Don't Speak Ill of Others Behind Their Backs: Receivers' Ostracism (Sender-Oriented) Reactions to Negative Workplace Gossip.

PURPOSE: Previous studies on negative workplace gossip have neglected the role of gossip receivers. The current study aims to explore the interpersonal interaction mechanism between gossip receivers and communicators. Drawing on social information processing theory, we propose a theoretical model for the relationships between negative workplace gossip, psychological safety, ostracism, and coworker-exchanges. PARTICIPANTS AND METHODS: Multi-wave data of 386 employees from eight service-oriented companies in China supported the proposed framework. Critical incident techniques and time-lag method were used for data collection. SPSS and Mplus were employed for hypothesis test.
RESULTS: The empirical results indicated that negative workplace gossip was positively related to gossip receivers' ostracism (sender-oriented), among which receivers' psychological safety (sender-oriented) played a mediating role. In addition, the coworker exchange relationship moderated the relationship between negative workplace gossip and ostracism, and the psychological safety of interviewees also plays a mediating role. Specifically, the effect of negative workplace gossip on psychological safety and the mediating effect of psychological safety were stronger when the coworker exchange relationship was higher.
CONCLUSION: Drawing on social information processing theory, the present study constructed a process model of the recipients' sender-oriented ostracism reactions to negative workplace gossip, which helps explain the cognitive psychological mechanism and the boundary conditions of the above "perception-interpretation-behavior" social information process model. The above framework contributes to both theory and practice.

Introduction

Gossip, which has been identified as a special and important interpersonal communication interaction, occupies about two-thirds of individual communication time.1 Workplace gossip, which refers to the positive or negative comments about an absent third party in the workplace,2,3 usually occurs in informal environments such as work breaks and dining.4 While workplace gossip is normally divided into positive gossip and negative gossip, negative workplace gossip tends to arouse greater employee interest.5 Therefore, the current study focused on negative workplace gossip.

Most previous studies on negative workplace gossip have examined the impact on the gossip target, and there has been a high degree of consensus regarding the negative effects. For example, it was found that negative workplace gossip negatively affected the organizational citizenship, proactive behavior, service performance and creativity of the gossip target;6,7 Moreover, it has a negative impact on the physical and mental health of the gossip target, such as increasing their emotional exhaustion and performance pressure and reducing their organizational-based self-esteem and organizational identity.8,9 However, regardless of these previous insights, there are still limited research concerns in several areas.

First, as a complex social phenomenon, the impact of negative workplace gossip can be multifaceted.4 Negative workplace gossip refers to the negative evaluation information of gossip target from the sender to the gossip receiver. There are three subjects in the event: the gossip sender, the gossip receiver, and the gossip target. However, even though the negative gossip observed by organizational members has been found to be greater than their personal experiences, the role of the gossip receiver has been largely ignored, with only a few studies focusing on receivers’ responses to negative workplace gossip.10 For example, some studies have found that the gossip receivers can learn from the normative organizational information contained in gossip, and carry out reflective self-improvement, self-promotion, and self-protection activities to improve their work performances.11,12 However, these studies focused on the impact of the negative workplace gossip on the gossip receivers themselves, while ignoring the employees’ response to the gossip senders after receiving the negative workplace gossip. As the main focus of gossip lies is its influence on the interpersonal relationships within organizations,4 investigating this problem could assist in understanding the bilateral “sender-receiver” interpersonal interaction processes during the gossip dissemination.

Second, most previous studies that from a gossip receiver perspective emphasized the learning effect of negative workplace gossip based on social learning theory.11,12 Therefore, although this research emphasized the role of learning motivation in the recipient employees’ response to negative workplace gossip, it also correspondingly ignored the psychological cognitive mechanism of employees’ reaction to this gossip. Employees tend to make reasoned judgments based on their colleagues’ behaviors, which then forms part of their psychological beliefs about these colleagues,13 with these individual beliefs further affecting their evaluations of and behavioral responses to their colleagues. Therefore, as the psychological cognitive mechanism of the recipient employee determines their understanding of their relationship with the gossip sender and the subsequent interpersonal interactions,14,15 exploring the psychological cognitive mechanism of the recipient employees towards the gossip senders could help open the black box between negative workplace gossip and the recipients’ behavioral responses toward the sender.

In order to better understand the behavioral interaction between recipient employees and gossip senders, as well as the psychological cognitive mechanism of recipient employees in the above process, we further introduce the social information processing theory, which suggests that individuals are adaptive and tend to adjust their attitudes and behaviors to suit the environment,16 to guide the theoretical construction of the current study. Based on the perspective of social information processing, we propose that the negative behaviors of colleagues (such as negative workplace gossip) provide employees with social information about the interpersonal insecurity (sender-oriented) of the interacting objects.17 Therefore, we propose that the negative behavior of gossip senders will encourage gossip receivers to implement workplace ostracism by reducing the psychological security of gossip receivers or the belief that the context is dangerous for interpersonal risk-taking.18,19 Further, drawing on social information processing theory, gossip receiver’s attention to the senders’ motivations will further affect the degree of influence that senders’ negative behaviors have on the recipient’s interpersonal interaction behavior.17,20 Therefore, we further suggest that when the degree of cooperation between the gossip receivers and the senders is low, that is, when the gossip senders and the gossip receivers have a low exchange relationship,21 the negative behavior of the sender will have a greater impact on the workplace ostracism of the gossip receiver. In summary, drawing on social information processing theory, this paper constructed a gossip recipients’ behavior response process model to elucidate the psychological cognitive mechanism with-in the gossip receivers and the boundary conditions for the above “perception-explain-behavior” social information processing process. Specifically, as negative workplace gossip provides negative social information, the recipient employees will act meaning construct about the gossip sender based on this information, and find the gossip sender is unpredictable and threatening. Therefore, they have a low perception of sender-oriented psychological safety, which prompts them to remove themselves from the generated threat by excluding the gossip senders in their subsequent behavior choices. However, high-quality bilateral coworker exchange relationships indicate that gossip senders have high value and may give receivers greater support in their future work, which would strengthen the mediating role of psychological safety and reduce the recipient’s workplace ostracism behavior. By constructing the above moderated mediation model, the present study further clarifies whether, why, and when negative workplace gossip affects the recipient’s sender-oriented behavioral responses such as ostracism.

The current paper makes the following theoretical contributions. First, by shifting the research perspective to the gossip receiver, the current study emphasizes that the need for gossip receivers to process negative social information of negative workplace gossip. The meaning construct is based on the above information to carry out the subsequent behavioral decision-making, which is a beneficial addition to the study of negative workplace gossip from the receiver’s perspective.22,23 Meanwhile, we not only respond to the previous criticism that most negative workplace gossip studies are conducted from a single research perspective, but also respond to the call for research from multiple perspectives in this field.10,11 Second, different from the few studies based on gossip receiver’s perspective, which only focused on the impact of negative workplace gossip on the gossip receivers themselves, this paper focused on the behavioral responses of receivers to senders. The results show that negative workplace gossip is a kind of negative social information. Gossip receivers may reduce the frequency of interaction with gossip spreaders through workplace exclusion and release another kind of social information to gossip spreaders, which means that spreading negative workplace gossip within the organization is not acceptable to colleagues. This enriches the bilateral interpersonal interaction model of “spreader-receiver” during the spread of gossip.4,11 Third, the current research takes psychological safety as the psychological cognitive mechanism between negative workplace gossip and the behavioral response of the receivers and emphasizes the meaning construction process of the receiver in responding to the negative workplace gossip. Furthermore, as high-quality bilateral coworker exchange relationship indicates the high value of the gossip senders, this kind of social information can enhance the role of psychological safety in the recipient employee’s behavioral responses. Therefore, based on the logic of “perception-interpretation-behavior”, this study constructed a moderated mediation model that enriches the mediating mechanism of recall-related negative workplace gossip and its subsequent impact on employee behavior24 and further enriches the antecedents of workplace ostracism.25

Theory and Hypotheses Development

Social Information Processing Theory

Social information processing theory claims that people are adaptive individuals who tend to adjust their attitudes and behaviors to suit their environments.16 As the environment in which individuals operate is often their source of information, by coding and interpreting this information, people respond accordingly. Moreover, different ways of information processing can result in different attitudes and behaviors.26 Therefore, the employee attitudes and behaviors in the workplace largely depend on their social environment and information processing in the specific environment rather than their individual needs.27 Specifically, employees construct their work behavior norms by observing and explaining others and shape their behaviors based on their decision-making process, which are influenced by their overall interpretation of the working environment. That is, individual’s attitudes and needs are cognitive products generated from their own social information processing, with their specific observations, past experiences and colleague comments on work, which all constitute potential social information sources.28

Workplace job characteristics, such as the supervisor’s leadership style, the colleagues’ job attributes and the work meaning, are often vague and difficult for employees to directly judge. Therefore, employees may use social information such as supervisors’ interpretations, peer reviews, and their own past experience to construct meanings for the job characteristics,29,30 which means that an employee’s judgment of things is affected by the vagueness of the judgment standard. When criteria are vague, the employees need to obtain diagnostic clues from the surrounding environment, with the resulting attitude towards the behavior of the other members of the organization and their working characteristics belonging to this vague standard.26 In short, social information is the information and suggestions conveyed by the individuals around the employees, while the attitude and work characteristics of the members of the organization belong to this fuzzy standard. Therefore, employees will make judgments based on their actual feelings regarding the job characteristics and social information and then choose appropriate behaviors based on these judgments.16,31 Accordingly, we construct the theoretical framework of this paper based on social information processing theory. The research model is shown in Figure 1.

Figure 1

Theoretical framework.

Negative Workplace Gossip

Negative workplace gossip occurs when an employee spreads negative comments about another member of the organization. Therefore, based on this definition, negative workplace gossip is one-way communication from the sender to the recipient, but does not involve any further interaction between these two parties.9,24,32 If the recipient continues to spread the negative workplace gossip, they receive, they then become senders in the new gossip dissemination round. The characteristics of gossip require that it must be aimed at a specific employee rather than an event or a related environment and the gossip content must be essentially negative and about third-party employees,9 which means that any dissemination of factual colleague information is not gossip. For example, commenting about an employee’s lateness is not negative gossip as it is an established fact. However, if the gossip sender says that “the employee is deliberately late and despises leadership and organizational discipline”, this is negative workplace gossip as it includes a negative evaluation of the third-party employee.33

Moreover, because of the destructive effect of negative workplace gossip on the victims, it has usually been classified as social mistreatment.5,34 However, as a particular psychosocial construct, negative workplace gossip could be seen to be different from other types of social mistreatment or informal communication. First, other types of mistreatments (such as bullying and abusive supervision) usually manifest as direct public mistreat.35 However, as negative workplace gossip occurs in the absence of the target employee, it is therefore implicit and indirect in nature, which means that it is an indirect form of social mistreatment that could lead to moral, psychological and behavioral responses by the bystanders (gossip receivers).36 Second, as mentioned above, negative workplace gossip involves negative evaluation. The target of gossip is organizational employees. In an environment where the gossip target is not present, information will flow between the sender and the receiver.32,37 However, informal communication (such as chatting) is usually unintentional, less evaluative and can occur in public places.38,39 Third, as the gossip target is generally unable to identify the origin of the negative workplace gossip to verify the content, the sender usually avoids any direct confrontation with the gossip target.32 Finally, negative workplace gossip tends to produce greater uncertainty than other types of social mistreatment and informal communication.40

Receivers’ Sender-Oriented Ostracism Reactions to Negative Workplace Gossip

Social information processing theory claims that the social environment provides the background for employees to construct meaning about what is acceptable to the organization and which behavior is more important.2 Meanwhile, employees’ attitude towards other members of the organization is characterized by vague judgments, so they need to obtain diagnostic clues from the work environment. Therefore, in the context of this study, negative workplace gossip that often occurs in the workplace as the important information source for the meaning construction of the gossip receivers towards the senders.

Specifically, because negative workplace gossip involves a negative evaluation of absent colleagues, it is a type of social mistreatment that can destroy team cooperation, violate organizational ethics, and threaten organizational stability.41,42 Therefore, as employees usually work together to achieve the organization’s vision,43 negative workplace gossip can violate this expectation as it is a social message that the gossip senders are not really concerned about the employees.44 Consequently, negative workplace gossip can destroy the goodwill among colleagues, which is a necessary condition to promote cooperation toward a common organizational vision;45 That means when employees spread negative workplace gossip, their willingness to work with the team is questioned and thus the recipient of the gossip will show lower the degree of trust in such employees.46,47 Further, the recipient employees believe that the gossip senders are violating the organizational norms and are therefore highly unpredictable and uncontrollable,48 which may lead to the workplace ostracism of the gossip sender.42

Furthermore, social information processing theory also claims that employees socially construct their work behavior norms through their observation and interpretation of others, which in turn informs their consequent behavioral decisions.16 Therefore, it is surmised that after receiving negative workplace gossip, the recipient employees interpret and evaluate the behavior of gossip senders, which then constructs their meaning as the spread of negative workplace gossip illuminates the unpredictability of the gossip senders. When the employee receives the social information, they are often reluctant to establish social contact with the gossip sender as they believe that the gossip sender is violating the organizational ethics.49 Specifically, it is further surmised that the gossip receivers escape from the gossip senders by practicing workplace ostracism.50 As this type of social exclusion could be seen as a low-cost punishment measure, it is usually used to punish employees who violate ethics and bring threat perception to organization members, with the purpose of promoting the behavior of violators to be consistent with social exclusions. Therefore, we suggest that the gossip receiver should reduce their interactions with the gossip sender through workplace ostracism, and uses it as a form of punishment to send a social information to the gossip sender that spreading negative workplace gossip within the organization is not acceptable.

Accordingly, the following hypothesis is proposed:

Hypothesis 1: Negative workplace gossip is positively correlated with the recipient employees’ workplace ostracism. (sender-orientation)

Mediating Mechanism of Psychological Safety (Sender-Oriented) by the Gossip Receivers

The social information processing theory recognizes that employees are not independent within an organization as the organizational environment provides various social information affecting their attitudes and behaviors. So, employees will make meaningful construction with the information of peers’ behaviors, norms, expectations and values in organization environment, which in turn guides their workplace behaviors.16 Therefore, when the gossip receiver carries out exclusive workplace behavior to the gossip sender, there should be an inherent psychological cognitive mechanism associated with the meaning construction processing of social information. Specifically, it is surmised that the perceived psychological safety (sender-oriented) of the recipient employees can explain the internal path between negative workplace gossip and workplace ostracism. Consistent with previous studies on interpersonal relationship, psychological safety is defined as a person’s perception of whether they can risk communicating with the target in the current situation;19,20,51 Therefore, in this paper, psychological safety is related to the employee’s assessment of their interpersonal workplace risk. Studies have found that a higher psychological safety can promote communication between colleagues, thus promoting knowledge sharing within the organization and improving organizational performance.52–54

However, as negative workplace gossip violates the mutual respect given in organizational ethics and harms positive work environments,55 when gossip receivers are exposed to social information, they doubt the personal qualities and professional ethics of the gossip sender. As the attitude of the recipient employees largely depends on their social information processing, negative information may cause their negative attitude, which gives rise to negative emotions towards the gossip sender, which may lead to negative emotions towards the gossip sender, thus causing the receiver to pay more attention to the negative evaluation of the gossip sender. The receiver may feel that the gossip sender is bringing uncertainty to their own careers and well-being,56 which reduces their perception of psychological safety towards the gossip sender.

Moreover, because of the difficulty to trace the origins of negative workplace gossip, it is hard for the gossip target to identify who is slandering them,40 which means that the gossip senders are able to arbitrarily deride their colleagues without having to worry about conflicting the positive relationships they may have with the gossip target.57 When a gossip receiver sees a colleague spreading negative workplace gossip, they may view the gossip sender as a hostile and threatening colleague, doubt the meaning of continuing to communicate with them, and feel at risk of becoming a gossip target.

Accordingly, we propose the following hypothesis:

Hypothesis 2: Negative workplace gossip is negatively related to the recipient employees’ psychological safety. (sender-orientation)

Psychological safety is a core psychological communication mechanism among employees in workplace.58,59 When individuals feel safe, they are more likely to share their opinions, suggestions and doubts with their colleagues;18,60 However, if they feel threatened by their colleagues, they are reluctant to communicate.61 Studies have found that when employees receive uncivilized information (such as negative workplace gossip), their behavioral cognition is affected that they tend to make reactions using uncivilized behavior (such as workplace ostracism) in subsequent social interactions.62 In other words, gossip receivers having a low perception of their psychological safety with the gossip senders may implement self-protective behaviors such as workplace ostracism for fear of becoming the target of negative gossip for receivers.63 This means that these behaviors depend on the interpretation and evaluation of the social information. As mentioned above, the recipient employees will choose to exclude the gossip sender to stay away from the threat perception state with low psychological safety for self-protective motivation.

Accordingly, the following hypotheses are proposed:

Hypothesis 3: The psychological safety of recipient employees (sender-oriented) is negatively related to their workplace ostracism behavior. (sender-oriented)

Hypothesis 4: The psychological safety of the recipient employees (sender-oriented) mediates the relationship between negative workplace gossip and workplace ostracism. (sender-oriented)

Moderating Role of the Coworker Exchange Relationship Quality

It is worth noting that not all employees who spread negative workplace gossip will be excluded by those employees who receive it. Drawing on the social information processing theory, individual’s past work experience is also an important information source in the process of employee meaning construction.64 Therefore, the quality of bilateral relationship is also an important diagnostic information for the receiver to interpret and evaluate the gossip sender. In view of this, we suggest that the quality of the bilateral relationship between the sender and the receiver will have a certain impact on the cognition and behavior decision-making of the gossip receiver.

High-quality bilateral coworker exchange relationships are based on positive interaction histories, lower interpersonal risks, higher expectations of social rewards and instrumentality (similar to reciprocity) between the two parties. A positive bilateral relationship represents a social information that the colleague’s future actions will be beneficial, helpful, or at least not detrimental to his or her interests.65 In other words, positive bilateral coworker exchange relationships focus attention on the positive characteristics of colleagues. If an employee does not believe that they will be rewarded for helping their colleagues, the employee is unlikely to continue to interact with those colleagues.66 Therefore, when the gossip sender-recipient exchange relationship is of high quality, the recipient seeks indirect social information regarding the negative gossip transmission to complete their meaning construction of the sender, which means that the negative gossip of the sender may only have a small negative impact on the psychological safety of the recipient. Conversely, a low-quality gossip sender-recipient exchange relationship indicates a history of negative contact and as the receivers cannot rationally interpret the behavior of the gossip sender, they are more inclined to give them excessive negative evaluations, which would have a stronger negative impact on the psychological safety of the recipient.

Accordingly, the following hypothesis is proposed:

Hypothesis 5: The quality of the bilateral coworker exchange relationship has a moderating effect on the relationship between the negative workplace gossip and the psychological safety of the gossip receiver. Specifically, when the relationship quality is high, the negative correlations between negative workplace gossip and the gossip receiver’s psychological safety is weakened, vice versa.

Only when colleagues have had mutually positive interactions in the past are they likely to have higher exchange relationship quality and further communication intentions.67 At the same time, as it is costly to exclude certain colleagues in the workplace, especially when the gossip sender and recipient need to work together.68 In this case, the gossip receiver needs to make a comprehensive judgment based on multiple information sources for meaning construct and choose the most suitable behavior. First, a higher exchange relationship quality means that the gossip senders are of high value to the receiver and may give them greater support in the future. Second, as the consequences of excluding the gossip spreaders with which the receiver has a high exchange relationship quality could threaten the receiver’s career development, the receiver does not view the dissemination of the negative gossip as the only information source to determine their attitude and actions. As a result, receivers are more inclined to judge from their past interactions, which means that they would be far less likely to ostracize the gossip senders in the workplace. Therefore, the gossip recipient’s meaning construction is based on the importance placed on the (past, present and future) interactions with the gossip sender and the related social information. Based on the above and relevant mediating and moderating effect studies, this study further proposes the following a moderated mediating hypothesis.

Hypothesis 6: The quality of the bilateral coworker exchange relationship moderates the mediating effect of psychological safety between negative workplace gossip and workplace ostracism. Specifically, the higher the relationship quality, the stronger the mediating role of psychological safety.

Methods

Samples and Procedure

Considering the overarching theory of our research is the Social Information Processing Theory, an within-person theoretical framework, which means that our model should be tested at the individual level.69 This study uses a combination of questionnaire surveys and critical incident techniques for data collection. Related research shows that the above method can effectively assess individual perceptions and responses to specific incidents.70 Previous studies have shown that negative workplace gossip occurs frequently in service-oriented industries.51,71 Therefore, the data of the present study were collected from 8 large service-oriented companies in china. With the help of the human resources directors of eight companies, our questionnaires were distributed through the company’s mail system. The survey plan is as follows: First, the researchers of our survey group discussed with the companies’ human resources specialist to determine the entire investigation process. Secondly, 450 employees were randomly selected to participate in our questionnaire survey based on the list of personnel of each company. One thing to point out here is that we have stated in the email that every employee has the right to refuse to participate in the questionnaire survey. In order to avoid the common method variance, we follow the previous studies adapting a time-lag longitudinal tracking research design for the sample collection, which means that our variables should be measured in different times.47,72,73 Specifically, our data were collected on three different time nodes, with two weeks between each time node, and the entire sample collection process lasted for four weeks.74

The specific sample collection process is as follows: Before the first questionnaire phase, the interviewed employees were asked to recall in as much detail as possible incidents in the past month that involved colleague “S” spreading negative workplace gossip about colleague “T”. Then in the formal questionnaire phase, the recipient employees in the first stage were asked to evaluate their perceived negative workplace gossip. Overall, 450 questionnaires were distributed and 423 returned. Two weeks later, the same recipient employees were again asked to evaluate their own psychological safety perceptions regarding colleague S (gossip sender) and their relationship quality with colleague S (coworker exchange), for which 423 questionnaires were distributed, and 401 returned. Then, two weeks later, the previous questionnaire was distributed again, and the recipient employees were asked to evaluate their ostracism of colleague S in the workplace and completed some background information, for which 386 valid questionnaires were returned. Table 1 gives information about the sample participants.

Table 1

Demographic Information (n=386)

Features	Category	Quantity	Percentage
Sex	Male	247	64.0
	Female	139	36.0
Age	25 years old and below	55	14.2
	26–35 years old	107	27.7
	36–45 years old	190	49.2
	Over 46 years old	34	8.8
Education	Senior high school and below	23	6.0
	Training school	103	26.7
	Undergraduate	214	55.4
	Postgraduate and above	46	11.9
Tenure	1 year and below	53	13.7
	1–2 years	43	11.1
	2–3 years	12	3.1
	3–4 years	13	3.4
	4–5 years	220	57.0
	Over 5 years	45	11.7

Measures

The study adopted the Likert five-item scale to measure the above dimensions. We mainly refer to the authoritative scales. The words were modified based on the pre-investigation.

(1) Negative workplace gossip

The scale developed by Brady, Brown, Liang24 was employed to measure the negative workplace gossip that employees perceived from their colleagues. Based on the gossip attributes (positive vs negative) and the gossip spread object (colleague vs supervisor), Brady, Brown, Liang24 divided the gossip into four dimensions; therefore, the “Negative Workplace Gossip About Colleagues” construct was employed to measure the research object in this paper, for which there were five items; For example, “Colleague S talked about another colleague’s disgraceful actions when talking to me” with the scale ranging from 1=strongly disagree to 5=strongly agree”. The Cronbach’s Alpha for this scale was 0.901.

(2) Psychology safety

In order to measure the psychological safety of the recipient employees against negative workplace gossip, we employ the 5-item scale developed by Liang, Farh, Farh,60 which was measured from 1=strongly disagree to 5=strongly agree. An example item was “I feel free to express my true thoughts to colleague S at work. The Cronbach’s Alpha for this scale was 0.897.

(3) Workplace ostracism

We reference the scale of Zhou, Fan, Cheng, Fan33 to measure the workplace ostracism behavior (sender-oriented) of gossip recipients. There are 6 items in the scale. One example item is “I only interact with my colleague S when I need it; 1=strongly disagree, 5=very agree”. The Cronbach’s Alpha of this scale is 0.883.

(4) CEX (coworker-exchange relationship)

We use the scale developed by Sherony, Green21 to measure CEX. The scale is mainly used to evaluate the exchange quality between the gossip sender and receiver. The scale has a total of 6 items, one of the sample items is “How would you describe the relationship between you and your colleague S? 1=very bad, 5=very good”. The Cronbach’s Alpha of this scale is 0.915.

(5) Control variables

Studies have found that employees’ background variables, such as age, gender, education level, and tenure, can affect employee social information processing.75–77 Therefore, gender (1 for male, 2 for female; age: coded from 1 to 4, respectively, for 25 years and below, 26–35 years old, 36–45 years old, 46–55 years old, and 56 years old and above), education level (1 = high school and below, 2 = college, 3 = undergraduate, 4 = postgraduate and above) and working years (1 = 1 year and below, 2 = 1–2 years, 3 = 2–3 years, 4 = 3–4 years, 5 = 4–5 years, and 6 = 5 years and above) were used as the control variables in this study.

Results

Common Method Variance

For the sake of controlling the common method variance, the sample was anonymized at the collection stage. However, even though the variables in the theoretical model are all reported by the gossip recipients, the relationship may still be affected by common method variance. In order to test the effect of common method variance, we used Harman’s single factor method to perform factor analysis on all variables and isolate the unrotated common factors. Six factors with a characteristic root greater than 1 were found, with the first factor accounting for 32.58% of the total load of all factors, and the total explained variance is 69.27%, which met the requirement that the maximum extracted variance should be less than the 50% critical value of the total explained variance and proved that the common method variance was in an acceptable range and could be statistically analyzed.

Confirmatory Factor Analysis (CFA)

Mplus was employed for the confirmatory factor analysis (CFA) to test the construct validity of the main variables: negative workplace gossip, psychological safety, workplace ostracism, and CEX. Based on the single-factor, two-factor, three-factor, and four-factor model comparisons, the fitting indicators for the four-factor model all passed the threshold standard and were significantly better than the other models; therefore, the variables had good discrimination validity. See Table 2 for the details.

Table 2

CFA to Examine the Factor Structure of the Scales Used in the Study (n=386)

Model	χ2	df	CFI	SRMR	RMSEA	Model Comparison Test
						Model Comparison	∆χ2	∆df
1. Four factors: NWG; PS; WO; CEX	647.867	203	0.907	0.057	0.075
2. Three factors a: NWG; PS+CEX; WO	1350.957	206	0.762	0.107	0.120	2 vs 1	703.090	3
3. Three factors b: NWG+WO; PS; CEX	1489.110	206	0.733	0.125	0.127	3 vs 1	841.243	3
4. Three factors c: NWG+PS; WO; CEX	1325.400	206	0.767	0.101	0.119	4 vs 1	677.533	3
5. Two factors: NWG+PS+WO; CEX	2079.289	208	0.610	0.122	0.153	5 vs 1	1431.422	5
6. Single factor: NWG+ PS+ WO+ CEX	2769.924	209	0.467	0.143	0.178	6 vs 1	2122.057	6

Note: “+” indicates combination of factors.

Abbreviations: NWG, negative workplace gossip; PS, psychological safety; WO, workplace ostracism; CEX, coworker exchange relationship.

Correlation Analysis

We use SPSS software to perform descriptive statistics and correlation analysis on data variables. There are two reasons for this: (1) If the correlation coefficient between variables exceeds 0.75, it indicates that there may be multicollinearity between the variables; (2) The significance of the correlation coefficient will also be affected by the sample. Therefore, this study calculated the mean and standard deviation of all variables and conducted correlation analysis between the variables. Table 3 shows the mean, standard deviation and correlation coefficient of each variable.

Table 3

Means, Standard Deviations, and Inter-Correlations Between the Study Variables (n=386)

Variables	Mean	SD	1	2	3	4	5	6	7	8
Sex	1.64	0.48	1
Age	3.52	0.84	0.039	1
Education	2.73	0.74	−0.051	−0.070	1
Tenure	4.14	1.64	−0.111*	−0.156**	−0.244***	1
NWG	1.74	0.71	−0.111*	−0.063	0.022	−0.047	(0.727)
PS	3.81	0.73	0.026	0.029	−0.036	−0.015	−0.451***	(0.709)
WO	2.38	0.77	−0.053	−0.085	0.229***	−0.080	0.372***	−0.352***	(0.649)
CEX	3.74	0.95	0.016	0.155**	−0.087	0.020	−0.366***	0.426***	−0.439***	(0.709)

Notes: *p < 0.05, **p < 0.01, ***p < 0.001; The diagonal brackets are the AVE value.

Abbreviations: NWG, negative workplace gossip; PS, psychological safety; WO, workplace ostracism; CEX, coworker exchange relationship.

According to the data presented in Table 3, we found that negative workplace gossip is significantly positively correlated with workplace ostracism (r=0.372, p<0.01); negative workplace gossip is significantly negatively correlated with psychological safety (r=−0.451, p<0.01); psychological safety is negatively correlated workplace ostracism (r=−0.352, P<0.01). It can be seen that the correlation coefficient between the variables in this study did not exceed 0.75, so the preliminary judgment is that there is no problem of multiple collinearities. At the same time, the correlation between these variables is basically consistent with the theoretically expected relationship of this study, and it also lays the foundation for the subsequent testing of mediation and regulation.

Hypothesis Testing

In order to test the influence of negative workplace gossip on the recipient’s employees’ workplace ostracism (sender-oriented), the mediating role of the gossip receiver’s psychological safety (sender-oriented) and the moderating role of coworker relationship (sender-receiver), SPSS was used to conduct stepwise regression analysis on the research data collected by the research group firstly. Next, we conducted a bootstrap analysis using Hayes’ PROCESS for SPSS to test the robustness of our model.78,79

First, a main effect test examined the impact of the negative workplace gossip on the recipient’s workplace ostracism (sender-oriented). After setting workplace ostracism as the dependent variable, model 4 as the regression model for the control variable against workplace ostracism, and model 5 as the regression model for workplace ostracism after adding the independent variables (negative workplace gossip), it can be seen in Table 4 that in Model 5, negative workplace gossip was found to have a significant positive impact on workplace ostracism (β=0.351, P<0.01). The maximum VIF (variance inflation factor) of each variable in the regression equation was 1.121, which was far less than 10, indicating that there was no multicollinearity; therefore, Hypothesis 1 was verified.

Table 4

Regression Test of Receiver’s Workplace Ostracism (n=386)

	Workplace Ostracism
	Model 4	Model 5	Model 5.1	Model 6	Model 6.1	Model 7	Model 7.1
Sex	−0.047	−0.006		−0.039		−0.013
Age	−0.077	−0.052		−0.070		−0.054
Education	0.207***	0.208***		0.195**		−0.199**
Tenure	−0.056	−0.031		−0.063		−0.043
NGW		0.351***	0.363***			0.250***	0.261***
PS				−0.334***	−0.343***	−0.221***	−0.225***
R2	0.062	0.182	0.131	0.173	0.118	0.221	0.172
Adjusted R2	0.052	0.172	0.129	0.162	0.115	0.209	0.167
F	6.245***	16.958***	58.103***	15.862***	51.130***	17.946***	39.672***
ΔR2		0.121	0.069	0.111	0.056	0.039	0.054
VIF (MAX)	1.115	1.121	1.000	1.116	1.000	1.286	1.257

Notes: **p < 0.01, ***p < 0.001.

Abbreviations: NWG, negative workplace gossip; PS, psychological safety.

The second mediation test examined whether psychological safety mediated the relationship between negative workplace gossip and workplace ostracism. Model 2 in Table 5 shows that negative workplace gossip has a significant negative effect on psychological safety (β=−0.458, P<0.01). As the maximum VIF of each variable in the regression equation was 1.121, which was far less than 10, there was no multicollinearity; Therefore, Hypothesis 2 was supported. In Model 6, psychological safety was observed to have a significant negative effect on workplace ostracism (β=−0.334, p<0.01). As the maximum VIF for each variable in the regression equation was 1.116, which was far less than 10, there was no multiple collinearity; Therefore, Hypothesis 3 was supported. After adding the mediator (psychological safety) to Model 7, the influence of negative workplace gossip on workplace ostracism (β=0.250, p<0.01) was lower than in Model 5 (β=0.351, P<0.01) but still significant and psychological safety still had a significant negative effect on workplace ostracism (β=−0.221, p<0.01). As the maximum VIF of each variable in the regression equation was 1.286, which was far less than 10, which means that there was no multicollinearity; Therefore, Hypothesis 4 was initially supported. To further verify the mediating effect of psychological safety on negative workplace gossip and workplace ostracism, the PROCESS program designed by Hayes was used to test the significance of the mediating effect (see Table 6). The 95% confidence interval was found to be [0.0491, 0.1647] (excluding zero), that is, psychological safety significantly mediated negative workplace gossip and workplace ostracism, which further supported Hypothesis 4.

Table 5

Regression Test of Receiver’s Psychological Safety (n=386)

	Psychological Safety
	Model 1	Model 2	Model 2.1	Model 3	Model 3.1
Sex	0.022	−0.031		−0.018
Age	0.022	−0.011		−0.048
Education	−0.038	−0.038		−0.018
Tenure	−0.020	−0.053		−0.067
NGW		−0.458***	−0.452***	−0.374***	−0.36***
CEX				0.308***	0.302***
NGW*CEX				0.102*	0.096*
R2	0.003	0.208	0.205	0.296	0.291
Adjusted R2	−0.008	0.198	0.203	0.283	0.285
F	0.275	19.991***	98.815***	22.702***	52.155***
ΔR2		0.205	0.202	0.088	0.087
VIF (MAX)	1.115	1.121	1.000	1.227	1.207

Notes: *p < 0.05, ***p < 0.001.

Abbreviations: NWG, negative workplace gossip; CEX, coworker exchange relationship.

Table 6

Bootstrap Results for the Mediation Effect (n=386)

Direct Impact of Negative Workplace Gossip on Receivers’ Workplace Ostracism
Effect	S.E.	t	p	LLCI	ULCI
0.2499	0.0514	4.8623	0.0000	0.1489	0.3510
Indirect Impact of Negative Workplace Gossip on Receivers’ Workplace Ostracism
	Effect	Boot SE	Boot LLCI	Boot ULCI
Psychological safety	0.1014	0.0293	0.0491	0.1647

Notes: LLCI and ULCI indicate the minimum and maximum values of the confidence interval. This study uses bootstrap for random sampling 5000 times.

Abbreviation: S.E., standard error.

The moderating effect test found that for Model 3, the interaction term (negative workplace gossip * CEX) had a significant impact on psychological safety (β=−0.102, p<0.05), indicating that CEX moderated the impact of negative workplace gossip on psychological safety. As shown in Figure 2, the higher the CEX, the weaker the negative impact of negative workplace gossip on psychological safety. The maximum VIF of each variable in the regression equation was 1.227, which was far less than 10, indicating that there was no multicollinearity; therefore, Hypothesis 5 was verified.

Figure 2

Moderating effect of the coworker exchange relationship.

Finally, there is a moderated mediation model test. To verify the moderated mediation model proposed in the article, we use the PROCESS plug-in and the Bootstrap method to test whether CEX moderates the mediation effect of psychological safety. It was found that the moderate mediating effect point estimate was 0.0204, and the 95% confidence interval was [0.0008, 0.0457], excluding zero. When these results were combined with Table 7, as CEX went from one standard deviation below the mean to one standard deviation above the mean, the mediating effect of psychological safety increased significantly. In other words, when CEX was at a higher level, the mediating effect of psychological safety was stronger, indicating that a mediating effect was established, which verified Hypothesis 6. Using Edwards80 Mediated Interaction Effect Drawing Method, the adjusted mediation effect diagram corresponding to Hypothesis 6 is shown in Figure 3.

Table 7

Bootstrap Results for the Moderated Mediation Effect (n=386)

	Conditional Indirect Effect				Moderated Mediator
	Estimate	Boot SE	BC 95% CI		INDEX	S.E.	BC 95% CI
Low CEX	0.0623	0.0219	0.0276	0.1132	0.0204	0.0112	0.0008	0.0457
Middle CEX	0.0827	0.0240	0.0409	0.1355
High CEX	0.1031	0.0304	0.0522	0.1739

Note: This study uses bootstrap for random sampling 5000 times.

Abbreviations: CEX, coworker exchange relationship, low CEX represents mean “−1”; SD, standard deviation, high CEX represents mean “+1” SD; S.E., standard error; BC, biased corrected; CI, confidence interval.

Figure 3

Conditional effect of negative workplace gossip on workplace ostracism at values of CEX.

Discussion

Based on the theory of social information processing, the process model of recipient employees’ ostracism (sender-oriented) reactions to negative workplace gossip was constructed to explain the cognitive psychological mechanisms of the gossip recipients in the above process and determine the boundary conditions for this “perception-interpretation-behavior” social information processing path. Specifically, we suggested that as negative workplace gossip provided negative social information, recipient employees would find that gossip sender unpredictable and threatening, which would result in a low perception of psychological safety. It was also surmised that the gossip recipient would choose to exclude these gossip senders to reduce their feelings of low psychological safety. However, if the recipient and the gossip sender had a high-quality coworker exchange, the gossip recipient may consider that the gossip senders could be useful in the future, which would reduce the recipient’s intention of ostracizing the gossip sender. The following contributions arise from the above findings.

Theoretical Implications

First, most previous studies on negative workplace gossip have examined the impact on the gossip target, and there has been a high degree of consensus regarding the negative effects.6,7 However, the above research ignores the role of the gossip receiver. By shifting the focus to the perspective of the gossip receiver, the current study emphasizes that the gossip receiver needs to process the negative social information of negative workplace gossip, and use it as a basis for meaning construct, which further determines their behavioral decisions-making. Therefore, the present research revealed the influence of negative workplace gossip on another important group: gossip receiver, which is not only a useful supplement to negative workplace gossip research from the recipient’s perspective, but also response the recalls for research from multiple perspectives.10,11 The proposed theoretical framework provides an understanding of whether, why, and when negative workplace gossip affects the recipient’s actions to implement workplace ostracism (sender-oriented).

Second, the empirical results proved that negative workplace gossip can result in the gossip recipients implementing workplace ostracism behaviors against the gossip senders. To date, limited literature has carried out preliminary explorations on negative workplace gossip from the perspective of recipients. For example, previous studies have shown that gossip recipients can learn from negative workplace gossip, especially when they may refer to the organizational normative information in the gossip, which further leads themselves for self-improvement, self-promotion, self-protection and performances improving through reflective learning.4,11 However, these studies focused on the influence of the negative workplace gossip on the gossip recipients themselves, ignoring the gossip recipient’s responses to the gossip senders after receiving the negative workplace gossip. Drawing on social information processing theory, the current study showed that negative workplace gossip was a kind of negative social information. Meanwhile, gossip recipients may reduce their interactions with gossip senders through workplace ostracism as punishment to inform the gossip senders that spreading negative workplace gossip is not acceptable. From the perspective of internal communication, we construct a bilateral interpersonal behavior interaction model of “send-receive“ in the process of gossip dissemination, which provides us with a better understanding of the influence of negative workplace gossip on the complexity of social networks in organizations.

Third, previous studies based on the perspective of gossip receivers mostly emphasized the learning effect of negative workplace gossip according to the social learning theory12 and emphasized the role of the recipients’ learning motivations and beliefs in response to the negative workplace gossip, ignoring the role of “sender-receiver” bilateral relationship. It is noteworthy that the current study theorized the psychological safety as a psychological cognitive mechanism between the negative workplace gossip and the receiver’s behavior and emphasized the meaning construction process of the gossip receiver in dealing with negative workplace gossip. Furthermore, high-quality coworker exchange relationships indicate that gossip senders have a high value, and this kind of social information will enhance the role of psychological safety in the recipient’s behavioral response process. Therefore, the proposed “perception-interpretation-behavior” model responds to recent appeals for mediation mechanisms that highlight the factors explaining the behavioral responses to negative workplace gossip.24

Fourth, the developed framework extended the research on the antecedent variables of workplace ostracism. Specifically, past studies mostly focused on the perspectives of personality traits and organizational climate, emphasizing that factors, such as the unstable emotional state of employees, uncivilized behavior and competitive organizational climate, will lead to workplace ostracism.81–83 Present research showed that when employees spread negative workplace gossip, this behavior may lessen their colleagues’ perceptions of psychological safety that may lead to their colleagues’ exclusion, which further enriches the antecedent variables for workplace ostracism in response to the call for Mao, Liu, Jiang, Zhang.25

Practical Implications

The research conclusions of this article also provide a useful reference for the improvement of management practices in the workplace. First of all, our research results provide empirical support for the following facts: employees who spread negative workplace gossip will be excluded by observers in the organization. Therefore, in the workplace, employees should try not to speak badly about their colleagues behind their backs; otherwise, they are likely to be squeezed out by colleagues, who receive above negative workplace gossip. At the same time, a large number of past studies have shown that workplace ostracism will bring a lot of negative effects on individuals and organizations, such as reducing the performance and work engagement of excluded employees.84,85 This means that speaking ill of colleagues behind the back may bring a series of continuous negative effects to oneself, which should be eliminated. Therefore, this study reminds managers that they should pay full attention to the role of gossip recipients in the organization, so as to avoid the spiral negative impact on the organization caused by negative workplace gossip.

Secondly, our test results of mediation mechanism show that: Negative workplace gossip will reduce the recipient’s employees’ sense of psychological security, leading to workplace ostracism. This result has important enlightenment for supervisors to conduct team management. In particular, supervisors should regularly hold team building activities to improve organizational cohesion and generate a cooperative atmosphere, so as to strengthen mutual understanding and trust among team members, promote employees’ psychological safety. Through the above measures, supervisors can reduce the negative impact of negative workplace gossip to a certain extent.

Finally, our test results of the moderating effect and the moderated mediating effect both show: higher-quality coworker exchange relationships can help alleviate workplace ostracism caused by negative workplace gossip. Therefore, the supervisors could enhance the understanding among team members through team building activities, so that employees can have a more objective and comprehensive understanding of their colleagues on the basis of bilateral interactions, rather than resorting to gossip spreading as their own information source. In short, if leaders pay attention to these three aspects, they could promote the long-term, healthy development of their organization. By doing so, the coworker exchange relationship quality between employees could be effectively improved, which may help to resist a series of negative effects brought about by negative workplace gossip.

Limitations and Future Research

This study had some limitations. First, the data were collected from eight large service industries. Future research could collect samples from other industries to further verify our theoretical framework.

Second, a multi-wave longitudinal approach was used, and all core variable measures were reported by the participants. Although the common method variance test (CMV) met the basic requirements, a multi-source data collection could be applied in the future research; for example, the core variables could be reported by both the gossip receivers and the senders, and multi-level regressions such as hierarchical linear regression or a random coefficient method could be used to further verify the theoretical model.

Third, an important theoretical contribution of this article was to examine the effect of the negative workplace gossip on the receivers’ meaning construction process and consequent behavior. However, the study failed to take any possible additional mechanisms into account. Some previous studies based on social learning theory conducted from the gossip receivers’ perspective emphasized the learning effect of the negative workplace gossip.12,86 Therefore, it is crucial for future research to identify the indirect relationships between negative workplace gossip and ostracism through the psychological cognitive mechanism is incremental to the learning mechanisms. Future research could further develop the social information processing theory model by controlling the above mechanism.

Fourth, the consideration of the boundary conditions for the receivers’ meaning construction process mainly emphasized the bilateral coworker exchange relationship (sender and receiver). However, negative workplace gossip involves a trilateral “sender-receiver-target” communication relationship. Therefore, future research could investigate the bilateral “sender-target” and “receiver-target” relationships, analyze the differences between the three types of negative workplace gossip bilateral relationships, clarify the reasons for the differences, and construct a three-sided negative workplace gossip dynamic interaction model based on the analysis.

Fifth, compared with previous studies, the current paper constructed a theoretical negative workplace gossip relationship model for the recipient’s workplace ostracism behavior based on social information processing theory and three-waves of data collected to improve the reliability of the results. However, due to the research method limitations, the results were unable to eliminate the possibility of reverse causality, that is, whether the negative workplace gossip induced the recipient employees to enact workplace ostracism behaviors or whether the recipients’ workplace rejection behavior induced the negative workplace gossip. Therefore, in the future research, an experimental method or a two-way interaction mechanism could be employed to examine the causal relationships more deeply and between the variables.

Conclusion

Drawing on social information processing theory, we constructed a process model of the recipients’ sender-oriented ostracism reactions to negative workplace gossip, which helps explain the cognitive psychological mechanism and the boundary conditions of the above “perception-interpretation-behavior” social information process model. Based on a longitudinal tracking survey of 386 service-oriented employees, it was found that negative workplace gossip had a significant positive impact on the gossip recipients’ workplace ostracism (sender-oriented); the recipients’ psychological safety (sender-oriented) played a mediating role in these relationships. In addition, the bilateral coworker relationship moderates the mediating role of psychological safety (sender-oriented). Specifically, the higher the bilateral coworker relationship, the greater the mediating role of psychological security.