Antecedents of turnover intention among Gen z in Vietnam: The mediating role of affective commitment

Abstract Turnover intention (TI) in many countries has been widely discussed, but research on the turnover intention by Gen Z in Vietnam is still in its infancy. This study examines the role of affective commitment (AC) as a mediator in the relationship between job satisfaction (JOS), emotional exhaustion (EE), and TI. A survey dataset with 602 observations was primarily collected from major cities in Vietnam, and partial least squares structural equation modeling (PLS-SEM) was applied. The results indicate that AC is closely related to EE. Moreover, EE has a significant direct impact on TI. However, unlike previous studies, no direct relationship was found between JOS and TI for Gen Z in Vietnam. Given the increasing number of new small and medium-sized businesses in Vietnam, this study contributes significantly to helping organizations plan more rational human resource policies.


Introduction
Human resources have always been a core aspect of any organization.Sustaining the workforce is a key competitive factor for businesses, enabling them to operate more efficiently than their counterparts (Macke & Genari, 2019;Nguyen, 2020;Vu & Tran, 2022).However, in a rapidly developing business society where job opportunities are abundant, accompanied by the younger generation's open-mindedness, turnover intention (TI) among employees is becoming a more common issue in almost all enterprises.Because the turnover intention is the antecedent of actual turnover, it is necessary to understand the reason why employees have such behavior and conduct preventive methods (Dubey, 2017).High TI rates would lead to a high level of employee turnover, which not only has a detrimental effect on the financial performance of a business, it also undermines the overall productivity of other employees, diminishes the quality of services offered by the organization, increases the recruitment, training and developing cost of new employees and adversely influences the company's reputation (Deniz, 2020;Memon et al., 2017).In contrast, a low employee TI rate helps organizations maintain their competitive edge and attain sustainable long-term goals.Consequently, TI among employees is an urgent problem that needs to be solved and also a research topic that attracts many scholars' interests.
Worldwide, the labor market is witnessing the entry of Gen Z employees born between 1997 and 2013 (Dimock, 2019).Thus, employers need to prepare for the arrival of Gen Z employees in their organizations (Schroth, 2019).The generations prior to Gen Z have been the focus of scientific study over the past few decades, but as they have aged and become more mature in the workforce (Dimock, 2019), potential gaps for future research have emerged.
Gen Z is a dynamic, experiential and exploratory generation, so they are unafraid to jump jobs to experience various different positions, environments and cultures.According to DePass and Tribune (2023), in the U.S., about 60% of 18-to 25-year-olds said they were likely to change jobs in early 2023, up from 53% last year.Similarly, the labor market in Vietnam is also witnessing the constant turnover of Gen Z.According to Anphabe (a recruitment and consulting company), research on the largest professional social networking platform LinkedIn with 922.3 million users worldwide as of April 2023, up to 260,000 of the nearly 4 million members in Vietnam have updated their status to "Open to Work" or "Looking for a new job" on their personal profiles, indicating they want to find a new job while being employed.This shows that TI among young people in Vietnam is extremely high.
This study examines the relationship between factors believed to profoundly influence the TI of Gen Z, namely emotional exhaustion and job satisfaction.Previous studies worldwide have verified these relationships (Öztürk Çiftci, 2021).However, this study contributes by investigating these relationships among the Gen Z population in Vietnam.This study also examines the mediating role of affective commitment in the relationship of the above factors to TI. Affective commitment, one of the three components of organizational commitment, is among the factors that receive the most attention from scholars because it closely affects organizational outcomes such as turnover and employee attitudes.The employees who don't feel satisfied with the support from their companies have a low degree of commitment and want to look for new job opportunities (Julio et al., 2022).
The authors have identified the aforementioned research gap and applied PLS-SEM in this study to explain the factors that influence the TI of the young Vietnamese workforce.This study comprises five parts.Part two provides the theoretical background of emotional exhaustion, job satisfaction, affective commitment, TI, the research model, and the hypotheses.Part three offers the methodology, including the study design and data analysis techniques.Part four explains the research results.Part five provides the discussion, managerial implications, and study limitations.

Generation Z
Gen Z (1997-2013) was born into a society with modern technological development, affecting this generation as they are avid social media users.They have a higher level of education (Fry & Parker, 2018), are creative, and their families and society provide numerous opportunities for them to express and develop themselves (Rue, 2018).Therefore, they have the space to freely develop and comprehend their self-worth, and they will choose a working environment that helps them to maximize their potential and fulfill themselves.At the same time, Gen Z likes experiences, so they enjoy attempting new things and engaging in diverse activities, indicating that long-term commitment to an organization is not a priority for these individuals (Chillakuri, 2020).Additionally, Gen Z is highly independent in all aspects, particularly financially.They may switch jobs if organizations offer higher salaries and greater flexibility at work (Chillakuri, 2020).Tett and Meyer (1993) explained this concept as an employee's self-reported likelihood of seeking alternatives in another organization.Similarly, TI refers to a person's inclination and eagerness to leave their current organization, the potential of seeking opportunities elsewhere, the aspiration to secure a new job, and the availability of options to work in alternative companies (Chen et al., 2021).In this study, TI is understood as an employee's voluntary desire to leave their current job and seek alternative job opportunities.

Turnover intention
TI is a strong predictor of employees' actual decision to switch jobs and higher turnover intention leads to high actual turnover rate (Sun & Wang, 2016).According to Bryant and Allen (2013), turnover is costly for organizations due to the process of finding, recruiting and training new employees.These expenses indeed surpass the total compensation amount for the position (Bryant & Allen, 2013).Therefore, research on employees' intention to change jobs will assist managers and businesses in building a sound human resource management system, anticipating the costs that may be incurred when an employee leaves the organization and planning to prevent or reduce their employees' intention to quit.

Emotional exhaustion & turnover intention
Among the three dimensions of the Maslach Burnout Inventory (MBI) (Maslach & Jackson, 1981), emotional exhaustion receives the greatest attention and examination from scholars.Emotional exhaustion is a destructive process that begins with the employee's emotional resources being depleted and ends in physical and mental harm (Ducharme et al., 2007).Gupta and Srivastava (2021) concur that emotional exhaustion in the workplace involves feelings of burnout, job-related negative affect, and emotional, mental, and psychological distress.The overlapping aspect of these definitions is that employees' emotional resources become exhausted; thus, they feel psychologically incapable of contributing to their work.In this study, emotional exhaustion is chosen as an independent variable because it can potentially signal many consequential issues and is considered the core of burnout syndrome.
TIs are one of the most apparent consequences of emotional exhaustion.Baba et al. (2009) say that individuals with high levels of emotional exhaustion tend to have a decline in work performance, making it harder for them to meet job demands with the resources they have.Avoidance is another method of coping with this condition to avoid further harm (Cropanzano et al., 2003).From this perspective, feeling emotionally exhausted opens the door to negative thoughts about one's work, resulting in the intention to quit (Salehi et al., 2020).Compatible with this viewpoint, the Conservation of Resources Theory (Hobfoll, 1989) posits that employees who feel that they lack the necessary resources to do their job or are not receiving adequate support or rewards may become emotionally drained and exhausted, triggering their TIs.Aligned with the COR theory, emotionally exhausted employees tend to prevent additional resource depletion, and if their efforts fail, they turn to coping mechanisms such as withdrawal to alleviate the emotional toll of exhaustion (Zhang et al., 2019).More specifically, Chen and Huang (2016) suggest that experiencing a reduction in resources or the possibility of resource loss is a source of stress, which can lead to dissatisfaction with one's job, increased anxiety, and thoughts of quitting one's job.Increasing empirical evidence suggests that emotional exhaustion is strongly linked to important workrelated attitudes among employees, such as TIs (Nguyen et al., 2022).The study conducted with 141 employees of different organizations and occupations in Canada concludes that emotional exhaustion positively impacts TI (Kyei-Poku, 2019).Investigating the correlation among them, Azharudeen and Arulrajah (2018) concur that a significantly strong positive relationship exists between emotional exhaustion and TI.Moreover, when an individual is exhausted emotionally, they develop negative attitudes and feelings towards their job and organization and find less selfefficacy in performing duties (Azharudeen & Arulrajah, 2018).Results from surveying the behavior of employees working at shopping centers in Turkey also indicate that there is a positive relationship between emotional exhaustion and TI (Öztürk Çiftci, 2021).In 2023, a study conducted with 243 individuals working in Malaysian SMEs in the information technology sector displays a similar indication that there is a significant positive impact of early-career employees' emotional exhaustion on their turnover intention (Saleh et al., 2023).This relationship becomes even more accurate when applied to the Gen Z demographic.It is studied that Gen Z are prone to mental health problems, which often results in decreased performance at work when they experience feelings of depression and anxiety (Schroth, 2019).They prioritize happiness over their career and do not place long-term loyalty to organizations as a primary concern, thus, Gen Z tend to have turnover intentions if they are unhappy or have negative feelings towards the job (Gaan & Shin, 2022;Randstad, 2022).
Aligned with the extant literature on the relations between the variables and the abovementioned explanations, this study formulated the following hypothesis: Hypothesis 1: Emotional exhaustion positively impacts employees' TIs.

Job satisfaction and turnover intention
Job satisfaction plays a critical role in industrial and organizational psychology.A person with a high level of job satisfaction has positive feelings about their job, while a person with a low level of job satisfaction (dissatisfaction) has negative feelings (Robbins & Judge, 2017).In this study, job satisfaction is understood as an employee's affective response towards their job, including both positive and negative aspects.
Age exhibits a correlative nexus with job satisfaction.Certain studies elucidate a linear association between the two, positing that job satisfaction augments as one advances in years.At times, this correlation assumes a curvilinear trajectory, with satisfaction ebbing in youth, plateauing during middle age, and resurging in later stages.
A recurrent discourse revolves around whether the predilections and career aspirations of Generation Z job seekers diverge across disparate national and cultural milieus.Ngoc et al. (2022) conducted a comprehensive inquiry amongst Vietnamese final-year students poised to enter the workforce.Their findings underscored the heightened emphasis Vietnamese Generation Z places on intangible facets of employment or organizational culture, such as the ambiance of the workplace and ethical considerations, over tangible attributes.
According to a recent report by McCrindle Research (2019), Generation Z job aspirants are drawn towards organizations predicated upon a synthesis of instrumental and symbolic attributes.This proclivity may be ascribed to Generation Z's nascent presence in the labor market, thereby rendering extant research on this cohort inchoate.According to the instrumental-symbolic framework, Generation Z job seekers concurrently esteem the concrete benefits proffered by organizations.Recent empirical investigations corroborate this orientation, indicating that Generation Z values a salubrious work environment, well-defined objectives, clear directives from superiors, insitu training, meaningful occupational engagements, performance evaluations, interpersonal connectivity, a work-life equilibrium, and avenues for continual learning and development.Predicated on these discernible traits of Generation Z, in concert with judicious human resources policies, it is predicted that Generation Z will exhibit a palpable degree of satisfaction with their extant vocations.
It is evident that in all labor market conditions, different characteristics of each country can result in varying impacts on job contentment and intent to switch jobs.Based on the above discussion, we predict that job satisfaction has a positive effect on reducing the turnover rate among workers.Hypothesis 2: Job satisfaction negatively impacts employees' TI.

The mediating role of affective commitment
Affective commitment is defined as "identification with, involvement in, and emotional attachment to the organization" (Meyer & Allen, 1997).Specifically, employees with high levels of affective commitment are likely to remain with the organization because they have the desire to continue contributing to it.In different studies, affective organizational commitment represents an employee's dedication to their organization and the degree of emotional connection they feel toward it (Dhar, 2015;Ismail, 2016).Scholars have suggested that affective commitment may better predict employee behavior than other aspects of organizational commitment like continuance commitment or normative commitment (Dhar, 2015).
Researchers have thoroughly investigated the relationship between affective commitment and TI.According to Social Exchange Theory (Cropanzano et al., 2017), social behaviors pertain to exchanging anything deemed valuable, whether tangible or intangible, such as feelings of honor or prestige.Thus, positive work experiences and a sense of value can lead to higher affective commitment among employees, decrease their TIs, and enhance overall job performance.Additionally, this correlation is evidenced by numerous empirical studies such as Albrecht and Marty's (2017) study of Australian organizations; Yousaf et al. (2015) study of educators in the Netherlands.These studies support the idea that affective commitment and TI have a relationship that goes in the opposite direction.Wong's and Wong (2017) case study of a foreign-invested manufacturing company in Guangdong, China, also demonstrated that employees' affective commitment significantly and negatively affects their intention to leave.Evidence from many different fields and locations worldwide has shown that affective commitment and TI are closely related.
Burnout as a psychological concept negatively impacts employees' emotional connection to their workplace (Sharma & Dhar, 2016).Workers may experience emotional exhaustion because of the work atmosphere and job requirements, leading to an inequity in social exchange relationships, ultimately resulting in reduced levels of emotional commitment (Chênevert et al., 2013).Previous empirical research has shown that emotional exhaustion strongly impacts affective commitment among employees, which means that mentally exhausted individuals will demonstrate weaker affective attachment to their organization (Cole et al., 2010;Thanacoody et al., 2013).Additionally, Kemp et al. (2013) study on the impact of stress and mental exhaustion on the attachment of professional truck drivers to their organization also concluded that emotional exhaustion negatively affects affective commitment.
However, this relationship also has different results depending on the research subject.Karatepe and Tekinkus's (2006) empirical research with data collected from front-line employees who work directly with customers at Turkish banks does not support a significant inverse relationship between emotional exhaustion and affective commitment.Instead, it only demonstrates that emotional exhaustion negatively impacts affective commitment through job satisfaction.
Despite limited empirical research on the mediating effects of affective commitment between emotional exhaustion and TIs, based on the Conservation of Resources Theory and Social Exchange Theory, it is logical to suggest that emotional exhaustion may result in employees' reduced affective commitment of towards the organization and an increase in their intentions to leave.In addition, with the research population of Gen Z in Vietnam in the economic sector, a generation prone to mental and work-related worries, it is proposed that emotionally drained employees will have a weaker affective connection to the organization, thus triggering their TIs.With the above hypothesis that affective commitment has an inverse relationship with TI, the following hypothesis was formed: Hypothesis 3: Affective commitment mediates the relationship between emotional exhaustion and employees' TIs.
Scholars have proposed that organizational commitment can be adopted as a moderator to interpret employee behaviors (Ghaderi et al., 2021;Nauman et al., 2020;Safdar & Liu, 2019).The three-component model proposed by Meyer and Allen (1991) indicates organizational commitment fully mediated the relationship between job satisfaction and TI, suggesting that individuals who were highly satisfied with their jobs were more likely to be committed to their organizations, which in turn reduced the likelihood of their intention to leave.This is similar to the results conducted by Kartika and Purba (2018).Contrary to the three-component model, Clugston (2000), in his study on government agency employees, found that a partially mediated model better fits the data for this study.Considerable work has established the mediating effects of organizational commitment on job satisfaction and intent to leave.Meanwhile, although affective commitment is affirmed to be the best predictor of employee behavior compared to other aspects of organizational commitment (Meyer & Allen, 1997), the research investigating the mediating role of affective commitment for the relationship between job contentment and TI is limited both in Vietnam and other countries.Among them, one prominent study conducted by Gessesse and Premanandam (2023) revealed that affective commitment completely mediated the association between job satisfaction and TI.
In a meta-analysis, Meyer et al. (2002) found a strong relationship between overall job satisfaction and affective commitment.However, most of the studies emphasized on the affective commitment-to-job satisfaction relationship (Dinc et al., 2018;Koo et al., 2019) rather than the opposite relationship of job satisfaction-to-affective commitment.Only a few have underlined the latter.Lima and Allida (2021) discovered that job satisfaction positively impacts employee's affective commitment.Similar result was also confirmed by Gessesse and Premanandam (2023) with data of 319 academic employees in higher education institutions in Addis Ababa.Thus, employees with high job satisfaction tend to have a greater emotional attachment to their organization.
Nowadays, Gen Z is said to be less satisfied with their work compared to other generations and has a lower commitment to their organizations (McKinsey, 2023).Similarly, Gen Z in Vietnam might have a lower emotional attachment to organizations.However, we predict this phenomenon can be reduced through their emotional attachment to the organizations.Meanwhile, from the aforementioned positions, we believe that affective commitment has an inverse relationship with TI.Therefore, the following hypothesis is proposed: Hypothesis 4: Affective commitment mediates the relationship between job satisfaction and employees' TI.

Methodology
This questionnaire was conducted across major cities in Vietnam, with respondents from small and medium-sized enterprises (SMEs) in various working fields, including retail, marketing, banking and finance, human resources, manufacturing, logistics, and others.According to the Ministry of Planning and Investment, as cited by the Vietnam Chamber of Commerce and Industry ( 2022), Vietnam has approximately 900,000 active enterprises, with over 97% being SMEs.This factor, combined with the selection of employees from diverse working backgrounds and geographical areas, is a potential condition for generalizing the results to the population.
Among the 602 successful respondents, 59.0% were female, 40.2% were male, and 0.8% were others.Most of the respondents at the time of the survey were between 21 and 25 years old (78.2%) and from 18 to 20 years old (21.8%) and were either in junior or intern positions.Responses from people not in the Gen Z age range were omitted to ensure the study was valid.People born between 2005 and 2012 were also excluded because they had not yet entered the job market.The average time that the interviewees had been associated with their jobs was from six to 12 months (45.2%).Table 1 shows the descriptive statistics of the participants' demographics.

Procedures
A questionnaire was developed based on the theoretical background.Four experts were consulted to ensure that the scales were applicable in Vietnam.These experts comprised two researchers from the National Economics University (NEU), one human resources executive responsible for internal communication and one chief human resources officer (CHRO) from a career orientation company.After interviews with the experts, the survey questions were adapted for the Vietnamese context.
The survey includes 38 questions (see Appendix) (except questions about the participants' demographic information) which were translated into Vietnamese.The translation process was supported and reviewed by English teaching experts and students majoring in English to ensure the translations' accuracy and comprehensibility.After several pre-interviews with 15 university students having working experiences from National Economics University (NEU), the authors agreed to modify some statements that were in negative forms into affirmative statements to create consistency among indicators.For example, the statement "I feel insecure about my salary" in the job satisfaction factor was modified to "I feel secure about my salary."This modification also appears in affective commitment item."" The data was collected between October and December 2022 through Google Forms.A total of 637 responses from employees were received.However, 35 invalid responses were removed (including biased answers and answers from those who were not our survey target), leaving 602 valid responses.The survey statements were constructed based on a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree).

Measures
The research model includes four variables, of which three are independent: emotional exhaustion (EE) and satisfaction (JOS); one dependent variable is TI; and one mediating variable is affective commitment (AC), as shown in Figure 1 below.Most of the items built were based on scales and models from previous studies.More specifically, almost all indicators of the TI factor were formed based on research by Ganesan and Weitz (1996), Kelloway et al. (1999), and Jung and Yoon (2013).Based on the theoretical background presented in Section 1, only one item was self-developed by the authors, that salary is a potential antecedent of TI (Chillakuri & Mahanandia, 2018).In comparison, items of AC were derived from research by Allen and Meyer (1990), and those of EE stemmed from Maslach and Jackson (1981).Studies by Koustelios and Bagiatis (1997), Schmidt (2004), and Peltier et al. (2006) were used to construct JOS indicators.

Data analysis techniques
Data collected and processed were analyzed using the PLS-SEM approach by the SmartPLS 3.0 software.Researchers have typically used structural equation modeling (SEM) to estimate causeeffect models with latent variables (Sarstedt et al., 2020), and it has gained significant popularity in business research and scientific fields (Sarstedt et al., 2019).There are two approaches to SEM, and one of them is partial least squares (PLS-SEM).Specifically, PLS-SEM is a prominent method for specifying models with complex interrelationships (Hwang et al., 2019;Khan et al., 2019).
According to Hsu et al. (2006); Henseler et al. (2009); Hair et al. (2011);and Hair et al. (2014), PLS-SEM is suitable for exploratory research, can process non-metric measurement, and can handle formative measurement models.Notably, this is the preferred and superior approach when estimating mediation (Sarstedt et al., 2020).In this study, we specified the relationship between latent variables POS, EE, JOS and the observed variable TI based on the formative model and developed and expanded the theory-the mediating role of AC towards TI.Hence, it is appropriate to use PLS-SEM for this research.
There are three main steps in PLS-SEM: measurement model evaluation, structure model evaluation, and impact of mediator variables evaluation.First, the measurement model was assessed by analyzing outer loading, reliability, convergent validity, and discriminant validity.Then, multicollinearity among constructs was checked with the variance inflation factor (VIF) to ensure insignificant construct interdependency.The bootstrapping method examined the t-statistics for the path coefficients.The R2 index, f2 index, and Q2 index were obtained to evaluate the model.Lastly, the role of the mediator variable in the model was tested by the bootstrapping method.The threshold allowed for each evaluation is shown in Figure 2.
The discriminant validity shows the difference between a construct and other constructs in the model.Henseler et al. (2015) used simulation studies to demonstrate that discriminant validity is assessed by the HTMT index that they developed.With the HTMT index, Garson (2016) suggested that the discriminant validity between two constructs is ensured when the HTMT index is less than  1.Meanwhile, Henseler et al. (2015) proposed that discriminant validity is ensured if this value is below 0.9.In this study, 0.9 was the maximum threshold for the HTMT index.Table 3 depicts the results of measuring the discriminant validity of the variables, and all met the allowed threshold.

Structural proposed research model evaluation
In the structural model evaluation process, the first step is to check for multicollinearity through the Collinearity Statistics (VIF) values of all independent variables in the structural model.Inner VIF values are utilized to evaluate the multicollinearity phenomenon between independent variables.Hair et al. (2019) proposed the following VIF evaluation thresholds: VIF > 5: high risk of multicollinearity; 3 ≤ VIF ≤ 5: multicollinearity is possible; VIF < 3: multicollinearity is unlikely.Table 4 shows that all VIF values were below the threshold of 3. Therefore, multicollinearity between the predictor variables in the model is not an issue, and we could proceed to the next evaluation steps.
Confirmatory Factor Analysis (CFA) is a statistical test specifically designed to evaluate the quality of observed variables and explore hypothetical constructs.In CFA, the observed variables included in the analysis are assumed to have already been identified as belonging to specific factors.The function of CFA at this point is to assess whether the observed variables within the same factor adequately fit the data and meet the established standards.CFA results of this research has been displayed in Figure 3.
The R 2 value was utilized in the PLS-SEM technique to evaluate the model's ability to explain variance.The objective is to obtain a high R 2 coefficient that can account for the endogenous latent variance.The R 2 values vary between 0 and 1, with higher values indicating a stronger explanatory power of the model (Cohen, 1988;Sawilowsky, 2009).Table 5 shows the R 2 value for AC was 0.441 and for TI 0.593.
The bootstrapping method is utilized to examine the t-statistics for the path coefficients to evaluate the hypothesized connections' importance (Efron, 1992;Yung & Bentler, 1994).P-values are characterized as follows: highly significant if p ≤ 0.01, marginally significant if p ≤ 0.05, and not statistically significant at conventional levels if p > 0.10 (Rice, 1989).
Table 6 summarizes path coefficients, including the original sample, T-statistics, p-values, and confidence intervals.This indicates a statistically significant relationship between EE and TI (p < 0.01).The correlation between JOS and TI, meanwhile, is not statistically significant.As a result, EE has a significant impact on TI.Therefore, hypothesis 1 was accepted, and hypothesis 2 was rejected.The f 2 coefficient indicates the strength of the independent variable's influence on the dependent variable.Cohen (1988) proposed a table of f 2 values to evaluate the importance of independent variables as follows.If f 2 <0.02, then the impact is extremely small or non-existent.If 0.02 ≤ f 2 <0.15, the impact is small.0.15 ≤ f 2 <0.35 means moderate impact.The impact is large if f 2 ≥0.35.Figures in Table 7 show that AC significantly affects TI; the other relationships have a small impact, except for the influence of JOS on TI.
The structural model can also be evaluated based on its ability to make predictions, measured using Stone-Geisser's Q 2 (Geisser, 1974;Stone, 1974).This measure requires the model to predict the indicators of each endogenous latent construct accurately.Hair et al. (2019) proposed that when 0 < Q 2 <0.25, it indicates a weak ability to predict data outside of the sample, whereas values in the range of 0.25 to 0.5 indicate a moderate level of predictability, and Q 2 >0.5 suggest a good ability to predict such data.Table 8 demonstrates that both AC and TI have moderate predictive strength, indicating the model's quality reached an acceptable threshold.

Impact of mediator variables evaluation
The core characteristic of a mediating effect is that it involves a third variable that plays an intermediate role in the relationship between the independent and dependent variables.The initial approach involved analyzing specific indirect effects to analyze each indirect association separately.As shown in Table 9, all specific indirect effects have p-values less than 0.05, thus indicating that AC plays a mediating role in the relationships between EE and TI and JOS and TI.The original sample of the indirect effect EE→AC→TI is 0.225, meaning that AC is a stronger mediator in this relationship.Consequently, hypotheses 3 and 4 are supported.

Discussion
To address the limitations of previous studies (Ahmad, 2018;Astiti & Surya, 2020;Güzeller & Celiker, 2019;Rawashdeh & Tamimi, 2019;Yasin et al., 2023), which have only investigated the mediating role of organizational commitment, our research provides evidence for the cognitive and emotional factors influencing an individual's intention to leave a job through affective commitment as a mediator.Mercurio (2015) emphasizes that AC is the core of organizational commitment.
Specifically, the findings show that affective commitment is a critical mediator in the relationship between other variables, including job satisfaction, emotional exhaustion, and employees' intentions to quit their jobs.This is a new and unexpected point in our study.Our results demonstrate that job satisfaction has a positive effect on affective commitment.This means that high levels of job satisfaction will increase Generation Z employees' affective commitment, making them less likely to quit.Consequently, satisfying job requirements can increase affective commitment.
Furthermore, emotional exhaustion negatively influences affective commitment.When employees feel emotionally drained, their emotional attachment to the organization will likely decrease, resulting in a greater intention to leave.This study's findings have contributed to the understanding that emotional exhaustion and affective commitment have mediated the effect of organizational trust on the intention to leave.These findings are primarily generalizable across all disciplines, emphasizing the significance of increasing employee affective commitment.
The results reveal that Gen Z employees constantly harbor intentions to seek new employment opportunities in the near future.Regarding direct impact relationships, we found that emotional exhaustion positively correlates with TI.Our findings suggest that the more tired and exhausted employees feel, the more they want to switch to another job.For Gen Z, this is especially important because they are more prone to stress problems than other generations (Rue, 2018).Simultaneously, they like comfortable workspaces, so they are more likely to jump jobs when having emotional problems with work.However, our study found that job satisfaction did not correlate with TI.While we found that job satisfaction has a strong and negative correlation with the intention to switch jobs, we pinpointed that job satisfaction has no impact on employees' TIs.This means that although their job satisfaction remains stable, they may still subconsciously want to leave.We could reason that Gen Z likes to experience and try new things, which shows that long-term commitment to an organization is not a priority for these individuals (Chillakuri, 2020).

Theoretical implication
The empirical findings of our study give us a clear understanding of the direct relationship between emotional exhaustion and job satisfaction on Gen Z's TI and how affective commitment mediates these relationships.It also complements existing research on TI and its predictors.
Our research pinpoints that emotional exhaustion influences TI.This result aligns with the Conservation of Resources Theory (Hobfoll, 1989).However, unlike previous studies, we found that two organizational factors, specifically job satisfaction, do not directly affect Gen Z's TI.This result is inconsistent with Mobley's (1977) turnover theory and other studies (Ramalho Luz et al., 2018;Vu & Nguyen, 2018).Our discoveries have unveiled fresh dimensions regarding the influence of job satisfaction and emotional exhaustion on TI.Furthermore, compared to previous studies, our study found that affective commitment mediates these relationships.Hence, affective commitment is essential for employees and organizations.This finding fits perfectly with our argument based on the Conservation of Resources Theory (Hobfoll, 1989) and the Social Exchange Theory (Hooman, 1958).Nguyen and Le (2022) argued that normative commitment mediates the effect of job satisfaction and emotional exhaustion on TI within the scope of research on working employees in the information technology field.These employees in technology companies, primarily in Ho Chi Minh City, Vietnam, are aged 22 or older and have previously worked for at least one IT company.Combining Nguyen and Le's research with our own study has helped provide additional empirical evidence in Vietnam for the theory of organizational commitment (Meyer & Allen, 1997) and the Three-Component Model of Organizational Commitment (Meyer & Allen, 1991) in two primary aspects of organizational commitment.Organizations should proactively take measures for emotional exhaustion and job satisfaction to encourage emotional bonding with employees and reduce their TI.

Practical implications
From a human resources perspective, one of the greatest expenses for an organization is employee attrition.Information from this study can potentially assist organizations and human resources departments in minimizing Gen Z's TI.Human departments and organizations can utilize the findings flexibly from this study and combine them with relevant literature regarding antecedents affecting TI to develop new organizational policies that will provide a comfortable workplace for employees and increase their affective commitment.This study provides organizations with insight into some factors influencing Gen Z's TI.Additionally, the study also brings some macro-effects, as follows: Efficiency in science and technology: updating employees' job expectations in the context of economic integration; application in developing a roadmap and software to assist employees in easily performing tasks and helping businesses monitor employees' emotional commitment, thereby predicting their intention to change jobs and taking timely measures.
Economic efficiency can be used to develop programs to increase employee engagement with the enterprise, thereby contributing to lower TIs, stabilizing the enterprise's human resources, improving high working efficiency, and avoiding the costs of new recruitment and training for HR positions that are vacant due to employee turnover.
Social efficiency enhances emotional commitment, helping employees stay with the enterprise long-term, stabilizing the labor market, and minimizing temporary unemployment.

Limitations and future research direction
Although this research makes theoretical and practical contributions, it still has several limitations that must be addressed in the future.First, the study was conducted using online survey sampling; hence many individuals inevitably took the survey randomly.Future studies could employ stricter methods, such as direct sampling or interviews, to minimize the number of random polls, which could skew results.
Second, our study collected data at one point in time, making it difficult to explore the evolution of TI, affective commitment, emotional exhaustion, and job satisfaction in the context of a market and company policies that change daily.Therefore, future studies could use a longitudinal design to investigate the causal relationships and observe if organizations alter their policies towards their employees and whether employees change their minds about switching jobs.
Finally, the data were collected from the Vietnamese Gen Z population, which might limit our study's general applicability.Given the considerable differences in the social and economic systems across countries, what works in one nation may not work similarly in another.Also, Gen Z employees in Vietnam might have different attitudes about their careers than those in other countries.Researchers in other countries could use our research model to further improve the external reliability of the findings.

Conclusion
The problem of turnover appears increasingly in today's organizations, which negatively impacts organizations.The findings of this study provide important managerial implications for organizations.Therefore, this study forwards the following suggestions for management practices.
First, the findings present empirical evidence of a strong link between emotional exhaustion and TI.Therefore, the organization needs to take timely actions to support its employees physically and mentally to decrease their work pressure.For example, an organization can make a scientific design of their working hours and allow employees to choose their own working environment, such as their home, while still ensuring efficiency and progress.
Second, since this research shows that affective commitment mediates emotional exhaustion, job satisfaction and TI, organizations need to monitor employees' affective commitment levels as it frequently affects TI.When employees are unsatisfied with their jobs and constantly feel tired while working, it reduces their loyalty to organizations, which means their affective commitment is low, and the intention to switch jobs will arise.Hence, to decrease the risk of actual employee turnover, organizations need to strengthen the emotional bond between them and the company.
Finally, our study particularly emphasizes the importance of affective commitment to TI, so organizations need to pay more attention to building employees' emotional attachment to the organization from the beginning when new employees join the organization.Therefore, when recruiting a new employee, the organization should search for the best fit regarding goals and values.Finding the right employee for the organization and the type of employee that fits the organization's environment is most strongly correlated with affective commitment (KRISTOF-BROWN et al., 2005).

Figure
Figure 2. PLS-SEM technique used in data analysis.