How and When Perpetrators Reflect on and Respond to Their Workplace Ostracism Behavior: A Moral Cleansing Lens

Introduction

With the rapid evolution of diversified organizational structures and fiercer competition, conflicting interests and interpersonal friction in the workplace are difficult to avoid, which leads to an increase of workplace ostracism. Workplace ostracism refers to an employee’s perception for being isolated or rejected by the colleagues in the workplace,¹ much of the existing research had examined its numerous negative outcomes to both organizations and individuals.^2–4 Particularly, due to the influence of guanxi that is centered on the “Chaxu”, workplace ostracism has become more persistent, more concealed, and more harmful in China.^5,6 This causes huge losses to Chinese firms.³ Therefore, scholars have devoted much attention to workplace ostracism in Chinese organizations in order to govern and control it more effectively. Prior research has generally concentrated on the factors that contribute to workplace ostracism from multiple perspectives, such as leadership, personality, and contextual characteristics, expecting to effectively interevent its occurrence by comprehensively understand the formation mechanism.³ However, it still remains a crucial and vital problem about how to govern, alleviate or even transition the harmful ostracism when it has already happened in the workplace. In other words, is there a theoretical possibility that workplace ostracism leads to some beneficial outcomes?

Following the seminal work of Williams and Sommer that proposing the workplace ostracism may induce individuals’ compensatory behavioral intentions,⁷ chief concerns of related studies fall on the potential positive consequences from multiple perspectives.^8–12 Especially, Haldorai has demonstrated in a collectivistic country that ostracized employees may conduct helping behavior for account of being reaccepted and maintaining harmony.¹³ Nevertheless, the stream has only investigated the results of being ostracized based on the victim-centric perspective,¹⁴ failing to notice the fact that a perpetrator’s subsequent act can be affected by his/her past behavior. Similarly in China that also highlight collectivism, the possibility of perpetrators committing altruistic behaviors for reaccepting others also exists. More importantly, Chinese moral norms emphasis that individuals should intimate and respect the others, which significantly impacts interaction and management of China’s employees.¹⁵ Yet to our reviewing, seldom explored the potential moral costs of workplace ostracism from “perpetrator-centric” perspective, even though it has been considered as a kind of unethical emotional abuse that violates the norms.¹⁶ A recent study has found a leader’s previous unethical behavior will trigger his or her subsequent moral behavior based on the moral cleansing theory.¹⁷ This raised our curiosity of whether employees turn to moral and engage in compensatory behavior after implementing in ostracism?

According to the moral cleansing theory, individuals may experience changes in both moral emotion (such as guilt) and moral cognition (such as perceived loss of moral credit) when they violate moral norms.¹⁷ We propose that, as a form of workplace emotional abuse, workplace ostracism not only harms victims and firms,³ but also leads to perpetrators’ negative psychological experiences. Specifically, when perpetrators realize that their unethical behavior unreasonably harms others, they may experience an emotion alongside remorse, pain, and self-blame, and that is guilt.¹⁸ Consequently, they may want to take some prosocial action to compensate for their mistakes.¹⁹ At the same time, engaging in workplace ostracism may undermine employees’ positive moral self-image. They may perceive a loss of moral credit as workplace ostracism is a violation of social norms,^2,17 and tend to reshape a positive self-cognition through subsequent compensatory behavior.¹⁸ Therefore, relying on moral cleansing theory, we integrate guilt and perceived loss of moral credit into the examination of the mediating mechanisms, in order to clarify how employees’ workplace ostracism affects their subsequent compensatory behavior.

What exactly is the specific compensatory action the perpetrator will take toward the victim after perpetrating workplace ostracism? Some scholars have pointed out that as a negative interpersonal targeted behavior, workplace ostracism will not only cause long-term psychological damage to the perpetrators (and reduce their internal demands for social participation),³ but it will also destroy their social exchange networks in the organization, which may induce a decline in individual and organizational performance.² Therefore, perpetrators may tend to adopt some extra-role behaviors, such as employees’ helping behavior, that can help others to improve their performance, and repair their damaged interpersonal networks as compensation for their workplace ostracism.^20,21 The helping behavior is a spontaneous and altruistic positive moral behavior; it can not only directly benefit the organization, but also highlight the employee’s benign moral image in the traditional Chinese culture context with high collectivism.^22,23 Thus, it can indirectly compensate for the damage caused by previous workplace ostracism. In short, combined with the possibility that unethical perpetrators may become ethical through the process of moral cleansing, we argue that helping behavior will be selected by them as a compensation after committing workplace ostracism.

Further, although perpetrators of workplace ostracism with high moral cleansing intention may trigger their own helping behavior due to guilt experience and loss of moral credits, it is contingent to what extent this process will occur.²⁴ Previous studies have shown that moral identity symbolization is a critical contextual condition for exploring the mechanism of individual moral cleansing. It is a personality trait that reflects individuals’ moral traits, and a desire to publicly express themselves in reality.^25,26 This is because helping behavior has been viewed as cooperation-oriented and identifiably prosocial,²³ and individuals’ guilt and perceived loss of moral credit are closely related to their prosocial behavior based on the relevant study.¹⁷ Thus, we propose that individuals with varying degrees of moral identity symbolization may have varying behavioral intentions for “morally cleanse”, and exhibit different helping behaviors.

In short, we investigated the relationship between employees’ previous ostracism behaviors and their subsequent interpersonal helping based on moral cleansing theory in the Chinese context, discussing the mediating roles of guilt experience and perceived loss of moral credits and moderating role of individuals’ moral identity symbolization (see Figure 1). This research has three novelties that contribute to the relevant research field. First, previous works on the positive consequences of workplace ostracism have mainly focused on the perpetrators’ harmful behaviors on victims’ compensatory actions. Our study, which draw its inspiration from the theory of moral cleansing, explores the effects of employees’ prior workplace ostracism on their helping behavior by turning the perspective from “victim-centric” to “perpetrator-centric”. Moreover, we gain a deeper understanding of the inconsistency and transition between positive and negative behaviors by focusing on the moral cleansing mechanism. By doing so, we enrich the relevant studies on the moral costs and behavior responses to workplace ostracism and illuminate both its possible benefits and potential drawbacks. Second, our dual-path mediating effects model contributes to extensive research on moral cleansing theory in the organizational behavior and human decision processes literature.¹⁷ Meanwhile, we examined the potential consequences of workplace ostracism as well as how it can be facilitated. We also aim to illustrate the mediating mechanisms of two key factors, guilt and perceived loss of moral credits, since moral cleaning theory posits that employees consider morally dubious and undesired behaviors as unacceptable when they had committed unethical behaviors. In regard to workplace ostracism, each variable has been studied respectively. The third novelty is related to our boundary condition. By employing moral identity symbolization as a moderator, this study responded to the call of more research on the distinction and specific role of moral identity.²⁷ This also adds to the theoretical boundary condition and empirical support of the moral cleansing mechanism since we adopted moral identity symbolization as a moderator of the processes.

Figure 1 Research model.

Theoretical Foundation and Hypotheses

Moral Cleansing Theory

Moral cleansing theory posits that people are usually eager to maintain an ideal moral self-image, but the image is unstable and will change with their behavior in the short term.^28,29 As a negative feedback mechanism of moral self-regulation, moral cleansing aims to eliminate the influence of previous unethical behavior on individuals’ moral self-concept. Specifically, the initiation of the moral cleansing mechanism involves two parts: the ideal moral self-image (the individual’s expectation of their moral image) and the actual moral self-evaluation (the individual’s objective cognition of their moral image).²⁸ In general, people will make a comparison between their actual and ideal moral self-image to maintain a balance. When individuals engage in unethical workplace behaviors like abusive supervision and counterproductive work behavior, the immorality of the behaviors destroys the balance between their actual and ideal moral self-image, thus activating moral cleansing.^17,24 At this point, perpetrators may suffer from negative experiences in terms of both moral emotions (such as guilt) and moral cognition (such as perceived loss of moral credit), which drives them to adopt compensatory behavior in order to eliminate the negative emotions or cognitive feelings, thus restoring balance. Therefore, we attempt to explore how workplace ostracism drives employees to adopt compensatory behaviors, such as helping behavior, by drawing on moral cleansing mechanism. Further, we argue that perpetrators achieve the balance between their actual and ideal moral self-image through two mediating paths of moral emotion (i.e, guilt) and moral cognition (i.e, perceived loss of moral credits).

Perpetrators’ Workplace Ostracism Leads to Their Guilt Experience and Perceived Loss of Moral Credits

Workplace ostracism is the action of a person or group not accepting members of another organization when it is socially appropriate.³⁰ As a kind of exclusive behavior, workplace ostracism specifically includes behaviors that disrupt interpersonal interactions, such as avoiding eye contact and conversation, as well as indifferent and silent treatment.³¹ Studies have shown that as a kind of emotional abuse in the workplace,³² workplace ostracism not only brings severe adverse effects to the victims and their organizations, but also leads to a negative influence on the perpetrators themselves.³³ For the victims, workplace ostracism may increase their psychological pressure,³ trigger their negative emotions and deviance,³⁴ destroy their work–family balance, and endanger their family life and subsequent work.³⁰ For the perpetrators, workplace ostracism may obstruct their psychological needs for autonomy and guanxi, thus threatening the harmony and stability of their interpersonal networks.²¹ In addition, workplace ostracism impedes firms’ normal operation and decreases performance.²

Furthermore, although the scholars have emphasized the fact that workplace ostracism is an unethical behavior that violates social norms, such as mutual respect and equal opportunity,³³ few have explored whether individuals can recognize the immorality of their behavior after engaging in workplace ostracism, from the perspective of perpetrator-centric. In fact, there is some evidence that actors can perceive the immorality of their ostracism after ostracizing others.¹⁸ Ferris, Brown, Berry, Lian¹ argue that unlike conflicts, ostracism can occur without provocation. In this sense, ostracism cannot be justified as a legitimate defensive behavior, and the actor tends to view the behavior as unethical.³⁵ Further, Bastian, Jetten, Chen, Radke, Harding, Fasoli³³ found through four experimental studies that employees are more inclined to depersonalize themselves in coping with illegal behavior and immoral treatment after ostracizing others. In other words, the perpetrators are able to recognize the immorality of their behaviors, and thus can consider workplace ostracism as an immoral behavior causing substantial harm and violation.

Given the harmfulness and immorality of ostracism, employees may experience the perception of moral imbalance after its implementation, prompting them to make changes in both their moral emotions and cognition.¹⁷ As an important “self-conscious” aspect of moral emotion, guilt refers to a negative emotional experience with remorse, pain, and self-blame that occurs when an individual consults their conscience, and feels responsible for the behavior which harms others or violates moral norms.³⁶ The moral cleansing theory holds that the activation of guilt involves the process of self-evaluation and reflection. Specifically, when individuals realize that their behaviors (such as workplace ostracism) violate social ethics, they may develop negative moral self-evaluation, and feelings of disgust and guilt may follow.³⁷ In addition, given the severe harm that workplace ostracism can cause to colleagues and firms, employees’ ostracism may trigger their internal attribution tendency, and they may believe that they are mainly responsible for the negative consequences of the behavior.³⁸ Guilt is essentially a negative emotional experience of self-blame.³⁹ At this time, a driving mechanism of moral cleansing—self-conscious moral emotion (guilt)—may also be initiated along with self-blame.²⁶ Accordingly, it is plausible that an employee whose behavior violates organizational rules and moral norms is likely to experience guilt.

Hypothesis 1. Perpetrators’ previous acts of workplace ostracism positively affects their guilt experience.

The moral cleansing theory suggests that an individual’s previous unethical behavior not only triggers the guilt experience, but also has an impact on their moral cognition, such as the perceived loss of moral credit.⁴⁰ Some scholars have indicated that the perceived loss of moral credit is an individual’s perception of the decrease in their moral credibility at a certain time.⁴¹ Specifically, the moral balance model deems that everyone has a “moral bank account”, and moral credits are the “deposit” in the “account.” When an employee’s “account” becomes “deficient” due to unethical behavior, he/she may perceive a loss of moral credits. As a result, they may actively adopt moral compensation behavior to make up for moral “liabilities” and achieve the moral balance.⁴² Therefore, as a crucial explanatory path of moral cleansing theory, the fluctuation of moral credit acts in a crucial manner in the evolution of subsequent behavior when an individual engages in workplace ostracism.

As mentioned above, workplace ostracism is an immoral interpersonal deviance that harms the moral self-image of the perpetrator. According to moral cleansing theory, the immorality of workplace ostracism can break the actual and ideal moral self-image balance of an individual, resulting in a perception of loss of moral credit.^29,40 Relevant studies have verified the cognitive process of moral cleansing by indicating that previous unethical behavior will lead to the perceived loss of moral credit, and this provides theoretical and empirical support for our study. For example, Jiang, Liang, Wang⁴³ found that as a special type of moral deviance, unethical pro-organizational behavior evokes the perceived loss of moral credit by damaging the moral self-esteem of perpetrators. Similarly, Liao, Yam, Johnson, Liu, Song¹⁷ found a positive correlation between leaders’ abusive supervision and their enhanced perceived loss of moral credits. In summary, we suggest that individuals may experience a loss of moral credit due to their previous workplace ostracism.

Hypothesis 2. Perpetrators’ previous acts of workplace ostracism positively affects their perceived loss of moral credit.

Perpetrators’ Workplace Ostracism and Their Subsequent Helping Behavior: Mediating Effect of Guilt and Perceived Loss of Moral Credits

According to moral cleansing theory, the individual who experienced guilt after engaging in workplace ostracism and perceives the loss of moral credit tends to alleviate their negative emotional and cognitive experiences by adopting prosocial behaviors (such as helping behavior) toward victims and their firms.^29,40 As important types of prosocial behavior, employees’ helping behavior is characterized as voluntarily helping colleagues to prevent or solve work-related problems, or as providing work information or resources for others to complete work tasks. It includes the type of problem-solving that directly help others address their problems and the type of knowledge-sharing that helps colleagues clarify how to solve the problems.⁴⁴ Employee helping has been widely seen as a crucial dimension of citizenship behavior, which has basic attributes such as spontaneity, voluntary but not compulsory extra-role, and altruism.²² At a time when the complexity of work content and the uncertainty of the work environment are increasing rapidly, helping behavior highlights the aggregation of value because of its mutual assistance and collaboration, which can reduce the workload and work pressure of employees, improve team cohesion, stimulate team vitality, and maintain interpersonal trust and harmony, thus promoting individual and organizational innovation and performance.⁴⁵ Therefore, for the perpetrators of workplace ostracism, helping behavior is undoubtedly an appropriate after-the-fact compensation behavior to help them ease their guilt and re-establish a benign moral self-image.

Specifically, individuals who experience guilt have a behavioral tendency to compensate for the harm caused to others and reducing their pain.³⁸ Schaumberg, Flynn⁴⁶ believe that guilt stems from empathy and is a recognition of the causes of pain in others. The relationship between guilt and responsibility is stronger when individuals realize that they have caused others pain, especially when individuals feel more in control of events.⁴⁷ Thus, they are more willing to extend a helping hand in the face of others’ help-seeking behavior and hope to mitigate the harm of their negative behavior toward colleagues and organizations by helping others when individuals experience guilt over previous negative behavior. Furthermore, prior research suggests that people often undertake moral compensation behavior after the occurrence of improper behaviors for alleviating the experience of guilt.¹⁷ These behaviors could include providing others with knowledge, performing work tasks, and carrying out interpersonal related help. From this perspective, helping behavior may be a self-serving means for the perpetrator of workplace ostracism to achieve their personal goals.⁴⁸ Thus, we suggest the following hypothesis in light of the aforementioned inferences:

Hypothesis 3. Guilt positively affects the subsequent helping behavior of the perpetrators of workplace ostracism.

Combined with hypotheses 1 and hypotheses 3, we believe that guilt mediates the relation between employees’ workplace ostracism and their subsequent helping behavior. Specifically, when employees ostracize colleagues in the workplace, the immorality of the behavior and its negative impacts on the victim can easily give rise to guilt experience, which evokes the behavioral tendency of individual moral cleansing. They hope to make up for the harm caused to others by helping behavior to achieve the purpose of reducing their guilt. In other words, workplace ostracism positively impacts helping behavior through the employees’ guilt. Thus, we propose the following hypothesis:

Hypothesis 4. Perpetrators’ workplace ostracism will exert an indirect positive impact on their subsequent helping behavior through the mediating effect of guilt.

Subsequently, the theory of moral cleansing shows that as a cognitive processing mechanism, perceived loss of moral credit usually stimulates individuals’ strong behavioral tendency toward moral compensation, prompting them to consciously perform morally commendable behaviors, such as helping other employees.

As previously mentioned, because of organizational citizenship behavior attributes, such as spontaneity and altruism,⁴⁹ employees’ helping behavior is often regarded as socially ethical behavior,⁵⁰ meaning that it may help restore an individual’s damaged perception of moral credit. On the one hand, when individuals perceive the loss of moral credit due to previous immoral behavior, they may regard it as an opportunity to accumulate “moral credits” in the face of colleagues’ help-seeking behavior, so as to eliminate the loss of their negative moral perception. On the other hand, given that helping behavior has the utility of creating a positive reputation and enhancing self-efficacy,⁵¹ the individual cannot only satisfy their need for autonomy, but also positively evaluated by others in the process of providing help to them,⁵² thus leading to the improvement of their organizational reputation and status.⁴⁵ Furthermore, a better organizational reputation and status have positive spillover effects on the individual’s internal perception of moral self-image,⁵³ which may be conducive to increasing the individual’s moral self-concept. In summary, we believe that helping behavior can repair a damaged moral self-image and can be used as an effective way for individuals to eliminate moral cognitive stains and gain moral credits. Hence, this study proposes the following hypothesis:

Hypothesis 5. Perceived loss of moral credits positively affects employees’ helping behavior.

Combined with hypotheses 2 and 5, this study posits that the effect of perpetrators’ workplace ostracism on their subsequent helping behavior is mediated by perceived loss of moral credits. Specifically, when individuals ostracize colleagues in the workplace, perceiving the immorality of the behavior can easily provoke a perceived loss of moral credits. This usually stimulates their strong behavioral tendency toward moral compensation. And the individuals hope to repair their moral self-image which has been damaged and eliminate the loss of moral credits by providing help to others. In other words, workplace ostracism positively impacts employees’ helping behavior because of the perceived loss of moral credits. Therefore, this study proposes the following hypothesis:

Hypothesis 6. Perpetrators’ workplace ostracism will exert an indirect positive impact on their subsequent helping behavior through the mediating effect of perceived loss of moral credits.

The Moderating Effect of Moral Identity Symbolization

Moral identity has two dimensions: moral identity internalization and moral identity symbolization.²⁵ The former demonstrates how individuals internally value a set of moral characteristics in their self-concept, while the other describes how willing people are to externalize their moral identity by their actual behavior. Previous studies have focused on the role of these two characteristics in terms of public acceptance. Particularly, individuals who have a high-level moral identity internalization is prone to act in a prosocial manner whether or not others notice or acknowledge it, while individuals who have a high-level moral identity symbolization will act in such behavior only when others can witness or notice them.26 Given that helping behavior is the act of providing help to others, the current research focuses on whether the relationship between perpetrators’ guilt experience (or perceived loss of moral credits) and their subsequent helping behavior will be moderated by moral identity symbolization.

Relevant research has shown the dimension of moral identity symbolization originates from the perspective of symbolic interactionism. It is related to the self-actualization of the social entity, and is often thought to reflect a vital source of motivation for participating in prosocial behavior. Generally speaking, employees with stronger moral identity symbolization degrees can validate themselves through the evaluation of others reflected in prosocial behavior.⁵⁴ They tend to communicate their commitment to specific moral goals to others by engaging in visible activities,⁵⁵ which means they will achieve their goals by performing prosocial behavior when others can witness and acknowledge this behavior.⁵⁶ Further, employees with a higher degree of moral identity symbolization may react more sensitively to the identifiability of their prosocial behavior (such as employees’ helping behavior) when making decisions on whether to implement such behavior. Accordingly, we speculate that employees’ moral identity symbolization can strengthen the relationship between their guilt experience and subsequent helping behavior and the relationship between their perceived loss of moral credits and subsequent helping behavior.

Specifically, employees’ helping behavior is highly identifiable as an interpersonal-targeted organizational citizenship behavior, which helps people to obtain higher potential psychological benefits from the reflection and evaluation of others’ observations.⁵⁴ Therefore, perpetrators with a higher degree of moral identity symbolization will be more strongly motivated to preserve their moral self-image, and they may be more proactive in helping others when they have an experience of guilt and a perception of moral credits loss after engaging in workplace ostracism. In this situation, they are more inclined to provide help to others to strengthen their moral identity in the face of others’ help-seeking. Conversely, when employees have a low level of moral identity symbolization, they will be less inclined to implement helping behavior no matter how they experience guilt or perceive a loss of moral credits. This is due to the fact that they do not demonstrate to others that they highlight morality in their own self-concept, therefore it is unnecessary for them to exhibit overtly moral conduct. In conclusion, the hypothesis below is proposed in this study.

Hypotheses 7. Moral identity symbolization will positively moderates the association between perpetrators’ guilt and perceived loss of moral credits and their subsequent helping behavior.

Combined with the relationship predicted by hypothesis 6 and 7, this study suggests that the intensity of the mediating effect from workplace ostracism through experiencing guilt and perceived loss of moral credits to employees’ subsequent helping behavior will also be different, mainly impacted by varying degrees of moral identity symbolization of employees. Specifically, when perpetrators have a higher degree of moral identity symbolization, the negative guilt experience and perceived loss of moral credits originated from previous workplace ostracism will stimulate employees’ stronger behavioral tendency toward moral compensation, as they hope to eliminate the negative feelings and enhance their moral self-image by helping behavior. Hence, this study proposes the following hypothesis.

Hypothesis 8. Moral identity symbolization will positively moderates the indirect effect of perpetrators’ workplace ostracism on their subsequent helping behavior through experiencing guilt and perceived loss of moral credits.

Methods

Samples and Procedures

This study has selected enterprises of a certain scale in Guangdong Province and Fujian Province in China as research subjects, and the research samples are mainly drawn from the tourism, manufacturing, and IT industries. After thoroughly explaining the purpose of the study to the responsible managers of HRM departments in each company, we obtained relevant permission and invited the managers to assist us with the distribution and subsequent data collection of the questionnaires. Besides, before the beginning of the survey, it was emphasized to the respondents that the survey was fully anonymous and used for scientific research only to eliminate any concerns of the respondents. Withdrawal from the survey at any time was also allowed. Following the work of Wang, Xiao, Ren,²⁶ we collected employee self-assessment data at two different time points separated by two weeks for reducing common method biases. At time point 1, demographic information from samples was acquired and demographic variables, the level of moral identity symbolization, workplace ostracism, guilt experience, and perceived loss of moral credits were further measured. Two weeks later at time point 2, we required employees to assess their helping behaviors. Moreover, we coded and matched all questionnaires obtained at two stages on Wenjuanxing (an online questionnaire collection platform similar to Amazon Mechanical Turk) by using the last four-digits of employees’ cell phone numbers as a reference. Finally, a total of 306 employees have completed both of two stages. And we ultimately collected 284 self-assessment samples with a response rate of 92.8% after removing invalid samples, such as those can not be matched or had incomplete information.

The age of the respondents was mainly concentrated between 28–37 years old, accounting for 48.6% of the sample, and there were 119 males comprising 58.06% percent of the sample. In terms of educational attainment, 75.4% held a bachelor’s degree or a graduate degree, and the proportion of those with just high school and junior college education was relatively small. As for work experience, 56.3% percent had worked for 6 to 9 years, and 19% had worked for 10 years. In addition, the weekly working hours of the sample were mainly concentrated on 5 working days and 6 working days; this accounted for 95.4% of the participants, which was in line with the normal sample information distribution.

Measures

Workplace Ostracism

Following the work of Ferris, Brown, Berry, Lian,¹ we assessed the workplace ostracism of employees by using a 10-item scale measure (α=0.92). Prior research in the Chinese workplace has also indicated a good validity of this scale.³¹ The original scale evaluates the feeling of ostracism from the perspective of the victim, but this paper emphasizes the actor-centered perspective of ostracism. Therefore, we modify the expression of these items to evaluate the behavior of employees themselves. The representative items were “I ignored some colleagues at work”, and “some colleagues greeted me, and I ignored them”.

Guilt

Individuals’ guilt was evaluated using Tangney’s 3-item measure (α=0.91).⁵⁷ An example item is “When you reflect on your recent workplace ostracism, to what extent do you feel the following?” and example questions are “Feel guilty/ashamed”, “Should be blamed”, “Feel regretful”.

Perceived Loss of Moral Credits

Using the 5-item scale developed by Liao, Yam, Johnson, Liu, Song (α=0.88),¹⁷ the participants need to answer, “When you recall the workplace ostracism you have recently made, to what extent do you feel?”. The example items are “I lost my moral credibility because of recent unethical behavior”, and “Recent unethical behavior will cause a moral reputation deficit”.

Moral Identity Symbolization

We used the symbolic dimension of the moral identity scale developed by Aquino, Reed to measure moral identity symbolization,²⁵ including 5 items (α=0.83). Sample items are “I often wear clothes that show these characteristics I have”, and “Things I do in my spare time (such as hobbies) clearly reveal that I have these characteristics”.

Employees’ Helping Behavior

We use the scale compiled by Harari, Parke, Marr.²⁰ It includes 10 items (α=0.91) and participants need to answer “When you recall your recent workplace ostracism and the negative emotions and cognitive experiences it brings, to what extent will you engage in the following behaviors?” The example items are “I will show initiative to help my colleagues before they ask for help”, “I will help my colleagues when they need my help”, etc.

Control Variables

According to prior research, the attitudes and behaviors of employees will be influenced by their age, gender, educational background, and tenure.⁵⁸ Therefore, in this study, these factors are treated as control variables and assigned separately. Specifically, we asked participants to fill in the information about age and work experience, then they need to choose the four academic categories of college degree or below, college degree, graduate degree, and postgraduate degree to control educational background.

Results

Confirmatory Factor Analysis

We employed AMOS 24.0 in this study to conduct confirmatory factor analysis. The evaluation indicators and model comparison are shown in Table 1. Our results indicate that all the fit indexes of the baseline model (ie, the 5-factor model) met the standard, with χ2/df = 2.073, CFI = 0.907, TLI = 0.898, IFI = 0.908, RMSEA = 0.062. It implies that our research model with five variables has a favorable structural validity. Additionally, our 5-factor model owns better fit indexes than that of the other alternative models (including the 4-factor, 3-factor, 2-factor, and single-factor model). It shows that there is good discrimination between the main variables of this study.

Table 1 Confirmatory Factor Analysis Results

Common Method Bias Test

Data were collected by using employee self-reported method. Although the use of a two-wave survey design can partially reduce the common method bias,⁵⁹ to enhance the preciseness of the conclusion, according to the suggestion of Podsakoff, MacKenzie, Lee,⁵⁹ common method variance (CMV) has been included into the confirmatory factor analysis to evaluate the common method bias more effectively. As shown by Table 1, comparing with the fit indexes of the baseline 5-factor model, the RMSEA, CFI, and TFI of the five-factor + CMV model (χ² / df = 2.491, RMSEA = 0.073, CFI = 0.876, TLI = 0.858) vary by less than 0.02, indicating that the common method bias of our study falls within a permissible range.

Descriptive Statistical Analysis

Table 2 shows that perpetrators’ guilt experience (r = 0.332, p < 0.001) and perceived loss of moral credits (r = 0.329, p < 0.001) are both positively correlated to their previous workplace ostracism. At the same time, employees’ helping behavior (r = 0.228, p < 0.001), guilt (r = 0.091, p < 0.01), and perceived loss of moral credits (r = 0.100, p < 0.01) are also positively and significantly correlated. Similarly, the level of employees’ moral identity symbolization is also positively and significantly correlated with their helping behavior (r = 0.493, p < 0.001). Although age, gender, and educational background can affect research variables of the study, the correlation analysis points out that the above demographic variables can be studied as control variables.

Table 2 Descriptive Statistics and Correlations Matrix

Test of Hypotheses

(1) Examination of the main and the mediating effects. As shown by Model 2 in Table 3, when controlling for the demographic variables, perpetrators’ guilt experience is positively and significantly associated with their prior workplace ostracism (β=0.470, p<0.001). Meanwhile, as shown by Model 7 in Table 3, perpetrators’ workplace ostracism exerts a positive impact on their perceived loss of moral credits (β = 0.390, p < 0.001), thus Hypothesis 1 and Hypothesis 2 are both supported. Besides, from Model 13 in Table 3 we can easily find that perpetrators’ guilt experience also has a positive relationship with their helping behavior (β=0.055, p<0.10), which supports Hypothesis 3. Additionally, Model 14 shows that there is also a positive “perceived loss of moral credits-helping behavior” relationship (β=0.093, p<0.1), thus supporting Hypothesis 5.

Table 3 Mediating Effect Test

Further, the 95% unbiased confidence interval is constructed to verify the mediating effects of guilt experience and perception of moral credits loss, using a combination of the stepwise method and bootstrapping method of SPSS with 5000 repeat samplings. As shown by Table 4, the “workplace ostracism → guilt → helping behavior” indirect effect is β = 0.084, with 95% CI of [0.046, 0.133]. Thus, the mediating effect of guilt is significant, supporting Hypothesis 4. Simultaneously, the “workplace ostracism → perceived loss of moral credits → helping behavior” indirect effect is β = 0.063, with 95% CI of [0.014, 0.115], indicating that it has a statistically significant effect and supports Hypothesis 6.

Table 4 Results of the Bootstrap Analysis of Mediation Effects

(2) Test of moderating effect. Our study examines the moderating effect by adopting the interaction term construction method. To inhibit the influence of multicollinearity, we first standardized all the main study variables before calculating the interaction terms. As shown in Table 5, the interaction term between moral identity symbolization and guilt positively affects helping behavior (β = 0.196, p < 0.001) with a 95% CI of [0.121, 0.271], suggesting that the higher the degree of perpetrators’ moral identity symbolization, the stronger the positive “guilt-helping behavior” association. Also, the “moral identity symbolization × perceived loss of moral credits” interaction term positively affects helping behavior (β = 0.172, p < 0.001) with a 95% CI of [0.081, 0.263], which indicates that the higher the degree of perpetrators’ moral identity symbolization, the stronger the positive “perceived loss of moral credits-helping behavior” linkage. Therefore, Hypothesis 7 is supported.

Table 5 Interaction Effects Test

In an effort to concretely demonstrate the interaction effect, this study used the mean of the moderating variables plus or minus one standard deviation for the moderating schematic. The impacts of perpetrators’ guilt experience and moral credits loss perception on their helping behavior at different levels of moral identity symbolization are shown in Figures 2 and 3.

Figure 2 Moderating effect of moral identity symbolization on guilt and employees’ helping behavior.

Figure 3 Moderating effect of moral identity symbolization on perceived loss of moral creditsand employees’ helping behavior. Abbreviation: PLRM, Perceived Loss of Moral Credits; MIS, Moral Identity Symbolization.

(3) Moderated Mediating Effect. The 95% unbiased confidence interval is constructed by the Bootstrap method with 2000 repeat samplings to determine whether a moderated mediation effect exists. Based on Table 6, the mediating effect of guilt is −0.055 with a 95% confidence interval of [−.099, −0.014] excluding 0 when the dependent variable is employees’ helping behavior and the moderating variable is taken at a lower level (M-1SD). When the moderating variable takes on a high level (M+1SD), the mediating effect value is 0.057 with a 95% confidence interval of [0.023, 0.097] not containing 0, and the overall Index value of the mediating effect model with moderation is 0.085 with a 95% confidence interval of [0.041, 0.133] excluding 0. Further, the mediating effect value of perceived loss of moral credits is −0.039 with a 95% confidence interval of [−.094, 0.001] containing 0 when the moderating variable is taken at a lower level (M-1SD). Meanwhile, when the moderating variable takes a high level (M+1SD), the mediating effect value is 0.045 with a 95% confidence interval of [0.011, 0.082] not containing 0. The overall Index value of the mediating effect model with moderation is 0.063 with a 95% confidence interval of [0.014, 0.115] not containing 0.

Table 6 Moderated Mediation Model Test

Therefore, the mediating effect differs when the moderating variables are taken at different levels and rises as the level of the moderating variable increases; thus Hypothesis 8 is verified by the data.

Conclusion

This paper verifies that when individuals become perpetrators of workplace ostracism, they will feel guilty about hurting others, and perceive the loss of moral credits. Then this negative experience will act on their subsequent helping behavior. Specifically, workplace ostracism can bring great harm to an organization and individuals in the form of emotional office abuse, and the perpetrators may also experience self-blame and guilt. At the same time, the immorality of workplace ostracism may also impair the moral image of themselves, resulting in a “loss of moral credits” perception. Further, the perpetrators’ moral cleansing mechanism is activated, triggering them to restore their damaged moral image by adopting helping behavior. In this situation, on the one hand, they will provide help to others because of the psychological compensation. On the other hand, they may also engage in proactive helping behavior to make up for the damage they have caused. Besides, we further examine the moderating role of employees’ moral identity symbolization: Perpetrators who own a guilt experience and a loss of moral credits perception have a stronger tendency to help others and are willing to perform more helping behaviors when they have a higher moral identity symbolization.

Specifically, individuals with high-level moral identity symbolization are more likely to show their moral behavior to others. This is because they regard morality as a symbol of their outward display, and hope that people will evaluate them based on their moral behavior. The interpersonal nature of helping behavior will convey the moral signals to others more significantly. Therefore, for those with guilt experience and loss of moral credits perception, the higher the level of moral identity symbolization, the more likely they will engage in helping others.

Theoretical Contributions

Our emphasis on the moral costs and responses to workplace ostracism from a perpetrator-centric viewpoint enhances knowledge of potential positive consequences to workplace ostracism literature. Indeed, Williams, Sommer⁷ has enlightened us that workplace ostracism may induces compensatory behaviors. Yet to date, relevant studies have mainly showed ostracized employees may commit pro-social behaviors for rebuilding the interpersonal relationship.^11,13 We broaden the sight by focusing on perpetrators and propose that perpetrating workplace ostracism may significantly affect employees psychology and their subsequent behaviors. As noted by Ferris, Chen, Lim,¹⁶ workplace ostracism was considered as an immoral behavior since it violates organizational norms and hurts others. However, current literatures has mainly paid their attention to relational mechanisms for exploring outcomes, neglecting its potential moral costs and resulting in little evidence supporting the argument. Building on moral cleansing theory,⁴⁰ we hypothesized that engaging in workplace ostracism may specifically affect guilt experience and perceived loss of moral credits. It is the reason of eliminating these two negative experiences that drives the employees to adopt subsequent helping behavior.³ The original findings of the work diverge from typical research and indicate that perpetrators can also suffer from the negative psychological experiences in addition to victims by taking moral effects into a comprehensive framework for explaining its positive consequences.

Additionally, our study contributes to moral cleansing theory by exploring both of its emotional and cognitive process. Although the mechanism of how perpetrators turning from “bad” to “good” have been discussed in the relevant literature, their demonstration and empirical evidence are still inadequate. Most of them have concentrated on one single path,^38,60 neglecting employees’ behaviors are motivated by an integration of emotion and cognition.⁶¹ We have responded to the call of Kipfelsberger, Herhausen, Bruch and established a mediating combination of moral emotion (guilty) and moral cognition (perceived loss of moral credits).⁶² The results provide an empirical support that employees’ compensatory actions are explained with their dual efforts of repairing the damaged interpersonal relationship and replenishing the consumed moral credits.²⁵ Therefore, this comprehensive perspective representing an important complement to moral cleansing literature.

Furthermore, we have also contributed to the study on moral identity by seeing employees’ moral identity symbolization as a boundary condition of the moral cleansing process. In fact, some scholars have called for more research on the distinction between and the different roles playing by moral identity internalization and moral identity symbolization,^54,55 however, most studies still explore and theorize the moderating effect of moral identity from a one-dimensional structure.²⁷ By zooming into moral identity symbolization, we help to clarify the actual role of a specific dimension and expound on its moderating effects in restoring individuals’ public moral self-image. Besides, according to some interpersonal conflict literature,⁶³ workplace ostracism is inherently a destructive interpersonal action, and some perpetrators may be reluctant to taking compensation. The study found that when individuals experience guilt and loss of moral credits, perpetrators with a high-level moral identity symbolization have a stronger prosocial tendency and are more likely to adopt identifiable prosocial behavior to restore the self-image of public morality.²⁶

Practical Implications

First and foremost, although our study found that the perpetrators of workplace ostracism may cause subsequent helping behavior, this study does not encourage workplace ostracism. On the contrary, as an emotional abuse in the workplace, we suggest that managers should comprehend that individuals’ moral behavior is a dynamic process (ie, the “good” and the “bad” can be transformed into each other). By exploring the moral costs of workplace ostracism from the “perpetrator-centric” perspective, managers can show the employees that engage in such unethical behavior may also induce negative emotional and cognitive experiences to themselves, which will stimulate perpetrators to carry out positive moral compensation actions not only for alleviating such experiences but also for repairing damaged interpersonal relationships. More importantly, demonstrating the compensatory behaviors of perpetrators can further allow victims to develop a more balanced view of the others. This will endow them with skills on coping with occasional workplace ostracism and establishing constructive interactions with colleagues, instead of consistently turning negative after being ostracized. In doing so, organizations can better regulate and govern the harmful workplace ostracism.

Second, our research indicates that the two negative experiences of guilt and perceived loss of moral credits may also result in beneficial outcomes, triggering individuals’ subsequent helping behavior. Managers should proactively take advantage of such potential effects by encouraging employees who feel guilt or perceive loss of moral credits to do more helping behavior in the workplace for mitigating their negative experience. For example, managers can emphasize that workplace ostracism can sometimes be inevitable, and employees will not be arbitrarily labeled “immoral” because of a momentary action as long as they can repair the injuries they have caused, compensate for violations of moral norms, and re-establish a favorable self-image. Besides, a friendly and inclusive moral atmosphere in the organization contributes to improve employees’ ability of moral self-regulation so that they can interpret their workplace ostracism as a correctable behavioral immorality. Consequently, they will be more likely to be impelled by guilt feeling and loss of moral credits perception, which increase their likelihood of subsequently engaging in compensatory helping behavior.

Finally, as shown by the moderation effect we investigated in the moral cleansing process, the possibility of an employee compensating for the victim can be influenced by the levels of his/her moral identity symbolization. Employees with a high level of moral identity symbolization can be previously selected by means of psychological and ethical tests or situational experiments when they applied for a position, which can strongly promote the importance of ethical behavior in the organizations. Organizations can also provide training programs to enhance employees’ ethical identity. For example, ethics training regularly requires actors to reinforce the value of moral traits in their self-cognition.²⁶ In addition, organizations can strengthen employees’ desire to express their moral traits to others through their behavior by developing policies that establish norms of apology and forgiveness, as well as eliminating the risks associated with actors admitting to ethical violations. These measures will give employees more positive feedback on their pro-social behavior.⁶⁴

Future Research Directions

First, all the variables in this paper are self-reported. Although the statistical analysis results do not find serious common method bias, this measurement method may have social desirability problems. Self-assessment will lead to a generally high score of employees’ helping behavior, which in turn causes data distortion and cannot reflect the effect of perpetrators’ workplace ostracism on their subsequent work behavior in real scenarios. Therefore, future research could use diverse approaches to better control social desirability and examine the differences in outcomes between other-reported and self-reported. In addition, although the survey data in this paper are collected from two stages, there are still limitations. This study investigates the causal relationships among employees’ workplace ostracism, guilt, perceived loss of moral credits, moral identity symbolization, and helping behavior based on relevant theories and literatures. Yet the persuasiveness of the causal relationship between variables remains insufficient since the data were collected at two stages. Therefore, future research should adopt more scientific methods for data collection and verification, such as the experience sampling method (ESM), and scenario-based experiments, to better verify the association among the variables in this study.⁶⁵

Second, in this research, we explored whether the “guilt- and -perceived loss of moral credits-employee helping behavior” relationship is conditional on moral identity symbolization, while ignoring the exploration of the boundary condition of the “workplace ostracism-guilt emotion and perceived loss of moral credits” association. In fact, it has been noted that moral attention will enhance individuals’ sense of guilt, that individuals with higher moral attention have higher perceptions of guilt and loss of moral credits, and that individuals owning high moral courage also engage in more compensatory behaviors.¹⁷ The present study considered the working characteristic boundary of moral identity symbolization, but did not control for other individual traits, such as moral identity internalization,²⁶ relativism, and guilt repair tendencies of personality that can also influence the occurrence of moral compensation mechanisms.^66,67 In particular, concerning the personality trait of moral identity internalization, previous scholars have noted that in contrast to moral identity symbolization, individuals with high-level moral identity internalization tend to engage in prosocial behavior regardless of social approval, whereas the activation and extraction of moral qualities in those with symbolized moral identity are contextually constrained and they are willing to do so only when they have social approval.⁵⁴ Brought into the research context of this study, is there a possible differential effect of internalized moral identity and symbolic moral identity on individuals’ tendency to engage in moral cleansing? The evidence offered by the existing studies is still insufficient. Therefore, future research should explore the impact of the interaction of the two dimensions of the internalized and symbolic moral identity of individuals on the path of employees’ moral compensation.

Third, employees tend to engage in helping behavior as a compensation for their unethical behavior after committing workplace ostracism. However, there may be a great distinction in the propensity to perform helping behavior, which is significantly due to differences in the driving mechanism. According to different behavioral motivations, helping behavior can be divided into proactive and reactive modes. Proactive helping behavior refers to the self-serving behavior of the helper to meet their own needs—such as taking the initiative to help new employees get familiar with the company environment in order to obtain a favorable interpersonal relationship. Meanwhile, reactive helping is altruistic behavior based on the reciprocal psychology of the helper, such as sharing knowledge with the help of others.⁵¹ As described by Duan, Wong, Yue,⁵² compared with proactive helping behavior, individuals are forced to perform reactive helping behavior because of reciprocal pressure, which usually requires them to make great efforts and consume large amounts of their working time. Therefore, for employees who feel guilty and perceive moral credits loss due to their workplace ostracism, there may be different restorative effects between proactive and reactive helping behavior. Future research can further investigate the differential impacts of employees’ moral cleansing mechanisms regarding these two different types of helping behavior to fill the gap in existing studies.

Finally, this study suggests that perpetrators may experience guilt and perceive loss of moral credits after conducting workplace ostracism behaviors and may also perform compensatory behavior including prosocial behavior and impression management, which is a useful according to past research. However, most of the existing studies view workplace ostracism as a simple scenario composed of the perpetrator and the victim. These studies only focus on the impact of workplace ostracism from a victim-centric perspective, ignoring other perspectives to a large extent.¹⁸ In fact, the formation of workplace ostracism can be due to a multi-party interaction; it not only involves the victim, but other participants (such as third parties), who will also have a significant impact. Studies have shown that the victim will regard third parties who interact well with the perpetrators as their associates,⁵⁶ which in turn engenders hostility and suspicion towards those who are accepted and recognized by the actor. The victim’s cognition and behavior may be significantly influenced by the interaction between the actor and the third party. Unfortunately, there is a lack of research on the perspective of actors and third parties in the existing literature, which makes us unable to fully reveal the interactive process of workplace ostracism while providing more useful guidance to the management in practice. It is urgent to adopt multiple perspectives to systematically explore the interactive mechanism of workplace ostracism. Therefore, future research could be based on a third-party perspective, exploring what factors influence individuals’ perceptions and behaviors and how they affect third parties’ willingness to intervene in workplace ostracism. From an interactive perspective, the nature of the relation between the perpetrator and the third party could be examined to determine whether it activates the moral compensation effect of the third party. Finally, it could also be possible to examine whether the relationship between the actor and the third party activates the moral compensation effect from the perspective of interaction.