Introduction
Over the last decades, human capital has emerged as the most important asset for organizations. After all, human capital and its retention are related to organizational performance, job satisfaction, and other positive work-related outcomes (Arokiasamy, Fu, Kumar Piaralal & Sarah Omar, Reference Arokiasamy, Fu, Kumar Piaralal and Sarah Omar2023). Additionally, high turnover rates may translate into important costs for companies, both financial and social (Rumawas, Reference Rumawas2021). For instance, when people leave an organization, not only does the organization lose the time and resources invested in their training and development, but also the understanding and knowledge the workers possess of technical and organizational issues. Replacing this knowledge can be a challenging and expensive process for companies (Charles-Leija, Castro, Toledo & Ballesteros-Valdés, Reference Charles-Leija, Castro, Toledo and Ballesteros-Valdés2023). Consequently, talent retention becomes crucial for organizations. Recently, companies have faced talent shortages due to globalization, technological advancement, and the dynamic and uncertain environment in which organizations operate (e.g., the Great Resignation,new generations in the workplace) (Serenko, Reference Serenko2023; Wan & Duffy, Reference Wan and Duffy2022; Winterton & Turner, Reference Winterton and Turner2019). Therefore, understanding what motivates people to remain in their jobs and companies is essential for organizations.
Turnover intention represents the psychological state an employee experiences before the actual turnover behavior. This mental state often culminates in the abandonment of the company (Bhatnagar, Reference Bhatnagar2012). Literature supports that employees who perceive their work as meaningful are less likely to experience a desire to leave their work and organization (Vermooten, Boonzaier & Kidd, Reference Vermooten, Boonzaier and Kidd2019). This is because meaningful work possesses an intrinsic motivational force that prevents employees from searching for alternative employment opportunities (Lysova, Allan, Dik, Duffy & Steger, Reference Lysova, Allan, Dik, Duffy and Steger2019; Vermooten et al., Reference Vermooten, Boonzaier and Kidd2019). Although this relationship has been studied widely in literature, understanding the underlying psychological mechanisms is essential for a comprehensive understanding of its dynamics. According to Shuck (Reference Shuck2019), employees who think about their work as meaningful are more likely to deeply involve themselves in their work and feel more engaged in it. Consequently, they are less likely to abandon their organization.
Literature has traditionally studied employee engagement from the perspective of work engagement (Farndale, Beijer, Van Veldhoven, Kelliher & Hope-Hailey, Reference Farndale, Beijer, Van Veldhoven, Kelliher and Hope-Hailey2014). However, Saks (Reference Saks2006) introduced organizational engagement as another component of employee engagement. Therefore, following Saks (Reference Saks2006), employee engagement would be formed of work engagement and organizational engagement. Academics have mostly focused on work engagement, while practitioners and consultors have directed their attention towards organizational engagement. This difference in focus highlights the importance of considering both types of engagement for a comprehensive and accurate understanding of employee engagement and its influence on organizational results. In this study, we propose employee engagement (both work and organizational engagement) as the mediating variable explaining the relationship between meaningful work and turnover intention. Therefore, taking into account the current importance for organizations to understand the dynamics and mechanisms that motivate their employees to remain with them, the objective of this study is to examine all the relationships between meaningful work, work and organizational engagement, and turnover intention through a serial mediation model.
This study pretends to fill three literature gaps: (a) there is currently no study that examines the relationship between meaningful work, work engagement, organizational engagement, and turnover intention. While Sanchez-Cardona et al. (Reference Sánchez-Cardona, Vera and Marrero-Centeno2023) tested a mediation model wherein meaningful work and work engagement mediate the relationship between job resources and turnover intention, organizational engagement was not included in this model. (b) While there are several papers defending the mediating role of work engagement between antecedents and work-related outcomes (Alhajaj & Ahmad, Reference Alhajaj and Ahmad2024; Kissi, Ikuabe, Aigbavboa, Smith & Babon-Ayeng, Reference Kissi, Ikuabe, Aigbavboa, Smith and Babon-Ayeng2024, Ramaprasad, Lakshminarayanan & Pai, Reference Ramaprasad, Lakshminarayanan and Pai2021; Yucel, Şirin & Baş, Reference Yucel, Şirin and Baş2023) and few studies have explored organizational engagement as a mediator in the relationship between personal and organizational antecedents and outcomes (Juhdi, Pa’wan & Hansaram, Reference Juhdi, Pa’wan and Hansaram2013; Saks, Reference Saks2006; Shah & Beh, Reference Shah and Beh2016; Torabi, Ghochani, Nadali & Fathi, Reference Torabi, Ghochani, Nadali and Fathi2019) there is still a gap in literature. Concretely, no study considers both types of engagement as serial sequential mediators between work antecedents and outcomes. (c) Although previous research has investigated the influence of meaningful work on work engagement (Canboy, Tillou, Barzantny, Güçlü & Benichoux, Reference Canboy, Tillou, Barzantny, Güçlü and Benichoux2023; Charles-Leija et al., Reference Charles-Leija, Castro, Toledo and Ballesteros-Valdés2023), no studies have explored its impact on organizational engagement. Saks, Gruman and Zhang (Reference Saks, Gruman and Zhang2022) made an approximation of this issue, but empirical evidence remains lacking. Therefore, the main contribution of this study is to expand knowledge and literature about organizational engagement and how it relates to other work-related variables.
The paper is structured as follows. First, we provide a revision of the existing literature and present the structural framework of the study, as well as develop the study’s hypotheses. Next, the methodology used and the results obtained are presented. Finally, results are discussed, as well as implications, limitations and future research directions. Our findings suggest that meaningful work exerts a negative influence on turnover intention, and work and organizational engagement mediate this relationship. From a practical perspective, this study can provide business practitioners with guidelines on what can make employees willing to stay in their current organization while they contribute to its success. By promoting meaningful work and reinforcing engagement, organizations can develop a work environment that not only reduces turnover intentions but also contributes to creating a healthy, happy, and sustainable workforce. Additionally, this study opens new research avenues by proposing the exploration of the reciprocal relationship between work and organizational engagement, as well as the inclusion of the potential moderating role of sector differences in shaping these effects.
Theoretical background and hypotheses development
Meaningful work
According to Steger (Reference Steger, Yeoman, Bailey, Madden and Thompson2019, p.218), meaningful work is defined as ‘work that people gladly, gratefully, and energetically give their best self and effort to’. It refers to the significance or value of work (Blustein, Lysova & Duffy, Reference Blustein, Lysova and Duffy2023; Martela et al., Reference Martela, Gómez, Unanue, Araya, Bravo and Espejo2021). When people experience meaning at work, they feel their work has purpose and meaning, allowing them to create a more significant benefit (Soren & Ryff, Reference Soren and Ryff2023; Steger, Reference Steger, Oades, Steger, Fave and Passmore2016). Meaningful work goes beyond the meaning that people attribute to work. It is also a source of personal growth and contributes to one’s ultimate life purpose, while serving a good that goes beyond oneself (Steger, Dik & Duffy, Reference Steger, Dik and Duffy2012). Meaningful work is beneficial both for workers and organizations. When people feel their work is meaningful, they show better psychological adjustment and qualities that are attractive and essential for organizations (Steger et al., Reference Steger, Dik and Duffy2012). For example, they manifest greater well-being, attribute a higher value to their work, and show higher job satisfaction (Allan, Batz-Barbarich, Sterling & Tay, Reference Allan, Batz-Barbarich, Sterling and Tay2019; Martela et al., Reference Martela, Gómez, Unanue, Araya, Bravo and Espejo2021). In addition, people who perceive their work as meaningful become happier, develop more positive relationships, and experience better health and more satisfaction with their overall lives (Allan et al., Reference Allan, Batz-Barbarich, Sterling and Tay2019). Meaningful work allows individuals to flourish both personally and professionally by experiencing feelings of competence and purpose in life (Kim & Beehr, Reference Kim and Beehr2020). On the other hand, high rates of meaningful work are associated with higher levels of organizational commitment and lower levels of turnover intentions (Blustein et al., Reference Blustein, Lysova and Duffy2023; Lee & Han, Reference Lee and Han2024). In other words, meaningful work is effective in creating and developing healthy and positive workplaces (Nasir & Waheed, Reference Nasir and Waheed2022).
Meaningful work has been considered by different theories in literature, like the Job Characteristics Theory (Hackman & Oldham, Reference Hackman and Oldham1976), which considers meaningful work as a positive reaction; and the JD-R model (Bakker & Demerouti, Reference Bakker and Demerouti2014), which considers meaningful work as a personal resource. Despite each theory being different and conceptualizing meaningful work differently, the effect is the same: meaningful work constitutes a relevant component for developing well-being experiences and positive workplaces.
In sum, meaningful work could be the key to developing healthy job experiences (i.e., employee engagement) and reducing negative organizational outcomes (i.e. turnover intention).
Employee engagement: work and organizational
Employee engagement literature is quite extensive, with several conclusions that have been drawn. However, with regard to the operationalization of employee engagement, two principal frameworks can be found in literature. On the one hand, Schaufeli et al. (Schaufeli, Martínez, Pinto, Salanova & Bakker, Reference Schaufeli, Martínez, Pinto, Salanova and Bakker2002, p.74) define work engagement as ‘a positive, fulfilling, work-related state of mind characterized by vigor, dedication, and absorption’. Most of the research on employee engagement has been performed through this framework, therefore only considering work engagement. However, on the other hand, Saks (Reference Saks2006) considers that employee engagement is specific to the role an employee is performing and defends that employees have at least two roles: their work role and their role as a member of an organization, leading to the distinction of work engagement and organizational engagement (Saks et al., Reference Saks, Gruman and Zhang2022). Therefore – according to Saks (Reference Saks2006) – employee engagement is composed of work engagement and organizational engagement. Despite limited research on organizational engagement, it can be defined as the extent to which employees are actively involved in the achievement of organizational goals, tasks, events, and projects assigned to them as members of an organization (Saks et al., Reference Saks, Gruman and Zhang2022). Alternatively, Farndale et al. (Reference Farndale, Beijer, Van Veldhoven, Kelliher and Hope-Hailey2014) define organizational engagement as a state in which employees are very positive about their organization, acting as ambassadors for the corporate brand.
Following Saks’ (Saks, Reference Saks2006) theory, the conclusion is that, despite being related, the two types of engagement (work and organization) are distinct constructs. Originally, Saks (Reference Saks2006) defended that there was a significant difference between the two concepts by examining their antecedents and consequences, and revealing different strengths of the relationship with other constructs (e.g., organizational support or intention to quit). Subsequent studies have supported this distinction between work engagement and organizational engagement (e.g., Saks et al., Reference Saks, Gruman and Zhang2022; Šakytė-Statnickė, Bilan & Savanevičienė, Reference Šakytė-Statnickė, Bilan and Savanevičienė2023). Suhartanto and Brien (Reference Suhartanto and Brien2018) argue that it is possible for employees to be engaged towards their organization but not towards their work, and vice versa. In light of this, it becomes crucial to adopt a multidimensional approach to employee engagement. Not doing so faces the risk of limiting our understanding of antecedents and outcomes of engagement (Farndale et al., Reference Farndale, Beijer, Van Veldhoven, Kelliher and Hope-Hailey2014). Work engagement is one lens through which employees interpret their workplace experiences. However, it is important to consider alternative engagement perspectives, like organizational engagement (Farndale et al., Reference Farndale, Beijer, Van Veldhoven, Kelliher and Hope-Hailey2014). Therefore, for this study, we will use Saks’ (Saks, Reference Saks2006) framework of employee engagement.
Given the explanation of both types of engagement, one may ask oneself: what is the relationship between work and organizational engagement? Research has demonstrated that work engagement predicts organizational engagement (Galingan, Reference Galingan2018; Rai & Chawla, Reference Rai and Chawla2022; Šakytė-Statnickė et al., Reference Šakytė-Statnickė, Bilan and Savanevičienė2023; Suhartanto & Brien, Reference Suhartanto and Brien2018). Suhartanto and Brien (Reference Suhartanto and Brien2018) explain this relationship through the spillover theory. This theory suggests that an individual’s experience in one area of his/her life can influence other areas (Sirgy, Efraty, Siegel & Lee, Reference Sirgy, Efraty, Siegel and Lee2001). To illustrate this theory, consider the following example: when people feel happy and satisfied at work, they are likely to experience satisfaction and happiness with their family. In the organizational context, the spillover theory implies that an employees’ interaction with one organization element will affect their experience and perception of other elements. For this study, employees’ engagement towards their work may extend to their engagement with the organization.
Meaningful work, work engagement, and organizational engagement
Meaningful work has been studied alongside a wide range of psychological constructs. One of these constructs is employee engagement, since both are highly intercorrelated (Allan et al., Reference Allan, Batz-Barbarich, Sterling and Tay2019). There is a substantial body of literature that sustains the positive relationship between meaningful work and work engagement (Canboy et al., Reference Canboy, Tillou, Barzantny, Güçlü and Benichoux2023; Faisaluddin, Fitriana, Nugraha & Hinduan, Reference Faisaluddin, Fitriana, Nugraha and Hinduan2024; Sánchez-Cardona, Vera & Marrero-Centeno, Reference Sánchez-Cardona, Vera and Marrero-Centeno2023; Vermooten et al., Reference Vermooten, Boonzaier and Kidd2019). When people find meaning in their work, they dedicate themselves to their tasks with more vigor, dedication and absorption. Therefore, they become fully engaged with the activity they are performing (Charles-Leija et al., Reference Charles-Leija, Castro, Toledo and Ballesteros-Valdés2023).
On the other hand, the relationship between meaningful work and organizational engagement has not been thoroughly explored so far. However, following the distinction Pratt and Ashforth (Reference Pratt, Ashforth, Cameron, Dutton and Quinn2003) made between ‘meaningfulness in work’ and ‘meaningfulness at work’, Saks et al. (Reference Saks, Gruman and Zhang2022) proposed ‘meaningfulness in work’ to be related to work engagement, whereas ‘meaningfulness at work’ may be an organizational engagement predictor. ‘Meaningfulness in work’ refers to the intrinsic motivation for the job. In contrast, ‘meaningfulness at work’ alludes to the organization’s goals, values and beliefs, and the people you work with (Pratt & Ashforth, Reference Pratt, Ashforth, Cameron, Dutton and Quinn2003, p. 314). This distinction matches the one between work and organizational engagement. Even more, an intervention field study conducted by Fletcher and Schofield (Reference Fletcher and Schofield2021) demonstrated that improvements in meaningfulness resulted in higher levels of organizational engagement.
At this point, it becomes evident the need to explore the relationship between these three variables.
Having the following assumptions in mind: (a) a study made by Rai and Maheshwari (Reference Rai and Maheshwari2021) demonstrated that work engagement mediates the relationship between job characteristics and organizational engagement; (b) work engagement usually acts as a mediator between work-related variables (Fahmi & Mohamed, Reference Fahmi and Mohamed2020; Gadi & Kee, Reference Gadi and Kee2020; Kissi et al., Reference Kissi, Ikuabe, Aigbavboa, Smith and Babon-Ayeng2024; Yucel et al., Reference Yucel, Şirin and Baş2023); and (c) work engagement serves as an antecedent of organizational engagement (Galingan, Reference Galingan2018; Rai & Chawla, Reference Rai and Chawla2022); the following hypothesis is proposed:
H1: Work engagement partially mediates the relationship between meaningful work and organizational engagement
Meaningful work, work engagement, organizational engagement, and turnover intention
Turnover intention constitutes an individual’s disposition to abandon their current job and search for better employment opportunities (Triningsih & Darma, Reference Triningsih and Darma2023). This represents a big problem for organizations, as it leads to financial and psychological costs and productivity losses (Charles-Leija et al., Reference Charles-Leija, Castro, Toledo and Ballesteros-Valdés2023). Employee turnover intention can be attributed to various factors. These include work that does not contribute to the employee’s life purpose, boring work, and work that does not allow employees to develop their strengths or represent challenges for them (Charles-Leija et al., Reference Charles-Leija, Castro, Toledo and Ballesteros-Valdés2023). Therefore, meaningful work could have the opposite effect, reducing employees’ turnover intention by providing purpose, challenges, and opportunities for personal growth (Blustein et al., Reference Blustein, Lysova and Duffy2023; Heath, Williams & Luse, Reference Heath, Williams and Luse2024; Sun, Lee & Sohn, Reference Sun, Lee and Sohn2019; Vermooten et al., Reference Vermooten, Boonzaier and Kidd2019).
In addition, the positive state of mind individuals experience through engagement may serve as an antidote to turnover intention. In other words, it can decrease individuals’ motivation to search for new employment opportunities (Vermooten et al., Reference Vermooten, Boonzaier and Kidd2019). In other words, highly engaged employees result in lower turnover rates (Alias, Noor & Hassan, Reference Alias, Noor and Hassan2014; Bowen BowenXue et al., Reference Bowen, Feng, Zhao, Li, Yang, Zhang and Luo2024; Saks, Reference Saks2006; Sharma, Sharma & Nanda Agarwal, Reference Sharma, Sharma and Nanda Agarwal2022; Sheehan, Tham, Holland & Cooper, Reference Sheehan, Tham, Holland and Cooper2019). On the other hand, research on the relationship between turnover intention and organizational engagement is still limited, but various studies support the negative relationship between them (Akingbola & van den Berg, Reference Akingbola and van den Berg2019; Juhdi et al., Reference Juhdi, Pa’wan and Hansaram2013; Malinen & Harju, Reference Malinen and Harju2017; Shah & Beh, Reference Shah and Beh2016; Torabi et al., Reference Torabi, Ghochani, Nadali and Fathi2019). In addition, the systematic review, made by Sinisterra, Peñalver and Salanova (Reference Sinisterra, Peñalver and Salanova2024), provides substantial evidence for these relationships, demonstrating that both work engagement and organizational engagement mitigate employee turnover intentions.
The relationship between work engagement, organizational engagement, and turnover intention can be explained by the assumption that organizational engagement mediates the relationship between the other two variables. For instance, when employees feel engaged towards their jobs, they may also feel involved and engaged towards their organization, as explained by the spillover theory (Suhartanto & Brien, Reference Suhartanto and Brien2018). Consequently, their inclination to abandon the company will decrease. Moreover, organizational engagement has also been shown to act as a mediating variable between work-related antecedents and employees’ turnover intention (Juhdi et al., Reference Juhdi, Pa’wan and Hansaram2013; Kundu & Lata, Reference Kundu and Lata2017; Nguyen & Ha, Reference Nguyen and Ha2023; Shah & Beh, Reference Shah and Beh2016; Torabi et al., Reference Torabi, Ghochani, Nadali and Fathi2019). Therefore, the following hypothesis is proposed:
H2: Organizational engagement partially mediates the relationship between work engagement and turnover intention.
Serial mediation model
At this point, the last step is to explain the relationship between the four study variables. To do so, we chose the Social Exchange Theory (SET; Cropanzano & Mitchell, Reference Cropanzano and Mitchell2005). SET suggests that relationships in organizations are based on reciprocal exchange, where obligations emerge from interactions between parties who depend on each other. Over time, these interactions evolve into reciprocity or repayment rules. In other words, when employees perceive that the organization provides valued resources, they reciprocate with higher engagement, motivation, productivity, or commitment (Nguyen & Ha, Reference Nguyen and Ha2023; Saks, Reference Saks2006). A significant body of literature has used SET to explain the relationship between human resource practices, employee engagement, and turnover intention (Anjum & Din, Reference Anjum and Din2022; Kakkar, Dash, Vohra & Saha, Reference Kakkar, Dash, Vohra and Saha2020; Otoo, Reference Otoo2022). According to SET, when employees perceive HR practices as valuable and contributing to their professional growth, they repay their organization by engaging with their work and organization, reducing the likelihood for turnover. For this study, we proposed that, when employees perceived their organization as providing meaningful work aligned with their strengths, aspirations, and purposes, they are more likely to involve themselves deeply in their work and organization, and therefore, less inclined to leave their job. Adding to that the antecedent role work engagement has on organizational engagement and the previous hypotheses proposed in this study, researchers propose the following hypothesis:
H3: Work engagement and organizational engagement are serial mediators in the relationship between meaningful work and turnover intention.
The proposed model is displayed in Fig. 1.
Proposed model.
Method
Sample
To test the proposed model, a web-based questionnaire was employed. Complete confidentiality and anonymity were ensured at the beginning of the questionnaire. Following the Helsinki Declaration, participants were informed in writing at the beginning of the questionnaire about the study’s objectives and methodology, the confidentiality and anonymity of the data, the voluntary nature of the study, and the participant’s right to withdraw from the study at any time. Moreover, this study has the approval of the Ethics Committee of the Jaume I University (Castellón, Spain). The voluntary participation in answering the questionnaire was considered an indication of their informed consent to participate in the study.
For this study, data were obtained from two separate sources and subsequently combined into a single dataset. The initial sample consisted of 288 employees from a chemical company based in Spain. These data were collected between 2023-07-07 and 2023-10-26. To supplement this sample, an additional 274 participants were recruited using the snowball sampling method. These data were collected between 2024-02-14 and 2024-03-26. The two datasets were then merged to form a comprehensive dataset for analysis. Therefore, the study sample consisted of 562 individuals (47.5% women; 52% men; 0.5% other gender identity) from different economic sectors and organizations.
The age distribution of the sample was as follows: the largest proportion of participants were within the 18–35 age range (41.1%), followed by participants aged 36–45 (26.7%) and 46–55 (21.9%). Finally, 10.3% of the sample consisted of individuals older than 55. Regarding organization tenure, the majority of the participants had been employed with their current organization for more than 15 years (31.5%). This was followed by those with organization tenures ranging 1–5 years (29.7%), 6–10 years (16%), less than 1 year (14.1%) and 11–15 years (8.7%).
Measures
The questionnaire of the study was developed by the authors, drawing upon validated scales and instruments available in literature. For those scales that had not been validated in Spanish, a back-translation was performed (Brislin, Reference Brislin1970). The questionnaire had two sections: (a) the first one dealt with demographic aspects of the respondents, including gender, age, and tenure with their current organization; (b) the second included questions about the study variables (meaningful work, work engagement, organizational engagement, and employee’s turnover intention).
Despite the limitations of self-report measures (e.g.,, social desirability bias, common variance bias) the variables used involve the evaluation of psychological experiences that cannot be captured using other methods (Kwak, Holtkamp & Kim, Reference Kwak, Holtkamp and Kim2019; Podsakoff, MacKenzie, Lee & Podsakoff, Reference Podsakoff, MacKenzie, Lee and Podsakoff2003). In addition, recent studies have demonstrated that self-reported measures can yield valuable results in behavioral research and that their limitations may be overstated (Corneille & Gawronski, Reference Corneille and Gawronski2024).
Meaningful work was measured using the Spanish version (Duarte-Lores, Rolo-González, Suárez & Chinea-Montesdeoca, Reference Duarte-Lores, Rolo-González, Suárez and Chinea-Montesdeoca2023) of the Work As Meaning Inventory scale of Steger et al. (Reference Steger, Dik and Duffy2012). The scale consists of 10 items, organized across three dimensions: Positive meaning in work (four items, e.g., ‘I found a meaningful career’); Meaning making through work (three items, e.g., ‘My work makes me understand myself better’); and Greater-good motivations (three items, e.g., ‘I know my work makes a positive difference in the world’). Participants were asked to indicate their agreement with the statements on a 5-point scale ranging from 0 (completely disagree) to 4 (completely agree).
Work engagement was measured using the Utrecht Work Engagement Scale-9 (Schaufeli, Bakker & Salanova, Reference Schaufeli, Bakker and Salanova2006). This scale consists of nine items assessing the three dimensions of work engagement: vigor (3 items, e.g., ‘At my work, I feel strong and vigorous’), dedication (3 items, e.g., ‘I am enthusiastic about my job’), and absorption (3 items, e.g., ‘I am immersed in my work’). Participants were asked to indicate how often they experience this feeling on a 7-point scale ranging from 0 (never) to 6 (every day).
Organizational engagement was measured using six items adopted from Saks (Reference Saks2006) and translated to Spanish using back-translation. An example of an item is ‘One of the most exciting things for me is getting involved with things happening in this organization’. This scale measures the psychological presence of employees in their organization (Saks, Reference Saks2006). Participants were asked to indicate their agreement with the statements using a 5-point scale ranging from 0 (completely disagree) to 4 (completely agree).
Turnover intentions of employees were measured using three items from Colarelli (Reference Colarelli1984), which were translated into Spanish using back-translation. An example of an item is ‘I frequently think of quitting my job’. Respondents were asked to indicate their agreement with the statements using a 5-point scale ranging from 0 (completely disagree) to 4 (completely disagree).
Control variables
As two different samples were used, ‘sample’ was used as a control variable to avoid any misunderstanding in the relationships of interest. Additionally, organizational tenure was included as a control variable, as a t-test revealed significant differences in responses based on tenure.
Data analysis
First, preliminary analysis such as sample size, descriptive statistics, CR, discriminant validity, convergent validity, and bivariate correlations were conducted for all the study variables. Self-report measures could be associated with common method biases (CMBs). In order to mitigate this risk of bias, Harman’s single-factor test was used using AMOS 28.0 (Arbuckle, Reference Arbuckle2010: Podsakoff et al., Reference Podsakoff, MacKenzie, Lee and Podsakoff2003).
Second, to test the study hypotheses, structural equation modelling (SEM) was carried out through AMOS 28.0 using the maximum likelihood estimation method. Based on Finney and DiStefano (Reference Finney, DiStefano, Hancock and Mueller2006), maximum likelihood estimation is an optimal method under conditions where data follow a normal distribution, sample size is adequate, and data include at least five response options. To test the hypotheses, five models were compared: M1, the partially mediated model; M2, the fully mediated model; M3, the combined partially-fully mediated model; M3r, the combined partially-fully mediated model with errors correlated; and M4r, the alternative model with errors correlated. The following absolute goodness-of-fit indices were used to compare the fit of the models: (a) the X2 goodness-of-fit statistic; (b) the Root Mean Square Error Approximation (RMSEA); (c) the Normed Fit Index (NFI); (d) the Tucker–Lewis Index (TLI); (e) the Incremental Fit Index (IFI); (f) the Comparative Fit Index (CFI); and (g) Akaike’s Information Criterion (AIC). A p-value higher than 0.05 for X2 indicates a good fit. However, as X2 is sensitive to sample size, the use of relative goodness-of-fit indices is recommended (Bentler, Reference Bentler1990). Therefore, RMSEA values lower than 0.08 to 0.05 indicate an acceptable and good fit, respectively (Hu & Bentler, Reference Hu and Bentler1999). For NFI, TLI, IFI, and CFI, values higher than 0.9 indicate a good fit, while values higher than 0.95 indicate a superior fit (Hu & Bentler, Reference Hu and Bentler1999). Finally, according to Kline (Reference Kline2011), AIC can be used to compare competing non-nested models, with the lower AIC index indicating a better fit.
Finally, to examine the indirect effects proposed by the study’s hypotheses, bias-corrected bootstrapped confidence intervals were calculated, with 5,000 samples. A significant indirect effect is observed when intervals do not include the 0.
Results
Preliminary analyses
Based on the recommendations for sample size calculations in SEM (Soper, Reference Soper2023), 137 is the minimum recommended sample size to test the exact model fit in a model with four latent variables and 15 observed variables; 0.3 anticipated effect size; 0.08 statistical power level; and a probability level of 0.05. The sample of the present study included 562 participants. Hence, the minimum sample size in SEM was achieved. Moreover, the normal distribution assumption was supported, based on results obtained from kurtosis and skewness, which were between −2 and +2 (George & Mallery, Reference George and Mallery2010). Therefore, parametric tests (e.g., SEM) can be conducted.
Descriptive analysis
Table 1 shows means, standard deviations (DT), Cronbach’s alpha coefficients, and bivariate Pearson’s correlations for all the study variables. Pearson’s correlations indicate that meaningful work is positively related to work engagement (r = .730, p < .01) and organizational engagement (r = .604, p < .01); and negatively associated with turnover intention (r = −.500, p < .01). Moreover, work engagement is positively related to organizational engagement (r = .681, p < .01) and negatively to turnover intention (r = −.603, p < .01). Finally, organizational engagement is negatively related to turnover intention (r = −.642, p < .01). Regarding internal consistency, all scales presented adequate Cronbach’s alpha indexes (Nunnally & Bernstein, Reference Nunnally and Bernstein1994).
Means, standard deviations, Cronbach’s alphas, and Pearson correlations for the study variables
** p < .01.
From the measurement model, CR, discriminant validity, convergent validity, and bivariate correlations of the study variables are presented in Table 2. The CR of study variables is above 0.70, which, according to Fornell and Larcker (Reference Fornell and Larcker1981), indicates an acceptable level of internal consistency and reliability. Regarding convergent validity, which evaluates the degree to which the indicators of the latent construct correlate with each other; results revealed that the AVE for all the latent constructs of the study is above 0.50, indicating that a substantial portion of the variance in the indicators is explained by the underlying constructs. On the other hand, discriminant validity of the study was also achieved, since the square roots of the AVE were all higher than the inter-construct correlations. Therefore, CR, convergent validity, and discriminant validity were achieved for this study.
Composite reliability, average variance extract, and correlations
Notes: CR = Composite Reliability; Value on the diagonal are square roots of the AVEs.
Since the data of this study came from self-reported data, the potential for CMB exists. Harman’s single-factor test was conducted in AMOS 28.0 to assess CMB (Podsakoff, MacKenzie & Podsakoff, Reference Podsakoff, MacKenzie and Podsakoff2012). The results of Harman’s test revealed that a one-factor model (meaningful work, work engagement, organizational engagement, and turnover intention) showed a poor fit to the data: [RMSEA = .188, NFI = .731, IFI = .741, TLI = .697, CFI = .740, AIC = 1925.073]. Results also showed that the four-factor model fits the data better than the one-factor model: [RMSEA = .076, NFI = .948, IFI = .960, TLI = .950, CFI = .960, AIC = 431.680]. The difference between the two models was also significant, in favor of the model with four latent factors (ΔX2 (6) = 1505.393, p < .001). Therefore, CMB does not represent an issue.
Hypothesis test
To test the proposed hypotheses, various structural equation models (SEMs) were computed, as presented in Table 3. Initially, a partial serial mediation model examining the impact of meaningful work on turnover intention through work and organizational engagement was tested (M1), building upon previous research. Although model fit was acceptable, there was potential for improvement in certain goodness-of-fit indices. Additionally, the direct path from meaningful work to turnover intention was found to be statistically nonsignificant. Consequently, a full serial mediation model (M2) was explored, demonstrating poor model fit in both the chi-squared value (ΔX2M1-M2 = 56.720, p < .001) and goodness-of-fit indices. Finally, a combined full and partial serial mediation model was tested (M3), in which the serial mediation of work and organizational engagement in the relationship between meaningful work and turnover intention was total (since the direct effect of meaningful work on turnover intention in M1 was found to be nonsignificant), but the mediation of work engagement in the relationship between meaningful work and organizational engagement, as well as the mediation of organizational engagement in the relationship of work engagement and turnover intention, were partial. Given that this model showed an acceptable fit and the change in the chi-square value between models was not significant (ΔX2M1-M3 = .42, n.s.) we opted to retain M3 for parsimony and to eliminate nonsignificant effects that do not contribute meaningfully. To improve model fit, modification indices were examined to identify potential adjustments, resulting in the correlation of two errors of the organizational engagement scale (r = .824, p < .01) and two errors of the turnover intention scale (r = .561, p < .01). As the correlated errors belonged to items from the same construct, these adjustments resulted appropriate to refine the model. The new model (M3r) showed a significant improvement in the goodness-of-fit indices and in the chi-squared value compared to the previous model (ΔX2M3-M3r = 93.443, p < .001). Therefore, M3r was chosen as the final model to test the validity of the hypotheses. The different models are illustrated in Fig. 2.
Different models proposed.
Goodness-of-fit indices for the SEM models
Note: M1 = Partial Serial Mediation Model; M2 = Full Serial Mediation Model; M3 = Full and Partial Serial Mediation Model; M3r = Full and Partial Serial Mediation Model revised; M4r = Alternative Model revised; χ2 = Chi-square; df = degrees of freedom; RMSEA = Root Mean Square Error of Approximation; NFI = Normed Fit Index; TLI = Tucker–Lewis Index; IFI = Incremental Fit Index; AIC = Akaike Information Criterion.
First, direct relationships were examined. Results showed a significant negative impact of meaningful work on turnover intention (β = −.574, p < .001), and a significant positive impact of meaningful work on work engagement (β = .820, p < .001) and on organizational engagement (β = .268, p < .001). Moreover, the path from work engagement to turnover intention was found to be negative and significant (β = −.359, p < .001), as was the path from organizational engagement to turnover intention (β = −.405, p < .001). Finally, it was found that work engagement positively and significantly influences organizational engagement (β = .494, p < .001).
Second, to test the study’s hypotheses, bootstrap analysis was performed using 5,000 bootstrapped samples with bias-corrected 95% confidence intervals. The indirect effect is significant if their 95% confidence intervals do not include the 0 value. The indirect effect of meaningful work on organizational engagement via work engagement was found to be positive and significant (indirect effect = .419, 95% CI [.268, .584]). Therefore, work engagement partially mediates the relationship between meaningful work and organizational engagement, since the direct effect of meaningful work on organizational engagement was also significant (H1 confirmed). Besides, the indirect effect of work engagement on turnover intention via organizational engagement was found to be negative and significant (indirect effect = −.210, 95% CI [−.328, −.123]). Consequently, organizational engagement partially mediates the relationship between work engagement and turnover intention, thus confirming hypothesis 2. Finally, serial mediation was tested. The results support an indirect effect of meaningful work on turnover intention, with work and organizational engagement as serial mediators (indirect effect = −.237, 95% CI [−.365, −.139]). Therefore, this result indicates a full serial mediation of meaningful work and turnover intention in the association between work and organizational engagement, confirming hypothesis 3. The model explains 54.9% of the variance in turnover intention. Table 4 shows all the indirect effects studied.
Indirect effects of the final serial mediation model
Finally, an alternative model (M4r) was tested to explore for possible non-arbitrariness of the causal process. In this model, the order of the mediating variables was reversed (Meaningful work → Organizational engagement → Work engagement → Turnover intention). Both models showed identical goodness-of-fit indices, making it impossible to make comparisons between them (see Table 4). However, prior research supports the proposed serial mediation model, since literature has provided evidence supporting the predictor role of work engagement on organizational engagement, but this relation has not been studied in the inverse direction. Moreover, the serial mediation effect of meaningful work and turnover intention through organizational engagement and then work engagement (M4r) (indirect effect = −.121) is weaker than the one of the proposed model (M3r) (indirect effect = −.237). Therefore, the proposed serial mediation model (M3r) is selected as the optimal model.
Discussion
The study hypothesized that employees’ turnover intention could be reduced by promoting meaning in work, through the mediating role of both work and organizational engagement. The interest in studying the relationships among these variables lies in the need for organizations to preserve their human capital within an ever-changing and unpredictable environment. However, mere retention is not sufficient. It is crucial for organizations that employees who remain with them are deeply engaged and involved in their work and their organization, fostering a healthy environment that yields mutual benefits for both the organization and employees. In addition, study results do not show high levels of work engagement (4 out of 6) and organizational engagement (2.55 out of 4). Moreover, turnover intention results indicate a tendency for some employees to consider leaving the organization (1.06 out of 4). These results highlight the relevance of understanding the mechanisms for reducing turnover intention, while developing a healthy, happy, and engaged workforce.
Findings show meaningful work as a strong antecedent of both work engagement and organizational engagement. While the association between meaningful work and work engagement, found in this study is consistent with previous literature (Faisaluddin et al., Reference Faisaluddin, Fitriana, Nugraha and Hinduan2024; Sánchez-Cardona et al., Reference Sánchez-Cardona, Vera and Marrero-Centeno2023), the positive relation between meaningful work and organizational engagement addresses a gap in literature that had not been explored so far. Despite the fact that some theories (e.g., Saks et al., Reference Saks, Gruman and Zhang2022) introduced this relationship, no study to date has directly investigated it. This study suggests that identifying meaning at work may enhance employees’ involvement in their organization. However, these relationships do not have the same predictive power: meaningful work exerts a stronger influence on work engagement (β = .820) than on organizational engagement (β = .268). This is aligned with Saks (Reference Saks2006), who argued that work engagement is more closely linked to job-related variables (e.g., meaningful work), while organizational engagement is strongly associated with organization-related variables. Furthermore, this finding supports the notion that work and organizational engagement are two distinct constructs with different strengths of relationship with other variables (Farndale et al., Reference Farndale, Beijer, Van Veldhoven, Kelliher and Hope-Hailey2014). Therefore, it is crucial to consider both types of engagement when trying to understand employees’ motivations and needs.
Following, it has been discovered that both work and organizational engagement – separately – negatively influence employees’ turnover intention. This is consistent with previous research (e.g., Sheehan et al., Reference Sheehan, Tham, Holland and Cooper2019; Sinisterra et al., Reference Sinisterra, Peñalver and Salanova2024) and with Saks’ (Saks, Reference Saks2006) model, where work and organizational engagement predicted various work-related outcomes, including turnover intention. Moreover, Saks (Reference Saks2006) found that organizational engagement is a much stronger predictor of work-related outcomes than work engagement. Results of this study partially support this assumption, since organizational engagement has a slightly stronger influence on turnover intention (β = −.405) than work engagement (β = −.359).
By the acceptance of its hypotheses, this study confirms that meaningful work impacts positively on organizational engagement through the mediating role of work engagement (H1), and that work engagement negatively affects turnover intention through the mediating role of organizational engagement (H2). Besides, the serial mediation role of work engagement and then organizational engagement in the relationship between meaningful work and turnover intention is confirmed (H3). This is consistent with previous literature, supporting the mediating role of both types of engagement (Juhdi et al., Reference Juhdi, Pa’wan and Hansaram2013; Kundu & Lata, Reference Kundu and Lata2017) and the predictor role of work engagement on organizational engagement (Galingan, Reference Galingan2018). This study makes a significant contribution by examining the underlying psychological mechanism linking meaningful work to turnover intention through both work and organizational engagement. While previous research has explored individual mediations, the integration of these two mediators in a sequential process has not been studied before and is the major finding and contribution of this study. These findings provide information about how meaningful work influences job attitudes – specifically turnover intention – through the pathway of work engagement, which subsequently fosters organizational engagement, and this sequence finally affects – negatively – turnover intention. Also, this sequential pathway is relevant because it suggests that employees first experience engagement towards their work and, then, this engagement translates into an involvement and connection towards their organization. This model sheds light on the intricate mechanisms through which meaningful work influences employees’ turnover intentions. Specifically, it emphasizes the consideration of engagement at multiple levels in order to develop a committed and sustainable workforce.
Finally, the alternative model (M4r), in which organizational engagement predicts work engagement, exhibited similar goodness-of-fit indices, chi-square value, and AIC to those of the final model tested (M3r). Nonetheless, we opted for the latter model (M3r), since previous literature supports work engagement to be an antecedent of organizational engagement. The results of this study also confirm the predictor role of work engagement on organizational engagement. However, the reverse causality – that is, organizational engagement influencing work engagement – has not been explored so far. According to the spillover theory (Sirgy et al., Reference Sirgy, Efraty, Siegel and Lee2001), which suggests that an individual’s experience in one area of their life may have an impact on another area of their life, organizational engagement can act as a predictor of work engagement. Therefore, we investigated this relationship within the full serial mediation model and found it statistically significant (β = .302, p < .000). This reciprocal relationship suggests that work and organizational engagement may function as reciprocal variables generating gain cycles. Such cycles entail a mutually reinforcing relationship, where gains in one variable lead to further gains in both variables (Topa & Aranda-Carmena, Reference Topa and Aranda-Carmena2022). Therefore, this study proposes that work and organizational engagement may reinforce each other. Different studies have demonstrated the existence of gain cycles among organizational variables. For instance, Llorens-Gumbau and Salanova-Soria (Reference Llorens-Gumbau and Salanova-Soria2014) found a positive gain spiral among work engagement, burnout, and self-efficacy. Similarly, Topa and Aranda-Carmena (Reference Topa and Aranda-Carmena2022) identified a gain cycle between job crafting and work engagement.
Theoretical and practical implications
This study has significant theoretical implications, expanding knowledge and literature in the fields of meaningful work, engagement, and turnover intention.
First, while various theories have been proposed to explain employee’s behavior and motivations (e.g., JD-R model, Saks model), none of them integrates and links all four variables examined in this study. Saks’ (Saks, Reference Saks2006) model of employee engagement antecedents and outcomes shows how different job characteristics influence work and organizational engagement which, in turn, impact on turnover intentions. However, this model does not consider meaningful work as an antecedent of employee engagement. This study extends Saks’ framework by introducing meaningful work as an important antecedent of employee engagement and turnover intention. Second, research in organizational engagement is still limited. None of the theoretical frameworks – with the exception of Saks (Reference Saks2006) – consider the dual aspect of employee engagement. This study pretends to expand knowledge in organizational engagement by showing its impact on turnover intention. Additionally, this study proposes a sequential pathway in which these variables interact with each other: meaningful work affects work engagement which, in turn, promotes organizational engagement which, ultimately, reduces turnover intention. Third, this study reaffirms that work engagement and organizational engagement are different constructs, each with different strengths and impacts on organizational variables. This finding aligns with the works of Farndale et al. (Reference Farndale, Beijer, Van Veldhoven, Kelliher and Hope-Hailey2014) and Saks et al. (Reference Saks, Gruman and Zhang2022), and supports the need to assess both forms of engagement to obtain a comprehensive understanding of employees’ behaviors and motivations. Fourth, this study supports the predictor role of work engagement on organizational engagement, supporting existing literature (Galingan, Reference Galingan2018; Rai & Chawla, Reference Rai and Chawla2022; Šakytė-Statnickė et al., Reference Šakytė-Statnickė, Bilan and Savanevičienė2023). In addition, it has also made an approximation to the fact that organizational engagement may also influence work engagement. This bidirectional relationship suggests a reciprocal relationship between work and organizational engagement, making an innovative contribution to literature by suggesting a reciprocal relationship between these two types of engagement.
Furthermore, important practical implications for organizations can be extracted from this study. Turnover intention is a challenge for many organizations worldwide. Not only do events like COVID-19 or the Great Resignation impact employee retention, but also new generations (e.g., Gen Z) are characterized by their tendency to switch jobs more frequently and leave a company if they are dissatisfied (Anh Do et al., Reference Anh Do, Diem Doan, Khanh Vu, Thi Le, Minh Tran and Linh Nguyen2023; Gaan & Shin, Reference Gaan and Shin2023). In this new paradigm, managers need to understand the psychological mechanisms that encourage employees to stay in their organizations. This study provides guidelines for managers on how to retain their employees healthily and happily. First, managers should focus on providing meaning to their employees’ work. Although the perception of meaningful work can vary between sectors and industries (Tan, Sim, Yap, Vithayaporn & Rachmawati, Reference Tan, Sim, Yap, Vithayaporn and Rachmawati2023), managers can promote meaningful work by providing their employees with challenging tasks, autonomy, task variety and task significance (Martela et al., Reference Martela, Gómez, Unanue, Araya, Bravo and Espejo2021; Sanchez-Cardona et al., Reference Sánchez-Cardona, Vera and Marrero-Centeno2023). Some ways organizations can provide meaning to their employees’ jobs include implementing job crafting interventions or redesigning jobs according to their employees’ strengths, values and personal goals (Lysova et al., Reference Lysova, Allan, Dik, Duffy and Steger2019; Sanchez-Cardona et al., Reference Sánchez-Cardona, Vera and Marrero-Centeno2023). Second, this study highlights the importance for organizations of implementing interventions focused on promoting both work engagement and organizational engagement. Although a huge body of literature has explored the effects of psychological and organizational interventions on work engagement (Knight, Patterson & Dawson, Reference Knight, Patterson and Dawson2017), more research is needed on the effects of psychological interventions for improving organizational engagement (Saks et al., Reference Saks, Gruman and Zhang2022). Based on the findings of this study, psychological interventions should be focused on involving employees in their work and tasks, as well as in their organizations. An example of how organizations could do this include to implement leadership training programs or interventions to train managers on how to provide employees with autonomy, recognition, and opportunities for professional growth or teach managers strategies to promote work and organization engagement (Peláez Zuberbühler, Salanova & Martínez, Reference Peláez Zuberbühler, Salanova and Martínez2020). Other interventions to promote meaningful work and employee engagement include mentoring programmes where senior employees help younger people to find meaning in their jobs, internal mobility initiatives or team development programmes (Gomez-Borges, Reyes-Lujan, Salanova & Llorens, Reference Gomez-Borges, Reyes-Lujan, Salanova and Llorens2024; Lin, Cai & Yin, Reference Lin, Cai and Yin2021).
Limitations and future research
While the findings of this study provide valuable contributions to organizational and human resource literature and practice, it is important to recognize its limitations in order to have a comprehensive interpretation of the findings. First, although the sample size is adequate for the study’s objectives, half of the participants come from the same organization. Despite the fact that the study controlled for differences between the two samples, this overrepresentation may limit the generalizability of the findings to other organizational contexts. Future research should use a more diverse sample to provide more representative and robust results. In addition, information about the sector or industry of participants of the external sample is not available. Since work meaning can vary between industries, future research should collect information about sector and industry to account for sector differences and analyze potential differences in how meaningful work, employee engagement, and turnover intention are experienced across different occupations. Future research should also analyze if the serial mediation model proposed holds across different industries and sectors to help generalize the results.
Second, when using self-reported measures (e.g., self-administered online questionnaires), different limitations could emerge. For instance, social desirability bias can occur, in which participants may overestimate or underestimate their responses depending on what they believe is expected of them or what is socially acceptable. However, there are variables related to the psychological experience that cannot be assessed by other methods. In fact, recent studies have defended the validity of self-report questionnaires, and its advantages against other types of measures such as implicit measures (Corneille & Gawronski, Reference Corneille and Gawronski2024). These advantages include greater reliability, flexibility, and predictive accuracy. In order to mitigate this bias, confidentiality and anonymity of answers were ensured at the beginning of the questionnaire (Podsakoff et al., Reference Podsakoff, MacKenzie and Podsakoff2012). Another example of self-report measures’ limitation is CMB. In order to identify and mitigate this bias, some actions were applied: (a) a single-factor Harman’s test using confirmatory factor analysis was conducted, where the four-factor model showed a better fit to the data than the one-factor model; (b) reliability, convergent, and discriminant validity analyses were also conducted, confirming the distinctiveness of each measure (Podsakoff et al., Reference Podsakoff, MacKenzie and Podsakoff2012). Finally, we would like to mention some highly cited studies that have provided relevant implications using self-report measures (Saks, Reference Saks2006; Schaufeli & Bakker, Reference Schaufeli and Bakker2004; Schaufeli, Salanova, González-romá & Bakker, Reference Schaufeli, Salanova, González-romá and Bakker2002). In addition, future research should combine different methodological approaches in order to mitigate CMB resulting from self-reported measures. For example, future studies could integrate quantitative (e.g., questionnaires) and qualitative (e.g., interviews) techniques. On the other hand, in order to reduce social desirability bias, future studies could combine subjective and objective measures (e.g., third-party assessments).
Third, this study used a cross-sectional design that limits the prediction of causal relationships among variables. Future research should use longitudinal designs to understand causal relationships among meaningful work, employee engagement, and turnover intention over time. In addition, future research could use intervention-based studies to test the causal relationships proposed. By conducting these types of studies it might be seen whether changes in one factor (i.e. meaningful work or employee engagement) directly lead to changes in turnover intention of employees.
Fourth, this study made an approximation to the reciprocal relationship between work engagement and organizational engagement. Future research should study this relationship longitudinally to verify the existence of gain cycles among these variables over time (Sonnentag & Meier, Reference Sonnentag and Meier2024).
Finally, future research could extend this model by including additional variables (e.g., job performance, job satisfaction) in order to analyze if the sequential mediation of work and organizational engagement also applies to other wellbeing and work-related outcomes (Breevaart, Bakker, Demerouti & van den Heuvel, Reference Breevaart, Bakker, Demerouti and van den Heuvel2015; Reissová & Papay, Reference Reissová and Papay2021).
Data availability
The authors confirm that the data supporting the findings of this study are available within the article.
Funding
Research co-funded by the Generalitat and the European Social Fund within the framework of the collaboration agreement between the Generalitat Valenciana, through the Department of Education, Culture, Universities, and Employment, and the Universitat Jaume I for the promotion of doctorates in collaboration with companies (PREDOC-DI/2023/01).
Conflict of interest
All authors declare that he/she has no conflict of interest.
Open access
