The impact of procedural justice on employee turnover intentions and the role of two mediators

Abstract

Purpose

While there is ample discussion in management studies and organizational behavior textbooks about the factors that impact organizational outcomes, such as employee retention, this research is focused on exploring the previously unexplored question of how procedural justice, job characteristics and meaningful work influence employees' intentions to leave their organizations. As such, this study aims to investigate the impact of procedural justice on employees' intentions to leave, both independently and in conjunction with job characteristics and meaningful work as mediators.

Design/methodology/approach

This study uses partial least squares structural equation modeling (PLS-SEM) to develop the research model and for hypothesis testing. The path model is assessed using critical model fit indices and measures of goodness of fit.

Findings

The results reveal a negative relationship between procedural justice and employees’ intentions to leave. This negative relationship persists and is strengthened when both job characteristics and meaningful work act as mediators. Although job characteristics only exerted a significant effect through indirect effects, meaningful work demonstrated a significant negative impact on the intentions to leave through both direct and indirect effects.

Originality/value

This study presents a new perspective on employee retention by proposing an original mediation-based path model. Through the testing of eleven hypotheses, the study reveals the intricate relationships between the four constructs examined. The findings provide valuable insights that can serve as a basis for future research in management studies and organizational behavior.

Citation

O'Callaghan, M. (2024), "The impact of procedural justice on employee turnover intentions and the role of two mediators", Organization Management Journal , Vol. 21 No. 2, pp. 75-87. https://doi.org/10.1108/OMJ-03-2023-1822

Introduction

Employee retention has emerged as one of the most critical challenges faced by organizations today. This is due to various factors, including changing employee expectations, increased mobility, and the growing complexity and dynamism of work environments. Retention, which refers to the low or no turnover of employees, has been a well-studied phenomenon in organizational behavior for over a century (Cotton & Tuttle, 1986). Numerous investigations have been conducted to explore the variables and constructs that influence employee turnover. Nevertheless, the actual employee turnover and employees' intentions to leave the organization are often correlated (Castle, Engberg, Anderson, & Men, 2007). Turnover intention, also known as intention to leave, is the most important indicator of the actual leaving behavior of the employees (Ajzen, 1991).

While several studies have explored factors that influence employees' intentions to leave, research on the specific impact of procedural justice, both independently and in conjunction with job characteristics and meaningful work, remains limited. Therefore, this study aims to investigate the impact of three essential factors, procedural justice, job characteristics and meaningful work, on employees' intentions to leave.

Procedural justice deals with fairness in decision-making processes concerning the mechanisms, methods and processes used to make decisions; researchers began to propose procedural justice due to the limitation of the distributive justice concept (Folger & Cropanzano, 1998; Akoh & Amah, 2016). In the absence of procedural justice, employees are more likely to quit their jobs.

According to the job characteristics theory, job characteristics are the elements of a job that make work motivating and meaningful for employees. The five characteristics or task attributes proposed by the theory are autonomy, task identity, task significance, skill variety and feedback (Hackman & Oldham, 1980; Miner, 2005). Job characteristics are significantly associated with employee turnover intentions and are considered a contributing factor to employee retention (Chang, Wang, & Huang, 2013; Vui-Yee & Paggy, 2020). Further, studies have consistently demonstrated that job satisfaction, which can be predicted by job characteristics, serves as a predictor for employee turnover (Iverson & Currivan, 2003; Mudor, 2011; Flowers & Hughes, 1973; Loher, Noe, Moeller, & Fitzgerald, 1985; Glisson & Durick, 1988; Steyn & Vawda, 2014). The evidence in these studies suggests that employee turnover can be decreased if job characteristics are improved.

Meaningful work deals with jobs that create a sense of joy and help employees connect to a larger good and things that they view as important in their lives (Duchon & Plowman, 2005). A thorough review of the literature in this study shows that meaningful work is attributed directly and as a mediator to many organizational outcomes, such as employee performance, satisfaction, engagement and retention.

The researcher's assumptions on the relationship between the four constructs under investigation were supported by a literature review. This, in turn, helped develop the conceptual model, which was tested with data collected through a survey on Prolific following hypothesis testing and model evaluation. The interrelationships among the constructs were examined using dependable partial least squares structural equation modeling (PLS-SEM). Before concluding the study, its implications and limitations are discussed in detail.

Theoretical background and hypotheses

Intent to leave is indicative of current dissatisfaction with one's employment and is found to be the strongest predictor of an employee's actual turnover decision (Knani & Fournier, 2013; Michaels & Spector, 1982; Johnsrud & Rosser, 2002). Job dissatisfaction can even lead employees to leave their core professions (Sasso et al., 2019). Existing literature shows a positive and significant effect of job characteristics on job satisfaction (Glisson & Durick, 1988; Bhuian & Mengue, 2002; Morgeson & Humphrey, 2006), which is inversely related to turnover intentions (Medina, 2012). Thus:

H1.

There is a negative relationship between job characteristics and employees' intentions to leave.

Job characteristics are considered a significant driver of meaningful work (Chen, Lee, Chen, & Wu, 2016; Renard & Snelgar, 2016). With the application of Hackman & Oldham (1975) job characteristics model, the researchers observed positive associations between meaningful work and job characteristics such as skill variety and task significance (Johns, Xie, & Fang, 1992; Schnell, Höge, & Pollet, 2013). In their study, Bailey, Yeoman, Madden, Thompson, & Kerridge (2019) cited at least 20 studies that examine meaningful work as a psychological state based on the job characteristics model, including research on employee engagement and psychological empowerment. Job characteristics are, in fact, used as an indicator of meaningful work according to some studies (e.g. Piccolo & Colquitt, 2006). Thus:

H2.

There is a positive relationship between job characteristics and meaningful work.

Further, meaningful work can help organizations create positive work outcomes for their employees, such as satisfaction, engagement, commitment, individual and organizational fulfillment, productivity, loyalty and retention (Geldenhuys, Taba, & Venter, 2014). Thus:

H3.

There is a negative relationship between meaningful work and employees' intentions to leave.

Organizational justice refers to people's perceptions of how fairly they are treated by their organizations (Levy & Norris-Watts, 2004). Early theories attempted to define and understand justice in three forms: distributive, procedural and interactional justice (Fortin, Cojuharenco, Patient, & German, 2016). Procedural justice is the most significant of the three justice components that can help organizations build a reputation and improve the leader-subordinate relationship (Akoh & Amah, 2016). Also, procedural justice is more significant in improving organizational commitment compared to distributive justice (Clay-Warner, Hegtvedt, & Roman, 2005). The goodwill of the organization depletes when employees feel a lack of procedural justice, and it results in employee turnover, among other consequences (Akoh & Amah, 2016). Procedural justice negatively affects the turnover intentions of the employees and has a statistically significant inverse association with turnover intent (Raza, Ul Hadi, & Mujtaba, 2022; Lambert et al., 2010). Thus:

H4.

There is a negative relationship between procedural justice and employees' intentions to leave.

Studies that specifically examine the relationship between procedural justice and job characteristics are scarce. Nonetheless, organizational justice can predict job satisfaction and moderate the relationship between job characteristics and job satisfaction; hence, organizational justice is associated with job characteristics (Montañez-Juan, García-Buades, Sora-Miana, Ortiz-Bonnín, & Caballer-Hernández, 2019). To address the existing knowledge gaps identified in the literature concerning the two constructs, the next hypothesis proposes:

H5.

There is a positive relationship between procedural justice and job characteristics.

There is a substantial shortage of studies that have examined the relationship between procedural justice and meaningful work, in line with the previous hypothesis. The literature, however, showed that organizational justice and meaningful work with empathy and professional identification could reduce burnout (Correia & Almeida, 2020), which can significantly increase work stress and hence, turnover intentions (Salama, Abdou, Mohamed, & Shehata, 2022). As previous studies have not specifically examined the relationship between procedural justice and meaningful work, the formulation of the following hypothesis can contribute to bridging the existing gap in the literature:

H6.

There is a positive relationship between procedural justice and meaningful work.

The following hypotheses involve mediation effects, but they are all based on the same constructs and themes. Previous studies have shown how job characteristics are associated with meaningful work (Hackman & Oldham, 1975, 1980) and that workers who view their jobs as significant and meaningful are more committed and less likely to leave (Wingerden & Stoep, 2018; Holbeche & Springett, 2004). Further, procedural justice can reduce employees' intentions to leave; it significantly and negatively influences turnover intentions (Edrees, Sobaih, Gharbi, & Abu Elnasr, 2023). Thus:

H7.

Job characteristics mediate the relationship between procedural justice and meaningful work.

In their study, Wang, Liu, Luo, Ma, & Liu (2016) showed that job security fully mediates the relationship between procedural justice and turnover intentions. As job security is not the same as job characteristics, nor is it considered one of the key job characteristics according to the job characteristics theory (Hackman & Oldham, 1980), Wang et al.'s study offers only partial support for the subsequent hypothesis in this study when we substitute job security with job characteristics. Specifically, H8 is proposed to examine if job characteristics can mediate the relationship between procedural justice and intentions to leave. Thus:

H8.

Job characteristics mediate the relationship between procedural justice and intentions to leave.

Agarwal & Gupta (2018) conducted a study where they found that the relationship between job characteristics and managers' turnover intentions was mediated by work engagement. Drawing assumptions from similar studies, H9 investigates if meaningful work, like work engagement, can mediate the relationship between job characteristics and intentions to leave:

H9.

Meaningful work mediates the relationship between job characteristics and employees' intentions to leave.

Meaningful work is a significant factor in retaining employees and reducing intentions to leave (Scroggins, 2008; Janik, 2015). The results of Bayarçelik & Findikli (2016) study showed that perception of procedural and distributive justice directly affected job satisfaction and decreased the intentions to leave. In their study, job satisfaction mediated the relationship between procedural and distributive justice and intention to leave. Bayarçelik and Findikli's study offers partial support for the subsequent hypothesis (H10) when we substitute job satisfaction with meaningful work and remove the distributive justice construct. While their study uses different constructs, the overall framework of relationships and the core idea that justice mediated by another construct can influence intentions to leave aligns with H10. Thus:

H10.

Meaningful work mediates the relationship between procedural justice and intentions to leave.

Employees' intentions to leave are highly sensitive to their perceptions of procedural justice (Daileyl & Kirk, 1992), and as explained through different studies included in this research, there is a significant relationship between these two constructs. Although there are studies (e.g. Poon, 2012) that used a mediation-moderation framework to study how and when distributive justice and procedural justice interact to predict turnover intention, no studies were found to investigate if job characteristics and meaningful work can mediate the relationship between procedural justice and intentions to leave. Thus:

H11.

Job characteristics and meaningful work mediate the relationship between procedural justice and intentions to leave.

Methods

Data collection and participants

Data were collected from 448 participants in the USA, recruited on Prolific, who were employed either full-time or part-time. The final sample consisted of responses from a total of 439 participants since nine participants who completed the study in less than 50% of the median time were removed to ensure the quality (Greszki, Meyer, & Shoen, 2015). The final sample comprised 49.20% women and 50.56% men, 0.22% undisclosed gender. In total, 73.12% of the participants reported their ethnicity as white, 6.60% as black, 10.70% as Asian, 5.46% as mixed and 4.10% as other. Participants' age ranged between 18 and 79. Additionally, 72.44% of participants were full-time, and 27.56% were employed part-time.

Measures

Procedural justice or PJ: Five items from Moorman's (1991) procedural justice scale were adapted to assess the extent to which employees perceive that their organization has provided procedural justice. A sample item is “Concerns of all those affected by the decision are heard.” The Cronbach's alpha for this measure was 0.86.

Job characteristics or JC: Two highest loading items were selected from Hackman & Oldham (1980) job characteristics model, and one item from Morgeson & Humphrey (2006) Work Design Questionnaire was adapted to measure how employees rate specific characteristics of their jobs. The sample item is “The job requires me to use a number of complex or high-level skills.” The Cronbach's alpha for this measure was 0.79.

Meaningful work or MFW: Steger, Dik, & Duffy (2012) Work and Meaning Inventory was adapted to measure the extent to which employees perceive their work as meaningful. The sample item is “The work I do serves a greater purpose.” The Cronbach's alpha for this measure was 0.89.

Intention to leave or ITL: Two items from Landau & Hammer (1986) study were adapted to assess employees' intentions to leave their organization. The items are: “I am actively looking for a job outside of my organization” and “I am seriously thinking about quitting my job.” The Cronbach's alpha for this measure was 0.87.

All the items are measured on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Results

Smartpls 4 software is used to assess the model's validity and reliability and test the hypothesis. Due to its capabilities of generating results for a higher number of model fit measures, IBM Amos was used for model fit assessment.

The measurement model

The values of Cronbach's alpha and composite reliability of all the constructs in this study are above the threshold of 0.70; the AVE (average variance extracted) values are above the minimum threshold of 0.50, confirming the convergent validity of the four constructs in this study (Hair, Risher, Sarstedt, & Ringle, 2019). The HTMT (Heterotrait-Monotrait) ratio of the correlation values for all four constructs is below the maximum threshold of 0.85 (Clark & Watson, 1995), providing evidence for the discriminant validity of the measurement model. The research model is shown in Figure 1.

The structural model and hypothesis testing

The consistent PLS algorithm with 5,000 bootstrap samples was used to assess the model's structural relationships and test the hypotheses. The VIF (variance inflation factor) values below the maximum threshold of 5 indicate that collinearity issues do not bias the regression results (Hair et al., 2019). For hypothesis testing, a 95% confidence level with an alpha value of 0.05 was used, accounting confidence interval values for ULC (Upper-Level Confidence) and LLC (Lower-Level Confidence) do not contain zero (Hayes, 2013). Hypotheses one to six are tested with direct effects path coefficients, as shown in Table 1.

The results showed an insignificant positive relationship between JC and ITL with β = 0.071, SE = 0.118, t = 0.601, LLC = −0.152, ULC = 0.306, p > 0.05, which does not support H1. Therefore, H1 is rejected. H2 was validated, showing a significant positive relationship between JC and MFW with β = 0.746, SE = 0.038, t = 19.788, LLC = 0.668, ULC = 0.817, p < 0.05. H3 was also validated, showing a negative relationship between MFW and ITL with β = −0.519, SE = 0.107, t = 4.834, LLC = −0.729, ULC = −0.310, p < 0.05. H4 was validated, showing a negative relationship between PJ and ITL with β = −0.148, SE = 0.060, t = 2.473, LLC = −0.265, ULC = −0.028, p < 0.05. H5 was validated, showing a positive relationship between PJ and JC with β = 0.440, SE = 0.051, t = 8.593, LLC = 0.332, ULC = 0.536, p < 0.05. Finally, H6 was also validated, showing a positive relationship between PJ and MFW with β = 0.131, SE = 0.052, t = 2.531, LLC = 0.024, ULC = 0.227, p < 0.05.

Mediation analysis

As shown in Table 2, mediation analysis was performed with indirect effects where H7 was validated, showing JC mediate the relationship between PJ and MFW (β = 0.328, SE = 0.042, t = 7.765, LLC = 0.248, ULC = 0.416, p < 0.05). H8 was rejected since the results did not support the hypothesis proposing JC mediate the relationship between PJ and ITL (β = 0.031, SE = 0.053, t = 0.591, LLC = −0.068, ULC = 0.140, p > 0.05). H9 was validated as results showed evidence for MFW mediating the relationship between JC and ITL (β = −0.387, SE = 0.087, t = 4.468, LLC = −0.567, ULC = −0.228, p < 0.05). H10 was validated with (β = −0.068, SE = 0.029, t = 2.363, LLC = −0.132, ULC = −0.017, p < 0.05), showing MFW mediates the relationship between PJ and ITL. Finally, there was strong evidence in support of H11 proposing JC and MFW mediate the relationship between PJ and ITL (β = −0.170, SE = 0.043, t = 3.981, LLC = −0.271, ULC = −0.100, p < 0.05); H11 was validated. Both direct and indirect effects were significant for H7, H10 and H11, signifying complimentary or partial mediation. Full or indirect-only mediation is reported for H9 due to significant indirect effects and insignificant or no direct effects. The indirect effects were found insignificant for H8 with significant direct effects, and therefore, it is appropriate to conclude that no mediation exists between the two constructs in this hypothesis (Hair et al., 2021).

Model fit assessment and model evaluation

The model fit was tested, and results indicate that the model is a good fit with a CMIN/df (Chi-square minimum discrepancy/degrees of freedom) value of 3.53, SRMR (standardized root mean square residuals) value of 0.075 and RMSEA (root mean square error of approximation) value of 0.076, which are all below their respective thresholds. The NFI (normed fit index) value of 0.93 and CFI (comparative fit index) value of 0.95 both meet or exceed their thresholds, indicating a good fit for the model (Dash & Paul, 2021).

The R² values for ITL, JC and MFW are 0.298, 0.194 and 0.660, respectively, indicating weak to moderate predictive power. In addition, the Q² values for all endogenous constructs (ITL = 0.092, JC = 0.13, MFW = 0.16) exceeded the minimum threshold of 0, establishing the model has the capacity to make predictions (Hair, Matthews, Matthews, & Sarstedt, 2017).

Discussion

The model in this study was based on the relationships between the four constructs of procedural justice, job characteristics, meaningful work and intention to leave. The study analyzed three job characteristics items: task significance, specialization and skill variety. The direct coefficient value between job characteristics and intention to leave was positive, but the p-value was not significant. This indicates that job characteristics have no significant positive or negative impact on employees' intentions to leave.

However, when meaningful work was included as a mediator between job characteristics and the intention to leave, the relationship became negative and significant. The results suggest that job characteristics have an insignificant relationship with intention to leave, but the construct has a significant impact on enhancing meaningful work, which was measured based on the extent to which employees think their work helps them make a positive difference, understand themselves, make sense of the world around them and serve a greater purpose.

The direct effect results indicate that meaningful work can significantly lower employees' intentions to leave. Additionally, the results of indirect effects for H9, H10 and H11 show that meaningful work effectively serves as a mediator between job characteristics and intention to leave, procedural justice and intention to leave, and as a joint mediator with job characteristics between procedural justice and intention to leave. In all cases, meaningful work has a significant negative impact on intention to leave.

The findings suggest that procedural justice can directly reduce employees' intention to leave, but it also significantly impacts job characteristics and meaningful work. The direct effects of procedural justice on intention to leave are significant, but the coefficients become more substantial (with higher negative values) in reducing intention to leave as the two mediators, job characteristics and meaningful work, were added.

Implications and limitations

This study has important implications for industry practitioners, academic researchers and those in management and organizational behavior education. Many studies have previously explored the impact of job characteristics on a variety of organizational outcomes, including employee satisfaction, performance and turnover reduction (Hackman & Oldham, 1975, 1980; Knani & Fournier, 2013; Michaels & Spector, 1982; Johnsrud & Rosser, 2002; Morgeson & Humphrey, 2006; Glisson & Durick, 1988; Bhuian & Mengue, 2002; Medina, 2012). This study, however, sheds light from a different angle on the relationship between procedural justice, job characteristics, meaningful work and employee turnover intentions.

While the study found that job characteristics may not directly reduce employee turnover, it can significantly improve the meaning aspect of work. Policymakers can use these findings to design jobs that promote meaningful work, which is found to have a negative impact on intentions to leave directly and as a mediator.

Similarly, procedural justice seems to impact all three endogenous constructs substantially. Therefore, leaders and managers must focus on making their decision-making processes fairer. Most organizational behavior textbooks discuss decision-making, but they often fail to emphasize the importance of procedural justice. The equity theory of leadership touches on fairness, but it is not equivalent to procedural justice. Justice in organizations affects both supervisory and organizational level outcomes, including job performance, which can be predicted by organizationally focused procedural justice (Rupp & Cropanzano, 2002; Masterson, Lewis, Goldman, & Taylor, 2000). Employee performance is directly related to turnover intentions. Compared to low performers, high performers are more satisfied with their work and therefore, are less likely to leave their jobs (Willyerd, 2014).

The model in this study facilitates the future expansion of research endeavors by adding, removing or modifying variables within the existing constructs. Furthermore, by using diverse methodologies for identifying data patterns, future researchers are likely to uncover additional constructs that impact employees' intentions to leave the organization, such as employees' personal and professional life circumstances. Consequently, this study serves as a foundational resource for scholars and educators, helping them explore more supplementary constructs that enhance the meaningfulness of work and those constructs directly or indirectly influenced by meaningful work, which stands as the foremost mediator within this investigation. These findings allow organizational policymakers to comprehend the significance of fostering meaningful work experiences to retain employees and fully leverage the organization's talent pool. It is essential to acknowledge that the research model used in this study prioritized quantitative accuracy, reliability and validity, potentially resulting in the omission of crucial factors, such as interpersonal relationships, in the context of job characteristics. While the study's sample size was sufficiently large to generate a reliable model, increasing the sample size in future research endeavors may yield improved outcomes.

Conclusion

This study found a complex relationship between job characteristics, procedural justice, meaningful work and employees' intentions to leave their organizations. Therefore, it recommends that organizations emphasize improving job characteristics to enhance meaningful work, which can substantially reduce employee turnover intentions. In addition, procedural justice must be ensured in decision-making since it reduces employees' intentions to leave and positively affects job characteristics and meaningful work.

The implications of this study are significant for industry practice, academic research and management and organizational behavior education. This research will pave the way for future researchers to create a more holistic model that reveals crucial hidden patterns and relationships between these constructs and other constructs that influence employee turnover intentions, including job satisfaction, organizational culture, compensation and benefits and employee engagement.

Figures

Figure 1.

Research model with hypotheses

Table 1.

Hypothesis testing with direct effects

Hypothesis	Relationship	Beta	SE	t value	p-value	LLC	ULC	VIF
H1	JC → ITL	0.071	0.118	0.601	0.548	−0.152	0.306	2.878
H2	JC → MFW	0.746	0.038	19.788	0.000	0.668	0.817	1.240
H3	MFW → ITL	−0.519	0.107	4.834	0.000	−0.729	−0.310	2.939
H4	PJ → ITL	−0.148	0.060	2.473	0.013	−0.265	−0.028	1.290
H5	PJ → JC	0.440	0.051	8.593	0.000	0.332	0.536	1.000
H6	PJ → MFW	0.131	0.052	2.531	0.011	0.024	0.227	1.240

Notes:

JC = job characteristics; ITL = intention to leave; MFW = meaningful work; PJ = procedural justice

Source: Table by author

Table 2.

Hypothesis testing with indirect effects and mediation analysis

Hypothesis	Relationship	Beta	SE	t value	p-value	LLC	ULC	Mediation type
H7	PJ → JC → MFW	0.328	0.042	7.765	0.000	0.248	0.416	Partial
H8	PJ → JC → ITL	0.031	0.053	0.591	0.554	−0.068	0.140	No mediation
H9	JC → MFW → ITL	−0.387	0.087	4.468	0.000	−0.567	−0.228	Full
H10	PJ → MFW → ITL	−0.068	0.029	2.363	0.018	−0.132	−0.017	Partial
H11	PJ → JC → MFW → ITL	−0.170	0.043	3.981	0.000	−0.271	−0.100	Partial

Notes:

JC = job characteristics; ITL = intention to leave; MFW = meaningful work; PJ = procedural justice

Source: Table by author