Master Thesis
The Relationship between Ethical Leadership, Procedural Justice and Employee Gossiping Behavior, Moderated by Work Stress
Fleur Gaby Louise Burghard University of Amsterdam
Author Note
The progress of this Master’s thesis was supervised by dr. Annebel H.B. de Hoogh, and is part of the master track Business Administration Leadership and Management at the University of
Statement of originality
This document is written by student Fleur Gaby Louise Burghard who declares to take full responsibility for the content of this document.
I declare that the text and the work presented in this document is original and that no
sources other than those mentioned in the text and its references have been used creating it. The Faculty of Economics and Business is responsible solely for the supervision of the completion of the work, not for the contents.
Table of Contents
1. Abstract ... 3
2. Introduction ... 4
3. Theoretical Background ... 7
3.1. Workplace gossip ... 7
3.2. Leadership and employee gossiping behavior ... 10
3.3. Leadership and procedural justice ... 13
3.4. The mediating role of procedural justice ... 15
3.5. Work stress ... 17
3.6. Conceptual model ... 21
4. Research Design ... 22
4.1. Sample and procedure ... 22
4.2. Measures ... 23
4.2.1. Ethical leadership ... 23
4.2.2. Employee gossiping behavior ... 23
4.2.2.1. Scale development ... 24 4.2.3. Procedural justice ... 25 4.2.4. Work stress ... 26 4.2.5. Control variables ... 26 5. Results ... 26 5.1. Correlations ... 26 5.2. Hypothesis testing ... 27
6. Discussion and Conclusion ... 35
6.1. Discussion ... 35
6.2. Limitations ... 39
6.3. Future research directions ... 41
6.4. Managerial implications ... 42
6.5. Conclusion ... 43
7. References ... 45
1. Abstract
While a large amount of studies have been done on leadership and individual behaviors, little is known about the relationship between leadership behaviors and employee gossiping behavior. In this quantitative, cross-sectional study, I investigated a moderated mediation model proposing the mediating role of procedural leadership in the relationship of ethical leadership with employee gossiping behavior and the moderating role of work stress in the relationship of procedural justice and employee gossiping behavior. The total sample of this study included 121 employees. In line with expectations, this study indicates that ethical leadership is positively related to procedural justice and negatively related to employee gossiping behavior. Also, as expected, work stress moderated the relationship between procedural justice and employee gossiping behavior such that this relationship was negative when work stress was low, but non-significant when work stress was high. In contrast to expectations, procedural justice did not mediate the relationship between ethical leadership and employee gossiping behavior. This paper concludes with limitations, practical implications and suggestions for future research.
Keywords: Employee gossiping behavior, ethical leadership, procedural justice, organizational justice, work stress.
2. Introduction
Gossiping is an extremely complex activity that is widely practiced in daily life and in organizations. Traditionally, gossip is seen as improper, unethical, overly subjective and has been defined as “idle chatter, chitchat, or the evil tongue” (Kurland & Pelled, 2000, p. 429). Private conversations and conversations at work are often dominated by social topics. According to a research done by Dunbar (2004) about the content of everyday conversation, gossiping takes approximately 65% of our speaking time. This may not be a surprise as we read magazines and newspaper gossip columns, talk about friends, relatives and acquaintances during our leisure time and love to talk about our bosses and colleagues at work. Especially within organizations, gossiping is one of the most rampant and pervasive activities we engage in. Earlier research indicated that more than half of the general conversation between employees at work is actually gossip and thus is related to social topics concerning talk about other people (Cole & Dalton, 2009). Also, in general gossip provides a channel of informal communication and information exchange about an absent third party, and can be either positive or negative (Grosser, Lopez-Kidwell, & Labianca, 2010; Noon & Delbridge, 1993).
From an organizational perspective, gossiping is generally known as a dysfunctional activity that needs to be controlled by top management because of its negative consequences for organizations, as gossip may lead to a decrease in productivity and can lower morale within an organization (Baker & Jones, 1996). However, from an employee perspective gossiping does have several beneficial consequences (Michelson & Mouly, 2004). An example of a positive function of gossiping for employees is stress reduction at work (Michelson & Mouly, 2004). Therefore, the present study will both take into account the negative consequences of gossiping for organizations and its managers and the positive consequences of gossiping for employees.
During the past decade, there have been a considerable number of studies that have helped researchers to better understand the role of gossip within organizations (Decoster et al., 2013; Elwardt, Labianca, & Wittek, 2012; Ellwardt, Wittek, & Wielers, 2012; Grosser, Lopez-Kidwell, Labianca, & Ellwardt, 2012; Kuo et al., 2015; Sommerfeld, Krambeck, & Milinski, 2008). However, none of them looked at the role of positive leadership behavior in understanding the organizational perspective. Ethical aspects in organizations are becoming increasingly important for leaders and research shows that ethical leadership can raise the ethical awareness in organizations and stimulate the ethical behavior of followers, thus possibly reducing gossip (Den Hartog, 2015; Ellwardt et al., 2012). For this reason, in the present paper I study ethical leadership in relation to gossiping with organizations.
Ethical leadership behavior is primarily concerned with the procedural aspects of fair decision-making (Brown, Treviño, & Harrison, 2005), indicating that procedural justice may play an important role. Earlier research has shown that justice within organizations positively affects organizational behaviors like organizational citizenship behaviors and rule compliance, suggesting that procedural justice is likely to increase positive organizational behaviors and decrease negative organizational behaviors like employee gossiping behavior (Wong, Ngo, & Wong, 2006). In the present study, procedural justice will be investigated as a mediator between ethical leadership and employee gossiping within organizations.
Even though it may be likely that ethical leadership and procedural justice within organizations will decrease the likelihood of employees participating in gossiping behavior, taking into account the positive effects of gossiping for employees individually may suppress this effect. One of the positive effects of gossiping for employees is stress reduction in times when significant work stress is experienced (Grosser et al., 2012). In times of high stress at work, employees will find it difficult to experience the benefits of an ethical leader and procedural justice at work. Work stress is likely to make employees lose their ability to
experience the positive effects of a work environment high in procedural justice (Summers, DeNisi and DeCotiis, 1989). This suggests a moderating role of work stress in the relationship between procedural justice and gossiping. To date, only little is understood about the beneficial consequence of gossiping for employees. Therefore, this study will also take into account the positive effects of gossiping for employees by using the concept of work stress.
As gossip is a well-known and well-practiced activity at work, it is important to understand the role of gossip within organizations. Understanding the patterns of gossiping at work offers a key to understanding organizational processes. From an organizational perspective, it is important to understand how leaders can influence and manage the gossiping behavior of their employees. It is also important for leaders to be aware of potential benefits of gossiping for their employees. This study extends the previous literature by focusing on both the organizational perspective and the employee perspective on gossiping.
The present research aims to answer the following research questions: Does ethical leadership influences employee gossiping behavior through procedural justice? How is work stress able to influence the relationship between procedural justice and gossip? In sum, I propose that ethical leadership is able to influence employee gossiping behavior through procedural justice, and that the relationship between procedural justice and gossiping is moderated by work stress. In the next section, I will discuss this rational in greater detail.
3. Theoretical Background
3.1. Workplace gossip
Workplace gossip is a well-known phenomenon, which we all engage in. Despite gossip being such a well-known phenomenon at work it remains an issue, which is deeply complex and not yet fully understood in scientific literature. In the literature, there is no complete agreement on a definition of workplace gossip. Early research about gossip treated gossip as something improper, unethical and overly subjective and traditionally defined it as “idle chatter, chitchat or the evil tongue” (Kurland & Pelled, 2000, p. 429). This definition has a negative connotation, which probably arises from the fact that gossip is often unreliable and seen as socially undesirable (Decoster et al., 2013; Grosser et al., 2012). A more neutral definition of gossip is the definition described by Nevo et al. (1993, p. 975) who described gossip as: “verbal or written communication that regards personal matters of a third party”. In the present paper, I draw upon one of the neutral conceptualizations by defining workplace gossip as “informal and evaluative talk in an organization about another member of that organization who is not present” (Kurland and Pelled, 2000, p. 429). This definition allows for both the functional and dysfunctional side of gossip.
Within organizations, communication usually flows through both the formal and informal network. The formal network refers to the way of communicating official information provided by the organization. In the formal network, the organization has power and control over the frequency, content, form and directional flow of the information communicated to its employees. In contrast, the informal network refers to channels of communication that emerge spontaneously and cannot be controlled easily by top management. The informal network is often called the grapevine and this informal channel of communication may either support or undermine the official channel of communication. The
informal network includes several informal communication systems among which workplace gossip is one of them (Michelson & Mouly, 2004; Kurland & Pelled, 2000).
In organizational literature, gossip is often seen as an activity that should be abolished by managers as it is destructive for organizations because it can lower an organization’s credibility, undermine productivity and lower morale within an organization (Baker & Jones, 1996; Difonzo et al., 1994; Decoster et al., 2013). Gossiping is seen as unethical and destructive work behavior because gossiping about one’s supervisor is seen as supervisor-directed deviance, which is not appreciated in any organization. Gossiping can also be interpreted as interpersonal deviance, when gossiping is directed towards colleagues or peers (Decoster et al., 2013). Employees who engage in gossiping behavior are said not to work as hard as they could and are not as productive as they could be for the organization. In contrast to the formal communication network within organizations, within the informal network leaders do not have power and control over the information communicated within the informal system, for example through gossiping. In addition, research showed that leaders in organizations see employee gossiping behavior as an activity that undermines their authority and threatens their position (Davis-Blake, 2000). For example, Ogasawara (1988) showed that women in Japanese organizations were able to gain control over their work environment and their formal supervisors by engaging in gossiping behavior, which could ruin their supervisors’ reputation and their supervisors’ chances for promotion. Michelson and Mouly (2004) also noted that gossiping in organizations may lead to a fear of loss of control by top management and may ruin reputations. Despite these negative consequences of employee gossiping for organizations and leaders, gossiping does have several beneficial consequences for employees in particular.
Michelson and Mouly (2004) showed that gossip indeed provides several positive consequences for employees within organizations. In their paper, they pointed out that not all
of the outcomes and consequences of gossip within organizations are shown to be harmful within organizations. In fact, gossip helps to foster intimacy and can preserve group solidarity at work, gossip reinforces the interaction between, and social bonds of, the employees and gossip reduces stress within the workplace. Furthermore, gossip can also foster cohesion between people and serves as a function to seek for information (Michelson & Mouly, 2004). Grosser et al. (2012) provide a more comprehensive explanation of the several positive functions of workplace gossip for employees. One of the beneficial functions of gossip in organizations that they explain is that gossip can provide informal information to employees which otherwise might not have been available through formal communication channels in the organization. Another beneficial function of gossip is that employees can use gossip as a way to cope with stressful situations by expressing their emotions, concerns and thoughts about these situations in the work environment. Gossiping can also intellectually stimulate employees because communicating informally serves as a way to keep employees minds active and provides employees with enjoyment. Gossiping is also found to foster cohesion and stimulate interpersonal relationships between employees because gossiping requires interpersonal trust and intimacy, which in turn strengthen bonds between employees (Elwardt et al., 2012; Grosser et al., 2012).
For the above stated reasons, from the perspective of leaders and top management within organizations it may be necessary and beneficial to know when and how they can limit the negative effects of gossiping on organizations. On the other hand, from an individual perspective, for employees gossiping may serve as way to cope with contextual stressors in the organization and may provide potential benefits.
3.2. Leadership and Employee Gossiping Behavior
The way in which leaders can influence employee gossiping behavior has not yet been studied thoroughly. There have been several studies that show how negative leadership behaviors may encourage their employees to gossip, which in turn has negative effects on the organizations. For example, Mitchell and Ambrose (2007) showed how abusive supervision can lead to workplace deviance and supervisor-directed interpersonal deviance, such as gossiping. Also, according to Kuo et al. (2015), abusive leadership behavior predicts organizational gossiping behavior of employees, which in turn predicts employee cynicism in the organization. However, there are no studies to be found that look at the way in which positive leadership behaviors can influence employee gossiping behaviors. As stated earlier, leaders may want to be able to prevent gossiping behavior of employees because it could undermine their authority and lead to a loss of control. Therefore, from an organizational perspective it might be interesting to see if and how a positive leadership style can influence employees gossiping behavior. In this study ethical leadership is chosen as a positive leadership style to examine whether ethical leadership behaviors can influence the gossiping behavior of employees.
Ethical leadership is a leadership style that became of increasing importance in the scientific literature in the past decade. Due to national and international corporate scandals, ethical leadership became more important not only in the scientific literature, but also among governments and practitioners. Ethical leadership is a broad construct, which includes many different behaviors. In their study, Brown, Treviño and Harrison (2005) define ethical leadership as: “the demonstration of normatively appropriate conduct through personal actions and interpersonal relationships, and the promotion of such conduct to followers through two-way communication, reinforcement, and decision-making” (p. 120). Ethical leaders are commonly seen as fair, principled decision-makers who promote their ethical
conduct to their employees by being considerate, open and trustworthy, resulting in a whole range of positive outcomes which include employee ethical decision-making, satisfaction and commitment (Brown & Treviño, 2006; Den Hartog, 2015). Den Hartog (2015) provides an elaborate review on ethical leadership, summarizing the ethical leadership research to date. In this review it becomes clear that (perceived) ethical leader behaviors are related to an extensive range of outcomes, including ethical norms and decisions, identification-based motivation like engagement, relational and social information like trust and psychological safety and obligation in terms of responsibility. Further outcomes of ethical leader behaviors include effort, performance and organizational citizenship behaviors. Trust is an important factor within the leader employee-relationship as trust is shown to cause job satisfaction, commitment and organizational citizenship behaviors (Cropanzano & Mitchell, 2005). Ng and Feldman (2015) also found that ethical leadership predicts organizational citizenship behavior. Organizational citizenship behavior is defined as “contributions to the maintenance and enhancement of the social and psychological context that support task performance” (Organ, 1997, p. 91). Scientific literature has highlighted the importance of ethics for leaders in organizations. For example, Kalshoven, Den Hartog and de Hoogh (2013) showed the importance of ethical leadership on employees’ helping and initiative. Also, Demirtas (2015) showed that ethical leadership can lead to enhanced work engagement and a decrease in organizational misbehavior.
It is clear that the literature on ethical leadership is extensive and that it studied a whole range of different attitudinal, behavioral, ethical and performance outcomes. However, to date no research studied the effects of ethical leadership on gossiping behavior. The present study will investigate the relation between ethical leadership and gossiping behavior by using a social learning perspective (Bandura, 1977), social exchange theory (Blau, 1964) and organizational justice to explain how ethical leadership influences gossiping behavior.
Social learning theory proposes that people learn much of their behavior by observation, modeling and imitation of others. Social learning depicts the interaction of an individual’s knowledge and experiences, the environment (e.g. the leader) and the individual’s behavior (Bandura, 1977). According to this theory, an individual first notices something in the environment (e.g. the ethical behavior of the leader), then remembers what was noticed, produces a behavior and finally the environment delivers a consequence (e.g. a reward or punishment) that can change the probability that the behavior will appear again. Modeling in the workplace promotes social learning via an interaction between leader and employee in which neither party may be consciously aware of the exchange (Hanna, Crittenden, & Crittenden, 2013; Brown et al., 2005). Brown et al. (2005) state that: “A social learning perspective on ethical leadership proposes that leaders influence the ethical conduct of followers via modeling” (p. 119). When confronted with an ethical leader, employees thus react to their leaders behavior by imitation, modeling, identification and observational learning (Brown et al., 2005).
The social exchange theory is based on reciprocal exchange of social behavior (Blau, 1964). This theory further explains the rational described above by stating that people are aware and sensitive of the valued outcomes they receive and that they are motivated to reciprocate the outcomes they receive. This theory is based on the norm of reciprocity, which proposes that if one individual does something beneficial for the other individual, that behavior then generates an obligation to reciprocate with good faith behavior. One of the basic principles of the social exchange theory is that because of this norm of reciprocity, relationships evolve over time intro trusting, loyal and mutual commitments (Cropanzano & Mitchell, 2005). According to this theory, positive social behavior thus is the result of an exchange process with the intent to maximize the benefits and minimize the costs. Specifically within organizations, because individuals return the benefits they receive, they
are likely to reciprocate helpfulness and trust towards the person with whom they have a positive social exchange relationship (Masterson, Lewis, Goldman, & Taylor, 2000). Ethical leaders are expected to influence the behaviors of their employees through socio-emotional exchange (i.e., exchange based on the interpersonal treatment) because ethical leaders are likely to engender high levels of trust and fairness.
As stated before, ethical leaders are fair, trustworthy, open and considerate principled-decision-makers. With this behavior and attitudes a leader signals the importance of ethical behavior towards his/her employees. Based on social learning theory and social exchange theory then, employees who interact with their ethical leader notice the fair and trustworthy behavior of their leader and in turn are motivated to behave in compliance with the ethical conduct by engaging in fair and trustworthy behavior themselves. Employees thus will reciprocate with the same positive behaviors when treated ethically and fair by their leader (Den Hartog, 2015). Gossip is generally seen as improper, unethical and able to lower morale within organizations. By signalling the importance of ethical behavior as a leader, and thereby showing that unfair and unethical behavior like gossiping is not encouraged, employees are less likely to engage in gossiping behavior. Thus:
Hypothesis 1: Ethical leadership is negatively related to employee gossiping behavior.
3.3. Leadership and procedural justice
The moral behavior of ethical leaders focuses mainly on evaluations of decisions in terms of their moral rightness, justice and fairness (Den Hartog, 2015). Therefore, it is likely that organizational justice may also play a role in the relationship between ethical leaders and gossiping behavior. Organizational justice refers to the perceptions of fairness in organizations (Greenberg, 1987). Together with ethics, justice also became increasingly
important in organizations and scientific literature. Organizational justice affects employees’ actions and behavior within organizations. Justice within organizations is important because employees simply care really deep about the way they are treated in their organization (Demirtas, 2015).
People within organizations generally evaluate fairness based on three different types of justice: distributive, procedural and interactional justice (Colquitt, 2001). Distributive justice refers to the fair distribution of desired outcomes and the outcomes that people receive from the organization, procedural justice refers to the justice of the processes that lead to the desired outcomes and interactional justice refers to way individuals are treated during the interpersonal exchanges at work (Colquitt, 2001). As distributive justice focuses mainly on the outcomes in organizations and providing consequences for unwanted behavior instead of the processes and treatment in an organization, this research will not focus on distributive justice. Earlier research already showed the positive relationship between ethical leadership and organizational justice (Demirtas, 2015). Neubert et al. (2009) also indicated that interactional justice can strengthen the relationship between ethical leadership and an ethical climate. However, as ethical leadership behavior is primarily concerned with the procedural justice in terms of listening and fair decision-making (Brown, Treviño, & Harrison, 2005), this research will specifically focus on procedural justice.
According to Leventhal (1980) people assess procedures in terms of six different dimensions. These dimensions are consistency (e.g. the consistent application of procedures across individuals), correctability (e.g. the possibility to correct decisions), ethicality or morality, information accuracy, lack of bias and the opportunity for all individuals to participate. When people assess decisions or procedures as high on these dimensions, they would be considered fair. Ethics are important for procedural justice because ethicality refers to a fair process that does not violate employees’ personal standards and morality within the
organization (Leventhal, 1976). Ethicality has been found to be particularly important for procedural justice (Tyler, 1988).
Leaders are able to influence organizational justice by the organizational rules and practices they apply (Colquitt, 2001). As stated before, ethical leaders are seen as fair, principled decision-makers who promote their ethical conduct to their employees. They act with fairness, respect and integrity and are also perceived as open and trustworthy by their employees (Den Hartog, 2015; Mayer et al., 2012). As employees generally perceive ethical leaders as fair and principled decision-makers, they will also trust their leader in being fair when it comes to the processes that lead to the desired outcomes. Also, because ethicality is important in the determination of procedural justice (Tyler, 1988), it is likely that in organizations with an ethical decision-making climate (e.g. organizations with an ethical leader), perceptions of procedural justice will be present.
Ethical leaders signal the importance of procedural fairness. Consequently, ethical leaders are expected to be able to influence procedural justice by their fair decision-making process. In sum, I expect:
Hypothesis 2: Ethical leadership is positively related to procedural justice.
3.4. The mediating role of procedural justice
Organizational justice has been shown to positively affect organizational attitudes like job satisfaction, trust in the organization, trust in supervisors and organizational behaviors like organizational citizenship behaviors, job performance and rule compliance (Colquitt, 2001; Wong, Ngo, & Wong, 2006; Crawshaw et al., 2013). To date, there has been quite some research showing that employees’ attitudes and behaviors at work are influenced by their perceptions of procedural justice within organizations suggesting that procedural justice is associated with positive employee attitudes and behaviors. For example, Lee (2006) found
that procedural justice is directly positively associated with organizational commitment, Loi, Hang-Yue and Foley (2006) show that procedural justice is negatively associated with turnover intention, Ma, Liu and Liu (2014) were able to relate procedural justice with extrarole behavior, Searle et al. (2011) found that procedural justice is positively related to employees’ trust in their employer and Tepper and Taylor (2003) show that procedural justice is positively related to organizational citizenship behavior.
As stated earlier, social exchange theory explains that individuals are motivated to reciprocate the outcomes they receive. This also fits with the fair process effect, which explains that satisfaction with the procedure (e.g. voice in decisions) results in greater satisfaction (Folger, Rosenfield, Grove, & Corkran, 1979). More specifically, Brockner et al. (2007) found that people reciprocate more favorably to others behavior and situations when they experience procedural justice. Social exchange theory and the fair process effect suggest that when organizations show greater degrees of procedural justice, employees develop a need to reciprocate to their organization in a positive manner because they are satisfied with the fair procedure. They will do this by showing more favorable behavior (e.g. organizational citizenship behavior or extrarole behavior), making it less likely that they will engage in gossiping behavior as this is generally seen as behavior which is encouraged by an unfair environment (Dineen, 2006; Mitchell & Ambrose, 2007). Thus, when employees experience a work environment in which procedural justice is highly valued, they are motivated to positively reciprocate to the fair treatment they receive by the organization and their leader, resulting in less deviant behavior like gossiping. Procedural justice thus discourages employees to engage in gossiping behavior.
Building on social exchange theory and the fair process effect, I expect that employees who experience a high degree of procedural justice will be less inclined to gossip because
they are motivated to positively reciprocate to their fair organization and the respectful treatment they receive.
Hypothesis 3: Procedural justice is negatively related to gossiping behavior of employees.
Furthermore, as ethical leadership is primarily concerned with the procedural aspects of fair decision-making (Brown, Treviño, & Harrison, 2005), this signals the significant and present role of procedural justice within ethical leadership. Earlier research also suggested that when employees have an ethical leader, they tend to judge acts of deviance (e.g. gossiping) as morally inequitable and unfair (Resick et al., 2013). This thus suggests that ethical leaders promote a work environment in which procedural justice is highly important, which in turn influences the deviant behavior of employees (e.g. less gossiping). Therefore, in the present study I propose that organizational justice may play a mediating role between ethical leadership and gossiping.
Hypothesis 4: Organizational justice mediates the relationship between ethical leadership and gossiping behavior of employees.
3.5. Work stress
Despite gossip having mostly negative consequences from an organizational perspective, gossiping may be of serious importance from an employee perspective as it has several beneficial functions for employees in organizations. Even though leaders might be able to prevent their employees from gossiping by showing ethical behaviors and promoting organizational justice, employees may still be inclined to gossip in certain circumstances because of its beneficial functions. The present study will therefore also take into account a
contextual variable, work stress, which may impact the relationship between organizational justice and gossiping behavior.
Work stress is defined as an individual emotional experience that is related to strain and anxiety resulting from a job or work situation (Cooke & Rousseau, 1984). Work stress occurs when work pressure is high. When there is a certain amount of workload that has to be accomplished in a specific period of time, work pressure increases and this can lead to enhanced feelings of stress at work (MacDonald, 2011; Reichenberg & MacCabe, 2007). Central to work pressure is the lack of fit between task demands and an employee’s ability to cope with these task demands, which causes stress (MacDonald, 2011). Omar et al. (2011) looked at the relationship between work stress and deviant work behavior and found that employees who experience negative emotions due to stress at work are more likely to show deviant behaviors. Additionally, more researchers have shown that stress at work is able to result in deviant work behavior (Radzali, Ahmad, & Omar, 2013; Sulksky & Smith, 2005).
As stated earlier, one of the beneficial functions of gossip is that employees can use it as a way to cope with stressful situations by expressing their emotions and concerns about the situations in the work environment (Grosser et al., 2012). Waddington and Fletcher (2005) found that gossip is able to play a role as a safety gate by providing a useful way for stress relief and emotional support in stressful situations. Gossiping at work about an especially tense relationship, a difficult person or a stressful work situation can serve as an emotional exhaust for that specific person. Gossip can thus serve as way to vent or let off some steam, which in turn can help to reduce stress, worrying and feelings of anxiety (Grosser et al., 2012).
Gossip is generally seen as deviant behavior (Dineen, 2006) and has the ability to serve as a way to reduce stress for employees. Therefore, I expect employees who experience work stress to be more inclined to gossip.
Hypothesis 5: Work stress is positively related to employee gossiping behavior.
As stated earlier, ethical leadership leads employees to experience their work environment as fair and leads to an environment in which procedural justice is highly valued. This fair environment is in turn expected to prevent employees from gossiping. However, research has shown that experienced work stress is also related to organizational justice and can lead to several attitudinal consequences.
Choi, Moon, Nae and Ko (2013) found that organizational justice was negatively related to work stress in South Korean organizations. This negative relationship between organizational justice and work stress was also found by Tziner and Sharoni (2014). With their study, Summers, DeNisi and DeCotiis (1989) showed that work stress leads to a decrease in extrinsic job satisfaction. Extrinsic job satisfaction refers to satisfaction with job conditions, such as pay, coworkers, manager and the work environment. They also showed that work stress leads to a decrease in intrinsic job satisfaction. This refers to the satisfaction with intrinsic job characteristics such as challenge, recognition and the work itself. These results suggest that employees who experience work stress cope with negative feelings that lead them to perceive their whole work environment as more negative. In sum, employees who experience work stress will experience difficulties in perceiving their environment as a positive one.
As stated before, procedural justice is positively associated with a whole range of positive individual and organizational work outcomes, including organizational commitment, extrarole behavior, trust in management, organizational citizenship behavior and job satisfaction (Lee, 2006; Ma, Liu & Liu, 2014; Searle et al., 2011; Tepper & Taylor, 2003). Employees working in an environment high in procedural justice are thus more likely to perceive their work environment as a positive one and are more likely to be satisfied with
their job and show behavior that is positive for their organization. However, based on the research by Summers, DeNisi and Decotiis (1989) and earlier research on the relationship between organizational justice and work stress, it is likely to expect that when employees, despite having a work environment high in procedural justice, still experience enhanced feelings of work stress, they won’t be able to benefit from the positive work environment that procedural justice creates. Under high stress, working in an environment high in procedural justice will not result in the usual positive outcomes of procedural justice like job satisfaction, organizational citizenship behavior and the feeling to reciprocate to the organization. Instead, employees experiencing high work stress will experience a decrease in both extrinsic and intrinsic job satisfaction (Summers, DeNisi & Decotiis, 1989). Work stress is likely to suppress the positive outcomes of procedural justice in organizations. The decrease in extrinsic and intrinsic job satisfaction then will not result in the feeling of reciprocating to the organization but it will lead them to experience the consequences of work stress.
As stated before, when employees are confronted with high work stress they will be inclined to gossip as this is a way to express their emotions and relief some stress. In times of stress, the positive outcomes of procedural justice for the employee and the organization are not likely to emerge because work stress will lead employees to feel less satisfied with their whole organization, despite the environment high in procedural justice. To cope with the enhanced feelings of stress at work, employees will then be more inclined to gossip. This suggests a moderating role of work stress in the relationship between organizational justice and gossiping.
More specifically, if work stress is experienced, employees are expected to be inclined to gossip as a way to cope with the stress because they are unable to experience any positive outcomes of procedural justice, which then thus will not result in any feelings of reciprocating positively to their organization. A work environment high in procedural justice will thus not
prevent employees from gossiping when they experience work stress. In contrast, when work stress is low, employees are less inclined to gossip because they are able to experience the positive outcomes of procedural justice, leading them to reciprocate positively to their fair organization. In this situation, procedural justice serves as a mean to prevent employees from gossiping. Based on this rationale, in this study work stress will be taken into account as a moderator between procedural justice and gossiping behavior. I hypothesize:
Hypothesis 6: Work stress moderates the relationship between procedural justice and gossiping, such that the relationship between procedural justice and gossiping is weaker when work stress is high.
Hypothesis 7: The indirect effect of ethical leadership on gossiping through procedural justice is weaker when work stress is high.
3.6. Conceptual model
To consolidate the five hypotheses and clarify the associations between the research variables, a theoretical research model has been developed (see Figure 1).
4. Research Design
4.1. Sample and Procedure
Data for the present study was requested from 174 supervisor-employee dyads working in a number of different organizations in the Netherlands, Germany, Greece, Spain, Belgium and Luxembourg. The study was administered in both Dutch and English. Data were collected through self-administered computer-based questionnaires. The final amount of completed questionnaires resulted in 99 dyads of employees and their matched direct supervisors, yielding an overall response rate of 56,9 %. However, for the present study only the employee surveys were necessary. In total, 144 employee questionnaires were returned. A number of 23 questionnaires were incomplete and were removed from the dataset. The final amount of completed employee questionnaires were 121, yielding an overall response rate of 69,5%. Of the employees’ sample, 53 were male and 68 were female. The mean age was 35 years (SD = 1). Of all the participating employees, 5.8 % only completed high school, and 94,2 % obtained at least a college degree. Employees had an average job tenure of 10 years (SD = 1).
We recruited participants for this study using two different non-probability sampling techniques. First, we recruited participants using a snowball sampling procedure in which, after the first sample participant, subsequent supervisor-employee dyads were identified by earlier sample participants (Saunders & Lewis, 2012). Employees and supervisors from different medium to large companies were thus contacted and were asked to invite other employees and their corresponding supervisors to complete the questionnaire. As it was difficult to obtain data, we also used convenience sampling to select people who were easy to get hold of by asking random people in our personal and professional network who worked at least 20 hours a week to participate in our study.
All participants (both leader and employee) received an information email and a link to the online questionnaire. The email explained the importance and purpose of the present study, explained the voluntary nature of participation, assured confidentiality and described what was expected of participants when filling in the questionnaire. As the present study was a dyad study, which means that manager and employee were linked to each other, both the manager and the employee were asked to agree upon a code consisting of at least 6 letters and 2 numbers and both filled in the very same code when completing the questionnaire.
The questionnaire was part of a master thesis project of six students in Business Administration of the University of Amsterdam. Therefore, several different constructs were measured in both questionnaires that were not used in the present study. For the present study, ethical leadership, gossiping behavior, procedural justice and work stress were all rated by the employee. Thus, for the present study only employee data was used.
4.2. Measures
4.2.1. Ethical leadership.
Ethical leadership was measured by using the scale developed by Brown et al. (2005). The scale consists of ten items and the Cronbach’s alpha (α) for this scale was 0.93. An example item is: ‘My supervisor makes fair and balanced decisions’. Responses were recorded using a seven-point Likert scale (1= strongly disagree, 7 = strongly agree). Higher scores represent a higher degree of ethical leadership.
4.2.2. Employee gossiping behavior.
For this survey, a new scale consisting of 9 items was developed to measure gossiping behavior of employees at work. Prior gossip studies were revisited to develop a new workplace gossip scale. Gossip was measured using one gossip item from the
counterproductive work behavior scale adopted from Cohen-Charash and Yochi (2007), one gossip item from the social desirable response tendency scale adopted from Steenkamp, Jong and Baumgartner (2010), one item from the counterproductive work behavior scale adopted from Bing (2007), one item from the deviant work behavior scale adopted from Dineen (2006) and five items from the gossip scale adopted from Wittek and Wielers (1998). In this scale, both gossiping about colleagues and gossiping about supervisors was measured from the perspective of the employee. Cronbach’s alpha for this scale was 0.91. An example item is: ‘I participate in gossip about colleagues that I don’t really like’. Another example item is: ‘I talk to others about the bad nature of my manager’. Responses were recorded using a seven-point Likert scale (1= strongly disagree, 7 = strongly agree). Higher scores represented a higher frequency of certain gossip participation.
4.2.2.1. Scale development
For the present study, a new gossip scale was developed to measure gossiping behavior of employees at work. This scale was developed by using several gossip items from different scales, as mentioned above. As I developed a new scale that was not validated, I could not say with complete certainty that this scale would measure gossiping at work perfectly. To test whether the developed gossip scale only extracted one factor, a principal axis factor analysis was conducted on the 9 gossip items with oblique rotation. The Kaiser-Meyer-Olkin measure verified the sampling adequacy for the analysis, KMO = .88 (‘meritorious’ according to Hutcheson & Sofroniou, 1999). An initial analysis was run to obtain eigenvalues for each factor in the data. Two factors had eigenvalues over Kaiser’s criterion of 1 and in combination explained 69,99 % of the variance. The items that cluster on the same factor suggest that factor 1 represents gossiping behavior. However, factor 2 represents no clear subscale. When analyzing the factor matrix, all three counter indicative
items seem to be the reason for the second extracted factor. This suggests that a certain amount of the participants did not interpret the counter indicative items correctly. Therefore I decided to eliminate a total of three items because they did not contribute to a simple factor structure. Thus, all three indicative items: ‘I do not gossip about my supervisor’, ‘I do not gossip about other people’s business’ and ‘I do not talk about my colleagues behind their backs’ were eliminated from the gossip scale.
A principal axis factor analysis was conducted on the 6 remaining gossip items with oblique rotation. The Kaiser-Meyer-Olkin measure verified the sampling adequacy for the analysis, KMO = .87 (‘meritorious’ according to Hutcheson & Sofroniou, 1999), and all KMO values for individual items were greater than .82, which is well above the acceptable limit of .5 (Field, 2013). Bartlett’s test of sphericity, x² (15) = 459,89, p < .001, indicated that correlations between items were sufficiently large for PAF. An initial analysis was run to obtain eigenvalues for each component in the data. One factor had eigenvalues above Kaiser’s criterion of 1 and explained 69,07 % of the variance. The 6 items that cluster on this factor suggest that this scale represents a measure of gossiping at work. This gossip scale had a high reliability, Cronbach’s α = .91.
4.2.3. Procedural justice.
Procedural justice was measured using a five-item scale adopted from Colquitt (2001). An example item is: ‘My supervisor communicates open and honestly’. Responses were recorded on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree). The coefficient alpha (α) fort his five-item measure was .91.
4.2.4. Work stress.
Work stress was measured by the four-item scale adopted from Motowidlo, Packard and Manning (1986). An example item is: ‘I feel a great deal of stress because of my job’. Responses were recorded on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree). The sum of the four items, with appropriate reversals, made up the work stress score. Cronbach’s alpha for this scale was 0.82.
4.2.5. Control variables.
Based on previous research, I controlled for several factors that have been shown to be related to one or more of the variables that are incorporated in my hypothesis. One factor that I controlled for is age, as Acquino and Douglas (2003) showed that age is related to deviant behavior, like gossip, in organizations. Henle, Giacalone and Jurkiewicz (2005) also showed that gender is related to interpersonal deviance in organizations. Therefore, I also controlled for employees’ gender. Tenure with the organization is another factor I controlled for in the present study because employees with less tenure are more likely to show deviant behavior and this may influence employee gossiping behavior (Ellwardt et al., 2012; Hollinger, Slora, & Terris, 1992). Finally, I controlled for in the present study is language as the survey was administered in both Dutch and English.
5. Results
5.1. Correlations
Table 1 shows the means, standard deviations, and correlations among the study variables. Consistent with our expectations, ethical leadership was negatively and significantly correlated with gossiping behavior, r = -.42, p < .01 and procedural justice was also negatively and significantly correlated with gossiping behavior, r = -.32, p < .01. Ethical leadership was positively and significantly related to procedural justice, r = .78, p < .01. Also,
stress was negatively correlated with procedural justice, r = -.19, p < .05, and positively correlated with gossiping behavior, r = .28, p < .01.
Table 1
Means, Standard Deviations, and Intercorrelations Between the Scales
Variables M SD 1 2 3 4 5 6 7 8 1. Gender 0.44 .50 - 2. Age 35.34 11.75 .06 - 3. Tenure 10.84 11.72 .08 .83** - 4. Language 0.44 .50 -.18 -.02 -.12 - 5. Ethical Leadership 5.29 1.06 .07 .02 .11 -.23* (.93) 6. Procedural Justice 5.33 1.14 -.01 -.02 .10 -.14 .78** (.91) 7. Gossip 3.42 1.49 -.10 -.12 -.19* -.16 -.42** -.32** (.91) 8. Stress 3.84 1.26 -.17 -.16 -.21* .30** -.18 -.19* .28** (.82) Note. N = 121. * p ≤ .05; ** p ≤ .01. Gender was coded as 0 = female, 1 = male. Language was coded as 0 = Dutch, 1 = English.
5.2. Hypothesis Testing
To test my hypothesis, the Process macro written by Andrew F. Hayes for SPSS was used. Hypothesis 1, 2, 3 and 4 were tested using Model 4 in Process. This model tested a mediation model, with ethical leadership as the independent variable (X), employee gossiping behavior as the outcome variable (Y) and procedural justice as the mediator (M). Gender, age, tenure and language were taken into account as control variables. Results are displayed in Table 2. The total effect of ethical leadership on gossiping behavior when procedural justice is not included in the model is c = -.68, meaning that ethical leadership is negatively related to employee gossiping behavior, b = -.68, t = -5.87, p < .001. (path c, Figure 2). This is in line with Hypothesis 1. The direct effect of .86 indicates that ethical leadership is positively
related to procedural justice, b = .86, t = 13.13, p < .001, R² = .626 (path a, Figure 3). This is support for Hypothesis 2, ethical leadership is positively related to procedural justice. In contrary to Hypothesis 3, procedural justice is not significantly related to gossiping behavior, b = .16, t = .92, p = .36, R² = .322 (path b, Figure 3). Therefore we cannot conclude that procedural justice is negatively related to employee gossiping behavior. When including procedural justice as a predictor as well, the direct effect of ethical leadership on gossiping, c’ = -.81, predicts that when employees perceive their leader as ethical, they are less inclined to gossip. Again, this is in line with Hypothesis 1. Because procedural justice does not explain significant variance in employee gossiping behavior, ethical leadership still predicts gossiping significantly when procedural justice is included in the model, b = -.81, t = -4.37, p < .001, R² = .322 (path c’, Figure 3). Finally, as the indirect effect is not statistically different from zero when looking at the 95 % BC interval, ethical leadership does not have a significant indirect effect on gossiping through procedural justice, b = .13, BCa CI [-.1246, .3559] (path c’, Figure 3). In sum, there is no significant indirect effect of ethical leadership on gossiping through procedural justice. Therefore, Hypothesis 4 is not supported.
Table 2
Consequent of the mediation analysis
M (ProJus) Y (Gossip)
Antecedent Coeff. SE p Coeff. SE p
Gender -.08 .14 .59 -.32 .25 .19 Age -.01 .01 .25 -.00 .02 .97 Tenure .01 .01 .25 -.02 .02 .27 Language .15 .14 .29 -1.04 .25 <.001 X (EthLead) a .86 .07 < .001 c’ -.81 .19 < .001 M (ProJus) - - - b .16 .17 .36 constant i1 1.06 .47 .03 i2 7.72 .85 < .001 R² =.626 R² = .322 F (5,108) = 36.19, p < .001 F(6,107) = 8.47, p < .001 Note. N = 114. Gender was coded as 0 = female, 1 = male. Language was coded as 0 = Dutch, 1 = English.
To test Hypothesis 5, hierarchical multiple regression analysis was used to examine the relationship between work stress and employee gossiping behavior, after controlling for gender, age, tenure and language. In the first step of the hierarchical multiple regression, employee gossiping was entered as a dependent variable together with the control variables gender, age, tenure and language. This model was statistically significant, F (4,109) = 2.97, p < .05, and explained 9.8 % of the variance in employee gossiping behavior. After entering work stress at Step 2 as a predictor, the total variance explained by the model as a whole was 18 %, F (5,108) = 4.77, p < .001. The introduction of work stress explained an additional 8 % variance in employee gossiping behavior, after controlling for age, gender and language (ΔR² = .08, F (1,108) = 10.89, p < .001. In the final model, two out of four predictor variables were statistically significant with work stress recording a higher Beta value (ß = .31, p < .001) than language (ß = -.33, p < .01). Results are displayed in Table 3. Even though work stress
only can account for approximately 8 % of the variation in employee gossiping behavior, the findings support Hypothesis 5. Work stress is positively related to employee gossiping behavior.
Table 3
Hierarchical Regression Model of Employee Gossiping Behavior.
R R² ΔR² B SE ß t Step 1 .31 .098* Gender -.38 .38 -.13 -1.37 Age .00 .02 .02 .13 Tenure -.03 .02 -.22 -1.36 Language -.74 .28 -.25** -2.65 Step 2 .43 .18*** .08*** Gender -.29 .27 -.09 -1.06 Age .01 .02 .06 .36 Tenure -.03 .02 -.19 -1.27 Language -.98 .28 -.33*** -3.54 Work Stress .37 .11 .31*** 3.30 Note. N = 114. * p ≤ .05; ** p ≤ .01; *** p ≤ .001.
Gender was coded as 0 = female, 1 = male. Language was coded as 0 = Dutch, 1 = English.
I then sought to examine whether work stress moderates the relationship between procedural justice and employee gossiping behavior. Therefore, Hypothesis 6 was tested using Model 1 in Process. This model tested a moderation model, with procedural justice as the independent variable (X), employee gossiping behavior as the outcome variable (Y) and work stress as the moderator (M). Gender, age, tenure and language were taken into account as control variables. The results are displayed in Table 4. The results support the predicted
CI [0.1477, 0.5489], t = 3.44 p < .001. Thus, this indicates that the relationship between procedural justice and gossiping is moderated by work stress. Moreover, this interaction effect accounts for 7.5 % of variance in gossiping behavior of employees. A closer inspection of the conditional effects indicates the relationship between procedural justice and gossiping is significant only when work stress is low, b = -.86, CI [-1.2104, -0.5095], t = -4.86, p < .001, or when work stress is moderate, b = -.42, CI [-0.6330, -0.2077], p < .001. The relationship between procedural justice and gossiping is not significant when work stress is high, b = -.02, CI [-0.2903, 0.3286], t = -.12, p = .90. These interactions were plotted in Figure 4. As expected, this figure suggests that employee gossiping behavior will decrease when procedural justice is high and when work stress is low. However, when work stress is high, the relationship between procedural justice and work stress is weaker and non-significant. These results provide support Hypothesis 6.
Table 4
Linear model of predictors of employee gossiping behavior
Coefficient SE t p
Intercept i1 12.90 2.49 5.19 < .001 Procedural Justice (X) b1 -1.77 .42 -4.22 < .001 Stress (M) b2 -1.59 .56 -2.83 .005 Procedural Justice x Stress (XM) b3 .35 .10 3.44 < .001
Gender -.43 .25 -1.74 .08 Age -.00 .02 -.16 .87 Tenure -.02 .02 -.93 .36 Language -1.12 .25 -4.39 <.001 R² = .33 F(7,106) = 7.57, p < .001
Note. N = 114. Gender was coded as 0 = female, 1 = male. Language was coded as 0 = Dutch, 1 = English.
To test my complete theoretical research model (Hypothesis 7), model 14 in Process was used. This model tested a moderated mediation model, with ethical leadership as the independent variable (X), employee gossiping behavior as the outcome variable (Y), procedural justice as the mediator (M) and work stress as the moderator (V) between procedural justice and employee gossiping behavior. Gender, age, tenure and language were taken into account as control variables. The results are displayed in Table 5 and indicate that the indirect effect of ethical leadership on employee gossiping behavior through procedural justice is moderated by work stress, b = .21, t = 2.02, p = .04. A closer inspection of the conditional indirect effects of ethical leadership on employee gossiping behaviors at different levels of work stress indicate that this indirect effect through procedural justice is only
significant when work stress is high, b = .31, CI [.0694, .5965]. The indirect effect of ethical leadership on employee gossiping behavior through procedural justice is not significant when work stress is moderate, b = .08, CI [-.2355, .3277], or when work stress is low, b = -.14, CI [-6374, .2336]. In contrast to Hypothesis 7, the indirect effect of ethical leadership on gossiping through procedural justice is only present when work stress is high. However, this indirect effect of ethical leadership on gossiping through procedural justice turned out to be a positive effect, instead of a negative effect, when high work stress was taken into account.
Table 5
Consequent of the moderated mediation analysis
M (ProJus) Y (Gossip)
Antecedent Coeff. SE p Coeff. SE p
Gender -.08 .14 .59 -.32 .24 .18 Age -.01 .01 .25 -.00 .02 .91 Tenure .01 .01 .26 -.02 .02 .28 Language .15 .14 .29 -1.27 .24 <.001 X (EthLead) a1 .86 .07 < .001 c’ -.69 .19 < .001 M (ProJus) - - - b1 -.71 .49 .15 V (Stress) - - - b2 -.81 .57 .16 MV (ProJus x Stress) - - - b3 .21 .10 .04 constant i1 1.06 .47 .03 i2 10.60 2.43 < .001 R² =.626 R² = .409 F (5,108) = 36.19, p < .001 F(8,105) = 9.07, p < .001
Work stress Unstandardized Boot SE Boot LLCI Boot ULCI
Conditional indirect effect at gossip for the levels of work stress
High .31 .13 .0694 .5965
Moderate .08 .14 -.2355 .3277
Low -.14 .21 -.6374 .2336
Note. N = 114. Gender was coded as 0 = female, 1 = male. Language was coded as 0 = Dutch, 1 = English.
6. Discussion & Conclusion
6.1 Discussion
Within organizations, employee gossiping behavior is generally known as an activity that provides negative consequences for organizations, for example to lower morale and lower productivity (Baker & Jones, 1996). Gossiping is therefore seen as a dysfunctional activity that needs to be controlled by top management. However, from the perspective of the employee, gossiping provides several positive consequences for employees within organizations like reducing stress. The aim of this research was to expand the literature on the relationship between ethical leadership and employee gossiping behavior at work. In the present study, procedural justice was taken into account as a mediator between ethical leadership and employee gossiping behavior. Furthermore, work stress was included as a moderator variable to study the role of work stress in the relationship between procedural justice and employee gossiping behavior. In sum, I proposed that ethical leadership is able to positively influence employee gossiping behavior through procedural justice, and that the relationship between procedural justice and gossiping is moderated by work pressure, such that procedural justice will not negatively influence employee gossiping behavior when work stress is high.
First, this paper extends the leadership literature by providing a better understanding of how a positive leadership style can influence employee behavior. Researchers to date have mostly emphasized how negative leadership behaviors, like abusive leadership, can lead to more organizational gossiping behavior of employees (Mitchell & Ambrose, 2007; Kuo et al., 2015). In the present study, it was found that ethical leadership is negatively related to employee gossiping behavior. This confirms my rationale, based on social learning theory and social exchange theory, that employees who have an ethical leader are motivated to behave in compliance with the ethical conduct by learning from the fair behavior of their leaders and by
reciprocating to their leaders with the same positive behaviors that they receive from them. Because gossiping is generally seen as improper and unethical (Kurland & Pelled, 2000), employees are less likely to engage in gossiping behavior when they interact with an ethical leader. The results thus suggest that a positive leadership style, like ethical leadership, is able to influence employee gossiping behavior as a result of social learning and social exchange within the organization.
Second, the results in this study showed that ethical leadership is positively related to procedural justice. This confirms earlier research showing how ethical leadership is primarily concerned with the procedural aspects of fair decision-making (Brown et al., 2005). Both procedural justice and ethical leadership include the fair treatment of team members by their team leader. This suggests that employees who perceive their leader as ethical will also perceive their organization as high in procedural justice because their principled leaders promote a work climate in which the process of decision-making should be fair.
Third, in contrast to what was expected, procedural justice was not significantly related to employee gossiping behavior and procedural justice also did not mediate the relationship between ethical leadership and employee gossiping behavior. Despite the significant negative correlation between procedural justice and employee gossiping behavior, procedural justice was not able to influence gossiping significantly and thus was not able to mediate the relationship between ethical leadership and employee gossiping behavior. This may be explained by the high correlation between ethical leadership and procedural justice, which indicates an overlap between both constructs. Research by Singer (1998) found that organizational justice is highly related to ethics. Additionally, it was found that justice explains about 60% of the variance in ethics (Singer, 2000). An option to reduce this overlap between ethical leadership and procedural justice could be to validate the scales used, to make sure that the items measure each scale perfectly. As can be seen in the moderation analysis,
procedural justice is negatively correlated to employee gossiping behavior. However, this effect is not strong enough when incorporated in the mediation analysis together with ethical leadership. Ethical leadership is a stronger predictor for gossiping at work, explaining more variance in employee gossiping behavior than procedural justice. This may be the case because procedural justice is only one part of organizational justice. As explained earlier, organizational justice includes distributive justice, procedural justice and interactional justice (Colquitt, 2001). Despite ethical leadership being primarily concerned with the procedural aspects of fair decision-making (Brown et al., 2005), ethical leadership is also partially concerned with the distributive aspect of providing consequences for unethical behavior and the fairness element of ethical leadership also relates to interactional justice (Mayer et al., 2012). Furthermore, interactional justice is the quality of the interpersonal treatment that individuals receive at work and it includes the fair and respectful treatment and open communication between managers and employees, which is similar to ethical leadership (Den Hartog, 2015). Perhaps, procedural justice alone is too narrow to explain the relationship between ethical leadership and employee gossiping behavior. As stated before, Singer (2000) showed that ethics and justice are highly related and showed how justice explains about 60 % of the variance in ethics. Also, research by Demirtas (2015) already showed that organizational justice partially mediates the relationship between ethical leadership and organizational misbehavior (e.g. employee gossiping). Due to the many linkages between ethical leadership and organizational justice, it is possible that when taking into account organizational justice as a whole (including distributive, procedural and interactional justice together), this broader concept is better able to (partially) explain the relationship between ethical leadership and employee gossiping behavior.
Fourth, this study contributes to the literature on gossiping within organizations by indicating that gossiping possibly serves as a way to cope with highly stressful work
situations within organizations. The results show that work stress is positively related to employee gossiping behavior. This provides support for the rationale that employees can use gossiping in organization as a way to cope with stressful situations. The results suggest that when employees experience stress at work, they can use gossip as a way to let off some steam about organizational issues or colleagues, which in turn can help to reduce worrying (Grosser et al., 2012). More specifically, this paper also showed how work stress moderates the relationship between procedural justice and gossiping, such that the relationship between procedural justice and employee gossiping behavior is weaker when work stress is high. This suggests that employees who experience work stress cope with negative feelings (e.g. a decrease in both intrinsic and extrinsic job satisfaction) that make it difficult for them to perceive the positive outcomes of their work environment (Summers, DeNisi and DeCotiis, 1989). Employees thus will not benefit from a work environment high in procedural justice because they are not able to experience the positive effects of it, as a result of the experienced work stress. To deal with the feelings of work stress, employees in turn will be inclined to gossip as a way to vent. Employees who do not experience any work stress are less inclined to gossip when their environment is high in procedural justice, suggesting that procedural justice is able to influence employee gossiping behavior as a result of the feeling to reciprocate positively to the trustworthy and fair organization. These findings contribute to the work stress literature by suggesting that work stress is able to influence employees is such a profound way that they are not able to benefit from the outcomes of a positive work environment anymore (e.g. an environment high in procedural justice). Earlier research showed how work stress can lead to a decrease in work motivation, less intrinsic job satisfaction, less extrinsic job satisfaction, less health, more anxiety and strain and an increase in the intention to leave the organization (Summers et al., 1989; Ganster & Rosen, 2013). The present study contributes to these findings by suggesting that work stress leads to the inability
