RECEIVING STARS: HOW DOES PERFORMANCE GAP AFFECT TEAM LEARNING?
Master thesis, MSc. Human Resource Management University of Groningen, Faculty of Economics and Business
27th of July, 2020
LIANNE KAMMINGA Student number: 3483878
Supervisor: Dr. Y. Yuan email: yingjie.yuan@rug.nl
Second assessor: Dr. T.A. de Vries
1
RECEIVING STARS: HOW DOES PERFORMANCE GAP AFFECT TEAM LEARNING?
Abstract
Firms tend to spend many resources on hiring exceptionally performing personnel, since star performers are regarded as bringing exceptional contributions to their organizations. However, studies show contradicting findings regarding the effects of a performance gap, between the star and other team members, on team learning. This study aimed to gain insight into how and when teams do (not) take advantage of performance differences within their team. I posited that status conflict experienced by team members towards star performers mediate a negative indirect effect between performance gap and team learning, which was not confirmed by the data. Moreover, I proposed that the negative indirect effect is most negative at high levels of team competitive environment, even as at high levels of narcissism among stars. Findings show that the negative indirect effect indeed was moderated by team competitive climate. However, the data showed a significant positive indirect effect at low competitive climate, but no significant effect at high levels. No moderated mediation effect existed at different levels of star narcissism. I have studied this matter by conducting questionnaires among team members and I asked their managers to evaluate team members’ performance and teams’ learning.
Keywords
Star performance; Performance gap; Team learning; Status conflict; Narcissism;
2
INTRODUCTION
For years, companies have focused on recruiting new and the ‘best and brightest’ personnel (Meester, 2019). An extensive amount of research focused on individuals who bring exceptional value to organizations (Lacetera et al., 2004) and how to enlarge their value (Groysberg & Lee, 2009).
Especially star performers are regarded as bringing exceptional contributions to the organizations they work in (Wright, Coff, & Moliterno, 2014). Firms tend to spend many resources such as time and financial input on hiring exceptional performing employees in order to advance team performance (Groysberg & Lee, 2009). Over the past years, there has been a shift in the structure of work towards teams instead of individual jobs (Kozlowski & Ilgen, 2006) and organizations are depending more on teams to perform complex tasks in a dynamic environment (Thamhain, 2013). It is highly valuable to know how to stimulate team learning because research has repeatedly shown that team learning is positively related to team performance (Edmondson, 1999; Bunderson & Sutcliffe, 2003). Hence, star performance and team learning are nowadays important and active research areas.
However, both in practice and literature, stars’ impact on team members seems of great variance and unknown mechanisms. On one hand, star performance is positively related to coworkers’
productivity (Zucker, Darby, & Armstrong, 1998), organizational productivity (Lacetera et al., 2004), and team creativity (Yuan, Ning, Jingzhou, Jingyu, & Harris, 2018). As just depicted, studies have mainly focused on positive effects of stars. Nonetheless, earlier practice showed that little is known about how to stimulate positive effects of stars on teams (Groysberg & Lee, 2009).
On the other hand, research on negative consequences of stars on teams is lacking. Li, Li, Li, and Li (2019) emphasized on this gap of existing literature about detrimental effects of star performers on team processes. They for instance found that a star in a central position reduced non-stars’ learning.
Even though stars’ interaction patterns affect non-stars’ learning behavior both positively and negatively (Li, et al., 2019), the underlying mechanisms that explain how star performance, or more specifically performance differences within teams, affect team learning remain unclear. The differences in performance between the star and the other non-star members of the team will be called
‘performance gap’ throughout this research.
3
Many researchers have made attempts to investigate several antecedents and consequences of performance differences within teams and team learning, and one of the theories that researchers applied is conflict theory (Lewis, French & Steane, 1997; Bendersky & Hays, 2012). However, research on conflicts mainly focused on how different kinds of conflicts promote or prevent team performance, both consciously and unconsciously (Sell, Lovaglia, Mannix, Samuelson & Wilson, 2004; Loch, Huberman & Stout, 2000). One prevalent type of conflict is status conflict, defined as controversy about individuals’ relative status in the team’s social hierarchy (Bendersky & Hays, 2012). Status in itself is omnipresent in both everyday life as in the work environment (Barkow, 1989) and we cannot deny that status influences the behavior of members in any organization (Loch et al., 2000) and possibly many teams. Because of the prevalence and important influences of both status and conflicts on team processes, but the missing knowledge about how status conflicts specifically influence the relationship between performance differences and team learning, I will investigate the mediating role of status conflict on this relationship.
Status is related to, and partially depending on, competence, cognitive ability and task performance (Huberman, Loch, Önculer, 2004; Knight, 2000; Bhattacharya & Dugar, 2012).
Therefore, it is likely that the high-performers have a certain status. Additionally, there are several advantages of being the highest performer and having a certain status within a team (Huberman et al., 2004; Call, Nyberg & Thatcher, 2015) and as Loch et al. (2000) stated, the pursuit of status is something in all of us. Due to the status and advantages that come with a better performer in teams, it is likely that status conflicts arise in teams with increased performance gaps.
At the same time, status-seeking reduces the collaboration between team members, since status seekers are more likely to compete (Wu, Loch, & Ahmad, 2011). Information sharing with team members is impeded when status conflict arises, because they cause competitive exchanges (Bendersky & Hays, 2012). Both collaboration and information sharing are important in enhancing learning (Cordes, 2016; Edmondson et al., 2007), so if status conflicts impair these activities, it is likely that status conflicts reduce team learning. Hence, I propose that status conflicts towards a star performer mediate the negative indirect effect of performance gap on team learning (hypothesis 1).
4
However, status conflicts do not always emerge automatically or in any environment. Previous research focused on teams and organizations in which conflicts occur. Competition is one of the possible triggers of conflicts and specifically status conflicts are more likely to emerge in teams where competition is important (Pai & Bendersky, 2020). A team competitive climate is characterized as a shared perception (Fletcher, Major, & Davis, 2008) in which members demonstrate and experience behaviors that are focused on individuals’ own performance and personal achievements (Maruping, Magni, Caporarello, & Basaglia, 2008). Nowadays, many companies create competition between team members for e.g. recognition (Steinhage, Cable & Wardley, 2017), probably because they believe this will be beneficial for the teams. Nevertheless, among other things, competition hinders team learning compared to cooperation (Van der Vegt, De Jong, Bunderson & Molleman, 2010). Conflicts and competition are likely to be triggered in certain teams, and competitive team climates seems to play a role for status conflict to arise. Within this study I propose that a competitive team environment moderates the negative indirect effect of performance gap on team learning through status conflict, in which the indirect effect is more negative when levels of competitive climate are high (hypothesis 2).
Besides team context, star traits possibly inhibit or encourage team learning too. High scores on the Narcissistic Personality Inventory (NPI) among younger generations increase (Twenge, Konrath, Foster, Campbell, & Bushman, 2008) and narcissism among leaders and celebrities is becoming more visible (Gluck, 2014), but the effect of narcissism among stars on teams has not been researched yet. Narcissistic people have an increased desire for admiration and grandiosity and generally demonstrate a lack of empathy (American Psychiatric Association, 2000). Increased presence and unclear effects of narcissism make narcissism a relevant topic to investigate. Since narcissists rather don’t share information (Campbell, Hoffman, Campbell, & Marchisio, 2011), which is important to enhance team learning (Edmondson, Dillon, & Roloff, 2007), and perceived narcissism increases the conflict tendency among employees (Aguinis & O’Boyle, 2014), I propose that when star narcissism is high, a negative indirect effect of performance gap on team learning will be more negative (hypothesis 3).
5
By means of this research, I aim to contribute to the existing literature by proposing and investigating some possible underlying mechanisms that explain how and when performances differences inhibit team learning. I add to literature by taking into account and combining different perspectives, namely group interactions, team climate and personality of the star. More specifically, I apply conflict theory to star- and performance literature, as well as to literature about team learning, where I focus on status conflict in particular. I expect status conflict to explain how performance gap impedes team learning. To better understand this relationship, I consider when teams want to learn or not (high team competitive climate) and additionally I consider when stars make learning accessible or impede team learning (high star narcissism).
THEORY AND HYPOTHESES
Star performance gap and team learning
To be able to get a better understanding of the phenomenon ‘star performance’, it is important to first understand when a member of a team is a star performer. As Call, et al. (2015) describe, there are many different definitions of stars. A star member can be an individual who’s contribution by means of output is disproportionately high compared to other members (Aguinis & O’Boyle, 2014) or who’s production is extraordinary (O’Boyle JR. & Aguinis, 2012). Additionally, these extraordinary performers are more visible in the labor market compared to other less distinctive performers (Groysberg & Lee, 2008). Call et al. (2015) distinguish stars by their disproportionately high 1.
Performance, 2. Visibility and 3. Relevant social capital. The latter means that stars are able to utilize the valuable relationships they have (Call et al., 2015; Oettl, 2012; Oldroyd & Morris, 2012). Within this study, I focus on the relative performance of a star (Aguinis & O’Boyle, 2014), compared to the rest of his/her team, also known as the star versus non-star performance gap, from now on called performance gap.
Team learning, as Kehoe and Tzabbar (2015) explain, can be viewed from different perspectives and from several research streams which provide different descriptions of the concept.
This current paper studies team learning as a group process, instead of team learning as an outcome
6
improvement or task mastery. This ‘group process’-research stream focuses on what drives learning behaviors in organizational workgroups and this is especially relevant within this research, since I studied multiple real working teams in several different organizations in a field setting (Kenrick et al., 2002). A comprehensive definition used in this research stream states team learning is a process in which teams take action, subsequently obtain and reflect on received feedback and eventually make changes to improve or adapt accordingly (Edmondson, 1999; Argote & Miron-Spektor, 2011), which is in line with Edmondson et al. (2007) and Zeigler-Hill, McCabe, Vrabel, Raby, and Cronin (2018) who define learning as a process, including information sharing and reflecting on experience.
As Groysberg and Lee (2009) stated, little is known about how to stimulate positive effects of stars on teams. Neither much is known about how to impede negative effects of stars on team learning.
Some examples are Li et al. (2019), who found that a central star affects non-star member’s learning behavior both negatively and positively. More specifically, they found stars’ centrality directly increased team creativity but reduced teams’ learning behavior. Additionally, in order to become an exceptional performer, stars might have acquired a self-improvement mindset because they had to rely on themselves in previous situations (Kozlowski et al., 2001). This self-focused orientation can lead to situations in which team members miss important learning opportunities (Call et al., 2015), and therefore team learning is impeded. Groysberg et al., (2008) highlight the importance of a stars’
coworkers and explain that stars require the support of coworkers to have most desired positive effects. This corresponds to the results of Kehoe and Tzabbar (2015). They found that a star’s success depends on the collaboration with coworkers. This collaboration within a team is one of the main components for team learning (Cordes, 2016).
As described, researchers found some substantial proof for effects of stars, or performance differences, on team learning, and they highlighted the importance of teams and collaboration between a star and non-star team members. Therefore, it is important to understand when teams want or do not learn from a star. I expect conflict theory to explain a negative relationship between performance gap and team learning, because conflicts have several important implications for many team processes
7
(Lewis, et al., 1997; Bendersky & Hays, 2012; Sell, et al., 2004; Loch et al., 2000). More specifically, I argue for a mediating effect of status conflict.
Status conflict as a mediator
Conflicts between people or parties are tensions resulting from differences, disagreements or incompatible desires (De Dreu, Harinck & Van Vianen, 1999; Sell et al., 2004). These conflicts can apply to tasks, relationship or procedures but also status (Bendersky & Hayes, 2012). Within research about team processes and especially this study, the latter is just as important as other types of conflicts, if not, more important, mainly because status conflicts have important implications for team learning.
Status conflicts are “disputes over people’s relative status positions in their group’s social hierarchy (Bendersky & Hays, 2012: 323)”. Compared to for instance conflicts over relationships, tasks or processes, status conflicts induce more competitive behaviors because they have long-term consequences (Black, 1990). Additionally, status conflicts involve team members more, because status conflicts depend on social relationships within a social hierarchy. This is because status is related to relative standing in the social hierarchy and therefore always needs other members to compare with (Bendersky and Hays, 2012).
Assuming that star performers have a certain status, because status is e.g. related to task performance, cognitive ability and competence (Bhattacharya & Dugar, 2012; Knight, 2000;
Huberman, et al., 2004), I expect star performers to induce similar effects as other status-holders.
Moreover, there are several features that characterize status-holders, which are also benefits of being a star. Other non-star people possibly aim for these benefits and this could lead to an increased status conflict.
For instance, stars are seen as more valuable and they are internally and externally more visible than non-stars (Oldroyd & Morris, 2012). Therefore they tend to receive more resources and opportunities (Call et al., 2015). Similarly, people high in social or hierarchical status are more likely to obtain future resources (Huberman et al., 2004) than people with a low status. Additionally, people strive for competence and control (Breakwell, 2012), which are both benefits of being a star (Call et
8
al., 2015) and of status holders (Huberman et al., 2004). Moreover, a positive efficacy spiral exists, implying that people who already experience advantages (of a certain life style) will presumably receive additional resources too (Lindsley, Brass, & Thomas, 1995). Furthermore, being in a star- position by itself brings a certain status and as Huberman et al. (2004) found, status is something which people pursue as an end in itself. When people want to attain a higher status position (Huberman et al., 2004), e.g. the position of the star performer, or when they want to maintain their current position (Brett et al., 2007), status conflicts may arise (Brett et al., 2007; Huberman et al., 2004). I expect the latter will be even more the case in teams with increasing performance differences, because resources, and possibly other benefits, are more equally distributed in teams without a star (Kehoe & Tzabbar, 2015). So in more equal teams or teams without a highly performing star, I expect less status conflicts.
At the same time, I expect that status conflict explains how performance gap impedes team learning because status conflicts enhance many negative consequences for teams and team learning.
To start, Wu et al. (2011) argued that status seeking reduces collaboration between team members.
Since collaboration is important for learning as a team (Cordes, 2016), this implies that status seekers hinder team learning by not cooperating. Moreover, status conflicts produce competitive exchanges which lead to less information sharing with team members (Bendersky & Hays, 2012). Since sharing of information is an important part of learning (Edmondson et al., 2007) even as knowledge transfer (Lawrence, Mauws, Dyck, & Kleysen, 2005), team members who experience a status conflicts are very likely to inhibit team learning by not sharing information or knowledge.
To summarize, advantages that come with being a star performer, with or without status, cause status conflicts to arise among team members towards the star. In turn, status conflicts inhibit team learning because status conflicts impede information and knowledge sharing, just as collaboration within teams, which are all important ingredients for team learning. Hence, I propose the following:
Hypothesis 1: Performance gap has a negative and indirect effect on team learning via status conflicts towards the star.
9 Team competitive climate as a moderator
Status conflicts do not emerge under all circumstances. Therefore, it is valuable to consider which contexts trigger status conflict and moreover it is important to understand in which contexts team learning is impeded most. Consequently it is logical to consider the role of context as a moderator, especially because context or climates influence individual behavior but also many team processes, extensively illustrated and explained by James and Jones (1974) and James et al., (2008)
Research conducted on the effects of competition in teams defined perceived intra-team competition (Brown, Cron, & Slocum, 1998) or competitive psychological climate (Gim, Desa, &
Ramayah, 2015) as the extent to which employees experience that organizational rewards are provided depending on their performance compared to other team members. Additionally Fletcher et al. (2008) distinguish competitive workgroup climate as a shared perception among team members, as I will within this study too. Within a team competitive climate, team members show and perceive behaviors focused on individual achievement and their own performance, whereas in low competitive climates, teams feel safe to take risks and experiment (Maruping, Magni, Caporarello, & Basaglia, 2008).
Moreover, in competitive team climates, team members focus more on tangible resources and taking advantage of favorable circumstances, via joint competitiveness (Erdogan, Liden, & Kraimer, 2007).
Finally, results of an experiment by Fonseca (2009) showed that inequality between team members’
skills, lead to higher efforts of high-performers, whereas the lower-performing members dropped out, possibly because they were less able to compete.
More specifically related to this study, Wu et al. (2011) stated that team members who are seeking status are more likely to compete. At the same time, status conflicts promote competitive behavior more than other conflicts (Zeigler-Hill et al., 2018). Furthermore, Ingersoll, Glass, Cook, Kari, and Olsen (2019) summarize that public status conflicts emerge when status disagreements arise in teams where competition for status is highly important. Due to the effects that competition has on status conflicts, I expect that when teams have a highly competitive climate, the positive effect of performance gap on status conflict is more positive. On the other hand, increased competition between team members hinders team learning (Wu et al., 2011) and Van der Vegt et al. (2010) found that
10
competition also hinders team learning more compared to low competitive climates. Due to the negative relationship I expect status conflicts to have with team learning, interacting with highly competitive climates, I expect the relationship between status conflict and team learning to be even more negative when team competitive climate is high. Hence, I propose:
Hypothesis 2: Team competitive climate moderates the indirect effect of performance gap on team learning, through status conflict, so that the negative indirect effect is more negative at high team competitive climate.
Star narcissism as a moderator
Besides the reaction of team members towards stars, it is also important to take into account to what extent the star-performer motivates or allows team members to learn from him or her. As Fletcher et al. (2008) proposed, team climate and characteristics of people interact to influence e.g.
performance. Therefore it makes sense to take into account the personality of the star. I will especially focus on the personality trait that I expect to inhibit team learning most, due to its many other negative consequences (Bushman et al., 1998; Young & Pinsky, 2006), namely narcissism.
From a psychiatric and clinical perspective narcissism is defined as a personality disorder (Campbell et al., 2011), including e.g.. a character who shows a pattern of a desire for admiration, grandiosity and a lack of empathy (American Psychiatric Association, 2000). However, narcissism is also known as a subclinical pattern of characteristics, recognized as narcissism as a trait. The trait narcissism has many similarities to narcissism as a personality disorder, except that the personality trait narcissism does not involve impairment or distress (Campbell et al., 2011). Some negative consequences are narcissists cravings for attention, their lack of empathy and them being overconfident (Young & Pinsky, 2006). Wallace and Baumeister (2002) summarize that narcissistic people have unusual high self-efficacy, self-confidence, and high self-expectations, and they think highly of their own abilities, captured by Freud as excessive self-love (Bushman et al., 1998).
Narcissists usually show little empathy (Watson, Grisham, Trotter & Biderman, 1984) and impede team performance in tasks for which social sensitivity is required (Woolley, Chabris, Pentland,
11
Hashmi & Malone, 2010). In addition, narcissism has been found to contribute to short-term benefits for the person him or herself, but causes disadvantages to other members on the long-run (Campbell, Bush, Brunell, & Shelton, 2005). Moreover, narcissists tend to be excessively competitive and poor team players (Bergman, Westerman & Daly 2010, Campbell et al., 2005). Possibly because they are highly sensitive to negative criticism and feedback (Beck et al., 1990), which is an important part of the team learning (Bell, Kozlowski & Blawath, 2012). Eventually, a positive relationship exists between the level of perceived narcissism in organizations and conflict tendency of employees (Aguinis & O’Boyle, 2014).
Since stars are more visible than other non-star team members (Call et al., 2015; Groysberg &
Lee, 2009), the behavior of the stars that are narcissistic at the same time, might also be more visible.
At the same time, a high level of narcissism of a star may affect that this star will be even more visible or desires to be more visible (Young & Pinsky, 2006). I expect the destructive and visible behavior of highly narcissistic stars interacts with performance gap and together positively influence status conflict more than performance gap by itself.
Moreover, stars high in narcissism tend to withhold rather than share information, authority, and other work resources because of their sense of specialness and entitlement (Campbell et al., 2005;
Campbell et al., 2011). However, recombining knowledge of stars and team members is an essential feature of team learning (Li et al., 2019; Taylor & Greve, 2006). These negative consequences of narcissism make it more difficult to cooperate with narcissistic stars at an interpersonal level (Young
& Pinsky, 2006) compared to people low in narcissism. Thiscould lead to a more positive relationship between performance gap and conflicts when the star is highly narcissistic. Therefore I expect that the interaction of performance gap and high star-narcissism lead to a more positive relationship with star status conflict, which in turn decreases team learning. Hence, I propose the following:
Hypothesis 3: High levels of star narcissism moderate the indirect effect of star-non star performance gap on team learning through status conflict, such that the positive effect of performance gap on status conflict is more negative when the star is narcissistic, which in turn lowers team learning.
12 Figure 1: Conceptual model
METHOD
Sample
In order to test the hypotheses proposed, I conducted a survey study. I and a fellow research student collected data from teams consisting of at least 3 team members. These teams were working in the public and private sector, from various organizations and sections in the Netherlands. The final sample consists of 26 teams, with a response rate per team of at least 60%. The team sizes ranged from three to nine team members (M = 4.85, SD = 1.78). Of 112 team members, two members did not report their demographics, but other than that, there were no missing demographics. 56.4% was male in this sample. When looking at the star-members, the distribution of males and females was exactly 50% and when considering the gender of the supervisors, 14 teams were rated by males and 12 teams by females. Team tenure ranged from 3 to 332 months (which is 9.33 years) (M = 55.06, SD = 63.56) and organizational tenure ranged from 3 to 470 months (which is 39.17 years) (M = 119.31, SD = 125.85).
Data collection
I sent out questionnaires via Qualtrics to collect separate responses from supervisors and team members. Specifically, supervisors reported performance and team learning, whereas team members
13
reported status conflict towards the star, self- reported narcissism and team competitive climate. I sent out questionnaires by email at the 20th of March and three or four reminders in the 6 – 8 weeks after the first invitation.
Measures
Star versus non-star performance gap. In order to indicate the performance gap, supervisors
reported the performance levels of all individual team members based on seven items adopted from Mohammed & Nadkarni (2011) and Ancona & Caldwell (1992). Example items included ‘Please rate the timeliness (adherences to schedules) by which the individual work of the team member is completed’ and ‘Please rate the team member’s innovation’. Items were rated on a seven-point Likert scale, ranging from 1 (“far below average”) to 7 (“far above average”). One item that measured client's satisfaction could be rated as ‘not applicable’, since it is likely that not all teams work with clients or have client ratings. The mean score was used to capture individual performance. The member scoring highest was then identified as the star performer. Cronbach’s Alpha was .92 for the measures.
Subsequently, I calculated the mean performance of all non-star team members, and distracted this mean performance from the star’s performance. This gave a statistic indicating the performance gap between the star and the others team members. In teams with multiple members with the same highest performance, which was the case in five teams, I considered all these members as a star and compared these two or three people with all other members.
Team’s status conflict with the star. I measured this variable with the inventory adapted from
Bendersky & Hayes (2012). Instead of reporting the four status conflict items separately for every member, I asked the team members to report the amount of status conflict they experienced towards all other members based on a short description of a situation, including all four items (see Appendix C).
All team members rated the status conflict they experienced towards every other team member on a seven-point Likert-scale, ranging from 1 (“strongly disagree”) to 7 (“strongly agree”). The average status conflict experienced by the other team members towards the star in each team was calculated. In case of multiple stars in one team, I used the average of the reported status conflict scores.
14
Team learning. In order to measure team learning, I adapted 5 items from Schippers, Hartog
and Koopman (2007). Supervisors reported the items on a seven-point Likert scale, ranging from 1 (“strongly disagree”) to 7 (“strongly agree”). Example items included ‘We talk about different ways in which we can reach our objectives’ and ‘In this team the results of actions are evaluated’. Due to a very low Cronbach’s Alpha (α = .61), I conducted an Explanatory Factor Analysis (EFA) in order to examine whether team learning in this data possibly exists of different factors.
EFA results of team learning. In order to conduct the EFA, the assumptions should be met,
meaning that the data of team learning should be normally distributed and the correlations are reliable, which was the case. Second, the Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) should be greater than .5, which was .58, and the Bartlett’s Test of Sphericity should be significant, which was also true. The latter indicates there is at least one significant relationship between at least 2 items.
Moreover I checked for communalities. I found an extreme low value for item 1 of .18, indicating that this measure indeed should be dropped from the analysis. Subsequently, according to the ‘Total Variance Explained’, two factors could be extracted from the data. After running the test again, I extracted a fixed number of two factors after which the ‘Rotated Component Matrix’ showed that items 2 and 3 could be combined into one factor and items 4 and 5 too. Items 4 and 5 mainly focused on evaluation of the results, whereas items 2 and 3 underscored team discussion about working effectively and effectively reacting to change. The latter I deemed more suitable for measuring the concept team learning as described in the theory section. Therefore I used items 2 and 3 to measure team learning (α= .63).
Team competitive climate. To measure team competitive climate, I used the scale of Fletcher
et al. (2008) adapted from Brown et al. (1998). Four items, among which ‘Everybody is concerned with being the top performer’ or ‘My manager frequently compared my performance with that of my team members’ were evaluated by every team members on a seven-point Likert-scale, ranging from 1 (“strongly disagree”) to 7 (“strongly agree”).. Next, I calculated the average of all individual scores and aggregated the average individual scores to the team-level (α = .69).
15
Star narcissism. All team members self-reported this variable, using the Narcissistic
Personality Inventory (NPI-16 , Ames, Rose, & Anderson, 2006). A sample set of items was “I like to be in the center of attention” versus “I prefer to blend in with the crowd”. Team members indicated which option they identified with most. Cronbach’s Alpha was .59. After removing statement 8, Cronbach’s Alpha increased to .615. Star narcissism without item 8 was used for the main analyses.
Control variables. Within this study I controlled for team size, since this may have an
influence on group processes (Ancona & Caldwell, 1992). Team size was indicated by team supervisors.
RESULTS
Normality tests and outlier detection
Before starting the analysis, I assessed the assumptions necessary to conduct the regression analyses. The assumptions I checked for were normality and outliers, which were both met. According to the Shapiro-Wilk test, I found that both the results of the pretest and posttest, did not give a significant score for normality, indicating that the data is normally distributed. Star status conflict (skewness = .03), team competitive climate (from now on called competitive climate, skewness = -.40) and narcissism of the star (skewness = -.07) were approximately symmetric. Performance gap (skewness = .65) and team learning (skewness = -.71) were moderately skewed. All Kurtosis levels were between -.40 (for competitive climate) and .65 (for performance gap), indicating a flat distribution of the data. I applied a 3.29 times Standard Deviation calculation to detect outliers. No outliers were found for all variables.
In Table 1 means, standard deviations and correlations between the variables are reported.
--- Insert Table 1 about here ---
16 Hypotheses testing
For all analyses I used PROCESS version 3.5 of Andrew Hayes. Due to the small sample size of 26 teams, I chose to use 50.000 bootstrap samples and a confidence interval of 90% for all parameter estimates. I used team size a control variable in all results reported below. Additionally, for all variables I used the standardized items.
I tested the first hypothesis with PROCESS model 4, positing that a negative indirect effect of performance gap on team learning is mediated by star status conflict. The results of these analyses are reported in Table 2. The coefficient showed a positive and non-significant effect (b = .05, LLCI = -.05, ULCI = .19) for the indirect effect via star status conflict. Hence, hypothesis 1 is not supported.
More specifically, as shown in Table 2, model 2, I found a negative but non-significant relationship between performance gap and star status conflict (b = -.13, p = .50), controlled for team size. Table 2, model 4 depicts the effect of performance gap on team learning, which was positive but not significant (b = .19, p =.34), controlled for team size and status conflict. Moreover, as can be seen in Table 2, model 5, star status conflict is negatively and moderately significantly related to team learning, after controlling for performance gap and team size. Table 2, model 5 explains 25% of the variance in team learning (R = .25, p < .10).
--- Insert Table 2 about here ---
Hypothesis 2 posited a moderated mediation model, suggesting that the negative indirect effect of performance gap on team learning via star status conflict is more negative when competitive climate is high. The regression results of moderated mediation in Table 3 show that the moderated mediation index for competitive climate is significant (b = -.15, LLCI = -.36, ULCI = -.00). This supports that there is a moderated mediation effect for competitive climate.
17
However, when looking at the conditional indirect effects of performance gap on team learning via status conflict in more detail, hypothesis 2 is partially supported. I found a positive and significant indirect effect at a low level (-1 SD) of competitive climate (b = .22, LLCI = .00, ULCI = .48), and a less positive and non-significant effect at the mean level (b = .08, LLCI = -.05, ULCI = .20), opposed to what I hypothesized. The indirect effect became negative at high levels of competitive climate (+1 SD, b = -.09, LLCI = -.31, ULCI = .05), which is in line with the hypothesized pattern, namely that the indirect effect is more negative at high levels of competitive climate. Considering that the pattern is as I hypothesized, but the low and mean levels of competitive climate are positive, this hypothesis is partially supported.
--- Insert Table 3 about here ---
Hypothesis 3 posited that the negative indirect effect of performance gap on team learning via star status conflict, is more negative when star narcissism is high. The moderated mediation index for star narcissism (Table 3) did not show a significant effect (b = -.14, LLCI = -.45; ULCI = .00).
Therefore, the data did not support hypothesis 3.
The conditional indirect effects at three different levels of star narcissism showed a positive non-significant indirect effect at low levels of star narcissism (-1 SD, b = .18, LLCI = -.01, ULCI = .46) and a less positive indirect non-significant effect at the mean level of star narcissism (b = .04, LLCI = -.08, ULCI = .15). This indirect effect was non-significant and negative at high levels of star narcissism (+1 SD, b = -.09, LLCI = -.48, ULCI = .05). This pattern is similar to the conditional effects of competitive climate, which is also positive at low and mean levels of star narcissism and negative at high levels of star narcissism. Yet, no effects were significant.
Finally, I conducted a robustness test for a complete overview of the results with PROCESS model 60, including both moderators. None of the indirect effects were significant at any level of competitive climate or star narcissism. However, the indirect effect was positive at low levels of
18
competitive climate and star narcissism, and the indirect effect was most negative at high levels of competitive climate and star narcissism. This indicates that the pattern of the effects is partially similar to what was hypothesized. A table reporting these results is available upon request.
Supplementary analysis
I captured team learning with two items based on the EFA results. In supplementary analysis, I further tested the model of two versus one factor of team learning with a Confirmatory Factor Analysis (CFA) to justify my choice of operationalization. Moreover, I conducted a supplementary analysis examine to what extent the hypotheses as posited in the theory are supported by the data of team learning items 4 and 5.
An alternative operationalization of team learning. I conducted a CFA on the four items
(items 2-5 in question). The results in Table 4 point out that the model fits better when items 2 and 3 are separated from items 4 and 5, instead of using one total measure. Both Goodness of fit indexes showed better results for separated items, compared to combined items. CFI (Comparative Fix Index) and TLI (Tucker-Lewus Index) were respectively 1.00 and 1.27 for the separated items and respectively .87 and .62 for the combined items. Additionally, the RMSEA (Root Mean Square Error of Approximation) was .00 for the two-factor model, compared to .23 for the one-factor model.
Finally, ANOVA showed at a 99% confidence interval that the models are indeed significantly different (χ2diff = .03, p < 0.01), confirming that the two-factor model is significantly better than the one-factor model. This justified my choice of operationalization.
--- Insert Table 4 about here ---
Hypotheses testing on the alternative measure of team learning. I further tested the hypotheses on this alternative measure of team learning with items 4 and 5 (a = .83), under the same conditions as the main analyses.
19
Hypothesis 1 posited that an indirect effect of performance gap on team learning is mediated by star status conflict. I tested the first hypothesis with PROCESS model 4 and the results of the analysis are shown in Table 5. The coefficient for the indirect effect via status conflict showed a positive and non-significant effect (b = .03, LLCI = -.06, ULCI = .14). Hence, hypothesis 1 is not supported.
Specifically, I found a negative but non-significant effect of performance gap on star status conflict controlled for team size (b = -.13, p = .50), which is depicted in Table 5, model 2. The relationship between status conflict and team learning was negative and non-significant (b = -.21, p = .38), controlled for team size and performance gap (Table 5, model 5). Furthermore, Table 5, model 5 depicts a non-significant negative relationship between status conflict and team learning, after controlling for team size and performance gap (b = .24, p = .29).
--- Insert Table 5 about here ---
Hypothesis 2 proposed that the negative indirect effect of performance gap on team learning via star status conflict, is more negative when competitive climate is high. Table 6 shows the regression results of moderated mediation, giving a non-significant moderated mediation index for competitive climate (b = -.08, LLCI = -.37, ULCI = .07) . Hence, hypothesis 2 is not supported by the data.
Similar to the main analyses, the conditional indirect effects at three different levels of competitive climate show that the indirect effect is positive and non-significant at a low level of competitive climate (-1 SD, b = .12, LLCI = -.12, ULCI = .39) and a less positive and non-significant effect at the mean level (b = .04, LLCI = -.07, ULCI = .13). At high levels of competitive climate (+1 SD) the indirect effect became negative (b = -.04, LLCI = -.27, ULCI = .05). Despite the indirect effects became more negative when levels of competitive climate increased, similar to the hypothesized moderated mediation pattern, these effects were not significant.
20
--- Insert Table 6 about here ---
At last, hypothesis 3 posited that the negative indirect effect of performance gap op team learning mediated by status conflict, is more negative when star narcissism is high. Table 6 reports the moderated mediation index for star narcissism, which showed a non-significant effect (b = -.07, LLCI
= -.34, ULCI = .07). Thus, hypothesis 3 was not supported by the data.
More specifically, looking at the conditional indirect effects at three levels of star narcissism, the indirect effect is positive and non-significant at low levels of star narcissism (-1 SD, b = .10, LLCI
= -.09, ULCI = .36) and at mean levels of star narcissism (b = .02, LLCI = -.07, ULCI = .11). The indirect effect is negative and not significant at high star narcissism, (+ 1 SD, b = -.05, LLCI = -.37, ULCI = .06).
Finally, I conducted a robustness test for a total overview of the indirect effects at low, mean and high levels of both moderators. None of these indirect effects were significant but the pattern was similar to what I hypothesized, namely that the indirect effect was most negative at high levels of star narcissism and competitive climate. Tables reporting the conditional indirect effects are available upon request.
DISCUSSION
Over the past years, companies have spent an increasing amount of resources to recruit ‘the best and brightest’ personnel with the expectation that star performers will bring exceptional contributions to the organizations they work in. However, the impact of stars on teams varies across studies, resulting in positive and negative (in)direct effects of stars on teams. This research aimed to investigate how negative effects of differences in performance between a star performer and the average performance of the team on the team’s learning can be explained.
21
First, a mediation effect of star status conflict was hypothesized, positing that performance gap has a negative indirect effect on team learning via an increased status conflict towards the star performer. No significant effect supported this hypothesis.
Second, I hypothesized that team competitive climate moderates the indirect effect performance gap on team learning via status conflict, so that the negative indirect effect is more negative at high competitive climate. The data showed a significant moderated mediation effect and the indirect effects at different levels of competitive climate showed the hypothesized pattern, showing the most negative indirect effect at high levels of competitive climate. However, when looking at the indirect effects at low, mean and high levels of competitive climate in more detail, I just found a significant positive indirect effect of performance gap on team learning via status conflict, when team competitive climate was low. This positive indirect effect was not hypothesized. Therefore, the data partially supported hypothesis 2.
Third, I hypothesized that the indirect effect as defined in hypothesis 1, is moderated by star narcissism, such that the negative indirect effect is strongest when star narcissism is high. However, no significant support for this moderation effect was found.
When comparing the results of the main analyses to the results of the supplementary analysis, the pattern of the effects are similar: the indirect effect via status conflict is positive, and the indirect effect is most negative at high levels of competitive climate and high star narcissism. Since no significant effects were detected, the data did not support any of the hypothesis when using team learning items 4 and 5.
Theoretical implications
Prior research investigated different contextual and individual concepts in an attempt to explain indirect and interaction effects concerning star performance, team learning, and conflict theory, but mechanisms behind the negative and positive indirect effects of performance gap on team learning have remained unclear. This study adds to existing literature about these topics in several ways.
22
As Li et al., (2019) indicated, current research has not widely focused on negative effects that stars possibly have on team members, but more on the exceptional value stars can possibly bring to the organizations they work in (Aguinis & O’Boyle, 2014). Some of the negative effects other researchers found are inhibited learning opportunities of non-stars due to a star’s self-focused orientation (Call et al., 2015), a negative effect of a stars’ centrality on non-stars’ learning (Li et al., 2019) and additionally, Call et al. (2015) suggested that stars’ arrogance could lead to inflated self-confidence and therefore to riskier behavior (Li & Tang, 2010). Even though no significant results supported that star status conflict indeed mediated the negative indirect effect of performance gap on team learning, results did support the negative influence of star status conflict on team learning. This research contributes to existing literature by applying conflict theory to star-literature and literature about team learning and thereby adds a new insight, namely the importance of star status conflict for team learning. The negative effect of star status conflict on team learning is an important implication because it invites future research to take into account other factors concerning conflict theory.
Previous research mainly focused on how task conflicts or relation conflicts influence team performance (Bendersky & Hayes, 2012; De Dreu et al., 1999; Sell et al., 2004), but research on the effects of different kinds of conflict on team learning is missing. Additionally it would be interesting and valuable for future research to focus on the different conflicts star’s provoke in teams, how and when these conflicts are induced, and what their consequences are for the team members.
Since star status conflict did not directly seem to stem from performance gap, it would be interesting to investigate when or how star status conflicts arise in teams that contain a star. This research adds to star literature and conflict theory by examining the moderating roles of team competitive climate and star narcissism. Results showed that the indirect effect of performance gap on team learning via status conflict was positive when the team members reported a low team competitive climate. More specifically, the indirect effect consisted of a negative effect of performance gap on status conflict and a negative effect of status conflict on team learning. This research did not confirm the negative influence of competitive climate on team learning as Van der Vegt et al., (2010) found.
This may imply that the effects of competitive climates are different for teams that contain or do not
23
contain stars or that effects of climate depend on the degree of performance difference. Moreover, status conflicts might not arise from the performance gap, as I expected, but other star-related factors may induce these or other kinds of conflicts. It would be interesting to investigate the role of specific features of stars, such as stars’ visibility (Oldroyd & Morris, 2012) or the resources and opportunities stars receive and non-star members might not (Call et al., 2015; Huberman et al., 2004) and how these benefits provoke conflicts.
Despite some research indicated that a star needs his or her coworkers to reach the desired success (Groysberg et al., 2008), the role of team members in the success of stars remains unclear.
This research contributed to current literature by focusing on personality of the star, by investigating the moderating role of star narcissism. However, the role of personality of team members continues to be unexplained and it would therefore be interesting to investigate its effects.
Practical implications
In the Netherlands, 91% of the Dutch companies apply a system of variable compensation. Of the companies with variable compensation 76% used individual bonuses to reward employees based on their performance and increased individual salary was mainly based on assessment of functioning (91%) and not so much compared to other salaries (53%) (Van Dorst & Schartman, 2020). This indicates that in many companies and many cases, companies highlight the importance of individual performance, which in turn might underscore the performance differences within teams. Even though the results of this research did not proof that performance gap in itself caused status conflict, other factors related to star-performance or performances differences might do so, such as rewards that star’s might receive, but non-stars might not (Call et al., 2015). Practitioners working in organizations where performance differences are emphasized especially companies that find learning within teams important should carefully consider the triggers that could possibly cause status conflicts to arise, because I found support that status conflict reduces team learning. Individual monetary awards can lead to less cooperation and this might be detrimental to team learning. However, in order to make decisions for these rewards, organizations should therefore clearly define its goals and decide whether these are individual or collective team goals and how rewards are connected to these goals (Wu, et al.,
24
2011). Instead, companies could consider using team rewards to prevent within-team conflicts (Gerritsen & Sijm, 2003).
Limitations and future research.
Just as other studies, this study is not without limitations which should be taken into account for interpreting the results. To start, sample size was very small, namely just 26 usable teams. This research was conducted in a very unusual time, in which most respondents were working from home due to the COVID-19 pandemic. Distraction potentially induced that people were less focused on the questions in the survey and it is very likely that many supervisors and employees were not able to fill in the survey at all due to these circumstances. A larger amount of teams could have led to more significant results, however I cannot state that with certainty.
Second I want to point out the measurement of several variables, starting with team learning.
Perhaps, the supervisors interpreted the questions different than intended or supervisors could have in different ways from each other, or possibly the questions were not translated well enough or not suitable for all teams or companies that participated in this research. Future research could for example provide a more clear explanation of the items to ensure the items are interpreted as intended and measure what is intended to be measured, thereby increasing validity. As shown by an explanatory factor analysis and a confirmatory factor analysis, team learning in this data-set seemed to consist of two factors. Based on this, I suggest future research to focus on how different dimensions of team learning are influenced by star performance and status conflicts.
Moreover, future research should consider different inventories or items to measure team learning. The items I used, focused on the discussion of processes, methods and procedures, and the evaluations of methods and results. However, as Argote and Miron-Spektor (2011) and Edmondson (1999) describe, besides discussion and evaluation, team learning also consists of taking action, adapting to changes and making improvements, to subsequently evaluate these actions accordingly.
These last elements were not captured in the items I used and it would be interesting to consider these more thoroughly in future research, in order to measure team learning more comprehensively.
25
Another limitation was the measurement of status conflict. In this research, status conflict towards all members was indicated by every team member, based on a description of all items in a short text. This caused aversion because not all team members felt comfortable to rate how they felt about their coworkers, therefore filling in ‘this is me’ or a ‘not agree nor disagree’, which influenced the true data. In addition, the total description was possibly interpreted as a general conflict situation, instead of conflict regarding status. In some cases, a high conflict rating towards a star, was also returned with a high conflict rating of the star towards the same members. This possibly indicates a conflict between two members and not a status conflict towards a star. These kinds of conflicts can similarly prevent team learning to arise, but this was not intended to be measured. Therefore, the way I measured status conflict is possibly not as valid as when I would have used the status conflict items separately. Future research should pay attention to the measurement of status conflict and can compare this with other conflicts that arise within teams. Moreover, the analysis of star status conflict in this model was less suitable. Initially, the interrater reliability was difficult to indicate because every star was rated by different members and compared to other coworkers too. Additionally, the amount of team members differed per team, ranging from 2 to 7 ratings for different stars. This caused that it was not possible to compare all raters but neither the ratings with each other.
A third point of attention lies in the measurement and interpretation of narcissism. In this research, self-ratings were used to indicate narcissism by the NPI. However, future research could investigate whether a difference between narcissism measured by a self-rating and perceived narcissism indicated by team members lead to different effects on status conflict or team learning.
South et al. (2003) stated that informant reports, compared to self-reports, predict behavior more accurately and target people, stars in this study, report lower personality pathology than informants (Klonsky, Oltmanns & Turkheimer, 2002). Hence, it is possible that increased perceived star narcissism by team members lead to stronger and different effects than self-reported narcissism such as measured by NPI. Additionally, narcissism literature studies distinguish between different narcissistic tendencies (covert and overt narcissistic tendencies, Hendin & Cheek, 1997) and between different dimensions (admiration and rivalry, Back, et al., 2013). The effects of these different
26
dimensions or tendencies on status conflict or team learning can vary. For instance, narcissistic rivalry causes more negative social consequences (Back et al., 2013), which could play a different role in conflict-emergence than a narcissistic need for admiration. Based on the research aims and hypothesized effects, future research should examine the various dimensions of narcissism and researchers should carefully consider which measurement or methods should be applied to test the hypotheses.
Notwithstanding that not all hypotheses were confirmed, this research does raise interesting questions and invites researchers to further investigate the role of personality traits and team climate.
Other personality traits of the star should be taken into account in order to investigate when status conflict is most or least likely to arise or which star traits have an effect on learning. By knowing the effects of different personality traits, practitioners can use personality tests in their selection procedures. Even though personality tests might not directly predict individual performance (Heneman, Judge, & Kammeyer-Mueller, 2015), personality traits of stars might indirectly influence the subsequent learning of the team members. Similarly, the circumstances under which status conflicts arise and when it has the most detrimental effects could be further investigated. Specifically, the effects of different kinds of climates could be investigated and compared to competitive team climates.
Despite not all results were significant, perhaps due to the small sample size, this study is still valuable because it invites researchers to investigate whether these hypotheses can be confirmed under different circumstances or by measuring the variables in other ways and this research encourages future research to further explore the underlying mechanisms of the positive and negative relations between star performance and team learning.
CONCLUSION
This study investigated different mediation and moderation effects that possibly explain negative effects of differences in performance between a star performer and the average performance of the team on the team’s learning. The findings of this research suggest that an indirect effect of
27
performance gap on team learning is not mediated by status conflict. However, in some contexts, namely when a certain team competitive climate is experienced by the team members, the mediation effect is moderated. The results showed that the indirect effect was positive at low team competitive climate. When looking at the personality of the star, specifically star narcissism, no moderation effect was found and no proof was found that the negative indirect effect of performance gap on team learning via high status conflict is more negative when star narcissism is high. This study adds to existing literature by clarifying a part of the complex process of the influence of performance gap on team learning and highlighted the importance of status conflict and its effects on team outcomes.
REFERENCES
Aguinis, H., & O’Boyle, E. (2014). Star performers in twenty-first century organizations. Personnel Psychology, 67(2): 313–350.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th edition). Washington, DC: American Psychiatric Association.
Ames, D. R., Rose, P., & Anderson, C. P. (2006). The NPI-16 as a short measure of narcissism.
Journal of Research in Personality, 40: 440–450.
Ancona, D. G., & Caldwell, D. F. (1992). Demography and design: Predictors of new product team performance. Organization Science, 3(3): 321-341
Argote, L., & Miron-Spektor, E. (2011). Organizational learning: From experience to knowledge.
Organization Science, 22(5): 1123–1137.
Azamfirei, L. (2016). Knowledge Is Power. The Journal of Critical Care Medicine, 2(2): 65–66.
Beck, A. T., Freeman, A., Pretzer, J., Davis, D. D., Fleming, B., Ottaviani, R., Beck, J., Simon, K. M., Padesky, C., Meyer, J., & Trexler, L. (1990). Cognitive therapy of the personality disorders.
New York: Guilford Press.
Bell, B. S., Kozlowski, S. W. J., & Blawath, S. (2012). Team learning: A theoretical integration and review. The Oxford Handbook of Organizational Psychology, 2: 859-909
28
Bendersky, C., & Hays, N. A. (2012). Status Conflict in Groups. Source: Organization Science, 23(2):
323–340.
Bergman, J. Z., Westerman, J. W. & Daly, J. P. (2010). Narcissism in management education.
Academy of Management Learning and Education, 9(1):119-131.
Bhattacharya, Y. & Dugar, S. (2012). Status incentives and performance. Managerial Decision Economics, 33(7-8): 549–563.
Black, D. (1990). The elementary forms of conflict management. M. J. Lerner, ed. New Directions in the Study of Justice, Law, and Social Control. Plenum, New York
Breakwell, G. M. (2012). Social representations and social identity. Cambridge University Press, 2(3), 118–134.
Brett, J. M., Olekalns, M., Friedman, R., Goates, N., Anderson, C., & Lisco, C. C. (2007). Sticks and stones: Language, face, and online dispute resolution. The Academy of Management Journal, 50(1): 85-99.
Brown, S. R., Cron, W. L., & Slocum, J. W. (1998). Effects of trait competitiveness and perceived intra-organizational competition on salesperson goal setting and performance. Journal of Marketing, 62(4): 88–98.
Bunderson, J. S., & Sutcliffe, K. M. (2003). Management team learning orientation and business unit performance. Journal of Applied Psychology, 88(3): 552–560.
Bushman, B. J., Baumeister, R. F., Burrell, S., Hunn, R., Mackey, D., Mellinger, T., Oien, C., Sasaki, Y., Sidari, D., Skuster, C., Steffen, M., & Stevenson, D., (1998). Threatened egotism, narcissism, self-esteem, and direct and displaced aggression: Does self-love or self-hate lead to violence? Journal of Personality and Social Psychology, 75(1):219-229.
Call, M. L., Nyberg, A. J., & Thatcher, S. M. B. (2015). Stargazing: An integrative conceptual review, theoretical reconciliation, and extension for star employee research. Journal of Applied Psychology. American Psychological Association Inc.
29
Campbell, W. K., Bush, C. P., Brunell, A. B., & Shelton, J. (2005). Understanding the social costs of narcissism: The case of the tragedy of the commons. Personality and Social Psychology Bulletin, 31: 1358–1368.
Campbell, W. K., Hoffman, B. J., Campbell, S. M., & Marchisio, G. (2011). Narcissism in organizational contexts. Human Resource Management Review, 21(4): 268-284.
Charness, G., Masclet, D., & Villeval, M. C. (2014). The dark side of competition for status.
Management Science, 60(1): 38–55.
Cordes, S. (2016). Knowledge management & e-learning virtual team learning: The role of collaboration process and technology affordance in team decision making virtual team learning. Knowledge Management & E-Learning, 8(4): 602–627.
Cropanzano, R., & Mitchell, M. S. (2005). Social exchange theory: An interdisciplinary review.
Journal of Management, 31(6): 874-900.
De Dreu, C. K. W., Harinck, F., & Van Vianen, A. E. M. (1999). Conflict and performance in groups and organizations. International review of industrial and organizational psychology, 14:369- 414.
Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2): 350–383.
Edmondson, A. C., Dillon, J. R., & Roloff, K. S. (2007). Three perspectives on team learning:
Outcome improvement, task mastery, and group process. The Academy of Management Annals, 1(1).
Erdogan, B., Liden, R. C., & Kraimer, M. L. (2006). Justice and leader-member exchange: The moderating role of organizational culture. The Academy of Management Journal, 40(2):
395–406.
30
Fletcher, T. D., Major, D. A., & Davis, D. D. (2008). The interactive relationship of competitive climate and trait competitiveness with workplace attitudes. Journal of Organizational Behavior, 29(7): 899–922.
Fonseca, M. A. (2009). An experimental investigation of asymmetric contests. International Journal of Industrial Organization, 27(5): 582– 591
Friedlander, F. & Margulies, N. (1969). Multiple impacts of organizational climate and individual value system upon job satisfaction. Personnel Psychology, 22(2): 171-183.
Gerritsen, M. & Sijm, R. (2003). Resultaatafhankelijk belonen in Nederland. Tijdschrift voor HRM
Gim, G. C. W., Desa, N. M., & Ramayah, T. (2015). Competitive psychological climate and turnover intention with the mediating Role of affective commitment. Procedia - Social and Behavioral Sciences, 172:658-665.
Gluck, S. (2014). Famous people with narcissistic personality disorder. HealthyPlace. Retrieved January 20th, 2020, from https://www.healthyplace.com/personality-disorders/narcissistic- personality-disorder/famous-people-with-narcissistic-personality-disorder
Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25(2): 161-178.
Groysberg, B., & Lee, L. E. (2008). The effect of colleague quality on top performance: The case of security analysts. Journal of Organizational Behavior, 29(8): 1123–1144.
Groysberg, B., & Lee, L. E. (2009). Hiring Stars and their colleagues: Exploration and exploitation in professional service Firms. Organization Science, 20(4): 740–758.
Harms, P. D., Grijalva, E., Newman, D. A., Tay, L., Donnellan, M. B., Harms, P. D., Grijalva, E., Newman, D. A., Tay, L., Donnellan, M., Brent, Harms, P. D., Robins, R. W., Taiyi, Y. &
Gender. (2014). Gender differences in narcissism: A meta-analytic review. Psychological Bulletin, 141(2): 261-310.
