IJCMA 33,3 The link between interpersonal conflict and knowledge hiding: mediated by relatedness need 494 frustration, moderated by Received 12 May 2121 Revised 18 September 2021 narcissistic rivalry 14 November 2021 15 December 2021 Dirk De Clercq Accepted 16 December 2021 Goodman School of Business, Brock University, St. Catharines, Canada Tasneem Fatima Faculty of Management Sciences, International Islamic University, Islamabad, Pakistan, and Sadia Jahanzeb Goodman School of Business, Brock University, St. Catharines, Canada Abstract Purpose - The purpose of this study is to investigate the relationship between employees' experience of interpersonal conflict and their engagement in knowledge hiding, according to a mediating effect of their relatedness need frustration and a moderating effect of their narcissistic rivalry. Design/methodology/approach - The tests of the hypotheses rely on three-wave, time-lagged data collected among employees in Pakistan. Findings - A critical reason that emotion-based fights stimulate people to conceal valuable knowledge from their coworkers is that these employees believe their needs for belongingness or relatedness are not being met. This mediating role of relatedness need frustration is particularly salient among employees who are self- centered and see others as rivals, with no right to fight with or give them a hard time. Practical implications - The findings indicate how organizations might mitigate the risk that negative relationship dynamics among their employees escalate into dysfunctional knowledge hiding behavior. They should work to hire and retain employees who are benevolent and encourage them to see colleagues as allies instead of rivals. Originality/value - This research unpacks the link between interpersonal conflict and knowledge hiding by explicating the unexplored roles of two critical factors (relatedness need frustration and narcissistic rivalry) in this relationship. Keywords Interpersonal conflict, Knowledge hiding, Conservation of resources theory, Narcissistic rivalry, Relatedness need frustration Paper type Research paper Introduction Employees who undertake counterproductive work activities are important sources of International Journal of Conflict Management concern for organizations, because these activities undermine the quality of the Vol 1 No. J. 2023 PP. 494-518 organization's internal functioning and its competitive strength (Carpenter et al, 2021; Emnid Publishing Limdied 1044-4056 Kundi and Badar, 2021). A critical type of counterproductive behaviors is knowledge hiding, DXO 101106 10314-05-2021-0172 defined as employees' purposeful attempts to conceal valuable knowledge from othermembers who ask for that information (Connelly et al, 2012; Serenko and Bontis, 2016). Narcissistic Knowledge hiding behaviors can create detrimental outcomes, at individual (e.g. lower job rivalry performance; Jahanzeb et al, 2020), team (e.g. lower team creativity; Bogilovic et al, 2017) or organizational (e-g. lower firm innovative capabilities; Xiao and Cooke, 2019) levels. So why do people adopt them? Adverse interpersonal relationships might direct employees toward negative activities, despite their inherent disadvantages (Malik et al, 2019; Yuan et al, 2021). In addition, as prior research indicates, interpersonal conflict (i.e. extent to which employees' interactions with coworkers are marked by emotion-based tensions or fights; Ilies et al, 495 2011) is related to knowledge hiding (Semerci, 2019). Other predictions anticipate that this link may be mediated, such as by employee cynicism (Akhlaghimofrad and Farmanesh, 2021), envy (Peng et al, 2021) or tarnished well-being (Losada-Otalora et al, 2021). With this study, we seek additional insights into why interpersonal conflict might escalate into knowledge hiding behaviors, as well as evidence regarding which personal circumstances might make this detrimental process more likely to materialize. We investigate two unexplored, relevant factors, namely, employees' relatedness need frustration or their sense of unmet needs with respect to "feeling close and connected to others, feeling a sense of belongingness, caring for and feeling cared for by others" (Longo et al, 2016, p. 296) [1], and their narcissistic rivalry or the extent to which they consider others unworthy rivals (Back et al, 2013). First, in line with conservation of resources (COR) theory (Hobfoll et al, 2018), we postulate that employees' exposure to resource-draining, emotion-based conflicts may translate into knowledge hiding activities because they express frustration about their unfulfilled relatedness needs (Van den Broeck et al, 2014). The focus on this critical path is theoretically relevant because it helps pinpoint how employees might cope with the depletion of a key resource that is highly valuable to their personal well-being: their emotional connectivity with others (Hobfoll, 2001). Second, and also consistent with COR theory, we propose that this process is invigorated when employees exhibit high levels of narcissistic rivalry, because this personal factor, and its associated tendencies for self- love, increases the sense of offense employees experience when they must deal with emotion- based quarrels (Leckelt et al, 2015). Contributions By addressing these issues, we make several contributions. First, we establish that the positive relationship between interpersonal conflict and knowledge hiding can be explained by a factor that goes beyond employees' general work-related well-being (Losada-Otalora et al, 2021) and instead taps into a specific individual need, that is, the need for connectivity or belongingness (Longo et al, 2016). This hitherto overlooked explanatory mechanism is worth exploring from a conceptual perspective, because it generates relevant insights about a critical channel, over which organizational decision makers have relatively limited control but that still feeds employees' unhappiness at work: their sense that they are left out by others (Trepanier et al., 2016). In doing so, we offer unique theoretical insights into the possibility of a negative spiral for employees. That is, they suffer from interpersonal conflict situations, and their associated frustrations about being abandoned lead them to conceal pertinent knowledge from coworkers, which threatens to exacerbate the problems in their already suboptimal coworker relationships (Serenko and Bontis, 2016; Xiao and Cooke, 2019) Second, we address calls for contingency approaches to the harmful effects of interpersonal conflict (Agotnes et al, 2018; Kundi and Badar, 2021). We specifically propose that employees' narcissistic rivalry may intensify their annoyance with coworkers who do not appear to make any effort to maintain high-quality relationships with themIJCMA (Back et al, 2013; Hobfoll and Shirom, 2000). The intensification then may activate their 33,3 relatedness need frustration and subsequent knowledge hiding, which they perceive as well-justified responses (Van den Broeck et al, 2014). The theoretical value of examining the contingent role of this specific dark personality trait stems from its ability to pinpoint a notable, indirect, dysfunctional effect, thereby extending traditional considerations of its direct harmful impact on work attitudes or behaviors (Helfrich and Diet], 2019; Leckelt 496 et al, 2015). In a related contribution, we address calls for novel insights into the influence of dark, competition-oriented personality traits in general on employees' behaviors (Hernaus et al, 2019; Pan et al, 2018). In particular, the moderated mediation dynamic that underlies our proposed model - with a mediating role of relatedness need frustration and a moderating role of narcissistic rivalry - provides pertinent theoretical insights into how the aforementioned counterproductive spiral might get reinforced among self-centered employees who tend to blame others for their hardships. From a positive angle, we offer organizational decision-makers recommendations for how to protect against the danger that emotion-based fights cause valuable knowledge to "disappear" within their ranks, by discouraging narcissistic perceptions of coworkers as negligible rivals. Importantly, our theoretical focus is on the moderating effect of narcissistic rivalry on the relationship between interpersonal conflict and relatedness need frustration, not on how this personal factor may increase the likelihood of interpersonal conflict in and of itself, nor its direct effects on relatedness need frustration or knowledge hiding. In a post hoc analysis (reported in the Results section), we do not find any empirical support for the presence of a sequential mediation logic, in which narcissistic rivalry spurs knowledge hiding behavior through interpersonal conflict and subsequent relatedness need frustration. Empirical context We investigate these topics in the non-Western setting of Pakistan, whose cultural features are highly relevant for this study. First, its high levels of uncertainty avoidance (Hofstede et al, 2010) imply that employees may experience great hardships when they encounter emotion-based tensions in peer relationships, which tend to generate significant uncertainty about how to perform job tasks (Liu et al, 2011; Shaukat et al, 2017). Therefore, they may react forcefully, with enhanced levels of relatedness need frustration and knowledge hiding. Second, the elevated levels of collectivism that mark this country (Hofstede et al, 2010) may leave employees strongly upset by exposure to interpersonal conflict, because such conflict undermines normative expectations about the importance of group harmony and collective well-being (Hook et al, 2009). Emotion-based quarrels might cause employees in collectivistic settings to experience a strong sense of abandonment. These two cultural features, in turn, may encourage narcissistic employees to feel even more offended by interpersonal conflict, representing an additional stimulus to respond forcefully and in counterproductive ways (Leckelt et al, 2015). To be clear though, our theoretical arguments are explicitly country-neutral, so the nature of the proposed relationships is unlikely to vary across countries, even if their strength might. In other words, our investigation of how (narcissistic) employees' exposure to interpersonal conflict may escalate into relatedness need frustration, and then knowledge hiding, is compelling in any country but especially so in uncertainty-avoidant, collectivistic settings, such as Pakistan, in which employees tend to respond vigorously to conflict-laden interpersonal relationships (Kundi and Badar, 2021). Theoretical background The effective management of knowledge is an instrumental facet of an organization's internal functioning, which informs both employee and employer success(Trivedi and Srivastava, 2021; Wang and Noe, 2010). Prior studies underscore the Narcissistic advantages of employees who are willing to openly exchange knowledge with one another rivalry (De Clercq and Pereira, 2020; Yang et al, 2020). According to this research perspective, organizations should invest in management systems to encourage knowledge sharing throughout their workforce (Men et al, 2019; Yun and Lee, 2017). In addition to the advantages of open communication though, it can generate costs for individual employees. For example, some employees may worry about a loss of power if they grant others easy access to their personal insights and expertise (Cabrera and Cabrera, 2002; Casimir et al, 497 2012). In response, they may engage in knowledge hiding behavior or deliberate efforts to deny knowledge requests by other members (Connelly and Zweig, 2015; Malik et al, 2019). For example, they might believe that knowledge requesters should earn access to such unique insights, which they cannot do if, in the focal employees' view, those coworkers display negative interpersonal treatment (Serenko and Bontis, 2016). The knowledge hiding concept is interesting, in that it entails three pertinent, interrelated components: evasive hiding or promising to share knowledge in the future without any intention to do so; playing dumb or pretending not to understand others' requests for knowledge; and rationalized hiding or giving false arguments for why open knowledge sharing is not appropriate (Connelly et al, 2012). Consistent with previous studies, we treat knowledge hiding as a comprehensive construct that includes all three dimensions (Fong et al, 2018; Jahanzeb et al, 2019). We thus adhere to the well-established approach that acknowledges that these dimensions constitute a single, encompassing type of behavior, in that "employees engage in knowledge hiding or not, irrespective of the specific form it takes" (Jahanzeb et al, 2020, p. 258). Existing evidence about the determinants of knowledge hiding behaviors is varied. Some studies point to the role of personal factors, such as employees' sense of psychological ownership (Peng, 2013), trait competitiveness (Hernaus et al, 2019) or proving goal orientation (Rhee and Choi, 2017). Others highlight factors that reflect the organizational context, such as the extent to which employees suffer from excessive time pressures (Skerlavaj et al, 2018), dysfunctional political games (Malik et al, 2019) or job insecurity (Serenko and Bontis, 2016). Another set of interpersonal factors includes exposure to workplace ostracism (Zhao et al, 2016), bullying (Yuan et al, 2021) or interpersonal conflict (Semerci, 2019). We focus on this latter factor. In contrast with task or process conflict O'Neill et al, 2018), interpersonal conflict is notably destructive and marked by emotion- laden quarrels that make employees feel annoyed or even angry in interactions with colleagues (Baillien et al, 2017; Pooja et al, 2016). Employees engaged in such energy- consuming quarrels perceive that their work efforts are not appreciated, so they tend to focus on their individual well-being rather than efforts to add to the success of their organization or its constituents (Lau and Cobb, 2010; Shaukat et al, 2017). Several negative responses may arise in interpersonal conflict situations, such as diminished creativity (De Clercq and Belausteguigoitia, 2021) and organizational citizenship behavior (Pooja et al., 2016) or enhanced quitting intentions (Li et al, 2011) and deviant work behaviors (Kundi and Badar, 2021). As mentioned in the Introduction, we identify three studies that cite a positive link between exposures to interpersonal conflict and knowledge hiding, as a specific type of organizational deviance; they seek to explain this link according to the mediating influences of employee cynicism (Akhlaghimofrad and Farmanesh, 2021), envy (Peng et al, 2021) or well-being (Losada-Otalora et al, 2021). Peng et al (2021) rely on affective events theory, whereas the other two studies are not anchored in any specific theoretical framework. More fundamentally, none of these studies addresses whether or how victims of interpersonal conflict might suffer a sense of abandonment that undermines theirIJCMA need for belongingness (Longo et al, 2016; Van den Broeck et al, 2014). To address this gap, 33,3 we seek to explicate a possible mediating role of employees' relatedness need frustration in the interpersonal conflict-knowledge hiding link. In addition, we propose that the explanatory role of relatedness need frustration might be especially powerful among employees who exhibit high levels of narcissistic rivalry (Leckelt et al, 2015). These employees maintain a grandiose sense of self and view others as 498 unworthy rivals who do not have the right to make their lives difficult (Back et al, 2013). In work contexts, employees marked by narcissistic rivalry see their coworkers as inferior entities who should refrain from any negative treatments of them (Helfrich and Diet], 2019). We investigate how this dark personality trait in turn might combine with and influence employees' responses to interpersonal conflict. If employees exhibit elevated levels of narcissistic rivalry, they may be particularly bothered by emotion-based fights, which undermine their self-image, so they react very strongly to the negative situation (Grijalva and Newman, 2015; Hobfoll, 2001). Formally, the mediation dynamic by which interpersonal conflict spurs knowledge hiding behavior, through relatedness need frustration, likely is invigorated by employees' tendencies to see their colleagues as unworthy rivals who do not have the right to bother them with emotion-laden fights. Conservation of resources theory We rely on COR theory as a conceptual framework to anchor our arguments about the mediating role of relatedness need frustration and moderating role of narcissistic rivalry (Hobfoll, 1989; Hobfoll et al, 2018). According to this broadband theory, employees' work- related thoughts and actions are largely informed by their motivation to protect their current resource bases and avoid additional resource losses when they face resource-draining work situations (Hobfoll and Shirom, 2000). For this study, we draw from two premises that reflect the rationale of COR theory and that appear in recent applications of this theory to explain employees' responses to adverse work conditions. That is, even if these premises were not formally proposed by Hobfoll and his colleagues, they are consistent with the logic that underpins this theory, and they have been empirically validated in prior research. The first premise is that the threat of resource depletion, caused by adverse treatments at work, tends to steer employees toward beliefs and behaviors that enable them to cope with such depletion (De Clercq et al, 2019, Haq et al, 2021). For example, employees' counterproductive work behaviors represent responses that help them cope with the difficulties that arise with resource-depleting, time-related work stress (De Clercq et al, 2019). Similarly, Haq et al. (2021) draw from COR theory to show how employees who experience job insecurity seek to cope by forming beliefs about work-induced mental health deprivation, which in turn diminishes their willingness to respect work-related deadlines. The second premise is that certain personal factors might catalyze this coping process, particularly those factors that make it more likely that adverse work conditions appear to cause significant harm to the quality of employees' work functioning (De Clercq et al, 2019, De Clercq, 2021). For example, in studying how materialistic employees respond to perceived career compromises with plans to quit, De Clercq (2021, p. 3) explicates how a personal characteristic such as materialism can "inform the severity [italics in the original] of the hardships experienced in the presence of resource-draining work situations and the associated need to release their frustrations." In a similar vein, time-related work stress more likely escalates into negative work behaviors among employees with Machiavellian, narcissistic or psychopathic tendencies, who experience such stress as more upsetting (De Clercq et al, 2019). In addition, COR theory conceptualizes resources broadly, as entailing any "objects, personal characteristics, conditions or energies that are valued in their own right or that arevalued because they act as conduits to the achievement or protection of valued resources" Narcissistic (Hobfoll, 2001, p. 339). A particular resource that employees strongly value, according to rivalry Hobfoll (2001, p. 342), is affection they receive from others, which speaks to emotional connectivity with coworkers. In their comprehensive overview of COR theory, Hobfoll et al. (2018, p. 104) also emphasize that employees who suffer from resource-draining work conditions likely "enter a defensive mode to preserve the self that is often aggressive and may become irrational." Consistent with the aforementioned first COR premise, we propose that employees' expressed relatedness need frustration, and their subsequent propensity to 499 conceal knowledge from coworkers, may arise from their exposure to interpersonal conflict, in that it represents a way to release their irritation about their depleted emotional connectivity resources (Haq et al, 2021). Employees' expressed relatedness need frustration, as conceptualized herein, accordingly does not capture a state of resource drainage, but reflects their response to experienced resource drainage. Moreover, we do not assess the emotional connectivity resources directly, but we theorize that their depletion, due to the experience of interpersonal conflict, informs employees' expressed relatedness need frustration and knowledge hiding behaviors, in that these responses function like coping tactics. These coping tactics, in essence, enable employees to vent their frustrations and protect themselves against the hardships that stem from emotion-based tensions (Hobfoll et al, 2018, Shaukat et al, 2017). In investigating the connection of perceived contract breaches with knowledge hiding behaviors, Jahanzeb et al (2020, p. 253) similarly draw from COR theory and conceptualize purposeful knowledge hiding "as a means to cope with and vent their frustrations with the hardships that come with broken organizational promises." According to the second COR premise, such self-protective responses should be invigorated by the extent to which employees possess personal characteristics that make the responses appear highly justified (De Clercq, 2021; Hobfoll and Shirom, 2000). The probability that employees react to interpersonal conflict with expressed relatedness need frustration, and then knowledge hiding, is greater if they consider other people inferior and unworthy of bothering them (Helfrich and Dietl, 2019). Such narcissistic rivalry raises an expectation that others should treat them with affection (Leckelt et al, 2015). When these expectations get violated by emotion-based quarrels, employees likely respond by complaining about their sense of abandonment (Longo et al, 2016). Their narcissistic rivalry increases the perceived offense of depleted emotional connectivity resources due to interpersonal conflict, which triggers their desire to express relatedness need frustration and conceal pertinent knowledge from colleagues. The proposed conceptual framework is in Figure 1, and its constitutive hypotheses are detailed next. Hypotheses Mediating role of relatedness need frustration We hypothesize a positive relationship between employees' suffering from interpersonal conflict and their expressed relatedness need frustration. As mentioned, COR theory posits that employees' work-related feelings in resource-draining situations reflect their desire to express themselves in ways that help them cope with the difficulties created by these situations (De Clercq et al, 2019). For example, the challenges that employees encounter in the presence of interpersonal conflict powerfully drain their emotional connectivity resources (Ilies et al, 2011; Pooja et al, 2016). Again, we do not investigate the link between interpersonal conflict and depleted emotional connectivity directly, but we study it indirectly by theorizing about the consequences of this link with regard to how employees feel about their work situation. In particular, when they are convinced that their coworkers do not care for their personal well-being, employees may develop a desire to counter theIJCMA drainage of their emotional connectivity resources by venting frustrations about their 33,3 unfulfilled need to belong (Longo et al, 2016; Trepanier et al, 2016). Employees complain about coworkers' failures to address their relatedness needs, in the presence of interpersonal conflict, because they associate the emotion-based quarrels with a sense of being left out (Hon and Chan, 2013; Wong and Chan, 2020). In a related way, these quarrels may create a conviction that their devoted work contributions are not acknowledged, so they interpret 500 this adverse situation as a sign they are isolated and not taken seriously by others (Kim and Beehr, 2018; Shaukat et al, 2017). Conversely, employees who enjoy harmonious interpersonal relationships hold more positive views about the concerns that their coworkers have for their work functioning (Chen et al, 2015), which likely has a positive impact on their sense of being emotionally connected. To the extent that employees do not suffer emotion-based tensions in interactions with coworkers, they should experience a lower desire to express frustrations about unmet relatedness needs: HI. There is a positive relationship between employees' exposure to interpersonal conflict and their expressed relatedness need frustration. The tenets of COR theory similarly indicate that employees who experience frustration about unmet relatedness needs may engage in knowledge hiding behaviors, despite the harmful consequences for their coworkers (Xiao and Cooke, 2019), because they regard these behavioral responses as justified solutions to channel their disappointment that coworkers seemingly do not care about their isolation at work (Hobfoll et al, 2018). Relatedness need frustration thus may lead to an increased probability that they deny coworkers valuable knowledge, because this response helps them cope with their sense of abandonment at work (Serenko and Bontis, 2016). In addition, COR theory predicts that employees' work behaviors are critically informed by their goal to gain resources they can use to counter hardships experienced in the presence of negative work-related feelings (Hobfoll, 2001; Hobfoll and Shirom, 2000). When employees feel frustrated because their coworkers seemingly do not care about their relatedness needs, they may seek to conceal valuable knowledge as a means to generate resource gains, in the form of a sense of deservedness, which also can generate personal satisfaction (Jahanzeb et al, 2020). That is, employees who experience abandonment try to even the score through reactions that offer them some personal fulfillment (Ryan and Deci, 2000, Xiao and Cooke, 2019). Finally, employees who express relatedness need frustration may conceal knowledge to conserve their valuable energy resources (Quinn et al, 2012; Warburton et al, 2020), especially if their frustration about unfulfilled relatedness needs limits their stamina to accommodate knowledge requests from Narcissistic rivalry Interpersonal Relatedness need Knowledge Figure 1. conflict frustration hiding Conceptual modelcoworkers (Jha and Varkkey, 2018; Longo et al, 2016). Therefore, they seek to safeguard Narcissistic their personal energy resources by refusing to share their insights or experiences (Hobfoll rivalry and Shirom, 2000): H2. There is a positive relationship between employees' expressed relatedness need frustration and their knowledge hiding- These arguments imply a critical mediating role of employees' expressed relatedness need 501 frustration. Their exposure to emotion-based quarrels translates into more knowledge hiding because it prompts employees to criticize their coworkers for failing to address their need for belongingness, as a means to cope with the intrusive conflict (Longo et al, 2016; Warburton et al, 2020). Prior research indicates a similar mediating role of employees' expressed general need frustration in determining how other adverse conditions, such as job insecurity (Van den Broeck et al, 2014) or suppressive emotional regulation (Benita et al, 2020), produce negative work outcomes. As a complement, we propose that employees' sense that their relatedness needs remain unfulfilled can help explain the escalation of their interpersonal conflict into knowledge hiding. Their irritation about being abandoned by others gets "out of hand," such that they purposefully deny coworkers access to valuable knowledge resources (Hobfoll, 2001). Notably, our logic implies partial, rather than full, mediation. That is, we predict that expressed relatedness need frustration represents a critical but not exclusive factor that explains the connection between interpersonal conflict and knowledge hiding behavior, consistent with previous studies that pinpoint mediating roles of other factors, such as cynicism (Akhlaghimofrad and Farmanesh, 2021), envy (Peng et al, 2021) and well-being (Losada-Otalora et al, 2021): H3. Employees' expressed relatedness need frustration mediates the relationship between their exposure to interpersonal conflict and their knowledge hiding. Moderating role of narcissistic rivalry We predict a moderating role of employees' narcissistic rivalry. According to COR theory, the forcefulness of employees' negative responses to adverse coworker treatments depends on the intensity of the experienced hardships (De Clercq et al, 2019). As a dark personality trait, narcissistic rivalry reflects the degree to which employees cannot stand it when others, whom they regard as inferior, are the center of attention (Back et al, 2013). When employees perceive others as unworthy rivals, they require and expect those others to maintain peaceful relationships with them (Leckelt et al, 2015). Employees with this personality trait may feel especially upset with coworkers who seemingly do not make any effort to avoid interpersonal fights and who thus make the depletion of their emotional connectivity resources more salient (Hobfoll, 2001; Trepanier et al, 2016). Conversely, employees who do not hold beliefs that others are unworthy likely experience a lower need to complain about being abandoned in the presence of interpersonal conflict (Helfrich and Diet], 2019). They recognize coworkers as peers, with just as much "right" as anyone to create conflict, so they are less likely to feel offended by emotion-based conflict. With their more nuanced expectations of how coworkers should behave, their annoyance with interpersonal conflict may lessen (Leckelt et al, 2015) and be less likely to escalate into expressed relatedness need frustration:IJCMA H4. The positive relationship between employees' exposure to interpersonal conflict and 33,3 expressed relatedness need frustration is moderated by their narcissistic rivalry, such that this relationship is stronger at higher levels of narcissistic rivalry. These arguments indicate the presence of a moderated mediation dynamic (Preacher et al, 2007), such that narcissistic rivalry functions as a contingent factor that invigorates the 502 indirect relationship between interpersonal conflict and knowledge hiding, through expressed relatedness need frustration. At high levels of narcissistic rivalry, which operates as a catalyst, frustrations about unfulfilled relatedness needs get more easily activated as explanatory mechanisms that channel resource-draining, emotion-based fights into enhanced knowledge hiding (Back et al, 2013). Their self-centered, rivalry-based tendencies toward others make it more likely that employees with high levels of narcissistic rivalry are offended by the interpersonal conflict that marks their interactions with coworkers (Judge et al., 2006). This personal factor increases their propensity to express their sense of isolation, in an attempt to cope (De Clercq et al, 2019), which drives them to conceal valuable knowledge (Hernaus et al, 2019). In summary, to the extent that employees experiencing interpersonal conflict also see others as negligible rivals, we predict that their propensity to hide knowledge can be explained more powerfully by their expressed frustrations about unfulfilled relatedness needs H5. The indirect positive relationship between employees' exposure to interpersonal conflict and their knowledge hiding, through their expressed relatedness need frustration, is moderated by their narcissistic rivalry, such that this indirect relationship is stronger at higher levels of narcissistic rivalry. Method Sample and data collection The research hypotheses were tested with survey data collected among employees of seven universities in Pakistan. This focus on one industrial sector helps account for unobserved industry differences that might inform the likelihood that employees seek to hide knowledge from coworkers (e.g. strong external competitive pressures that drive employees to protect potentially precarious job positions). The surveys were written in English, which is the formal language of educational communication in Pakistan. One of the authors leveraged professional connections to identify pertinent universities, then made personal visits to the sites to distribute the surveys through the universities' internal mail systems. We applied a carefully prepared design to avoid biases that might arise with survey- based data collections. We collected the data in three rounds, with a time lag of three weeks between each round. The first survey asked employees about their interpersonal conflict with coworkers and their own narcissistic rivalry, the second gauged their relatedness need frustration and the third survey assessed their knowledge hiding behavior. The temporal separation between rounds helps reduce concerns about reverse causality, as well as the risk of an expectancy bias that arises when participants figure out the research hypotheses and adapt their responses accordingly, which can be more likely if all questions appear in a single survey (Malhotra, 2010). The three week period also is short enough to diminish the chances of bias due to disruptive external events that might occur during the research window and that would affect the theorized relationships. To avoid recall bias, we asked respondents to reflect on their work situation and work-related beliefs and behaviors over the prior six months (Malhotra, 2010).Moreover, several applied measures helped diminish the probability of acquiescence and Narcissistic social desirability biases, by assuring participants that their rights were fully protected. In rivalry particular, each survey was preceded by a statement that clarified the completely voluntary basis of their participation and the complete confidentiality of their responses. Employees were also informed that their organizations would not know who participated or not and that any research output would only include general patterns, based on aggregate data (Jordan and Troth, 2020). They deposited the completed surveys in a secure box to which only the researchers had access. To mitigate framing and representativeness biases, the 503 surveys emphasized that there were no good or bad answers, that it was normal that different participants would give varying answers to specific questions and that it was important for the relevance of the study that everyone provided truthful answers, pertaining to their personal situation (Malhotra, 2010). The complete lists of employees of all seven participating universities served as the sample frame, from which we selected 400 employees for possible participation, using a random digit generator. Of the 400 surveys distributed, we received 345 completed responses in the first round, 290 in the second and 275 in the third round. After excluding incomplete responses, the final sample consisted of 261 employees, for a response rate of 65%. Among the respondents, 58% were men and 42% were women; 92% had at least a master's degree; they had worked for their university for an average of eight years; and 49% were faculty and 51% were administrators. In each survey round, we found no significant differences in construct values between early and late respondents, indicating that systematic response bias was not a concern (Armstrong and Overton, 1977). Measures The focal constructs were assessed with measurement items drawn from prior research, applying five-point Likert anchors (1 = strongly disagree; 5 = strongly agree). Perceptions of interpersonal conflict. To assess employees' experience of emotion-based conflict in their interactions with coworkers, we relied on a four-item scale of relationship conflict (Jehn, 1995). Two sample items were, "There is lots of interpersonal friction between my coworkers and myself" and "There is lots of emotional tension between my coworkers and myself" (Cronbach's alpha = 0.75). Narcissistic rivalry. We assessed the extent to which employees regard other people as unworthy rivals with a nine-item scale of narcissistic rivalry (Back et al, 2013). For example, participants rated whether "Other people are worth nothing," "I enjoy it when another person is inferior to me" and "I can barely stand it if another person is at the center of events" (Cronbach's alpha = 0.95). Relatedness need frustration. To measure the extent to which employees expressed concerns about being abandoned by coworkers, we applied a three-item scale of experienced frustration with unfulfilled relatedness needs (Longo et al, 2016). In light of this study's conceptual arguments, we adapted the wording of the original items to reflect employees' beliefs about their coworkers, instead of other people in general. The three items were "Sometimes, I feel rejected by my coworkers," "I feel a bit alone when I'm with my coworkers" and "On occasions, I feel my coworkers are a bit cold toward me" (Cronbach's alpha = 0.70). Knowledge hiding. We measured employees' propensity to conceal knowledge from their coworkers with a 12-item scale of knowledge hiding (Connelly et al, 2012). The items captured the three dimensions of evasive hiding, playing dumb and rationalized hiding. To diminish social desirability bias, the survey emphasized that it was normal that people would vary in their answers and that it might not always be feasible for employees to shareIJCMA all of their knowledge with coworkers. Moreover, the survey items were preceded with a 33,3 statement that asked the respondents to reflect on how they tended to react when coworkers ask them for information. Three example items were: "I sometimes agree to help them even if I don't really intend to" (evasive hiding); "I sometimes say that I do not know, even though I do" (playing dumb); and "I sometimes tell them that my boss does not let anyone share this knowledge, even if this is not the case" (rationalized hiding). The Cronbach's alpha values 504 exceeded the 0.70 threshold for each of the three dimensions (0.79 for evasive hiding, 0.78 for playing dumb and 0.82 for rationalized hiding), and the reliability of the 12-item measure was high too (Cronbach's alpha = 0.89). A confirmatory factor analysis (CFA) indicated that the three dimensions loaded strongly (p 0.3.84) than the fit of their six constrained counterparts, in which the correlation between constructs was fixed to equal 1 (Lattin et al, 2003) Common method bias. We checked for the presence of common method bias with two statistical tests. First, we performed Harman's one-factor test (Huang et al, 2020; Podsakoff and Organ, 1986) with an exploratory factor analysis to determine if a single factor captured most of the total data variance. The first extracted factor accounted for only 29.7% of the total variance, which reduces concerns of common method bias. Second, we performed a CFA to compare the fit of a one-factor model, in which all the measurement items loaded on a single construct, with the fit of the aforementioned four-factor model. The one-factor model generated significantly worse fit (x (6) = 848.30, p