine employee satisfaction. According to Price (2001), employee satisfaction negatively affects the intention to quit and is positively correlated with organizational commitment. According to Tett and Meyer (1993), the intention to quit can be interpreted as a response of employees to unfavourable working conditions, in which economic factors, play an important role. In their view, the intention to withdraw goes hand in hand with the search for new alternatives and the ongoing assessment of options. Van Dierendonck et al. (2016) similarly concluded that dissatisfied employees will look for new opportunities and if they find a better alternative, they will leave their current job. Consistent with the ideas of (Dewi et al. 2023), the turnover process begins with employee dissatisfaction, the thought of quitting, job search, and assessment of prospects, and culminates in the decision to quit. The intention to quit is associated with a deterioration in work ethic, more frequent absenteeism, lower levels of productivity, and reduced customer satisfaction. Choi et al. (2011), Frye et al. (2020), and Osuji et al. (2014), find that coworker relationships, leadership, financial incentives, professional development opportunities, and recognition reduce employees’ intention to quit, while an unfavourable work environment, demoralizing workplace climate, and inadequate leadership increase employees’ intention to quit.

It is concluded that employee wellbeing and retention are key factors for sustainable, long-term employment. Employee wellbeing is a state in which employees feel satisfied with their jobs and the work they do there and feel that they are listened to and supported by their employer, and employee retention refers to the employer’s ability to retain valuable employees and avoid their leaving the company. The link between employee wellbeing and employee retention is such that employers who listen to their employees and support their health, happiness, and satisfaction significantly increase their chances of retaining employees. Engaged employees tend to be more loyal and productive and less likely to change jobs.

3. Materials and Methods

3.1. Sample and Data Collection

The research is based on an online survey of 406 employees. The data were collected in Hungary, with the participation of workers who are currently employed. Since the primary objective of the sampling was to reach an employee sample with a heterogeneous organizational background, the research focus was not narrowed down to, for example, sector or organizational size. In this context, the questionnaire was designed to be interpreted by respondents working in different jobs (physical, intellectual), in different sectors (services, industry, agriculture), in different domains (public, market, nonprofit) and different positions (subordinate, team leader, middle manager, senior manager). Two criteria were set for participation in the survey: participants must have at least one year’s work experience and be currently employed.

Sampling was carried out using the snowball method at a two-month interval between April and May 2022 in order to reach as many respondents as possible. The survey was published online. The questionnaire started with a cover letter describing the purpose of the study and the conditions for participation in the survey. Participants were informed that participation in the study is completely voluntary, that they can stop answering questions at any time without giving reasons, and that there is no financial remuneration for participating in the study. Participants were also informed about the anonymity of the survey and the strict and confidential treatment of data and information. At the same time as completing the survey, the respondents declared that they had reached the age of 18, that they were familiar with the conditions of participation in the survey and that they voluntarily agreed to participate.

The questionnaire was based on statements that had previously been used effectively in international research and published in scientific journals. The translation of the statements made in English in international research was carried out with the help of a professional translator. After the translation, a small sample of employees (n = 8) was included to test the clear wording of the questions and statements. The number of respondents required to pretest the questionnaire was based on the recommendation of Reynolds et al. (1993), who found that between 5 and 10 respondents were sufficient to test a questionnaire. Based on the results of the pretest, there was no need to modify the questionnaire. In addition to testing the questionnaire, in order to ensure the reliability of the results, the survey started with two screening questions, which were: do you have at least one year of work experience? Are you currently in active employment? Based on the screening questions and missing data, the results of 39 questionnaires had to be deleted, resulting in a total of 406 complete responses (91% response rate).

The questionnaire consisted of two parts, the first part related to the key dimensions of the study (

Table 1. Models on which the survey is based.

Dimensions	Factors	Number of Items	Sources
Employee wellbeing	Pay, Promotion, Fringe benefits, Contingent rewards, Supervision, Coworkers, Operating procedures, Nature of work, Communication	36	Spector (1985); Park and Martinez (2022)
Employee retention	Normative commitment	8	Allen and Meyer (1990); Kim et al. (2022)
	Organizational commitment	7	Kim et al. (2016)
	Turnover intention	7	Sjöberg and Sverke (2000); Newman et al. (2011); Wayne et al. (1997)

Source: created by the authors.

), and the second part assessed the background data of the respondents. Based on demographic data, it can be concluded that the majority of respondents (67.2%) are women. In terms of highest educational attainment, most respondents have tertiary education: 48.5% of respondents have a bachelor’s degree, 18.2% have a master’s degree, and 4.4% have a PhD. Among respondents with no more than secondary education, those with a high school diploma were in the majority (24.6% of the total sample). The share of respondents with vocational education or technical school and with up to eight years of primary education is negligible. Participants in the survey can be divided into four generations based on their date of birth. Members of Generation Y made up 43.6% of the respondents (born between 1980 and 1994), 23.6% were born in 1995 or later (Generation Z), and 23.6% are members of Generation X (born between 1965 and 1979). The Boomer generation of respondents born between 1946 and 1964 accounted for only 1.7% of the total sample.

The following statements can be drawn from the respondents’ current jobs. Nearly 60% of the respondents work in the private sector, 30% in the public sector, and 10% in the nonprofit sector. Most respondents work in the service sector (68.7%), 26.6% in industry, and only 4.4% in agriculture. The majority of respondents (51.7%) work in large companies, 33% of the respondents are currently employed by medium-sized companies, and 9.9% and 5.4% are employed by small or microenterprises. In terms of jobs, the preponderance of intellectual jobs (84.7%) is observed compared to physical jobs (15.3%). Regarding the current position of the respondents, the majority (73.6%) are in a subordinate position, 12.1% in middle management, 10.6% in a team leader position, and only 3.7% in senior management.

3.2. Measures

In addition to questions about demographics, respondents completed a structured questionnaire with 58 statements, where they could provide their answers on a Likert scale (1- Strongly disagree, 5- Strongly agree). All the statements in the questionnaire came from previously applied surveys (Allen and Meyer 1990; Kim et al. 2016; Newman et al. 2011; Sjöberg and Sverke 2000; Spector 1985; Wayne et al. 1997). The survey, which included 58 statements, consisted of two main headings (employee wellbeing, and employee retention), and additional subheadings. The factors and dimensions examined during the data collection are presented in Table 1.

The first part of the questionnaire, designed to measure employee wellbeing, was based on Spector’s (1985) Job Satisfaction Survey, which examined the topic through nine factors, with a total of 36 statements (four statements per factor, either positive or negative). Related factors are pay, promotion, fringe benefits, contingent rewards, supervision, coworkers, operating procedures, nature of work, and communication. Although the JSS was originally developed for application in human services in public and nonprofit organizations (Spector 1985), Spector (1997) finds that the JSS is also suitable for general use and not restricted to a specific organization (Park and Martinez 2022).

The second part of the questionnaire aims to examine the issue of employee retention. The related subchapters were normative commitment, organizational commitment, and intention to quit. The normative commitment questionnaire was based on Allen and Meyer’s (1990) survey, which included a total of 8 statements (also positive and negative), though four of the eight statements were deleted due to low factor loadings (the survey statements, their factor loadings and the deleted statements are presented in Appendix A. Organizational commitment was measured with seven statements based on Kim et al. (2016). Statements to measure intention to exit based on several surveys (Newman et al. 2011; Sjöberg and Sverke 2000; Wayne et al. 1997). Of the seven statements, six showed adequate factor loading.

3.3. PLS-SEM

The PLS-SEM (partial least squares-structural equation modelling) method is a widely used tool in management research (Hair et al. 2011). PLS-SEM is a modelling process designed to maximize the variance of latent dependent variables.

According to Hair et al. (2011), PLS-SEM consists of two components. The first is the inner model and the second is the outer model. Within the inner model, there are paths (relationships) between latent variables. We can separate the exogenous and endogenous parts. The exogenous part describes the relationships between latent variables between which there is no structural-path connection. The endogenous part describes structural relationships between latent variables that are affected by relationships between other variables. Within the outer model, the indirect relationships between the observed (indicator) and latent variables can be described. The method can handle both reflective and formative indicators at the same time. Reflective indicators can be interpreted as functions of latent variables. If there is a change in the latent variable, it is also reflected in the reflective indicators. Formative indicators cause latent variables. If the formative indicators change, it also attracts the change of the latent variable.

4. Conducting Research and Results

4.1. Measurement Model

Examining the measurement model shows that it includes internal consistency reliability and convergent and discriminant validity. Internal consistency reliability assesses the extent to which the items measure a specific latent construct. As recommended by (Hair et al. 2017), we investigated the internal consistency reliability by ensuring that Cronbach’s α and the composite reliability (CR) are higher than 0.70 and below 0.95. The results provided in Table 1 indicate that Cronbach’s α and CR values for all the constructs were above the cut-off value of 0.7—thereby specifying the high internal consistency of the measures.

Convergent validity is the second measure to assess the measurement model, which assesses the extent to which a measure correlates positively with alternative measures of the same constructs (Hair et al. 2017). Convergent validity was examined by generating 5000 bootstrapping samples in PLS. The assessment of convergent validity requires checking the outer loading values of the items and the average variance extracted (AVE). As recommended by Hair et al. (2017), indicators with weaker outer loadings can be retained if other indicators with high loadings explain at least 50 percent of the variance. In total, seven items were deleted, two items from the coworkers construct (i.e., Cow2 and Cow4), and four items from the normative commitment construct (i.e., Nc1, Nc2, Nc3 and Nc7). The deleted items are shown in Appendix A. As detailed in

Table 2. Internal Consistency Reliability and Convergent Validity.

Constructs	Measurement Item	Loadings	AVE	α Value	CR
BENEFITS (Ben)	Ben1	0.796	0.619	0.790	0.865
	Ben2	0.820
	Ben3	0.890
	Ben4	0.616
PROMOTION (Pro)	Pro1	0.752	0.653	0.822	0.882
	Pro2	0.859
	Pro3	0.736
	Pro4	0.876
SUPERVISION (Sup)	Sup1	0.815	0.656	0.825	0.884
	Sup2	0.757
	Sup3	0.780
	Sup4	0.883
PAY (Pay)	Pay1	0.868	0.722	0.872	0.912
	Pay2	0.803
	Pay3	0.854
	Pay4	0.874
CONTINGENT REWARDS (Cr)	Cr1	0.797	0.701	0.857	0.903
	Cr2	0.840
	Cr3	0.837
	Cr4	0.874
COMMUNICATION (Com)	Com1	0.771	0.650	0.820	0.881
	Com2	0.784
	Com3	0.839
	Com4	0.829
NATURE OF WORK (Now)	Now1	0.661	0.691	0.846	0.898
	Now2	0.864
	Now3	0.892
	Now4	0.887
COWORKERS (Cow)	Cow1	0.963	0.931	0.926	0.964
	Cow3	0.967
NORMATIVE COMMITMENT (Nc)	Nc4	0.782	0.563	0.739	0.836
	Nc5	0.773
	Nc6	0.813
	Nc8	0.617
ORGANIZATIONAL COMMITMENT (Oc)	Oc1	0.830	0.696	0.927	0.941
	Oc2	0.894
	Oc3	0.797
	Oc4	0.763
	Oc5	0.869
	Oc6	0.789
	Oc7	0.888
TURNOVER INTENTION (Ti)	Ti1	0.802	0.650	0.922	0.939
	Ti2	0.866
	Ti3	0.908
	Ti4	0.854
	Ti5	0.798
	Ti6	0.858

Source: PLS-SEM generated results.

, the AVEs of the latent variables were between 0.536 and 0.931, all greater than the 0.5 standard value, indicating that the reflective measurement variables had favourable convergent validity. The operating-procedures construct was deleted due to the low Cronbach’s α value.

Discriminant validity refers to the extent that the constructs used in the model are distinct from one another (Hair et al. 2017). Two methods were applied to assess discriminant validity. The first method was the Fornell and Larcker (1981) criterion, which was used to compare the correlation between the constructs and the square root of AVE for that construct. For discriminant validity, the square root of the AVE for each latent variable must exceed the correlation value of the same construct (Fornell and Larcker 1981). As we can see in

Table 3. Discriminant Validity (Fornell–Larcker Criterion).

Constructs	Ben	Pro	Sup	Pay	Cr	Ti	Com	Now	Cow	Nc	Oc
Ben	0.787
Pro	0.415	0.808
Sup	0.279	0.400	0.810
Pay	0.622	0.532	0.377	0.850
Cr	0.592	0.557	0.503	0.719	0.837
Ti	−0.338	−0.457	−0.440	−0.456	−0.555	0.849
Com	0.352	0.373	0.470	0.399	0.540	−0.536	0.806
Now	0.203	0.325	0.350	0.311	0.343	−0.566	0.439	0.831
Cow	0.292	0.357	0.446	0.337	0.469	−0.447	0.483	0.449	0.965
Nc	0.169	0.251	0.202	0.241	0.290	−0.487	0.239	0.378	0.223	0.750
Oc	0.407	0.454	0.408	0.474	0.541	−0.738	0.553	0.687	0.543	0.556	0.834

Source: PLS-SEM generated results.

, the square-root value of AVE for a specific latent variable is higher than the correlation values provided in the rows and columns (Fornell and Larcker 1981)—thus confirming an adequate discriminant validity.

The second method was the Heterotait–Monotrait Ratio (HTMT), which was used to confirm discriminant validity. Much research relies only on the Fornell–Larcker criterion and cross loadings when investigating discriminant validity (e.g., Hair et al. 2012), however, Henseler et al. (2015) have shown that these criteria perform poorly in terms of disclosing discriminant validity problems. Instead, researchers should use the HTMT criterion. High HTMT values indicate a problem with discriminant validity. Based on simulation and previous research, Henseler et al. (2015) recommend that HTMT values should not exceed 0.90 if the path model includes constructs that are conceptually similar. When the constructs are conceptually more distinct, a more conservative threshold value of 0.85 is recommended. Based on

Table 4. Discriminant Validity (HTMT criterion).

Constructs	Ben	Pro	Sup	Pay	Cr	Ti	Com	Now	Cow	Nc	Oc
Ben
Pro	0.508
Sup	0.337	0.474
Pay	0.741	0.620	0.443
Cr	0.712	0.654	0.595	0.826
Ti	0.383	0.504	0.499	0.503	0.619
Com	0.427	0.441	0.561	0.469	0.638	0.615
Now	0.242	0.382	0.414	0.364	0.404	0.638	0.534
Cow	0.336	0.402	0.501	0.375	0.527	0.480	0.543	0.506
Nc	0.210	0.301	0.237	0.287	0.355	0.570	0.296	0.462	0.259
Oc	0.463	0.507	0.455	0.524	0.605	0.783	0.625	0.768	0.588	0.659

Source: PLS-SEM generated results.

, it is evident from the HTMT value that the present study confirmed all the assumptions of discriminant validity.

Multicollinearity

Before assessing the structural model, in addition to validity and reliability, multicollinearity must be checked. Multicollinearity can be assessed through the variance inflation factor (VIF). According to Burns (2008), a VIF value greater than 10.0 indicates the issue of multicollinearity. Hair et al. (2014) recommend a cut-off value of 5.0 for multicollinearity, though there are researchers who suggest even lower values such as 3.33 (Diamantopoulos and Siguaw 2006). The VIF values for all items are presented in

Table 5. Inner VIF values.

Constructs	Ben	Pro	Sup	Pay	Cr	Ti	Com	Now	Cow	Op	Nc	Oc
Ben				1.565	1.288		1.252				1.290
Pro	1.146			1.474	1.405		1.388	1.219			1.386	1.146
Sup		1.248			1.473	1.477	1.371		1.000
Pay
Cr				1.880		1.774				1.133
Ti
Com					1.544	1.750		1.388			1.533
Now						1.949				1.133	1.505
Cow	1.146	1.248			1.460		1.333	1.370			1.490	1.146
Op						1.462					1.198
Nc						3.019
Oc				1.565	1.288		1.252				1.290

Source: PLS-SEM generated results.

. In our case, all the VIFs of the indicators were below 3.33, indicating no issue of multicollinearity between the latent constructs.

4.2. Structural Model

This study assessed the structural model using the method of 5000 bootstraps in Smart-PLS software.

Table 6. Bootstrapping Report.

Relationship between Latent Factors	Original Sample	Sample Mean	Standard Deviation	T Statistics	p Values
Benefits → Pay	0.281	0.281	0.041	6.942	0.000
Benefits → Contingent rewards	0.346	0.351	0.041	8.464	0.000
Benefits → Communication	0.153	0.156	0.050	3.074	0.002
Benefits → Organizational commitment	0.145	0.144	0.035	4.140	0.000
Promotion → Benefits	0.356	0.357	0.050	7.074	0.000
Promotion → Pay	0.157	0.153	0.039	4.005	0.000
Promotion → Contingent rewards	0.232	0.228	0.045	5.187	0.000
Promotion → Communication	0.104	0.104	0.048	2.138	0.033
Promotion → Nature of work	0.131	0.130	0.047	2.783	0.006
Promotion → Organizational commitment	0.082	0.083	0.037	2.208	0.028
Supervision → Promotion	0.301	0.297	0.050	6.020	0.000
Supervision → Contingent rewards	0.170	0.169	0.038	4.445	0.000
Supervision → Turnover intention	−0.077	−0.076	0.039	1.981	0.048
Supervision → Communication	0.257	0.257	0.052	4.998	0.000
Supervision → Coworkers	0.446	0.447	0.044	10.179	0.000
Contingent rewards → Pay	0.465	0.466	0.044	10.660	0.000
Contingent rewards → Turnover intention	−0.159	−0.159	0.039	4.090	0.000
Contingent rewards → Normative commitment	0.182	0.175	0.050	3.629	0.000
Communication → Contingent rewards	0.196	0.199	0.039	5.034	0.000
Communication → Turnover intention	−0.107	−0.109	0.046	2.320	0.021
Communication → Nature of work	0.256	0.257	0.057	4.481	0.000
Communication → Organizational commitment	0.158	0.156	0.039	4.063	0.000
Nature of work → Turnover intention	−0.100	−0.096	0.050	2.027	0.043
Nature of work → Normative commitment	0.316	0.321	0.044	7.198	0.000
Nature of work → Organizational commitment	0.378	0.381	0.041	9.239	0.000
Coworkers → Benefits	0.165	0.163	0.046	3.583	0.000
Coworkers → Promotion	0.222	0.226	0.047	4.688	0.000
Coworkers → Contingent rewards	0.114	0.111	0.040	2.836	0.005
Coworkers → Communication	0.287	0.287	0.046	6.224	0.000
Coworkers → Nature of work	0.279	0.277	0.060	4.652	0.000
Coworkers → Organizational commitment	0.160	0.159	0.039	4.128	0.000
Normative commitment → Turnover intention	−0.127	−0.130	0.040	3.166	0.002
Normative commitment → Organizational commitment	0.294	0.294	0.035	8.488	0.000
Organizational commitment → Turnover intention	−0.421	−0.422	0.057	7.410	0.000
Benefits → Pay	0.281	0.281	0.041	6.942	0.000
Benefits → Contingent rewards	0.346	0.351	0.041	8.464	0.000

Source: PLS-SEM generated results.

represents the results of the bootstrapping procedure.

Table 7. Saturated Model Results.

Construct	R2	Adj. R2	Q2	SRMR (sat. Model)	SRMR (est. Model)
Benefits	0.196	0.192	0.117	0.062	0.063
Promotion	0.200	0.196	0.127
Pay	0.592	0.589	0.423
Contingent rewards	0.579	0.574	0.398
Turnover intention	0.609	0.603	0.425
Communication	0.352	0.346	0.217
Nature of work	0.280	0.275	0.190
Coworkers	0.199	0.197	0.181
Normative commitment	0.172	0.168	0.093
Organizational commitment	0.694	0.689	0.478

Source: PLS-SEM generated results.

represents the saturated model results. Standardized root means square (SRMR) values were used to examine the model fitness. As per Henseler et al. (2016), the SRMR should be <0.08. This study exhibits an adequate level of model fitness since the SRMR for the saturated model was 0.062 and for the estimated model was 0.063. In the next step, we evaluated the R² and Q² values, according to Chin (1998), the acceptable values of R² must be >0.1 or zero, and the acceptable values of Q² must be >zero. Based on the results, the values of R² and Q² for this study show that both are >0.1. According to Chin (1998) recommended R² values for endogenous latent variables based on: R² < 0—very weak; 0.19 ≤ R² < 0.33—weak; 0.33 ≤ R² < 0.67—moderate; R² ≥ 0.67—substantial. Based on this, our constructs can be classified into the following categories: very weak (normative commitment); weak (benefits, coworkers, promotion, nature of work); moderate (communication, contingent rewards, pay, turnover intention) and substantial (organizational commitment) correlation.

Figure 1 represents the employee wellbeing factors, whose affecting employee retention, while Figure 2 represents the relationship between employee wellbeing factors and employee retention.

5. Discussion

The aim of the study is to create a new model using the PLS-SEM method, with which the retention of employees can be explained by objective factors. The literature review revealed that although many authors had previously examined employee retention, these studies did not use the PLS-SEM method. Previous publications contained a small number of variables due to methodological constraints. This research gap has led to the definition of a new model from the different variables used in previous research, comprising of 12 different variables.

Based on the path modelling, the correlations do not reach the 0.5 level in any of the cases. The strongest correlation (r = 0.465) can be seen between contingent rewards and pay, which leads to the conclusion that the more satisfied the employee is with the performance-based rewards, the more satisfied they will be with their salary. Based on the correlation between supervision and coworkers (r = 0.446), satisfaction with leadership is partly associated with satisfaction with coworkers. The correlation between the two variables does not reach the level of 0.5, which leads to the conclusion that employees can separate their satisfaction with their managers from their satisfaction with their colleagues. It implies that if perceptions of leadership improve, this will not automatically result in a change in satisfaction with other colleagues. Based on the negative correlation between turnover intention and organizational commitment (r= −0.421), if organizational commitment increases, the probability of termination decreases. However, the correlation is of moderate strength, so an increase in organizational commitment can only partially reduce the probability of quitting.

Both the discriminatory and convergent validity of the model is appropriate. The theoretical model fits the empirical data well, as the SRMR value is below the criterion level. Organizational commitment has the largest explained variance of latent variables (R² = 0.694). The variables examined account for nearly 70% of the variance in organizational commitment. The explained variance of turnover intention is also high (R² = 0.609), i.e., 60% of the variables in the model explain employees’ intention to leave their job. The explained variance of satisfaction with pay (R² = 0.592) is also high, which means that our model can explain almost 60% of the variance of this variable. We can also explain almost 60% of the variance of the variable contingent rewards (R² = 0.579), while the model can explain only 35% of the variance of the variable communication satisfaction (R² = 0.352). Within the model, there is a low explained variance for the following variables: normative commitment (R² = 0.172) and benefits (R² = 0.196). This raises the question of whether the normative commitment and benefits variables should be kept in the model since both latent variables can be interpreted as subvariables of other latent variables. Normative commitment can be interpreted as a subscale of organisational commitment and benefits as a subscale of contingent rewards. Therefore, in the future, we should empirically examine whether these two variables can really be interpreted as independent variables or whether they are only subdimensions. If this assumption were confirmed, it would also be necessary to examine the effect of the omission of these two variables on the model.

In our model, a total of 34 significant relationships between latent variables can be detected using the bootstrap procedure. Organizational commitment has an impact on Turnover intention. Organizational commitment is influenced by six other variables: benefits (r = 0.145), promotion (r = 0.082), communication (r = 0.158), nature of work (r = 0.378), coworkers (r = 0.160), normative commitment (r = 0.294). It follows that if the organizational commitment within an organization changes, the intention to quit will change. In the case of changes in the perception of benefits, promotion, communication, nature of work, coworkers, and normative commitment, there will be an impact on organizational commitment. In other words, these variables are critical for organizational commitment, especially when employees perceive them negatively, resulting in a drastic decrease in commitment. According to research, the level of pay is not related to organizational commitment. The results refute the popular assumption that pay increases commitment to the organization and support the argument of Pfeffer (1998) that pay only has a role in attracting employees and is not sufficient in itself to retain them. At the same time, we contradict the studies of Trevor et al. (1997) and Williams et al. (2008), who argue that increasing wages increases the ability of organizations to retain them.

Turnover intention is affected by the following variables: organizational commitment (r = −0.421), normative commitment (r = −0.127), nature of work (r = −0.100), communication (r = −0.107), contingent rewards (r = −0.159), and supervision (r = −0.077). As a result, the intensity of the intention to quit decreases when organizational and normative commitment reaches a higher level, or satisfaction increases for the nature of work, Communication, contingent rewards, and supervision variables. This result supports Price’s (2001) suggestion that employee satisfaction negatively affects the intention to quit. However, it is worth noting that in our case, the intention to quit has a negative relationship not only with organizational commitment but also with normative commitment. Normative commitment is when an employee feels a moral obligation to be a member of an organization (McDonald and Makin 2000). In other words, with the decrease in normative commitment, the moral commitment between the employee and the organization is eliminated, which increases the turnover intention. The results reflect those changes in communication, job nature, and perceptions of normative commitment variables all affect employees’ organizational commitment and intention to quit, though the direction of the effect is univariate. In addition, the results also support that workers with higher normative commitment have lower levels of intention to leave and higher levels of organizational commitment.

Pay is associated with only three variables: benefits, promotion, and contingent reward. It follows that satisfaction with pay can increase if satisfaction with the three variables mentioned above improves. We were unable to find a link between payment and commitment and payment and intention to quit. This refutes the results of Bryant and Allen (2013), who, however, argue that wage dissatisfaction has a weak correlation with the intention to quit, since in our case, during the bootstrap procedure, there is not even a weak link between pay and intention to quit. However, this result supports the conclusion of Pfeffer (1998), who argues that pay only plays a role in attracting workers and is not sufficient, in itself, to retain them.

The results of the research will provide new theoretical and practical conclusions. The main theoretical contribution is that a complex model has been developed that can explain retention through the factors of employee wellbeing. This is a new academic achievement since previous research has not taken a holistic approach to the study of this topic. Previously, the direction and magnitude of the effects of the 11 factors examined in our model on labour-force retention have not been identified. Our results demonstrate that pay has no significant effect on loyalty, though benefits increase employee engagement and rewards reduce the intention to quit. Therefore, one of the main theoretical contributions of this paper to the retention issue is that it highlights the motivational role of performance-related financial rewards as a primary determinant of pay. Nevertheless, in the context of retention, we emphasise that instead of developing compensation systems, it is recommended to focus on the development of working conditions that allow for a higher level of individual performance based on appropriate employee relations and challenging and responsible tasks.

The most important practical relevance of the research (for both managers and HR professionals) is that it identifies the factors that can be developed to increase the effectiveness of organisations’ retention capacity. Moreover, the research is not limited to identifying key factors, though the results also allow for the wellbeing factors that influence employee retention to be ranked in terms of their impact on employees’ organisational commitment and intention to leave. The results suggest that improving the nature of work is the most effective way to retain employees. Employers are therefore advised to pay particular attention to providing meaningful, interesting, and challenging tasks for employees, as this can increase not only their engagement but also their performance. In addition to the nature of the job, satisfaction with contingent rewards, manager, and promotion opportunities, the effectiveness of organisational communication and the quality of employee relations and benefits also have a significant impact on employee retention. Employers are therefore advised to pay attention to these key factors when developing their retention strategy. Finally, employers should place greater emphasis on measuring the normative commitment of prospective employees during the selection process, as this may have a significant impact on employees’ organisational commitment and intention to leave in the future.

6. Conclusions

The aim of the study was to create a model to identify the key determinants of employee wellbeing and labour retention. As a result of the research work, the PLS-SEM methodology was used to create a theoretical framework to model the effects of individual employer measures on labour retention. Based on the research results, similarities can be identified between the factors shaping the intention to quit and the organizational commitment, though they differ in their orientation.

If the employee’s normative commitment is at a higher level or is satisfied with the benefits, promotion, communication, and nature of work factors, then the organizational commitment is strengthened, otherwise, the dominance of the intention to quit prevails. It can be seen that cash benefits do not reinforce the intention to quit or the organizational commitment. The role of financial incentives in labour retention is not as decisive as in attracting labour. However, research on pay systems has highlighted the employee paradigm that satisfaction is triggered by pay systems in which individual performance plays a dominant role in wage differentiation.

Finally, it can be concluded that broadly defined workplace wellbeing factors play a decisive role in labour-force retention. If the workplace is organised in such a way that its processes are transparent and understandable through effective internal communication, and the employee has a clear vision of their career (promotions, nature of work, etc.), then pay systems in the retention strategy can be seen as support functions rather than as a pillar of retention.

7. Limitations and Future Research Suggestions

The study sought to answer the question of what wellbeing factors could be the basis of a retention strategy, which has been answered in detail in the previous chapters. In order to answer the research question, a holistic model was developed based on literature sources and tested on a sample of a heterogeneous group of employees in Hungary. This approach has helped to validate the measurement scales, though there is no empirical evidence yet on the use of the scales with homogeneous groups of workers. Thus, in the future, a research topic can be identified in which the research is carried out among different groups of employees after the optimization of the measurement scales. The increasing inflationary effects in the international economic environment justify a multidimensional analysis of the relationship between wage and labour retention.