Does bullying increase compliance?

Abstract

Social ostracism can be a form of bullying at the workplace (a process of frequent and repeated acts of hostile communication and humiliation of an employee). Previous findings suggest that experimentally evoked ostracism leads to compliance. The aim of these studies was to examine willingness to comply among bullying targets. It was found that being subjected to bullying is connected with lower proneness to comply with various requests of coworkers (the first study, N = 197). A drop in the self-reported compliance rate occurred among those bullied participants who were presented with a description of various types of social exclusion at the workplace (second study, N = 309). It is argued that long-term rejection and maltreatment diminishes victims' self-regulation and tendency to fortify threatened needs.

Keywords:

• Bullying at the workplace
• Ostracism
• Social exclusion
• Compliance
• Social influence

Workplace bullying

The process when indirect forms of aggression appear repeatedly and systematically against an employee and he or she is unable to defend him or herself from these negative behaviors is called bullying (Einarsen, Hoel, Zapf, & Cooper, 2003). Bullying may develop from personal focused disagreement or a conflict over a specific argument and escalate over time, which means that such negative treatments turn from subtle to more direct and cruel ones (Björkqvist, Osterman, & Hjelt-Back, 1994; Leymann, 1996).

Studies on bullying targets show that they have lowered self-esteem (e.g., Harvey & Keashly, 2003; O'Moore & Kirkham, 2001) and experience anxiety, negative emotionality, depressive symptoms, including suicidal tendencies (Leymann, 1996; Matthiesen & Einarsen, 2004), and stress symptoms (Hoel, Zapf, & Cooper, 2002; Marchand, Demers, & Durand, 2005). Victims feel helpless, and lack a sense of power and control over their situation (Einarsen et al., 2003; Lewis, 2004; Salin, 2003). They report chronic fatigue, problems with concentration, sleeping, and health generally (Hansen et al., 2006; Høgh, Mikkelsen, & Hansen, 2011).

It has been shown that during the process of maltreatment, targets of bullying display a dynamic series of reactions. In Zapf and Gross's (2001) semi-structure interview study, targets declared that as soon as they realized the bad intentions and hostility of a bully, they began with constructive conflict-solving solutions, but the integrating, task-oriented strategies were found to be ineffective (see also Rayner, 1997). Victims tend to resort to other strategies at different stages in the process. Any attempt to find understanding and support at the workplace proves to be ineffective, merely increasing the feeling of shame (Lewis, 2004). Høgh and Dofradottir (2001) found that workers subjected to bullying used humor and avoidance strategies more often and problem solving less often than nonvictims. If there appear any forms of resistance, they are covert because of fear of being punished. In the study by Lutgen-Sandvik (2006), one of the forms of resistance displayed by victims was retaliation. Reciprocation took the form of hostile gossip or fantasies of physically harming or killing the bully. A very common behavior was “talking behind the abuser's back”, derogating the bully (Lutgen-Sandvik, 2006). It seems that hostility toward others, which is a kind of reaction of cognitive nature combining a sense of resentment and suspicions (Palmer & Thakordas, 2005), is one of the responses of victims of bullying to maltreatment (Gamian-Wilk, Bjørkelo, & Hauge, in preparation, (2013); Ireland & Archer, 2002). It seems that bullying as a process of long-lasting maltreatment and rejection results in experiencing stress and impairment in many areas including social functioning: it is difficult for a victim to respond with pro-social behaviors when having to bear constant humiliation.

Workplace ostracism as one of the negative bullying activities

Bullying is interpersonal by nature and social isolation has been regarded as one of the most frequent forms of negative activities reported by victims and differentiated in empirical studies (e.g., Einarsen, 2000; Leymann, 1996; Vartia, 1993), although bullying is a broader construct that can involve personal bullying (making insulting remarks, excessive teasing, spreading gossip or rumors, persistent criticism, and isolating), work-related bullying (excessive monitoring of work, assigning unreasonable deadlines and unmanageable workloads, assigning meaningless tasks, or no tasks at all), and physical intimidation (various forms of threatening, physical abuse) (Einarsen et al., 2011). It has been demonstrated that ostracism is a common experience in the workplace context and may have different forms: being ignored at work, being unanswered when giving greetings, being left alone during lunch breaks, not being looked at, being shut out during conversation, or not being invited for informal meetings (Ferris, Brown, Berry, & Liam, 2008). In a survey conducted by Fox and Stallworth (2005), it was found that 66% of workers had experienced being ignored or rejected during the past 5 years. Although ostracism has been perceived as one of the core bullying activities and causes similar devastating consequences, research has shown that workplace ostracism per se is conceptually separate from bullying (Ferris et al., 2008).

Workplace ostracism often aims at expelling a victim from the workplace community (Einarsen et al., 2003). This is an extreme form of social rejection but ostracism is also represented by more subtle behaviors, e.g., using silent treatment or avoiding eye contact (Williams, 2007). Ostracism, which leaves a feeling of being invisible or of not existing, communicates symbolically to the target that he or she is dead and meaningless for the source (Williams, 2007). Therefore, even short episodes of social rejection lead to immediate negative consequences, such as hurt feelings (distress, anger, sadness) and threat to social self-esteem, need of belonging, need of control, and sense of meaningful existence (Smart Richman & Leary, 2009; Williams & Zadro, 2005; Zadro, Williams, & Richardson, 2004). These emotional responses occur immediately after rejection, but behaviors which accompany these reactions vary considerably, from socially desirable to withdrawal or antisocial responses. Smart Richman and Leary (2009) in a multimotive theory provide a conceptual framework explaining different outcomes to various kinds of rejection circumstances. Socially appealing responses are possible if exclusion is connected with high perceived costs, if one expects that a relationship may be repaired or if a relationship is highly valued. Conversely, in the case of such chronic rejection as workplace bullying where victims perceive the exclusion as unfair and have no hope of repairing the relations, they tend to behave aggressively or they withdraw (e.g., Maner, DeWall, Baumeister, & Schaller, 2007; Twenge, 2005; Williams, 2007; Williams & Zadro, 2005). Therefore, it is reasonable to predict that long-term bullying should reduce people's tendencies to respond to rejection in a pro-social manner.

Social exclusion and compliance

In general, if possible, people seek acceptance and tend to restore their positive self-esteem and sense of belonging after rejection by behaving in a socially desirable way, e.g., by engaging in a collective task rather than in an individual task (Williams & Sommer, 1997), conforming to others (Williams, Cheung, & Choi, 2000) or mimicking other persons' behavior (Lakin & Chartrand, 2005). In line with these findings, Carter-Sowell, Chen, and Williams (2008) demonstrated that regardless of the tactic employed (the foot-in-the-door, the door-in-the-face, and/or a single target request), experimentally evoked ostracism increased compliance. It is therefore important to deepen the knowledge on the link between being ostracized and susceptibility to social influence in the context of real-life situations. The aim of this study was to investigate whether long-term social exclusion in the form of bullying an individual leads to his or her compliance.

It seems therefore that the bullying phenomenon may represent boundary conditions, that is, circumstances in which anti-social rather than socially desirable responses are more likely to occur after exclusion. As compliance is perceived as a socially desirable behavior displayed after being ostracized (Carter-Sowell et al., 2008), a drop in compliance among bullying targets was predicted. As in the Carter-Sowell et al. study (2008), it was shown that rejection leads to greater compliance rates regardless of the social influence technique used; it is reasonable to hypothesize that bullying victims will not be compliant with single requests. Therefore, the first hypothesis stated that being bullied at work would result in lower proneness to comply in comparison with not being bullied situation (H1). It was also predicted that the drop in the tendency to comply would be even greater among bullied workers after remembering being exposed to ostracism by other coworkers (H1a). As it was earlier proved experimentally, evoked ostracism resulted in greater compliance. It was anticipated that the mere memory of being exposed to ostracism at work would result in increased proneness to comply (H2). In previous studies, it was shown that both experimentally evoked ostracism and being rejected at the workplace cause a significant drop in needs satisfaction and mood. Thus, the third hypothesis stated that being subjected to bullying at the workplace would lead to lower needs satisfaction and lowered mood (H3). Moreover, it was predicted that the drop in needs satisfaction and mood would be even greater among bullied targets after remembering being exposed to ostracism by other coworkers (H3a). Finally, it was anticipated that the mere memory of being exposed to ostracism at work would result in a decrease in needs satisfaction and mood (H4).

To test these predictions, one questionnaire study and one quasi-experimental study were conducted. In the first study, only the first hypothesis was tested. The second study verified all of the four predictions.

Study 1

Overview

The first study aimed at investigating the level of compliance among bullying victims. Participants representing various professions completed two questionnaires: the Negative Acts Questionnaire to diagnose bullying symptoms and the Compliance Scale measuring the general tendency to agree to requests asked by coworkers.

Procedure and sample

The study was conducted among 197 employees, recruited among mature part-time university students taking a further education course in pedagogy. The study was an integrated part of an academic course and participants were not rewarded for their cooperation. The selection criteria were that the participants were in full-time positions and in contact with their superiors or subordinates at least three times a week (recruitment procedure is according to Glasø & Einarsen, 2008).

The participants' age range was from 20 to 51 (M = 26.13, SD = 6.22) and 81% of the sample were female. Nineteen participants were leaders and the others were subordinates. The majority of the participants worked in the public sector (76%), mainly in education (37%) and health services (26%).

Instruments

Bullying

Workplace bullying was measured by two different methods. First, workplace bullying was measured by a Polish version of the Negative Acts Questionnaire – Revised (NAQ-R) (Warszewska-Makuch, 2007), developed by Einarsen, Hoel, and Notelaers (2009). The NAQ-R consists of 22 items and describes different behaviors which may be perceived as bullying or harassment if they occur on a regular basis. All items are formulated in behavioral terms, with no reference to the phrase “bullying and harassment.” The NAQ-R contains items referring to both direct (e.g., open attack) and indirect (social isolation, slander) behavior. It also contains items referring to personal as well as work-related forms of bullying. For each item, the respondents were asked how often they had been exposed to the behavior at their present workplace during the last 6 months. Response categories were (1) “never,” (2) “now and then,” (3) “monthly,” (4) “weekly,” and (5) “daily.” The NAQ-R showed good internal consistency in the present study (Cronbach's α = .92).

After the NAQ-R was listed in the questionnaire, the respondents were asked to indicate whether they had been exposed to bullying at work during the last 6 months according to a formal definition of bullying at work (see Einarsen & Skogstad, 1996). The response categories were (1) “No,” (2) “Yes, but occasionally,” (3) “Yes, now and then,” (4) “Yes, once a week,” and (5) “Yes, several times a week.”

Groups of bullying targets (N = 32) and nonvictims (N = 30) were distinguished on the basis of Leymann's criteria (one has to be exposed to at least one negative act per week over a period of at least 6 months to be considered as a bullying target)1 on the sum of 22 NAQ-R items.

Compliance

The Compliance Scale consisted of 15 items connected with agreeing to requests proposed by supervisors or coworkers. It contains items referring to both task-related (e.g., performing some additional duty, explaining work problems, and borrowing money) and social (e.g., spending time with colleagues) requests. In some statements, the difficulty or irrationality of requests is stressed. For each item, the respondents were asked to what extent they agreed with a particular request. Response categories were (1) “does not describe me,” (2) “rather does not describe me well,” (3) “hard to say,” (4) “describes me rather well,” and (5) “describes me well.”

Factor analysis of the data obtained in the first study using Principal Axis Factoring and Promax rotation with Kaiser nominalization revealed two meaningful factors: compliance with task-related requests (11 items, e.g., “I agree to do a task which is not my duty when a work mate asks me to”) and compliance with social requests (four items, e.g., “I agree to go for a lunch with my friends when they invite me even if I do not have time”). The total percentage of variance for the two factors was 62.18% (54.91% of variance for task-related requests and 7.28% for social requests). The initial eigenvalues were, respectively, 8.24 and 1.09. The task-related compliance scale showed good internal consistency in the present study (Cronbach's α = .93). The social compliance scale showed satisfactory consistency (Cronbach's α = .83). Both task-related and social compliance scales were highly correlated, r = .74, p < .001, d = 2.20. Reliability of the total compliance scale was Cronbach's α = .94.

Results

Table 1 shows descriptive statistics for bullying, the total score of the compliance scale, task-related compliance, and social compliance. The hypothesis which stated that bullying targets would report a low level of compliance was confirmed: bullying targets were generally less compliant, both with social and task-related requests (Table 2).

Table 1 Descriptive statistics for negative activities, compliance in general, task-related compliance, and social compliance

	M	SD	Min	Max
Negative acts	40.09	15.77	22	103
Compliance (total score)	33.07	13.72	15	71
Task-related compliance	24.38	10.65	11	51
Social compliance	8.69	3.83	4	20

Table 2 Differences in general, task-related, and social compliance between bullying targets and nonvictims

	Bullying targets		Nontargets
	M	SD	M	SD	t	p	Hedges g
Compliance	26.59	10.11	36.93	13.89	3.33	.002	.84
Task-related compliance	19.03	7.63	27.37	10.49	3.59	.001	.90
Social compliance	7.56	3.43	9.57	3.87	2.15	.04	.54

Discussion

In general, the results confirm the main prediction (H1): being subjected to negative treatment at work is connected with low compliance with various requests of coworkers. These results suggest that although single acts of ostracism, as shown in experimental designs (Carter-Sowell et al., 2008), lead to compliance as a sign of socially desirable behavior, long-term rejection results in completely opposite responses. Bullied participants were reluctant to agree to both task-related and social requests: they tended to refuse to do additional work, stay extra hours, lend money or personal things, give work materials and also spend time with work mates, have lunch with them, or go for informal meetings. If they have no time or they have other urgent duties, they do not decide to comply with coworkers' requests. On the one hand, such responses appear to be assertive and reflexive. However, being reluctant to agree to colleagues' requests means not meeting social expectations and may result in further exclusion.

These findings are in line with Smart Richman and Leary's (2009) multimotive model of reactions to rejection: long-lasting ostracism at the workplace assessed as an unpleasant experience leads to a drop in socially desirable behaviors. The results are also consistent with data suggesting that bullying victims tend to withdraw rather than to cope constructively in the long term (e.g., Zapf & Gross, 2001). Moreover, research on bullying targets' profile indicates that employees who have been exposed to bullying at work lack social competence (Matthiesen & Einarsen, 2001, 2004), are less social and talkative, as well as less likeable, understanding, and diplomatic (Glasø, Matthiesen, Nielsen, & Einarsen, 2007; Lind, Glasø, Pallesen, & Einarsen, 2009).

However, it is highly possible that it is not the victims' personality features but the process of workplace ostracism and maltreatment which generates bullying targets' socially undesirable responses. Unfortunately, research has shown that social rejection deteriorates self-regulation, which is connected with a significant drop in cognitive functioning, and lower resistance to temptations (Baumeister & Dewall, 2005). Worsened self-regulation may result in limited capacities of proper social functioning. As Baumeister and Dewall (2005) state, many people experience a downward spiral in which social exclusion may lead to socially disvalued behavior which may, in turn, elicit further rejection. Bullying victims' poorer self-regulation caused by social exclusion may constitute a pitfall: it seems extremely difficult to escape from the downward spiral and regain social acceptance.

It is, however, important to ponder whether bullying targets have suitable opportunities to agree to coworkers' requests. In the workplace where bullying flourishes, that is, in a climate of conflicts, fear, and low job satisfaction, social interactions may be deteriorated (Hauge, Skogstad, & Einarsen, 2007). It seems difficult to be compliant with coworkers who bully or do not provide support. Victims are often left alone, isolated, which means they do not receive any requests. Therefore, by having no possibilities to be compliant, they may have no chances to fulfill their threatened needs and to regain social acceptance.

Moreover, the decrease in compliance rate may not be a result of social exclusion as being bullied means not only being rejected from a group but also being a subject of gossip, negative communication, work overload, and other forms of maltreatment. In the second study, it was assessed whether the rejection itself causes a decrease in needs satisfaction and mood and a drop in compliance. In this study, participants declared their compliance with requests of coworkers staying at the University, which is in a quite different context from their workplaces. It is possible that being in an alternative group distant from unpleasant work conditions influenced the results. In the second study, the context of workplace has been highlighted.

Study 2

Overview

The aim of the second study was to highlight the context of workplace rejection. To meet this purpose, after completing the questionnaire concerning being bullied at work, employees did a recall task of being either included or excluded by their coworkers. Then they completed the measure of compliance, manipulation check, and a questionnaire measuring the level of satisfying needs and the level of mood index.

Procedure and sample

The study was conducted among 309 employees, recruited among mature part-time university students taking a further education course in pedagogies, special pedagogies, banking, management, finance, and marketing. The study was an integrated part of an academic course and participants were not rewarded for their cooperation. The selection criteria were the same as in the first study.

The participants' age range was from 21 to 48 (M = 27.64, SD = 6.18) and 69% of the sample were female. Eighty-six participants were leaders and the rest were subordinates. The majority of the participants worked in the public sector (69%), in education, health services, social care sector, production, trade, administration, finance, banking, police, prisons, and nonprofit organizations.

Instruments

Bullying

Polish version of the NAQ-R (Warszewska-Makuch, 2007) described in the first study was used. The NAQ-R showed good internal consistency in the present study (Cronbach's α = .94). Groups of bullying targets (N = 48) and nonvictims (N = 46) were distinguished on the basis of Leymann's criteria on the sum of 22 NAQ-R items.

Exclusion versus inclusion at workplace

Participants were asked to think about their workplace. Previous research has shown that recall and re-living procedures occurred to be successful in generating negative ostracism consequences (Chen, Williams, Fitness, & Newton, 2008; Craighead, Kimball, & Rehak, 1979). In the exclusion condition it was stated that in each organization coworkers sometimes neglect or reject others either in delegating tasks, passing messages, inviting for meetings, or by not answering phones or emails. Participants were to remember and describe situations from their own work when other coworkers ostracized them in any way. In the inclusion condition, it was stated that in each organization coworkers sometimes invite each other to participate in collective tasks or meetings, become involved in spreading messages, and answer emails or phones immediately. Participants were to remember and describe situations from their own work when other coworkers included them in any way.

Compliance

The compliance scale described in the first study was used. On the basis of factor analysis of the data obtained in the second study using Principal Axis Factoring extraction and Promax rotation with Kaiser nominalization, it was decided to treat the compliance scale as a homogeneous measure (although factor analysis revealed four factors with initial eigenvalues of 5.26, 1.45, 1.37, and 1.15, the factor loadings in the first factor were not lower than .62, whereas in two others they were lower than .43; the scree plot indicated one significant factor explaining 35.05% of variance, whereas the other three factors explained 9.67%, 9.15%, and 7.65% of variance; and analysis of the three other factors' items did not show any logical structure). The compliance scale showed good internal consistency in this study (Cronbach's α = .92).

Fundamental needs and mood

After describing being excluded or included at the workplace and completing the compliance scale, participants were asked to provide self-reports concerning their satisfaction levels with belonging, self-esteem, meaningful existence, and control on 7-point scales (after van Beest & Williams, 2006; Cronbach's α = .82 in this study). Next they were asked to assess their emotional state at work also on 7-point scales. This mood index contained three items assessing positive emotions and three items assessing negative emotions (after van Beest & Williams, 2006; Cronbach's α = .82 in this study). Negative emotions were reverse-scored. Both needs satisfaction and mood items referred to workplace context: participants were to evaluate the level of their needs fulfillment and mood at their workplace. Finally, to check the exclusion versus inclusion manipulation, participants were asked in three items on a 7-point scale to evaluate the extent to which (a) people in general are ostracized (one item) and (b) they themselves feel rejected by their coworkers (three items on 7-point scales) (in accordance with van Beest & Williams, 2006; Cronbach's α = .73). It was also checked whether participants after reading the manipulation message described any situation.

At the end of the study, all of the participants were thanked and debriefed.

Results

Manipulation check

The manipulation was found to be successful. An (inclusion vs. exclusion manipulation) analysis of variance (ANOVA) on the sum of three items measuring the level of feeling rejected and ignored at the workplace showed that participants who were to remember and describe the situation of being rejected at their workplace felt significantly more rejected, F (1,307) = 6.15, p = .01, η _p ²  = .02. The fact that participants described a situation after remembering being included or excluded at the workplace influenced the manipulation check results, increasing the effect. A 2 (inclusion vs. exclusion manipulation) × 2 (described a situation vs. did not describe any situation) (ANOVA) on the sum of three items measuring the level of feeling being rejected at the workplace indicated a main effect of manipulation, F (1,119) = 12.15, p < .001, η _p ²  = .04, a main effect of description, F (1,57) = 6.13, p = .01, η _p ²  = .02, and an interaction effect, F (1,57) = 4.79, p = .03, η _p ²  = .02. Therefore, although only 120 participants described their own experiences of being included (85 participants) or excluded (35 participants) at the workplace, still merely reading the information was found to have an impact on the feeling of being included or rejected at work, respectively, to the manipulation information.2

Means and standard deviations of the compliance index, needs satisfaction, and mood are given in Table 3.

Table 3 Descriptive statistics for negative activities, compliance, positive and negative emotions, and fundamental needs

	Overall descriptives				Inclusion condition				Exclusion condition
	M	SD	Min	Max	M	SD	Min	Max	M	SD	Min	Max
Negative acts	36.28	14.15	22	106	35.94	14.69	22	106	36.80	13.86	22	82
Compliance	44.41	10.48	15	70	44.19	10.51	15	70	44.66	10.47	12	70
Emotions index	30.66	6.34	10	42	31.11	6.28	10	42	30.14	6.42	11	42
Belongingness	27.47	5.62	14	36	28.02	5.67	14	36	26.75	5.51	15	35
Self-esteem	26.39	5.20	10	35	26.27	4.75	16	35	26.45	5.68	10	35
Control	22.89	6.45	8	51	22.95	6.32	8	35	22.72	6.61	8	51
Meaningful existence	28.37	5.38	13	35	28.67	5.25	15	35	28.00	5.55	13	35
Need satisfaction index	105.4	17.06	68	140	106.46	15.83	72	140	103.92	18.24	68	140

Compliance rates

A 2 (inclusion vs. exclusion manipulation) × 2 (bullied vs. not bullied) ANOVA on compliance yielded a main effect of being bullied, F (1,89) = 16.08, p < .01, η _p ²  = .16, a main effect of exclusion manipulation F (1,89) = 6.18, p = .02, η _p ²  = .07, and a significant interaction effect, F (1,89) = 14.16, p = .001, η _p ²  = .14 (Figure 1).

Figure 1 The influence of inclusion versus exclusion manipulation and being bullied at the workplace on compliance.

The analysis of contrasts indicated that there is a significant difference in proneness to comply between bullying targets and nontargets in the inclusion condition (F (1,89) = 36.91, p < .001, η _p ²  = .42). The difference was not significant in the exclusion condition (F (1,89) = .03, p = .87).

This result suggests that although remembering a single situation of being ostracized at work produces compliance among nonvictims, it seems that it may not be the situation of exclusion itself which causes a drop in compliance among bullying victims. To investigate the possible predictors of compliance, further analysis of contrasts among nonvictims and among victims comparing needs satisfaction, mood, and feeling of being excluded as predictors (dichotomized based on median split) on compliance was conducted. It was found that only mood was a significant predictor of compliance (F (1,24) = 7.56, p = .01, η _p ²  = .25).

In conclusion, although in the second study bullying targets were generally less reluctant to comply with coworkers' requests (H1 was only partially confirmed), they were less willing to comply after focusing on an exclusion situation at work (H1a has been confirmed). H2 has been confirmed: the mere focusing on a situation of being rejected at work produces greater proneness to comply among nonvictims.

Fundamental needs

A 2 (inclusion vs. exclusion manipulation) × 2 (bullied vs. not bullied) ANOVA on needs yielded only a strong main effect of being bullied, F (1,93) = 86.34, p < .001, η _p ²  = .49, such that nonbullied participants had much more satisfied needs (M = 112.52, SD = 14.74) than bullied participants (M = 86.94, SD = 18.37).

Mood

A 2 (inclusion vs. exclusion manipulation) × 2 (bullied vs. not bullied) ANOVA on mood yielded only a strong main effect of being bullied, F (1,93) = 83.54, p < .001, η _p ²  = .51, such that nonbullied participants were much more positive (M = 36.36, SD = 3.68) than bullied participants (M = 24.52, SD = 6.51).

In conclusion, H3 has been confirmed: bullied participants experienced lower levels of needs satisfaction and mood at the workplace. Moreover, H3a has not been confirmed: bullying targets did not display lower mood when focused on being excluded at work. However, although the manipulation of focusing on inclusion versus exclusion at work was successful, H4 was not confirmed: reading information about possibilities of being rejected or included by a group and remembering such situations influenced neither mood nor needs satisfaction regarding work context.

General discussion

Previous research has shown that experimentally evoked ostracism leads to greater compliance rates (Carter-Sowell et al., 2008). Greater compliance after ostracism was due to lowered needs satisfaction. The aim of this study was to check whether this pattern of results would be replicated in the context of long-term workplace rejection.

First of all, the pattern of results obtained by Carter-Sowell et al. (2008) has been confirmed in the second study: even the mere focus on social exclusion produces greater compliance. Consistent with other studies, the threat of being rejected is information which generates socially valuable responses (e.g., Lakin & Chartrand, 2005; Williams & Sommer, 1997). Here, reading the examples of social ostracism at the workplace was a sufficient threat to generate counteraction and report being more compliant. Surprisingly, although reading a message pointing out various kinds of social exclusion at work caused significant feelings of being rejected at work, it decreased neither mood nor needs satisfaction. The possible explanation of these results is that such single experiences of social ostracism at work did not impact on the participants' general job satisfaction as the items in the mood index and needs fulfillment referred generally to workplace context. It is even possible that participants' immediate socially desirable responses such as compliance with work mates' requests fortified threatened social needs and even resulted in more positive emotionality. Maybe if asked about certain episodes of rejection, participants would report lowered mood and needs satisfaction (if a particular experience of ostracism caused negative emotionality and threatened needs), which would mediate compliance. These suggestions need to be examined further.

Most importantly, the present studies are the first studies investigating compliance among bullying targets. The first study has shown that being subjected to workplace bullying is connected with unwillingness to comply with both task-related and social requests of coworkers. The second study has proven that being bullied causes a dramatic drop in needs satisfaction and mood. Moreover, being bullied leads to a lower level of compliance, but only after focusing on the situations of being excluded at work. The findings obtained in both studies are therefore consistent.

Interestingly, it was neither the feeling of being excluded nor lowered needs satisfaction which was found to predict compliance among bullying targets. Victims' compliance was explained only by lowered mood. Perhaps it is not social exclusion which is so harmful and, therefore, decreases the willingness to perform socially desirable behaviors but there may be other negative activities of coworkers which result in targets' negative emotionality and unwillingness to comply and to affiliate with others.

Second, the findings suggest that in the case of bullying targets, the threat of social exclusion is not a motivation to undertake socially desirable behaviors. As social ostracism indicates that one has not behaved in a socially valuable way, nonvictims seem to respond to this cue and change their behavior. On the contrary, victims' responses are much less differentiated than nonvictims' reactions to inclusion or exclusion information. Bullying targets appear not to take advantage of such a social cue as exclusion to self-regulate and display socially desirable behavior.

Future studies should focus on replicating the obtained results using behavior-based compliance measures as it is traditionally evoked and observed (e.g., Pratkanis, 2007). Like in other fields of bullying research (Monks & Coyne, 2011), it is important to implement various methods of assessing both negative workplace activities and targets' responses. In this study, compliance was diagnosed by self-reports, which is very common in workplace bullying research (implementing surveys, questionnaires, or semi-structured interviews). Behavior observation or in-depth case studies are rather rarely conducted.

In this study, bullying was evaluated outside of the workplace, which is to some extent beneficial. Participants could freely and safely answer questions without being afraid of being punished. We may assume that self-report measures are more credible when completed outside the workplace. However, future studies should take organizational context (climate, values, and leadership) into account. Bullying at work is a complex phenomenon which develops over time. There are a number of organizational, group, and individual antecedents promoting bullying (Coyne, 2011). Moreover, responses of bullying victims to negative activities should be analyzed in the larger context of time. Future studies should focus on investigating the sources and dynamics of bullying targets' compliance.

Notes

¹ Although the NAQ (NAQ-R) results of individual items may be summed and the sum scores may be included in correlation analyses or regression analyses, the behavioral experience approach may also be used to distinguish between different groups of respondents (targets and nontargets). The common method of separating targets from nonvictims is to apply an operational criterion (Nielsen, Notelaers, & Einarsen, 2011). This approach seems to fit most to the theory of the bullying phenomenon which is not a “continuous” experience but a phenomenon of several criteria (e.g., Leymann, 1996). It is, therefore, more reasonable to separate victims from nonvictims. On the other hand, the operational criterion has several limitations. Nielsen et al. (2011) recommend latent class cluster analysis as the best method of identifying different groups of respondents (e.g., personal bullying victims, work-related bullying victims, and occasional victims). The purpose of this study was to compare the strategies of workers being bullied (according to the definition) and not bullied.

² Approximately 40% of the narratives (N = 48) were selected randomly and subjected to a word count. There were no significant differences in the number of words participants wrote between included (M = 22.82, SD = 20.27) and excluded (M = 30.13, SD = 20.26) conditions, t = 1.27, p = .21. The narratives contained mainly descriptions of facts. Single participants who were to remember situations when they were included at work stated the consequences (positive emotions, relationships deepened).