Abstract
The review of employee well-being literature in the context of human resource management and organisational behaviour advocates that well-being is a major area of consideration for every organisation. Thus, based on job demand resource (JDR) theory, the study aims to examine the influence of job autonomy on the psychological well-being of employees with promotive voice behaviour playing the role of a mediator. Structural equation modelling (SEM) is used to analyse the cross-sectional data of 206 employees from the Indian service sector. The findings support the direct relationship between job autonomy, employee promotive voice and psychological well-being and further confirm the mediating role of promotive voice between job autonomy and psychological well-being. This study suggests that organisations should delegate authority to their employees to carry out their job roles, as it can improve their voice behaviour and further their psychological well-being. To the best of our knowledge, this is the first study to investigate the mediating effect of promotive voice behaviour between job autonomy and psychological well-being.
Introduction
Employees have been recognised as the organisation’s most valuable asset (Pfeffer & Veiga, 1999), and the quality of human resources (HR) determines its success and failure. In today’s competitive and technology-driven era, where organisations continuously seek new ways to improve, it is very important to nurture and retain manpower, as machines cannot always replace the capabilities of humans. Concerning the same, the significance of employee well-being cannot be overlooked (Guest, 2017) because ‘the well-being of employees is in the best interest of communities and organizations’ (Harter et al., 2003, p. 2). We have witnessed employers’ growing interest in their employees’ well-being owing to its varied benefits. For example, Aon India Primary Care Report 2023, which mapped 250 organisations in India, highlighted that 80 per cent of the employers had raised their well-being spending in 2023 compared to 2019 (Verma, 2023). Much evidence has linked employee well-being to several organisational outcomes, including productivity (Jada et al., 2023), job performance (Kundi et al., 2020), affective commitment, job satisfaction, extra-role behaviour (Turban & Yan, 2016) and employee productivity (Grant et al., 2007). In light of this, pursuing a healthy workforce is vital both as an end and as a route towards organisational productivity goals. Therefore, the identification of workplace factors that facilitate employee well-being is crucial.
Researchers have identified three key dimensions of employee well-being: Physical, social and psychological. Previous research indicates that psychological well-being (PWB) is a significant variable associated with individual and organisational level outcomes, such as employee health, employee absenteeism and turnover (Harter et al., 2003) and organisational productivity (Jada et al., 2023). PWB is a multidimensional phenomenon; therefore, researchers have utilised various approaches to explain it. For example, Diener and Seligman (2002) conceptualise it as an emotional state of an individual where they feel contented, happy and strong in their social relationships. Whereas, according to Guest (2017), a distinction between hedonic and eudemonic well-being is often made while considering employees’ PWB. Here, the hedonic aspect is concerned with the subjective feeling of happiness (i.e., job satisfaction), whereas the eudemonic aspect is more concerned with fulfilling human potential and purposeful work. As noted by Harter et al. (2003), ‘the workplace is a significant part of an individual’s life that affects his or her life and the well-being of the community’ (p. 2). Thus, it is important to investigate what kind of workplace characteristics might enhance employee well-being.
Recognising this, employee PWB became a top priority for service management scholars (Anderson & Ostrom, 2015). For instance, authors like Grant et al. (2007) and Guest (2017) stated that employees with better PWB are more committed and involved in work-related tasks. Similarly, we have considered the psychological dimension of employee well-being among private service sector employees in India. The service sector in India is one of the prominent sectors as it contributes over 50 per cent to the nation’s gross domestic product (GDP). Moreover, in 2022, the service trade surplus amounts to $107.52 billion, which is vital in fostering the nation’s economic development (IBEF, 2023) (
Guest (2017) has proposed five sets of HR practices that play a significant role in improving employee well-being, with job autonomy (JA) and employee voice (EV) as the core components of these practices. Previous studies have proved that autonomy is an important factor in positively influencing employees’ mental well-being (Park & Searcy, 2012), subjective well-being (Wheatley, 2017), PWB (Clausen et al., 2021) and also has a beneficial effect on many job attitudes, that is, EV (Dedahanov et al., 2019; Kao et al., 2021), reduced turnover intention (Dysvik & Kuvaas, 2013) and organisational commitment (Parker et al., 2001).
Further, Guest (2017) has also highlighted that ‘the job-demand resource model gives primacy to autonomy as a key resource’ (p. 37). Thus, autonomy is the vital element of job characteristics, as it allows employees to control their work and how they perform it (Hackman & Oldham, 1974). It is one of the motivating components of the job that is desirable for achieving beneficial job-related outcomes (Hackman & Oldham, 1976). Previous studies posit that individuals with higher autonomy experience greater well-being than individuals with lower levels of autonomy. Choi (2007) also suggested that individuals who experience autonomy at work might feel responsible for the change resulting in extra-role behaviour, that is, voice.
EV behaviour is a significant component in the organisational context as it suggests new and improved ways to function (Kim & Leach, 2020). Also, employees’ knowledge and views are pivotal to ensure the success and survival of the business in today’s competitive and dynamic environment. Morrison (2011) defines EV as ‘discretionary communication of ideas, suggestions, concerns, or opinions about work-related issues with the intent to improve organisational or unit functioning’ (p. 3). Further, employees who raise their voices and give suggestions are seen as the benefactors of the organisation (Chamberlin et al., 2017); yet, in some cases, it could be seen as deviant behaviour (Morrison, 2011). Moreover, employees may be reluctant to express their views, particularly in collectivist countries (Prince & Rao, 2020), where individuals exhibit implicit behaviour and generally avoid conflict (Jahanzeb & Fatima, 2018). A recent study by Mannan and Kashif (2020) provided empirical evidence indicating that individuals in collectivist cultures tend to choose silence in the face of workplace violence. Thus, it can impede one’s PWB, which has received less attention from researchers and is seldom prioritised by managers in the workplace (Hasan & Kashif, 2020). The study of Kao et al. (2021) highlights the significance of certain psychological needs that must be fulfilled to anticipate favourable expressions from employees. Therefore, it is essential to understand what factors foster voice behaviour in the organisational setting, considering both positive and negative outcomes.
In this regard, autonomy serves as a powerful medium in empowering individuals to voice their problems, propose solutions and enhance their well-being in the workplace. Thus, autonomy is a crucial factor influencing employees’ tendency to express their opinions and ideas, as stated by Dedahanov et al. (2019). This way, our study justifies the need for an investigation to assess the impact of JA on employees’ promotive voice and PWB. Past studies have identified voice and autonomy as well-being oriented HR management practices (Cooper et al., 2019; Guest, 2017). Raising voice enhances employee morale by instilling a sense of importance and the potential to bring positive change, thereby eliciting positive emotions (Hasan & Kashif, 2020). According to Avey et al. (2012), when individuals express their thoughts, it enhances their self-confidence and contentment. These studies have proposed investigating whether voice is associated with PWB. The present study addresses the underexplored association between voice and employees’ PWB, thus contributing to the existing literature (Xu et al., 2021). Therefore, this study has the potential to assist organisations in implementing suitable measures to enhance the well-being of their employees.
The present study further intends to contribute to the existing literature by testing the mediating role of promotive voice behaviour between autonomy and employees’ PWB. As per the author’s knowledge, it is the first study to check the role of promotive voice as a mediator in the aforesaid relationship. In addition, this study draws conclusions based on job demand resource (JDR) theory to deepen our understanding of how autonomy influences employees’ PWB. We believe autonomy is a valuable job resource that influences employees’ promotive voice behaviour and fosters their well-being. Previous studies also highlight that these variables are positively associated with individual and organisational-level outcomes (Choi, 2007; Park & Searcy, 2012; Parker et al., 2001). Further, this study makes a valuable contribution to literature by focusing on a non-western context, specifically in the South Asian region, where literature is surprisingly limited (particularly in India) (Jada et al., 2023). Jaiswal and Arun (2020) emphasised that well-being is a ‘culture-specific phenomenon’ and determinants of well-being exhibit variations across different country cultures. Therefore, it is crucial to test and validate the association between these variables using Indian employees’ data for better insight (Subhakaran & Dyaram, 2018).
Furthermore, the outbreak of COVID-19 presented new challenges at work, hampering employee well-being. Thus, it is important to reconfirm the previously established association in the present time. Also, the growing importance of India’s economic significance and contribution to global output constitutes the relevance and originality of the study. We believe this will help the management better understand how JA fosters promotive voice behaviour and enhances the employees’ well-being. Based on the above discussion, we propose the following hypothesised model, as presented in Figure 1, that depicts the association among the variables used in this study.

Theoretical Background
According to the JDR theory, employee well-being is the product of balancing job demands and job resources (Bakker & Demerouti, 2007). Job resources are the intrinsic and extrinsic motivating tools for employees: (a) Intrinsically because they help people grow, learn and develop and (b) extrinsically because they help to achieve work goals. They also assist in mitigating the effect of various job-related demands (Bakker & Demerouti, 2007). These two broad categories include several variables. Bakker and Demerouti (2014) argue that certain variables, such as autonomy, work pressure and social support, are relatively general, whereas others are more influenced by the particular work domain. Based on previous research, autonomy is an important job resource that helps in enhancing employee well-being (Clausen et al., 2021). The present study focuses on employees’ ability to make job-related decisions and how it affects their proactive voice behaviour and well-being. The rationale is that employees with higher autonomy are more likely to participate in work-related issues by offering suggestions because they believe they can handle organisational problems through their own actions (Kao et al., 2021; Tangirala & Ramanujam, 2008). Using similar logic, this study considers autonomy as an important job resource. In addition, the study also acknowledges EV as a vital job resource within JDR theory and investigates its mediating effect as suggested by Almeida et al. (2020).
Furthermore, based on self-determination theory, autonomy is a basic psychological need that must be fulfilled to perform effectively (Gagné & Deci, 2005). In line with this, the job characteristic model has also characterised autonomy as one of the important job dimensions having positive outcomes (Hackman & Oldham, 1976). Accordingly, this study suggests autonomy as an important job resource positively associated with employees’ promotive voice and PWB.
Hypothesis Development
JA, Employee Promotive Voice and PWB
JA is a pivotal component of the workplace (Breaugh & Becker, 1987). It is ‘the degree to which the job provides substantial freedom, independence, and discretion to the employee in scheduling the work and determining the processes to be used in carrying it out’ (Hackman & Oldham, 1976, p. 9). When an individual has autonomy at work, they feel more responsible towards the outcome of their job (Hackman & Oldham, 1976). Autonomy can be present in different facets of the job, and here, it is conceptualised as the sum of method, decision and schedule autonomy. Here, method autonomy means the degree of freedom employees have over the procedure they follow to do their job; scheduling autonomy means the extent of freedom employees have to decide about the timing of the job they perform (Breaugh, 1985), whereas decision-making autonomy means the extent of control they have over the decisions related to their job (Morgeson & Humphrey, 2006).
JA is a key element that leads to work enrichment, and organisations that promote employee autonomy have higher employee involvement and performance (Wood & Wall, 2007). Logically speaking, when individuals have control and freedom over their work, they feel free to put forward their ideas and suggestions more openly in the organisations. Hence, employees develop a sense of independence, which leads to more self-directed actions in resolving work-related issues. Further, past research has demonstrated that autonomy is an important job characteristic as it has implications for EV behaviour (Kao et al., 2021) and their well-being (Briner & Walshe, 2015; Clausen et al., 2021; Gardner, 2020; Wheatley, 2017).
In this study, we respond to Subhakaran and Dyaram’s (2018) call for management research on EV using data from Indian organisations because India differs from Western contexts in terms of socio-cultural diversity. Thus, it would be interesting to see how JA can influence the promotive voice behaviour of employees, especially in a collectivist and high-power distance country (India), where subordinates are supposed to take instructions and orders from their supervisor, and supervisors are less likely to accept the discretionary behaviour (Hui et al., 2004). As a result, raising voice could be seen as deviant behaviour (Subhakaran et al., 2020). However, past studies have shown that employees with work autonomy are intrinsically motivated and more likely to exhibit discretionary behaviour (Kao et al., 2021). Thus, JA can be an antecedent to promotive voice behaviour, as past studies suggest that JA is an important element that encourages EV behaviour (Dedahanov et al., 2019) and fosters PWB (Gardner, 2020). Thus, investigating these links in a distinct contextual work environment might lead to a better understanding of the relationship between the study variables. As a result, the following hypothesis is proposed:
H1: JA positively relates to promotive voice.
H2: JA positively relates to PWB.
Employee Promotive Voice and PWB
PWB refers to employees’ positive emotions at work and the degree to which they perceive meaning and purpose at their workplace (Robertson & Cooper, 2010). This definition gives a holistic understanding of the meaning of PWB, as it represents both aspects: (a) Hedonic well-being representing the emotional state of mind which is an integral part of PWB (Ryan & Deci, 2001) and (b) eudemonic well-being representing meaning and purpose in life that is fundamental to an individual PWB (Ryff, 1989).
Past studies have documented a positive association of well-being with performance and affective commitment (Kundi et al., 2020). Further, a plethora of researchers have also identified various antecedents that influence employees’ PWB, such as JA (Gardner, 2020), voice (Xu et al., 2021), psychological empowerment, safety and meaningfulness (Hasan & Kashif, 2020). In line with this, voice is an important variable that positively influences employees’ PWB (Avey et al., 2012; Xu et al., 2021). Past research has demonstrated that employees are more likely to feel positive when they express their views and concerns in accordance with their beliefs and values (Avey et al., 2012).
Hasan and Kashif (2020) suggested that employees’ voice behaviour stimulates PWB when used to bring positive changes to the workplace. Hence, we believe that employees feel more satisfied and happier when they freely express their ideas and views at the workplace. Given the extensive research conducted in the Western context on the relationship between EV and its outcomes (Nawakitphaitoon & Zhang, 2021; Subhakaran et al., 2020), it is important to examine the effect of promotive voice on employee PWB in a non-Western context. As a result, the hypothesis given below is proposed:
H3: Promotive voice is positively related to PWB.
Employee Promotive Voice as Mediator
In today’s competitive environment, employees’ views and ideas are considered to make better decisions and foster organisational growth (Tian et al., 2018). One way of representation of ideas and suggestions is EV. It tends to improve the organisation by giving suggestions and proposing new projects that benefit the organisation (Lin & Jhonson, 2015). Interestingly, Detert and Burris (2007) identify EV behaviour as discretionary, whereas Morrison (2011) identifies it as challenging because it aims to challenge the status quo. It is challenging because it may damage the existing social relationships by giving suggestions that bring changes in already established procedures with which others may disagree (Mowbray et al., 2015; Van Dyne & LePine, 1998). Consistent with this, Liang et al. (2012) have categorised voice into promotive and prohibitive. Liang et al. (2012) define promotive voice as ‘employees’ expression of new ideas or suggestions for improving the overall functioning of their work unit or organization’ (p. 4). Whereas, prohibitive voice is defined as ‘employees’ expressions of concern about work practices, incidents, or employee behavior that are harmful to their organization’ (p. 5).
In this study, we considered promotive voice behaviour of employees as we are more interested in knowing which workplace factors influence employees to voice their opinions and suggest innovative ideas for the betterment of organisational functioning. Liang et al. (2012) suggest that promotive behaviour is more impactful than prohibitive voice as the latter is more concerned about pointing out ‘factors that are or could be harmful for the organisation, without necessarily providing clear solutions to the concern voiced’ (p. 5). So, promotive voice is one of the extra-role behaviours that points out existing problems and suggests better ways for doing things (Van Dyne & LePine, 1998).
Drawing on JDR theory, we argue that JA can significantly and positively affect employees’ PWB, and promotive voice enables this process. When employees have decision-making authority over their work, they are more likely to engage in voice behaviour (Kao et al., 2021). We argue that the employees engaged in promotive voice behaviour will likely benefit from JA and achieve better PWB. This study attempts to contribute to the voice literature; hence, we assume the mediating role of promotive voice to be significant in the relationship of JA and PWB. As a result, the hypothesis given below is proposed:
H4: Promotive voice mediates the relationship between JA and PWB.
Methods
Sample and Procedure
The study uses a cross-sectional survey to investigate the relationship among JA, promotive voice and PWB. A self-reported questionnaire in English was used, and variables were measured using a Likert scale of five points. The scale measured the responses on a range of 1 (strongly disagree) to 5 (strongly agree).
Convenience and snowball sampling methods were used to collect the data from corporate sector employees working in the service industry (India), as used by previous studies (Prince & Rao, 2020). This is because the data were collected during the COVID-19 pandemic, and it was quite challenging to get permission from the higher management to visit their offices. Moreover, most of the employees in the service industry were working from home, so it was not easy to contact them. Thus, a reference-based approach was used to collect the data. Respondents of the study mainly belonged to the banking, finance, insurance, information technology (IT) and information technology enabled services (ITES) sectors. Employees were directly contacted through social media platforms like LinkedIn to fill out the questionnaire, and a Google Forms link was shared with them. Before sharing the questionnaire, individuals were informed about the study, and consent was taken. There was no incentive for participation.
Further, reminders were sent to fill the questionnaire, and a request was also made to further share it with their colleagues. Also, some of the data were personally collected from various private banks and insurance employees. For offline data, branch managers were contacted and informed about the study’s purpose. Once permission was granted from the branch managers, hard copies of the questionnaire were shared, and participants’ consent was also obtained directly. The responses were collected over 4 months, from April to July 2021. A total of 350 questionnaires were sent via both online and offline modes. Out of these, we received 260 filled questionnaires, constituting a response rate of 74 per cent. However, 54 responses were eliminated during data cleaning because of missing data and respondents’ duplicity. Finally, analysis was done on 206 responses. For sample size determination in a scientific manner, we followed the recommendation of Miller and Kunce (1973). It states that at least 10 respondents must be for each predictor in the study.
The study followed the ethical principles of research that included informed consent of the respondents, data confidentiality and limiting the use of information gathered (only for research purposes). Out of the 206 respondents, 152 were male, and 54 were female. With respect to age, the percentage of respondents belonging to various age groups were as follows: 39.3 per cent were below the age of 25, 46.1 per cent were in the age range of 26–30 years, 12.6 per cent were in the age range of 31–40 years, whereas only 2 per cent of the respondents were above the age of 40 years. Further, 36.4 per cent of the respondents were married, whereas 63.6 per cent were unmarried. And, regarding educational qualification, 56.8 per cent of the respondents were graduates, 32.5 per cent were post graduates and the rest 10.7 per cent had professional degrees. Coding for demographic variables used in this study was: Age (under 25 years = 1, 26–30 years = 2, 31–40 years = 3, 41–50 years = 4, 51–60 years = 5), educational qualification (graduation = 1, post-graduation = 2, professional degree = 3, others = 4), gender (male = 1, female = 2) and marital status (unmarried = 1, married = 2).
Table 1 represents the mean, standard deviation and correlation among the three variables used in this study. As presented in Table 1, the alpha coefficient (α) value of all three scales was above the threshold limit of 0.70, as suggested by Nunnally (1978), which confirms the high reliability of the scale used in this study (Streiner, 2003).
Mean, Standard Deviation, Correlation and Reliability.
JA: Job autonomy; PV: Promotive voice; PWB: Psychological well-being.
Measures
JA
A revised version of Breaguh’s (1985) JA scale was used (Morgeson & Humphrey, 2006). It consists of nine items. The scale consisted of three sub-constructs having three items for each construct. The sample item of each sub-construct was—‘the job allows me to make my own decisions about how to schedule my work’ (work scheduling autonomy), ‘the job provides me with significant autonomy in making decision’ (decision-making autonomy), ‘the job allows me to make decisions about what methods I use to complete my work’ (work methods autonomy) (α = 0.932).
Employee Promotive Voice
Liang et al. (2012) scale, consisting of five items, was used to measure employee promotive voice behaviour. A sample item was—‘I proactively develop and make suggestions for issues that may influence the organization’ (α = 0.868).
Employee PWB
Diener et al. (2009) scale, consisting of eight items, was used to measure the PWB of the employees. According to Diener et al. (2009), the scale includes key components of life such as meaning and purpose in life, good relationships and feelings of competence. A sample item was—‘I am optimistic about my future’ (α = 0.892).
Control Variables
Variables like participants’ age and gender were taken as control variables to reduce the impact of exogenous variables in the present study. The decision was made on the basis of past studies, in which the effect of these variables was found to be significant on employee well-being (Jaiswal & Dyaram, 2018; Xu et al., 2021).
Common Method Bias and Multicollinearity Assessment
Because the data were gathered only from single source in a one-time survey, the possible occurrence of common method variance bias cannot be ignored. To prevent this, we used procedural and statistical methods (Podsakoff et al., 2003). Considering the procedural method, respondents were initially informed about the purpose of the study, and then consent was obtained to participate. Further, confidentiality and anonymity of the information provided were ensured. Regarding the statistical procedure, Harman’s single-factor test was applied. The result shows that the total variance extracted by one factor was 41.446 per cent, which was below the qualifying limit of 50 per cent as given by Podsakoff et al. (2003). It indicates that a single factor does not explain the majority of the variance in our study. Hence, problem of common method bias has been ruled out in our study data.
Further, to ensure that there is no multicollinearity issue in our study, the variance inflation factor (VIF) and tolerance value was checked for each predictor. VIF checked the degree to which an independent variable is correlated to another independent variable. In our study, the VIF and tolerance value (Table 2) was within the suggested range (Kock & Lynn, 2012). It means there is no collinearity issue in our study variables.
Confirmatory Factor Analysis (CFA) Loadings and Multicollinearity Analysis.
VIF: Variance inflation factor.
Normality Assessment
To ensure the normality of the data, skewness and kurtosis indices were checked. The values of both lie within the acceptable range between –1 and +1 (Decarlo, 1997).
Result and Analysis
Structural equation modelling (SEM) technique is used to analyse the data with the help of the AMOS™ (v.21) software package. We chose SEM technique because it is one of the powerful tools that evaluate the overall empirical fit. It allows the testing of causal relationships involving both direct and indirect effects (as in mediation). Also, in comparison to regression, using SEM is comparatively easier and error-free for mediation analysis. In addition to this, it allows correction for measurement errors (Orlizky & Frenkel, 2005).
Confirmatory factor analysis (CFA) was performed to ensure the model fitness of the study. The model includes three constructs, namely JA, promotive voice and PWB. CFA was conducted on 22 items, which include JA (nine), promotive voice (five) and PWB (eight). CFA loadings of all the items were above 0.50 (Table 2), so all 22 items were retained for performing further analysis (Hair et al., 2010). Further, the CFA model fit and goodness of fit (GFI) of the model were examined by considering a range of indices (Hu & Bentler, 1999; Schermelleh-Engel et al., 2003). The results are as follows: Chi-square statistics (χ2) was 442.740, degree of freedom (df) was 206, χ2/df was 2.149, comparative fit index (CFI) was 0.915, root mean square error of approximation (RMSEA) was 0.075, Tucker–Lewis index (TLI) was 0.904 and standardised root mean squared residual (SRMR) was 0.0547. Thus, the CFA model indicates that the model used in our study is a good fit of the measurement (Hu & Bentler, 1999; Schermelleh-Engel et al., 2003).
We further evaluated the reliability and validity of the data. The composite reliability of the data in this study was above the threshold limit of 0.70 (Hair et al., 2016), which confirms the reliability of the measurement model. Average variance extracted (AVE) was assessed to evaluate the element of convergence further. The value of AVE was above 0.50, which proves the convergent validity of the model (Hair et al., 2010). Later on, the discriminant validity of the model was assessed, where the square root of AVE marked in bold (Table 3) was higher than the correlation mentioned in the non-diagonal matrix (Hair et al., 2016), which confirms the discriminant validity.
Reliability and Validity Analysis.
SEM method was used for hypothesis testing. The structural model’s adequacy was assessed by comparing the GFI of the model. Table 4 represents the result of hypothesis testing. First, the model was run with control variables. It was found that control variables (age and gender) were not significantly related to PWB in our study. Thus, following the suggestion given by Becker et al. (2016), results without control variables are reported.
Result of Hypothesis Testing.
CFI: Comparative fit indices; TLI: Tucker–Lewis index; RMSEA: Root mean square error of approximation; significance was checked at p < .01 level; JA: Job autonomy; PV: Promotive voice; PWB: Psychological well-being.
The direct relationship between JA and PWB was checked and has been found significant (β = 0.529, p = .000 < .01). Fit indexes were as follows: χ2 = 275.627, df = 118, χ2/df = 2.3358, GFI = 0.856, CFI = 0.926, IFI = 0.926, TLI = 0.914 and RMSEA = 0.081, explaining good model fitness with the available data as suggested by Hu and Bentler (1999). Therefore, H1 is supported. Squared multiple correlation value is 0.280.
Further, the direct relationship between JA and promotive voice has been found significant (β = 0.591, p = .000 < .01). Fit indexes were as follows: χ2 = 180.676, df = 76, χ2/df = 2.3773, GFI = 0.881, CFI = 0.942, IFI = 0.943, TLI = 0.931 and RMSEA = 0.082, the result shows good model fitness in our study as suggested by Hu and Bentler (1999). Therefore, H2 is supported. Squared multiple correlation value is 0.497.
Consequently, the direct relationship between promotive voice and PWB has been found significant (β = 0.599, p = .000 < .01). Fit indexes were as follows: χ2 =162.001, df = 64, χ2/df = 2.5312, GFI = 0.892, CFI = 0.928, IFI = 0.929, TLI =0.912 and RMSEA = 0.86, which shows the good model fitness in our study as suggested by Hu and Bentler (1999). Therefore, H3 is also supported. Squared multiple correlation value is 0.36.
According to the results (see Table 4), JA and promotive voice are directly and significantly related to employees’ PWB and thus meet the conditions laid by Baron and Kenny (1986) for a variable to act as a mediator between any relationship. In social science research, the causal procedure method developed by Baron and Kenny (1986) has been widely used to test the mediation effect (Domhardt et al., 2021). Along with this, we further used AMOS™ (v.21) to determine the mediation effect of promotive voice between JA and PWB. Bootstrapping was also used by requesting 5,000 resamples at 95 per cent of confidence intervals to understand the significance of mediation analysis results obtained through Baron and Kenny’s (1986) approach, as suggested by Abugre and Nasere (2020). The direct effect of JA on PWB was found β = 0.529, whereas, in the presence of promotive voice as a mediator, the value of β is reduced to 0.210 from 0.529. Furthermore, the presence of zero between both lower bound and upper bound intervals [lower limit of confidence interval (LLCI) = –0.079, upper limit of confidence interval (ULCI) = 0.429] confirms that the direct effect was insignificant in the presence of the mediator. The result thus cross-validates the findings obtained through Baron and Kenny’s approach and supports the role of mediator between the independent variable (JA) and the dependent variable (PWB). The result supports the full mediation of proactive voice between JA and PWB. Hence, H4 is also accepted. Table 5 depicts the result of the mediation analysis.
Mediation Result of Promotive Voice Between the Relationship of Job Autonomy and Employee Psychological Well-being.
Discussion
There is an ongoing debate among management scholars and practitioners on how to improve employee’s well-being, particularly in light of observed business environment where things are rapidly changing. Moreover, employees are under continuous pressure to adapt to these new changes, for instance, transition to remote work and the adoption of hybrid work models due to COVID-19. There are studies positing that empowered employees effectively cope with these changes, thus leading to better performance. It is because of the fact that autonomy and proactive voice at work are characterised as the psychological state that instils a sense of pride and accomplishment when employees experience the direct contribution of their efforts on the success of the organisation. In line with the same, the study aimed to contribute to the literature on employee well-being and to explore the mediating role of promotive voice on the relationship between JA and PWB. The results suggest four findings.
First, the study supports the positive influence of autonomy on employees’ PWB. The finding is consistent with Nielsen et al.’s (2017) meta-analysis study. It supports the idea of autonomy as an important resource which is positively related to the PWB of employees. The findings suggest that when employees have control over their work, they tend to perform better, which improves their well-being. Drawing on JDR theory, the study provides evidence that autonomy is a significant job resource that has a favourable impact on employees’ PWB.
Second, the study shows the positive association between JA and promotive voice behaviour. The study employs job demand–resource theory and provides a novel perspective to understand the aforesaid relationship. The findings support past results that claim a positive association between autonomy and voice behaviour (Dedahanov et al., 2019). The study suggests that individuals with higher level of autonomy eventually offer ideas that are beneficial for the functioning of their work unit and express concerns about issues that could potentially lead to significant losses. Moreover, autonomy allows individuals to exercise their own judgments without the fear of reprimand from their supervisors (Sheoran et al., 2022). This allows employees to engage in extra-role behaviour, for example, voice. According to Morrison (2023), organisations that adhere to best practices for cultivating a highly engaged workforce, such as offering autonomy, are likely to create an environment that promotes greater EV. Accordingly, our results also support autonomy as an important factor in employees’ promotive voice behaviour and PWB.
Third, the finding reveals that promotive voice enhances employees’ PWB. Consistent with past studies, our study also shows that employees benefit psychologically when they express their thoughts (Xu et al., 2021). Past studies also indicate a positive association between promotive voice and a person’s positive emotion (Lin & Jhonson, 2015). In this regard, management needs to cultivate a work environment where employees feel they are being heard and their ideas and opinions are appreciated. Raising voice involves expressing dissatisfaction in the workplace (Wood & Wall, 2007), which may have potential losses (Detert & Burris, 2007), due to which individuals often hesitate to put forth their ideas. These studies demonstrate that employees are more likely to voice their opinions and concerns when they feel valued and empowered (Morrison, 2011), and it can contribute to the enhancement of their PWB.
Fourth, the study also supports the mediating role of promotive voice between JA and PWB. This means that when employees enjoy higher autonomy at work, they are more likely to participate in promotive voice behaviour that is beneficial for the organisation, regardless of situational factors. It will help them to sort out any work-related problems, furthering their well-being. The full mediation effect of promotive voice behaviour signifies its importance in the service industry. The findings highlight promotive voice as an important link between JA and PWB. Relying on the same, we suggest that managers should encourage employees to be more proactive in raising their voices, as it improves their well-being.
Limitations
Even though every empirical study contains flaws, it is vital to point them out so that future studies can improve and further replicate the findings of current study with caution when drawing conclusions (Avey et al., 2012). Our study also suffers from few limitations. First, the association between JA, employees’ promotive voice and PWB is examined using cross-sectional data, which may result in common method biasness (Podsakoff et al., 2003). However, in our study, common method biasness was not an issue as depicted by Harman’s single-factor test. Second, the study made use of only one type of voice behaviour, that is, promotive; hence, future study may include other type of voice behaviour, such as prohibitive voice. As prohibitive voice is different from promotive voice in terms of ‘behavioral content, their function and implications for others,’ (Liang et al., 2012, p. 5). It may also influence the association between JA and PWB. Third, the study made use of employees’ perspectives only, while the perspective of their managers/supervisors is missing to understand the influence of JA on employees’ promotive voice and PWB. So, in future, research studies should collect data from both employee–supervisor dyad. Further, for the generalisation of the results, future studies may be conducted in other contextual areas (on basis of geographical location and demographic profile of the respondents).
Theoretical Implications
The present study offers a novel perspective on the association between JA and PWB through promotive voice as a mediator. It also contributes to the existing literature on job demand theory. We observed that although previous studies have used JDR theory to examine how JA influence employees’ well-being, relatively little attention was paid to the mediating role of promotive voice behaviour. The positive association between these variables suggest that organisations should provide more flexibility to employees via JA in order to enable promotive voice behaviour. This study also responds to Dedahanov et al. (2019) call to investigate the effect of voice on individuals. In line with this, we hypothesised and found a significant effect of promotive voice on employees’ PWB. Our study suggests that when employee raises their opinion and views, their well-being is improved. This study further responds to Subhakaran and Dyaram (2018) call to conduct management research using indigenous data from Indian enterprises in order to elicit employee perspectives on their voice experiences. Thus, the findings of our study reinforce the relationship of JA, promotive voice and PWB in the new geographical context, that is, India, whereas previous studies were more focused in the Western context (Nawakitphaitoon & Zhang, 2021).
Practical Implications
The results of the present study offer important implications for management practitioners. There is no denying that employee well-being is a key factor for the overall well-being and smooth functioning of any organisation. Further, in the context of the service industry, the survival of organisations depends on their ability to deal with the changes happening in the business world. A significant source to deal with these changes is the flow of information, and here, EV could play an important role. Through their voice and opinions, employees can suggest better and more innovative ways to improve organisational agility. This study suggests that managers should try to develop a work environment where employees feel valued, empowered and secure to speak up. This is important, considering the fact that employees who raise their voices at workplace often do this at the risk of their interpersonal relationships, especially in a collectivist country like India where people tend to follow others (Subhakaran & Dyaram, 2018). We believe employee’s control over their work (autonomy) and how they perform could help them to feel safer and more empowered in raising their ideas and suggestions. The underlying rationale here is that individuals with higher autonomy will actively seek opportunities to fulfil their needs. This can foster employees’ promotive voice behaviour, leading to improved well-being. This study will help management to understand how employees with autonomy tend to have higher PWB and how promotive voice can strengthen this association. Also, by empowering employees through JA, management can exhibit commitment towards the well-being of its employees.
Conclusion
Through this study, the importance of JA on promotive voice and PWB has been highlighted. We hope that this study will contribute to managers’ understanding that giving freedom to employees is an important practice in developing a workplace where employee well-being is ensured. Drawing on JDR theory, we posit that JA is an important job resource that can help employees improve their PWB. Further, it could be seen that voice behaviour induced by JA improves their well-being. Thus, organisations need to foster a positive work culture where employees feel comfortable expressing their opinions on job-related issues.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.
