The relationship between interpersonal conflict and workplace bullying

Purpose – The purpose of this paper is to examine the role that conflict management styles play in the relationship between interpersonal conflict and workplace bullying. Design/methodology/approach – A survey study was conducted among 761 employees from different organizations in Spain. Findings – Results suggest that an escalation of the conflict process from task related to relationship conflict may explain bullying situations to some extent. Regarding conflict management, attempts to actively manage conflict through problem solving may prevent it escalating to higher emotional levels (relationship conflict) and bullying situations; in contrast, other conflict management strategies seem to foster conflict escalation. Research limitations/implications – The correlational design makes the conclusions on causality questionable, and future research should examine the dynamic conflict process in more detail. On the other hand, to the authors’ knowledge, this is the first study empirically differentiating interpersonal conflict and workplace bullying. Originality/value – This study explores how conflict management can prevent conflict escalating into workplace bullying, which has important implications for occupational health practitioners andmanagers.


Introduction
Workplace bullying is an emergent phenomenon; it refers to "a social interaction through which one individual (seldom more) is attacked by one or more (seldom more than four) individuals almost on a daily basis and for periods of many months, bringing the person into an almost helpless position with potentially high risk of expulsion" (Leymann, 1996, p. 168).In this regard, research has shown that receiving unwanted behaviors of a psychological nature has severe negative consequences not only for the target of bullying but also for the organization as a whole (e.g., Nielsen et al., 2012;Topa-Cantisano et al., 2007).
Thus, there has been growing interest over the last few years in exploring the antecedents of workplace bullying (Einarsen et al., 2011).The predominant theoretical framework used is the so-called 'work-environment hypothesis', which conceives workplace bullying as an extreme social stressor that results both from inadequate working conditions and other organizational factors (e.g., Giorgi, 2009;Hauge et al., 2007;Notelaers et al., 2010).Although this has resulted in some valuable suggestions, recent trends have demonstrated the need to move forward and obtain a more in-depth knowledge of workplace bullying by focusing on its underlying interpersonal mechanisms (e.g., Glaso et al., 2009;Neuman and Baron, 2011).
To meet this challenge, we follow Einarsen et al. (2011), who argue that workplace bullying is "an escalating process in the course of which the person confronted ends up in an inferior position and becomes the target of systematic negative social acts" (p.15).Thus, considering that this escalating process is driven by the existing interpersonal conflict between parties and their preferences for managing conflict, the purpose of this study is both to examine the relationship between interpersonal conflict and bullying at work, and assess the role that conflict management styles play in such conflict-bullying relationships.Findings will shed some light on the underlying mechanisms of workplace bullying, and therefore help managers take more effective steps to prevent bullying at work.

Conflict escalation and workplace bullying
According to Van de Vliert (2010), conflict researchers can shed some light on the underlying mechanisms of bullying since the two concepts share some definitional Coping with conflict and bullying at work 4 elements that allow these interrelated arenas to be linked and additional well-established theories and practical experiences to be used.On one hand, both concepts focus on the perception of incompatibility in the interaction between two parties, "with at least one of them experiencing obstruction or irritation by the other party" ( Van de Vliert, 2010: p. 87).On the other hand, whereas time is a key characteristic of bullying, it is not so important in conflict (i.e., conflict does not have to be repeated but can be a one-off incident).
In this regard, Leymann (1996) proposed a model of bullying based on case studies highlighting the fact that bullying behaviors are the result of an escalated conflict that was not satisfactorily solved.Similarly, Einarsen (1999) differentiated between "dispute-related bullying" and "predatory bullying".Whereas the former originates from highly emotional interpersonal conflicts between workers, the latter is a non-ethical mechanism used by some workers to maintain their status and/or to get rid of stress and frustration at work by exerting negative acts on other coworkers.Indeed, Zapf and Gross (2001) conducted a study based on semi-structured interviews with 20 victims of bullying as well as a survey comparing the coping strategies of victims of bullying with those of a control group not exposed to negative acts at work.They concluded that bullying situations are congruent with Glasl's conflict escalation model (1994): the situation started with disagreements about the content of the conflict (i.e., what conflict researchers describe as a socio-cognitive or task-related conflict, which refers to disagreements concerning the content of inter-related tasks, including differences in their views about the distribution of resources or the procedures they have to follow: De Witt et al., 2012).Then, this conflict turned to more personal issues in which both parties polarize their positions and differences (i.e., conflict concerning perceptions of interpersonal incompatibilities and hostility, which is more akin to a socio-emotional or relationship conflict: De Witt et al., 2012).Finally, as "the relationship between the parties has become the main source of tension" (Zapf and Gross, 2001: p. 502), the conflict became destructive since the party with more power tried to destroy the opposite party's reputation and self-esteem (i.e., workplace bullying: Leymann, 1996).
In conclusion, as Ayoko et al. (2003) pointed out, more intense and long lasting conflicts cause negative behaviors and a variety of emotional responses that constitute Coping with conflict and bullying at work 5 workplace bullying.Moreover, personal-oriented conflicts "are likely to lead to longlasting relationship conflicts since they contain a high degree of emotionality, which may manifest itself in negative behaviors such as raised voices, hostility toward others, and threats and intimidation" (Greer et al., 2008: p. 281).Thus, the intensity of the conflict seems to be the mechanism or process underlying the association between interpersonal conflict and workplace bullying.On this basis and considering that workplace bullying and interpersonal conflict are distinct but interrelated phenomena, we hypothesize that relationship conflict is a mediator of the relationship between task conflict and bullying.
Hypothesis 1: Relationship conflict mediates the effect of task conflict on workplace bullying.

The role of conflict management styles
The way disputes are managed plays a pivotal role in the (de)escalation of conflict.Broadly speaking, conflict management refers to what the parties (individuals, groups or organizations) who experience conflict intend to do and what they actually do (Van de Vliert, 1997).According to the Dual-Concern Model (see De Dreu et al., 2001), interpersonal conflict at work is managed in accordance with an individual's concern for the self (competition) and for the other party (cooperation).Thus, when concern for both the self and the other is high, problem solving is a more likely strategic choice (e.g., when two employees do inter-related tasks in which performance depends on their joint outcomes).In contrast, if concern for both self and the other is low, inaction or avoiding is more likely (e.g., when an issue is trivial and other issues are more important or when the potential cost of confronting the conflict outweighs the benefits of addressing it).Moreover, if concern for one's own outcome is high but concern for the other is low, this leads to the use of forcing strategies (e.g., when a manager imposes a deadline on subordinates according to his/her own priorities without taking into account the others' preferences or possibilities); on the other hand, if concern for self is low but concern for the other is high, this results in yielding strategies (e.g., when employees know that the manager has more power and they have to forego personal interests).
Coping with conflict and bullying at work 6 Moreover, conflict management styles are considered stable traits of individuals (i.e., types of behavior or generalized behavioral orientations) that affect conflict escalation and therefore individual and group outcomes (Behfar et al., 2008;Janssen and Van de Vliert, 1996).For example, managing conflicts in a cooperative and active way (i.e., problem solving) promotes productive conflict (i.e., integrative or win-win solutions) and strengthens trust among parties, thereby reducing the level of conflict present and facilitating team performance (e.g., Chen et al., 2005;Hempel et al., 2009;Janssen and Van de Vliert, 1996).In contrast, using forcing and yielding styles is related to conflict escalation because it can lead to a deterioration in the parties' relationship even though results may satisfy one party in the short run (Behfar et al., 2008;Janssen and Van de Vliert, 1996).Finally, conflict avoidance entails increased negative emotion and a higher probability of conflict escalation because it also leaves conflicts unresolved (e.g., Desivilya and Yagil, 2005;Dijkstra et al., 2009).
Hypothesis 2: Problem solving is negatively related to relationship conflict, whereas forcing, yielding and avoiding are positively related to relationship conflict.
As for research on workplace bullying, Baillien et al. (2009) explored 87 bullying incidents and concluded that three processes can explain the development of bullying: (a) organizational factors that constituted fertile soil for bullying; (b) reactions to workplace conflicts; and (c) inability to cope with stress and frustration.Thus, as Ayoko et al. (2003) concluded: "it may not be the conflict events per se that trigger workplace bullying, but rather the duration and intensity of the conflict, as well as reactions to conflict events" (p.297).Indeed, Leymann (1996) claimed that bullying emerges when conflict is poorly managed or not satisfactorily resolved.Similarly, some studies have shown that strategies used by the targets of bullying can be related to the escalation or de-escalation of the bullying situation (e.g., Baillien et al., 2011;Baillien and De Witte, 2009;Zapf and Gross, 2001): the use of both dominating (forcing) and passive (such as yielding and avoiding) conflict management styles are positively related to the escalation of conflict and a higher number of negative behaviors received by the target; whereas problem solving strategies are associated negatively with bullying.
Coping with conflict and bullying at work 7 Hypothesis 3: Problem solving is negatively related to workplace bullying, whereas forcing, yielding and avoiding are positively related to workplace bullying.
Finally, based on the above mentioned considerations and findings, we hypothesize that the mediating effect of relationship conflict on the association between task conflict and workplace bullying depends on the conflict management styles used to deal with conflict (i.e., moderated-mediation model, in which conflict management styles moderate the mediation of relationship conflict on task conflict and workplace bullying, see Figure 1).In particular, conflict management styles will moderate the path between task conflict and relationship conflict as well as the path between relationship conflict and workplace bullying.
Hypothesis 4: Conflict management styles moderate the mediating effect of relationship conflict on task conflict and workplace bullying.

Procedure and participants
Data were gathered in three Spanish organizations distributed across sectors in Andalusia: a large-size organization from the public administration sector, a mediumsize company from the service sector, and a medium-size company from the manufacturing sector.Participation was voluntary and confidential.Indeed, surveys were administered to groups of workers in company time with a research assistant present to answer any questions.Participants placed their completed questionnaires in a sealed box to ensure the anonymity of responses.
A total of 762 valid questionnaires were returned (response rate of 54.4%).Most of the participants were men (61.4%) with job tenure of more than five years (86.2% vs. 13.8% who reported job tenure between two and five years, and 4.4% reporting less than two years).Their ages ranged from 21 to 68 years (M = 41.62;SD = 7.42).
Coping with conflict and bullying at work 8

Measures
Exposure to Workplace Bullying was measured using the reduced Spanish version of the Negative Acts Questionnaire-Revisited (NAQ-R: Einarsen et al., 2009) developed by Moreno-Jimenez et al. (2007).Participants scored the frequency (response categories were 1: Never, 2: Now and then, 3: Monthly, 4: Weekly, and 5: Daily) with which they had been exposed to 14 specific negative acts (bullying behaviors) over the last six months (e.g., gossiping or having information withheld).Cronbach's alpha of the NAQ-R was .88.
Interpersonal Conflict was measured with a 9-item scale that includes both taskrelated conflict (e.g., 'How often are there disagreements about the task you are working on?') and relationship conflict (e.g., 'How often do you experience hostility at work?') (see Benitez et al., 2012).All items were rated on a 5-point Likert-type scale ranging from 1 = 'almost never' to 5 = 'very often'.Cronbach's alpha for task-related conflict was .76,and for relationship conflict was .92.

Conflict Management Styles were measured with the Dutch Test for Conflict
Handling (DUTCH: see De Dreu et al., 2001).This measure was translated to Spanish using the standard back-translation procedure.The scale has 16 items, four items for each conflict management style measured: forcing (e.g.'I fight for a favorable outcome for myself'), avoiding (e.g.'I avoid a confrontation about our differences'), problem solving (e.g.'I stand up for both my own and the other's goals'), and yielding (e.g.'I concur with the other party').Items were rated on 5-point Likert-type scale ranging from 1 = 'almost never' to 5 = 'very often'.Cronbach's alpha for each dimension was: problem solving .75;forcing .75;avoiding .72;and yielding .76.

Statistical analysis
Prior to forming the various scales for regression analyses, we conducted a confirmatory factor analysis using asymptotic covariance matrix and maximum likelihood estimation (since the data were not normally distributed) to assess the discriminant validity of the substantive constructs measured in this study.Based on earlier research and theoretical notions, we expected three underlying factors from the 23 items that made up our measures: task conflict, relationship conflict, and workplace bullying.
Coping with conflict and bullying at work 9 We then examined a simple mediation or indirect effect model (Hypothesis 1): the degree to which employees perceive relationship conflict (M) mediates the effect of task conflict (X) on workplace bullying (Y).We tested this mediation hypothesis using a SPSS macro provided by Hayes (2013) that facilitates estimation of the indirect effect (ab) with a bootstrap approach to obtain confidence intervals.The application of bootstrapped confidence intervals (CIs) outperforms the normal theory Sobel tests since it avoids power problems introduced by asymmetric and other nonnormal sampling distributions of an indirect effect (MacKinnon et al., 2004) and makes Type I error less likely (Preacher and Hayes, 2008).
Finally, we predicted that the conflict management styles used to handle conflict (W) moderate the first (path task conflict -relationship conflict) and second (path relationship conflict -workplace bullying) stages of the mediating effect of relationship conflict on task conflict and workplace bullying (Edwards and Lambert, 2007;Preacher et al., 2007).This moderated-mediation model (see Figure 1) was tested using the above mentioned SPSS macro (see model 58: Hayes, 2013), which allows combining moderation and mediation analyses or conditional indirect effects (Preacher et al., 2007): the strength of the hypothesized indirect (mediation) effect is conditional on the value of the moderator (i.e., conflict management styles).The SPSS macro facilitates the implementation of the recommended bootstrapping methods and permits the probing of the significance of conditional indirect effects at different values of the moderator variable.
Given that it is better to view conflict management styles as a set of related dimensions than as a superordinate construct, we tested each dimension separately to enable investigation of specific questions on the association of each dimension with a possible escalation of conflict to workplace bullying (i.e., we conducted four moderated-mediation models, one for each conflict management style acting as moderator).However, the effect of the remaining conflict management styles was controlled as they were introduced as covariates in each moderated-mediation analysis to reduce the likelihood of Type I error (i.e., familywise error rate when performing multiple tests).
Coping with conflict and bullying at work 10

Results
First, we compared the fit of two competing models (unifactorial model vs. three-factor model) to our data (Kelloway, 1998;Marsh et al., 2004).Models were based on the polychoric correlation matrix, and asymptotic covariance matrix was estimated since the data were not normally distributed.
--Please insert Table 1 here-Results showed that an overall measure (unifactorial model) was associated with non-acceptable goodness-of-fit statistics even after reducing the number of parameters in the model as suggested by Modification Indices (see Table 1).According to Modification Indices, error correlations between NAQ items 8 ("Intimidating behaviors such as finger pointing, invasion of personal space, shoving, blocking your way") and 14 ("threats of violence or physical abuse or actual abuse") were set free to be estimated since these two items represented the same kind of negative behavior (physically intimidating bullying: see also Einarsen et al., 2009).Hence, the proposed three-factor structure demonstrated good fit with the data, suggesting that task conflict, relationship conflict, and workplace bullying were distinct constructs (see Table 1).Thus, results allowed us to compute the various constructs by taking the average of their respective items.
We then tested our hypotheses.Table 2 reports means, standard deviations, and correlations among the main variables of the study.
--Please insert Table 2 here-Regarding the simple mediation model (hypothesis 1), results revealed that relationship conflict partially mediates the association between task conflict and workplace bullying (see Table 3).Furthermore, the index of mediation was .28(bootstrapped 95% CIs of .23 to .34) and the ratio of the indirect effect to the total effect of task conflict on workplace bullying was .65 (bootstrapped 95% CIs of .50 to .81), which means that 65% of the increase in exposure to bullying behaviors was due to the increase in relationship conflict among employees (see Mackinnon, 2008;Preacher and Kelley, 2011).Similarly, Preacher and Kelley's kappa-squared (2011) revealed a modest indirect effect (k 2 = .24;bootstrapped 95% CIs of .20 to .29).
--Please insert Table 3 here-Coping with conflict and bullying at work 11 Results from the moderated mediation models partially supported hypotheses 2 and 3 since main effects were found; whereas results did not support hypothesis 4 since there were no significant interaction effects.Bootstrap CIs corroborated these results as the indirect and positive effect of task conflict on workplace bullying through relationship conflict was observed independently of the levels of the conflict management style used (see Table 4).First, problem solving is negatively related to relationship conflict (B = -.36,p < .001)and workplace bullying (B = -.08,p < .01).

Discussion
This paper addresses workplace bullying as a conflict escalation process and explores the role played by conflict management styles in the association between interpersonal conflict and workplace bullying.Results from a confirmatory factor analysis indicate that interpersonal conflict and workplace bullying are different but related constructs.This is in line with previous theoretical assumptions that have noted the differences and commonalities between interpersonal conflict and workplace bullying and other counterproductive behaviors at work (e.g., Raver and Barling, 2008;Van de Vliert, 2010).
Our results suggest a conflict escalation process as workplace bullying develops since relationship conflict mediates the association between task conflict and bullying at work.This is congruent with previous theoretical assumptions and findings (e.g., Einarsen, 1999;Leymann, 1996;Zapf and Gross, 2001), and allows us to connect the literature on conflict and workplace bullying to obtain a more in-depth knowledge of the underlying interpersonal mechanisms of bullying at work.Indeed, our results underline that bullying can be conceived as a conflict escalation process, or a longstanding conflict, which is developed over a certain period of time.
Coping with conflict and bullying at work 12 According to this conflict escalation perspective, specific conflict management styles may encourage or discourage conflict and bullying at work.In this regard, following the Dual-Concern Model (see De Dreu et al., 2001), we explored the role of several conflict management styles on different conflict escalation stages (i.e., path task-relationship conflict and path relationship conflict-workplace bullying).Our results revealed that active attempts to manage conflict through problem solving and forcing seem the best strategy to prevent task conflict escalating to relationship conflict; in contrast, trying to avoid conflicts may lead to the escalation of conflict to more emotional issues (relationship conflict).Overall, these results are in line with previous studies that have indicated that the most effective way of dealing with task conflicts is using problem solving in combination with forcing (i.e., reframing the conflict from our owns perspective and then working through difference by exchanging accurate information to find compromise consensus: Van de Vliert et al., 1997); in contrast, although the avoiding approach aims at diminishing conflict and achieving harmony (i.e., postponing an issue until a better time, or simply withdrawing from a threatening situation: De Dreu and Van Vianen, 2001), it reflects a lack of conflict resolution and usually results in negative emotions and conflict escalation (Desivilya and Yagil, 2005;Dijkstra et al., 2009).
As for workplace bullying, the only strategy associated negatively with workplace bullying was problem solving or integrating both parties' interests and points of view about the conflict.On the other hand, yielding is associated positively with workplace bullying.Additionally, it should be noted that we did not find moderation effects of conflict management styles on the mediating effect of relationship conflict on the task conflict -workplace bullying association.These results are in line with those of Baillien and De Witte (2009), who reported a negative association between problem solving and bullying and found no moderation effects of conflict management styles on the relationship between the occurrence of conflict and bullying at work.Problem solving is an assertive and cooperative way of dealing with conflicts and involves an attempt to work with the other person to find a solution which fully satisfies the concerns of both parties; therefore it helps to reduce the intensity and hostility of conflict (Rognes and Schei, 2010).This seems the most suitable strategy for resolving interpersonal conflicts like workplace bullying since other styles may be Coping with conflict and bullying at work 13 dysfunctional, particularly yielding and accepting the situation (i.e., giving into others' demands in bullying situations is associated with higher levels of victimization and severe detrimental consequences for health: Zapf and Gross, 2001).

Limitations and further research
This paper has some methodological limitations that can influence our results and explain the lack of moderation effects.First, our findings are based on self-report data from a cross-sectional study; this could lead to common method variance although we offered variations in the response format and instructed the participants that there were no correct or incorrect answers (for a discussion, see Brannick et al., 2010).Indeed, the cross-sectional nature of the data and the use of self-report measures make it difficult to infer causality.Thus, further research should overcome these limitations by using a longitudinal design to capture workplace bullying as a conflict escalation process.Moreover, future studies may benefit from considering workplace bullying as a gradual process rather than an all-or-nothing phenomenon, thereby exploring the intensity of conflict in each bullying stage or sub-group (Leon-Perez et al., 2013).
Finally, as our sample was not representative of the Spanish workforce, the results cannot be generalized.Moreover, considering that Spain has a more collectivist culture than other European countries (Hofstede et al., 2010), "avoiding conflict can be undertaken to support relationships and promote the goals of both protagonist" in conflict (Tjosvold and Sun, 2002: p. 144).Thus, future studies should consider cultural variables in exploring the role of conflict management (e.g., under what circumstances is avoiding functional) in the escalation of conflict towards bullying situations.

Managerial implications
Despite the limitations inherent in the study design, our results also have implications for companies' conflict management and anti-bullying practices.Our first recommendation is that companies should focus on developing conflict management systems to prevent conflict escalation.For example, it seems that the adoption of problem solving strategies (or integrating both parties' interests and points of view about the conflict) helps de-escalate the intensity of a conflict and prevent workplace bullying.Low-intensity conflicts can be constructive or positive under certain Coping with conflict and bullying at work 14 circumstances, e.g. in jobs where some controversy among employees on aspects related to their tasks may encourage a climate of creativity and innovation (e.g., De Witt et al., 2012;Medina et al., 2005).However, when conflict reaches a higher intensity, it produces negative emotional reactions such as increased stress, decreased job satisfaction, and fear of social rejection (e.g., De Witt et al., 2012;Friedman et al., 2000).Thus, companies can reduce conflict escalation to improve employees' health and well-being and also productivity, corporate image and organizational brand by providing training on effective conflict management (e.g., De Dreu and Van Vianen, 2001;De Witt et al., 2012;Dijkstra et al., 2009).
In addition, we recommend pairing strengthening individual strategies with structural interventions (i.e., organizational strategies) since the development of antibullying policies or alternative dispute resolution systems may contribute to creating a constructive conflict resolution culture in the organization (e.g., Giorgi, 2010;Heames & Harvey, 2006;Leon-Perez et al., 2012).

Conclusion
This paper gives opportunities for bridging conflict and workplace bullying research arenas.Our results suggest that workplace bullying can be conceived of as a conflict escalation process that is perceived as stressful and threatening, leading employees to experience negative emotions.Thus, a preventive approach seems to be more appropriate to counteract bullying at work in this conflict management framework.