Does conflict stimulate or hinder
creativity?
The effects of gender diversity and conflict on
creativity in teams
Esther Huffener 10247467 18-08-2017
MSc. in Business Administration- Leadership and Management Final Version Master Thesis Supervisor: Dr. Claudia Buengeler
STATEMENT OF ORIGINALITY
This document is written by Student Esther Huffener who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
ABSTRACT
This study aims to explain the relation between team gender diversity and creativity by a team process, in this case, intragroup conflict. There is no agreement about when a conflict has benefits and when it leads to problems. By adding contingency factors and using another way of measuring creativity, the present study tries to explain these inconsistent results. In this study, videotaped meetings are used to require a more objective way of creativity measurement. Additionally, the videos allow observation of actual interactions between team members during their meetings and thus provide insights into the interactions, behaviors, and statements of team members that are needed to successfully use diversity in teams. The results show that task conflict is positively correlated to relationship conflict. Additionally, the results suggest that the perception team members have about conflict in the team is not always in line with the actual interpersonal interactions. Therefore, more objective ways of measuring should be used in research to determine the reality instead of the perceptions.
Key words: Gender diversity; Intragroup conflict; Task conflict; Relationship conflict; Creativity; Time of conflict; Debate; Inclusive climate.
TABLE OF CONTENTS
INTRODCUTION..………... 5
LITERATURE REVIEW AND HYPOTHESES..……….... 8
Creativity…..………... 8
Diversity………. 8
Intragroup conflict………. 10
The moderating role of time……….. 12
The moderating role of relationship conflict………. 13
The moderating role of interpersonal interactions………. 14
Debate……… 15 Inclusive climate……… 16 METHODOLOGY……… 18 Sample...……… 18 Data collection………... 18 Coding……… 18 Measures……… 19 Gender diversity………. 19 Intragroup conflict ……… 20 Time of conflict....……….. 20 Creativity ...………... 20 Debate……… 21 Inclusive climate……… 22 Control variables………... 22 RESULTS……….. 24
Bivariate and descriptive statistics………. 24
Hypotheses testing………. 24
Mediations………. 24
Moderations………... 25
DISCUSSION AND CONCLUSION………... 30
REFERENCES……….. 34
INTRODUCTION
People like other people who are similar to them. So, when they are free to choose their own teammates, they tend to form homogenous groups (Barsade, Ward, Turner, & Sonnenfeld, 2000). But is this the right choice if the team has to be creative? Creativity is becoming increasingly relevant in today’s businesses (Santos, Uitdewilligen & Passos, 2015). It is commonly believed that teams will be better able to come up with ideas, find solutions, and adapt to changing market environments than individuals (Gilson, 2015).
Teams in organizations are becoming more and more diverse, both demographically (e.g., with respect to age, gender, or ethnicity) and functionally or educationally (van Knippenberg & Schippers, 2007). A substantial body of research has studied the relation between diversity and team outcomes (Stahl, Maznevski, Voigt, & Jonson, 2010; van Dijk, van Engen, & van Knippenberg, 2012). However, these studies have found mixed and contradictory results. Different dimensions of diversity have been shown to be positively, negatively, or not at all related to performance outcomes (van Dijk et al., 2012). The existing literature suggests that team processes mediate the diversity-performance link and contextual variables moderate this relation (Stahl et al., 2010).
When it comes to creative performance, working in diverse teams is the norm in today’s businesses, but diversity in teams also creates significant challenges (van Dijk et al., 2012). A high level of creativity can be reached through the integration of knowledge and ideas of different team members (Chen, 2006). However, tensions and conflict rather than the desired information exchange occur in teams (de Dreu & Weingart, 2003).
Trust, knowledge exchange, and the connection between team members are key ingredients for successful collaboration and creativity (Chen, 2006). Thus, conflict within a team seems counterproductive. However, with the growth of research on the link between conflict and performance, it has become clearer that this is not always the case. On the contrary, research shows that a certain amount of task conflict, for example “conflicts about the distribution of resources, procedures and policies, and judgment and interpretation of facts” (de Dreu & Weingart, 2003, p. 741), could sometimes be beneficial for team performance (Jehn, 1995; Chen, 2006; Farh, Lee, & Farh, 2010). But studies show mixed and contradictory results (Hülsheger, Anderson, and Salgado, 2009;O’Neill, Allen, and Hastings, 2013).
On the other hand, research confirms that interpersonal or relationship conflict, that is, personal conflicts between team members, negatively affects performance and creativity
(Jehn, 1995; de Dreu & Weingart, 2003; Santos, et al., 2015). The problem is that it is difficult to completely separate the two types of conflict because they are positively correlated (Simons & Peterson, 2000; Chen, 2006; de Dreu & Weingart, 2003). One possible explanation is that team members take task-related conflicts personally. These misattributions could lead to relational conflicts (Simons & Peterson, 2000).
There is no agreement about when a conflict has benefits and when it leads to problems. This study tries to explore the role of conflict in relation to teams’ creativity. By adding contingency factors such as when in a team’s life cycle (earlier or later) the conflict takes place and how the two types of conflicts relate to each other, the present study tries to explain these inconsistent results.
This study aims to explain the team diversity-creativity link by a team process, in this case, intragroup conflict. Intragroup conflict is focused on because conflict plays a role in the two most common theories that explain the effects of diversity (Nishii, 2013). According to the social categorization theory, people form positive attitudes towards others in their own group (based on different characteristics) and negative opinions towards the outsiders. This can lead to relationship conflict and lower the task performance (Pelled, Eisenhardt, & Xin, 1999). On the other hand, according to the information-processing theory, diverse teams could have a positive effect as well, because they could provide more diverse information (Schneid, Isidor, Li & Kabst, 2015).
So, diversity can both positively and negatively affect team performance. Both perspectives should be considered simultaneously to understand this relationship (van Knippenberg, de Dreu & Homan, 2004). In so doing, it is possible to clarify the underlying mechanism of gender diversity and team performance. Here, I focus on diversity in terms of a team’s gender composition for two main reasons. First, gender diversity is a universal issue and the differences in status based on gender are still relevant today (Nishii, 2013). Second, gender is one of the most visible differentiating characteristics. This is one reason this type of diversity can easily trigger conflict (Pelled, 1996; Nishii, 2013)
This study takes into account factors that could influence the positive or negative side of diversity. These factors stimulate or discourage interactions between team members that take place during team meetings. The level of intragroup conflict of diverse teams is likely dependent on what happens during their meetings. If team members think that they are excluded so that they cannot say anything without being judged, criticized, blamed, or interrupted, diverse teams are likely to experience more conflict, especially relationship conflict. On the other hand, if people listen actively to others, discussing their disagreement
or giving feedback, this can strengthen the effect of diversity in teams in a positive way. So, the positive or negative effect of gender diversity could depend on the interpersonal interactions that affect information-processing and social categorization (van Knippenberg & Schippers, 2007; Schneid et al., 2015). Furthermore, little is known about the influence of interactions between team members that take place during their meetings. So, this study provides insights into the interactions, behaviors, and statements of team members that are needed to successfully use diversity in teams.
This study also answers the call to overcome the problem of subjective performance ratings by providing a more accurate assessment of the work of demographically diverse teams. Van Dijk et al. (2012) showed that many studies in the field of diversity and performance outcomes are biased because of the use of subjective performance ratings. This is especially problematic in demographically diverse teams. Minorities in teams (e.g., women, ethnic minorities) get lower subjective performance ratings because of the stereotypes or prejudices of the raters. This study can overcome this problem by not using self-reports for measuring creativity.
This study has practical relevance as well. Conflicts in the workplace are often seen as harmful, which is why people try to avoid conflicts by, for example, working with others who are similar to them. As mentioned earlier, however, some research shows that conflict and diversity in teams can be beneficial for teams’ creativity (Jehn, 1995; Chen, 2006; Farh, et al., 2010). So, the easiest way of collaborating is not automatically the best way. Escalation of a conflict or a relationship conflict in the workplace has negative consequences, like stress, high turnover, absenteeism, and lower creativity (Moriarty, 2007). This study explores the functional and dysfunctional sides of conflict, which will help practitioners encourage the positive consequences and avoid the negative consequences of conflict.
LITERATURE REVIEW AND HYPOTHESES
Creativity
Most studies in the creativity field have researched the individual creative competence of innovators instead of team dynamics in reaching creativity (Kurtzberg & Amabile, 2001). However, team creativity is not the sum of individuals’ creativity. Rather, teams can reach creative synergy (Kurtzberg & Amabile, 2001). Teams can generate creative output that cannot be achieved by an individual. Meetings provide the opportunity to produce team creativity (Reiter-Palmon & Sands, 2015). Generated ideas are developed through the interactions between team members with different viewpoints and ideas. Additionally, the forming of creative ideas in the mind of the individual team member can be influenced by the other team members in the room. In this way, ideas could be formed, shared, inspired, and adapted by others (Kurtzberg & Amabile, 2001).
Team creativity is an interesting aspect of performance to study, because interaction between team members with different points of view influences the creative process (Østergaard, Timmermans, & Kristinsson, 2011). A meta-analysis supports the relationship between diversity and innovative performance. This study found no evidence for a relation between diversity and how well a team did the task (van Dijk et al., 2012). Additionally, generating ideas is a functional meeting behavior and prior research shows that it is positively correlated with other indicators of performance, like meeting satisfaction and productivity (Kauffeld & Lehmann-Willenbrock, 2012).
The section that follows provides a closer look at the concepts of diversity and intragroup conflict. Both diversity and conflict affect the interactions in teams and so can influence teams' creative process.
Diversity
The commonly agreed definition of diversity focuses on differences in attributes that could result in the perception that another is different from the self (Roberge & van Dick, 2010). There are many forms of diversity. Some studies distinguish the demographic dimensions of diversity from job-related dimensions of diversity (van Knippenberg & Schippers, 2007; Joshi & Roh, 2009; van Dijk et al., 2012). The demographic dimension, including age, gender, and ethnicity, is often related to negative team performance and the job-related
attributes such as function, education, and tenure, positively affect teams (Joshi & Roh, 2009).
However, a meta-analysis has shown that this proposition is highly biased by subjective performance measurements (van Dijk et al., 2012). Studies use subjective ratings as a proxy for objective performance, which could be problematic in the diversity field. For example, minorities in the business environment get lower performance ratings because people tend to evaluate people who are similar to them higher (van Dijk et al., 2012). So overall, there is little evidence that demographic related diversity negatively affects team performance and that job-performance diversity shows positive results. By using a more objective way of measuring creativity, this study can overcome the biases of subjective performance measurements.
Gender diversity is used in this study as an indicator of diversity. Gender diversity is a type of surface-level diversity. This means that the differences between team members are visible or easy to find out. The visibility is an important aspect of this type of diversity, because it can have immediate effects without the people knowing each other well (Harrison, Price, Gavin, & Florey, 2002). In this study, teams were composed temporarily, so gender is one of the characteristics that first becomes apparent. Additionally, prior research has shown that the visibility of gender is more likely to trigger conflict, because it leads to an easy and quick categorization between team members (Pelled, 1996; Nishii, 2013).
There are different perspectives on how diversity influences performance. First, the similarity-attraction paradigm holds that people prefer to cooperate with others who have the same values, beliefs, and attitudes (Byrne, 1997; Barsade et al., 2000). Second, the social identity theory, or social categorization theory, holds that people tend to categorize people from their own group and see others as outsiders. People treat members of their own group with favoritism and manifest negative attitudes to others based on their stereotypes and group traits (Tajfel, 1982; Stahl et al., 2010; Joshi & Roh, 2009). These two mechanisms assume negative effects of diversity on creativity (O'Reilly, Williams, & Barsade, 1998).
The third mechanism, the information-processing theory, or group information elaboration, underlines the positive effect of diversity. Diversity brings different knowledge, skills, views, and opinions to teams. A diverse team thus has a wider pool of resources, more diverse information, and a broader range of perspectives, which provide creative benefits (van Knippenberg et al., 2004; Dahlin, Weingart, & Hinds, 2005; Stahl et al., 2010).
Various underlying mechanisms could play a role in how diversity influences a team (Poling, Woehr, Arciniega, & Gorman, 2006). It is important to note, that studies with more
than one type of diversity show that even diversity types in the same category could have different results (Østergaard et al., 2011; Jehn, Northcraft, & Neale, 1999). Østergaard et al. (2011) found a positive relation between innovation and gender diversity on an organizational level, but a negative relation of age diversity and no relation of ethnicity diversity. Additionally, other studies show that diversity relates to creativity only under certain conditions, like intragroup safety, managing conflicts effectively and information elaboration (West, 2002; Hoever, van Knippenberg, Ginkel, & Barkema, 2012). Team members in a diverse team need a feeling of psychological and psychosocial safety in the group to discuss, exchange, and integrate ideas and knowledge related to the task (West, 2002; Hoever et al., 2012).
Therefore, when diverse team members have difficulties communicating and reaching a common understanding regarding the task, diversity could likely result in conflict (Kurtzberg & Amabile, 2001). The next section describes the types of conflict, how conflict aids or harms creativity in teams, and the contradictory results about the effect of conflict.
Intragroup conflict
Intragroup conflict occurs when team members perceive that their values or interests do not align with those of other team members (de Wit, Greer, & Jehn, 2012). Research examining the relation between intragroup conflict and (creative) performance outcomes in teams shows mixed and contradictory results (de Wit et al., 2012). Some research suggests that these conflicts negatively affect creativity because they cause incompatible goals and tensions among team members. Intragroup conflicts thus disturb the process of generating new ideas (Santos et al., 2015). Additionally, research suggests that openness is important in the process of creativity, therefore, conflict should reduce the production of creative ideas (Kurtzberg & Amabile, 2001). However, research also shows that a certain type of conflict can positively affect a team’s creativity. This is because conflict can lead to a discussion about the task what makes relevant information visible, stimulates critical thinking, and could result in more creative ideas (Jehn, 1995).
Prior research distinguishes task conflict and relationship conflict and both types of conflict have divergent effects on team outcomes (Jehn, 1995). Task conflict involves discussions related to the team’s task results stemming from differences in ideas, opinions, and viewpoints about the content of the task, whereas relationship conflict means that team members personally clash with each other independent of the task's content. This type of
conflict has an affective component resulting from tension, annoyance, and personal dislikes among team members (Chen, 2006).
Research shows that relationship conflict consistently has negative results. It limits the process of information exchange, because people focus less on the task and more on each other (Simons & Peterson, 2000). Relationship conflict leads to lower performance, commitment, and well-being (Greer, Saygi, Aaldering & de Dreu, 2012).
On the contrary, research shows that a certain amount of task conflict can have positive effects (Jehn & Bredersky, 2003). Discussion about the task could provide new insights, stimulates critical thinking, and promote integration of different viewpoints (Jehn, 1995). However, a meta-analysis found a negative relation between task conflict and team performance (de Dreu & Weingart, 2003). It is important to note that this meta-analysis uses decision quality, product quality, production quantity, and team effectiveness, rather than creativity as indicators of team performance. De Dreu and Weingart (2003) recognize the possibility that some level of conflict can positively affect creativity. Additionally, task conflict has a less negative effect on team performance than relationship conflict and some task conflict was even positively related to team performance.
Similarly, the empirical findings of the effect of task conflict on creativity are inconclusive. The meta-analysis of Hülsheger et al. (2009) did not support a relationship between the concepts. On the contrary, the meta-analysis of O’Neill et al. (2013) found a positive relation between task conflict and creativity. Moderating variables might influence this relationship, which will be discussed in the next section.
It seems that relationship conflict could never be beneficial for creativity. However, the existing research is unclear about the effect of task conflict regarding whether task conflict positively or negatively affects creativity and to what extent task conflict is beneficial.
Team processes are crucial to unlocking diversity's potential for team creativity. Research shows that diverse teams only reach higher creativity if team members can take the perspective of the other and integrate diverse viewpoints (Hoever et al., 2012). Therefore, the information-elaboration perspective of diversity is related to task-related aspects of team processes, whereas the social categorization perspective concentrates on the relational aspects (van Knippenberg et al., 2004).
Diversity can lead to interpersonal conflict, which can harm the creativity. On the other hand, diversity enables team members to discuss ideas and integrate diverse viewpoints.
It also leads to task conflict, which could positively affect team creativity. Based on theoretical arguments and existing research, I hypothesize:
Hypothesis 1: Task conflict positively mediates the relation between gender diversity and team creativity.
Hypothesis 2: Relationship conflict negatively mediates the relation between gender diversity and team creativity.
The moderating role of time
A possible explanation of the contradictory findings regarding the relation between conflict and organizational outcomes is that interactions between team members change during a project life cycle (Gersick, 1988). The relation between conflict and creativity is more complex than prior research has proposed. Conflict is a dynamic process instead of a static concept (Jehn & Mannix, 2001). Therefore, the effect of conflict on team creativity can change during a project (Farh et al., 2010).
Chen (2006) finds that project life cycle moderates the relation between conflict and creativity. The relation between conflict and creativity changes in different stages of the project. In the early phase, team members focus on how they will complete the task. This is the phase of discussion between team members about ideas, issues, and strategies (Gersick, 1988). In this stage, task conflict is functional for creating ideas. The group development literature establishes that storming, the intragroup conflict as members resist constraints, is one of the early stages of team development and so a requirement for good group functioning (Shaw & Barrett-Power, 1998). After the storming stage, team members establish a sense of social cohesion and group identity, which positively affects the performance of a team. However, in the later stages of the project life cycle, teams do not have enough time to manage conflicts, because team members have to complete the task. In this stage, task conflict hinders creative outcomes (Farh et al., 2010).
In contrast to the group development literature, Jehn and Mannix (2001) suggest that relationship conflict is destructive no matter the time and amount. Their research focused on when and what type of conflict high-performance teams have. Relationship conflict harms performance the most if it increases during the project life cycle; this is an indication of an escalating conflict.
Task conflict shows a more complex pattern. If relationship conflict is managed well, task conflict occurs the most in the middle of the project, because this is when people can discuss the content of the task (Jehn & Mannix, 2001). It is necessary that teams only have a task-related conflict that is beneficial for team performance, without friction or tension between team members. Otherwise, they will take the task conflict personally. Additionally, task conflicts negatively affect performance if they occur right before the project deadline. Contradicting other research, task conflicts could be detrimental in the earliest stages as well, because relationship conflict has a longer time to arise (Jehn & Mannix, 2001).
Prior research is inconsistent about the benefits of task conflict in the earliest stages of a project, but both task and relationship conflict appear to have more negative results for creativity when they occur right before the deadline, so I hypothesize:
Hypothesis 3a: The time when conflict occurs positively moderates the relation between task-related conflict and creativity, such that when conflict occurs in the beginning of the project, task-related conflict will have a stronger positive effect than if conflict occurs just before the deadline.
Hypothesis 3b: The time when conflict occurs negatively moderates the relation between relationship conflict and creativity, such that when conflict occurs in the beginning of the project, relationship conflict will have a weaker negative effect on creativity than if conflict occurs just before the deadline.
The moderating role of relationship conflict
In this context, it is interesting to find out how relationship conflict and task conflict relate to each other. Several studies show a positive relation between task conflict and relationship conflict (Simons & Peterson, 2000; Chen, 2006; de Dreu & Weingart, 2003; de Wit et al., 2012; Pelled et al., 1999). Various underlying mechanisms could explain the relation between the two types of intragroup conflict (Simons & Peterson, 2000).
The first mechanism is the process of misattribution. This mechanism is highly supported by research. Team members take a task-related conflict personally, especially if there is low level of trust in the team (Simons & Peterson, 2000) Second, if a debate is highly emotional or team members use aggressive language, a task-related conflict could easily be perceived as personal (Pelled et al., 1999). Third, if a task-related conflict occurs between
two members of the team for a long time, there is a greater possibility that these members perceive it as a personal conflict. However, there is less empirical evidence for this mechanism, because it is hard to test (Simons & Peterson, 2000). It is even possible that a relationship conflict sometimes leads to a task conflict. If a team member has a personal dislike for an individual, this team member could tend to dispute the ideas of this individual, take a contradictory perspective or be more aware of mistakes the individual makes (Pelled et al., 1999).
Two meta-analyses confirm that a high association between task and relationship conflict negatively affects the relation between task conflict on team performance (de Dreu & Weingart, 2003; de Wit et al., 2012). The meta-analysis of de Wit et al. (2012) found that task conflict could positively affect group outcomes under certain circumstances, like the absence of relationship conflict. This indicates that if task conflict is not managed effectively, conflict can escalate into an emotional relationship conflict (Kurtzberg & Amabile, 2001). Therefore, if task-related conflict occurs without relationship conflict, it will less negatively or even positively affect team performance (de Wit et al., 2012).
Hypothesis 4: Relationship conflict negatively moderates the relation between task conflict and creativity, so there is a less positive relation between task conflict and creativity for higher values of relationship conflict.
The moderating role of interpersonal interactions
As mentioned before, task-related conflicts could have positive effects, while relationship conflict shows only negative results. For reaching the full potential of diversity, it is important to research the interactions between team members during their meetings that reduce the relationship conflict and benefit productive task conflict in diverse teams. The information-elaboration perspective states that diverse teams have access to a greater pool of resources and information (Schneid et al., 2015). On the other hand, the social categorization perspective proposes that diverse teams cause problems, because team members are more positively inclined towards others who are similar to them than to those who are not (Stahl et al., 2010; Joshi & Roh, 2009). Therefore, the advantages or disadvantages of diversity in the form of relation or task-related conflict lie not in the conflict per se, but in the processes of how diversity leads to critical thinking and the creative use of diverse viewpoints and information instead of a process of exclusion (van Knippenberg et al., 2004).
The concepts of debate and inclusive climate are used to explain the role of interpersonal interactions during team meetings. Both aspects influence the processes (information-elaboration and social categorization) through which a diverse team could positively or negatively affect the form of a relational or task-related conflict (Nishii, 2013; Hoever et al., 2012). First, debate during a team meeting is beneficial for the information-elaboration process of diverse teams through the open discussion between team members, which allows the exchange of information (Hoever et al., 2012). Second, Inclusive climate could be linked to the social categorization theory (Nishii, 2013). In an inclusive climate, all members are treated equally. There are no inequalities based of differences in attributes between team members. Therefore, exclusion based on social categorization would be less the case in a team with an inclusive climate (Nishii, 2013).
Normally, both debate and inclusive climate are measured by self-report questionnaires, but because the concepts are associated with behavioral aspects, they can be observed during team meetings. This is beneficial, because it overcomes the biases of self-report questionnaires. In diverse teams, if team members hold negative biases against minorities in the team, the rating and perception of inclusive climate and debate would be different from reality (van Dijk et al., 2012). For example, measuring debate by watching the actual interactions that take place during team meetings enables the researcher to focus more on specific behaviors rather than on the general perception of differences or problems of team members (Simons, Pelled, & Smith, 1999).
Debate
Debate is a desirable and positive way of communicating in the form of an open discussion about task-related topics (Amason & Sapienza, 1997). In a debate, team members do not avoid differences in opinions, but there is an interaction between team members with opposing views and different approaches when solving a problem (Simons et al., 1999). Debate enables teams to benefit from diversity, because it allows integration of diverse viewpoints and opinions. Without debate, diversity in teams is not useful (Amason & Sapienza, 1997). Additionally, debate strengthens the information-elaboration process of diversity and thus the positive side of diverse teams. If debate takes place during the team meeting, the group will discuss, exchange, and integrate ideas and knowledge related to the task, which is a requirement for a successful diverse team (Hoever et al., 2012).
Active listening and voice are indicators of a debate. If team members have the idea that the others do not listen or that they do not have the right to speak, it could lead to
frustration and friction (Amason & Sapienza, 1997). This could provoke team members to try to avoid differences and thus avoid task-related conflict.
To get the full potential of diverse teams, availability of diverse viewpoints is not enough, because teams also need to use this diverse information (van Knippenberg et al., 2004). A team can only profit from diversity if it integrates diversity in the way the team works, communicates, and makes decisions (Simons et al., 1999; Ely & Thomas, 2001). If team members have a debate with each other, they have access to new information based on differences between team members. This will likely lead to a constructive task-related conflict. Therefore, debate is the condition for a constructive use of diversity in teams (Williams & O’Reilly, 1998; Simons et al., 1999).
Hypothesis 5: Debate positively moderates the relation between diversity and task conflict, so there is a more positive relation between diversity and task conflict for higher values of debate.
Inclusive climate
I propose that inclusive climate moderates the relation between gender diversity and conflict (Nishii, 2013). An inclusive climate means creating an environment that includes all employees. This climate involves personalized contacts, openness, and encouraging diverse perspectives even if they upset the status quo. Contradicting or conflicting viewpoints should be rated as valuable. An inclusive climate enables team members to feel safe to express themselves (Nishii, 2013; Ensari & Miller, 2006).
Indicators of an inclusive climate are that all team members have their voices, feel appreciated as part of the team, and are involved in the team process (Wasserman, Gallegos, and Ferdman, 2008). An inclusive climate means more than a sense of belonging. Team members are encouraged to show their unique characteristics and feel like part of the team at the same time (Shore et al., 2011). Based on the social categorization theory, people have negative attitudes to dissimilar others, which causes exclusion of individual team members (Joshi & Rho, 2007). An inclusive climate could overcome this phenomenon. This is especially important in a diverse team, because team members in a diverse team need a psychological and psychosocial safety in the group (West, 2002). Moreover, higher interpersonal trust increases the communication and connection between team members and decreases conflict (Chen, 2006).
Additionally, an inclusive climate could overcome evaluation apprehension, which is the fear that ideas will be evaluated negatively. This seems damaging for creativity, because how a team reacts to an idea influences the development of that idea and the generating of new ideas (Kurtzberg & Amabile, 2001). Moreover, research shows that encouraging participation of all team members weakens the effect of gender diversity on relationship conflict (Mohammed & Angell, 2004). Therefore, an inclusive climate could avoid the difficulties associated with diversity (e.g., relationship conflict). There will be fewer categorizations based on identity differences like gender, which is why especially in diverse teams, these interactions related to an inclusive climate will decrease relationship conflicts (Nishii, 2013).
On the other hand, an inclusive climate allows a team to benefit from diversity. According to the learning-and-effectiveness paradigm, teams could benefit from diversity by assigning value to diversity and using team members’ unique characteristics (Ely & Thomas, 2001). Belongingness also plays a role in this paradigm, because in an environment where differences are appreciated, all team members are involved in the work and so are part of the team (Ely & Thomas, 2001). Therefore, I hypothesize:
Hypothesis 6: Inclusive climate negatively moderates the relation between diversity and relationship conflict such that there is a weaker relationship between diversity and relationship conflict for higher than for lower inclusive climate.
Figure 1: conceptual model Gender diversity Relationship conflict Task conflict Creativity Time of conflict
+
+
+
-Debate Inclusive climate+
-+
-METHODOLOGY
Sample
A total of 204 students, divided into 41 teams, participated in this study. Each team consisted of four to six students and the teams had to work on a course-related assignment. Thirteen teams participated in a pre-master course and had to write a paper and 29 teams participated in a first-year bachelor course and had to deliver a presentation. Students from both courses were free to decide whether they wanted to participate in the study. Students that successfully completed all parts of the study got a bonus of 0.5 on their final grade. In this study, 31 teams were examined, because of technical problems with the videos and the lack of time to code all of them. The final sample consists of 153 students, 65.4% women and 34.6% men. The age of the students was between 17 and 32 years old, with a mean of 21,4 (SD = 2.581).
Data collection
The data gathering consisted of three questionnaires and two videotaped team meetings of approximately one hour each. Before the first team meeting, students filled in an online pre-questionnaire and at the end of each team meeting, students completed a paper-based questionnaire. The first team meeting was scheduled in the beginning of the course and the second just before the deadline of the assignment. The meetings were videotaped. Participating students were instructed to use the meeting to work on their team assignment.
Coding
Meeting interactions were coded with the act4teams coding scheme (Kauffeld & Lehmann-Willenbrock, 2012). INTERACT software of Mangold was used to code the content of the videos. Four types of interactions were coded: problem-focused statements, procedural statements, socio-emotional statements, and action-oriented statements. These four types are subdivided into 44 categories (see Appendix A). In this study, categories were selected from the problem-focused, socio-emotional statements, and action-oriented columns. These categories were chosen to measure debate and inclusive climate in the team. The behavioral codes that are relevant for the present research are shown in Table 1 and are explained in the next section.
Interrater reliability is calculated to ensure reliable coding by different raters. Nine videos were coded twice by two raters. The first rater subdivided the videos into units and
coded each unit with one of the 44 act4teams categories. The second rater encoded the same video using the subdivision of the first rater, but without seeing the provided codes of the first rater. Therefore, both raters independently coded each sense unit. The videos reached an interrater reliability value of Cohen’s κ between 0.739 and 0.865.
Table 1: Used codes of the Act4teams coding scheme (Kauffeld & Lehmann-Willenbrock, 2012)
Problem-focused statements Socio-emotional statements Action-oriented statements
Cross-linking a solution Positive socio-emotional statements Positive statements promoting action
Problem with a solution (PS)
objection to a solution
Encouraging participation (E)
e.g., addressing quiet participants
Providing support (A)
agreeing to suggestions, ideas, etc.
Active listening (AL)
signalizing interest (“hmm”, “yes”)
Humor (H)
e.g., jokes
Expressing feelings (F)
mentioning feelings like anger or joy
Offering praise (OP)
e.g., positive remarks about other people
Interest in change (IN)
signalizing interest in ideas, options, etc.
Negative socio-emotional statements Negative statements inhibiting action Criticizing/running someone down
(CR)
disparaging comments about others
Interrupting (I)
cutting someone off while speaking
No interest in change (NI)
e.g., denial of optimization opportunities
Terminating the discussion (END)
ending or trying to end the discussion early
Measures
Gender diversity
Blau’s index is the most commonly used measurement for diversity in research, with the perspective that heterogeneous groups outperform homogenous teams because of the access to more diverse resources (Harrison & Klein, 2007; Biemann & Kearney, 2010). However, a limitation of this index is that the scores of Blau’s index increase when there are more team members. In this research, where the number of team members differs, Blau’s index could lead to biases. Therefore, the BlauN index is used. This index accounts for varying group sizes in a sample (Biemann & Kearney, 2010). The formula is:
𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝑁𝑁 = 1 − �𝑁𝑁𝑁𝑁(𝑁𝑁 − 1)𝑖𝑖(𝑁𝑁𝑖𝑖− 1)
Where Ni is the absolutely frequency of group members in the ith category and N is the
number of group members. The score of the BlauN index can be between 0 and 1, where 0 means no diversity at all and 1 means all team members belongs to a different category. In
the case of gender diversity, a score of 1 is only possible in a team of two. The highest score in this present study is 0.67, with a mean of .376 (SD=.245).
Intragroup conflict
The level of intragroup conflict is measured with the 8-item Likert-scale of Jehn (1995). Four items measure relationship conflict and four items measure task conflict. An example of an item that measures relationship conflict is: How much friction was there among members in your team? and an example of a task conflict item is: How often did people in your team disagree about opinions regarding the work being done? The items were measured on a 5- point Likert scale, where 1=never/none and 5=very often/very much. These questions were asked after the first (T1) and the last meeting (T2). Table 2 shows the Cronbach’s alpha of the scales of the first and second meeting on the individual level.
The individual scores on these four variables were converted into team scores by taking the mean score of all the team members. Finally, the mean scores of T1 and T2 of task conflict and of T1 and T2 of relationship conflict were calculated.
Table 2: Reliability of the scales—Cronbach’s alpha
Time of conflict
The time of conflict is measured by the difference in perceived intragroup conflict in T1 and T2. The amount of task conflict at T2 was subtracted from the team score of task conflict at T1. The same calculation was performed for relationship conflict. A negative score on time of conflict means that the team perceived more conflict on time point 2 than on time point 1 and a positive score means less perceived conflict on time point 2.
Creativity
After coding the videotaped meetings, all solutions and ideas the students came up during their second meeting were rated on creativity. The technique used is derived from the Consensual Assessment Technique (CAT) developed by Amabile (1982) and transformed to
Variable # items Cronbach’s alpha
Task conflict T1 4 .836
Task conflict T2 4 .881
Relationship conflict T1 4 .797
fit the current study. The definition of workplace creativity used as starting point is: the production of ideas, solution, and processes that are novel or original and are (potentially) relevant or useful (Oldham & Cummings, 1996). The validated technique of Amabile (1982) assumes that judges who are familiar with the theme have a common understanding of creativity and so could reach agreement of their ratings.
For the present study, a framework was developed to guideline the ratings (Appendix B). The procedure of rating creativity is as follows. First, all the ideas and solutions mentioned in the videos are written down. Second, each idea and solution is ranked on a five-point scale on both the novelty and usefulness. Third, the mean score of the novelty and usefulness of each idea and solution is calculated. Finally, the number of ideas and solutions with an average score higher than four is counted. This is the team’s creativity score.
Debate
Most studies measure debate with a self-reporting questionnaire. In this study, debate was measured by observed behaviors during the video-taped meetings. As mentioned before, these behaviors were coded with the act4teams coding scheme. Debate was operationalized in concrete indicators, namely the frequency that certain statements were made during the team meetings.
Statements used to operationalize debate are part of the problem-focused category and the action-oriented category. The problem-focused category consists of statements about problem or solution related aspects of the meeting and the action-planning category consists of communication related to the team members’ willingness to improve their work and their planning to take action to reach this improvement (Kauffeld & Lehmann-Willenbrock, 2012). Simons et al. (1999) developed a 4 item Likert-scale to measure debate in a decision-making context in top management teams (TMT). Their items are related to problem, solution, and action taking related aspects of a team meeting. For example, an item of this scale is: ‘In discussion of the issue, executives stated clear disagreement with each other’. This is comparable with stating a problem with a solution, which is part of the act4teams coding scheme.
Another important part of a functional debate is an open discussion without avoiding differences. This is the opposite of ending a discussion early, which is coded during the team meetings. Additionally, interest in change (e.g., signaling interest in ideas and options) can be an indicator of a functional debate. The opposite, namely no interest in change, is damaging for a debate. Examples of no interest in change are statements that propose everything
remaining the same, trying to not confront problems, or showing little trust in new innovations.
The number of sense units during both meetings was counted for the two positive statements (stating a problem with a solution and interest in change) and the frequency of the two negative statements (ending a discussion early and no interest in change) is subtracted from it to operationalize a functional debate.
Inclusive climate
As mentioned in the theoretical section of this study, an inclusive climate means an environment where everyone feels free to say what they want and all the team members feel that they are appreciated (Wasserman et al., 2008; Nishii, 2013). Statements reflecting an inclusive climate are part of the social-emotional communication category of the act4teams coding scheme and consist of the relational aspects that occur during team meetings. Statements falling within this category include encouraging participation, providing support, active listening, and offering praise. These statements indicate the opportunity that team members can bring their unique characteristics to the table. On the other hand, the negative statement of interrupting each other is a dysfunctional meeting behavior and obstructs the openness of the team. Additionally, personalized contacts and a sense of belonging are part of an inclusive climate (Ensari & Miller, 2006) and the statements humor and expressing feelings of the act4teams coding scheme reflect this. On the other hand, evaluation apprehension, the fear that ideas presented will be negatively evaluated, indicates a poor inclusive climate (Kurtzberg & Amabile, 2001). The statement of criticizing/running someone down could lead to evaluation apprehension.
The operationalization of inclusive climate is as follows: the number of functional sense units reflecting inclusive climate (encouraging participation, providing support, active listening, offering praise, humor, and expressing feelings) minus the number of dysfunctional indicators of an inclusive climate (interrupting each other and criticizing/running someone down).
Control variables
In this study, the team size and the course the team participated in were controlled for. The team sizes varied from four to six team members.Early research shows that if teams grow in size the performance reduces. Moreover, group size is related to complexity and thus larger teams will experience more problems to manage diversity (Stahl et al., 2010). Additionally,
there might be difference, both in terms of team outcomes and team processes, between students who participated in a pre-master course and in a first-bachelor course (i.e., students could differ in experience in team work). Moreover, the teams of both courses had different assignments. Therefore it was necessary to control for team size and course in all the analyses.
RESULTS
Bivariate and descriptive statistics
Table 3 shows the descriptive statistics and the results of the bivariate correlation analysis. These statistics provide a first impression of the data and the relations between the variables. Additionally, multicollinearity (i.e., too much correlation between two predictors) was checked (Field, 2013). Table 3 shows that task conflict and relationship conflict correlate highly with each other (ρ = .822***). This could be problematic, but because of the point of interest in this study, namely how these two constructs relate to each other, I kept these constructs separated.
The correlation analysis shows some remarkable results. First, the amount of observed debate positively correlates with the time of task conflict (ρ = .552**) and the time of relationship conflict (ρ = .449*) such that if teams perceived more conflict in the early stages and reduce this in the second meeting, debate is higher. No correlation was found between debate and conflict. Additionally, no correlations were found at the main points of interest, namely between diversity and creativity, diversity and conflict, and conflict and creativity.
Hypotheses testing Mediations
To test the first two hypotheses, model 4 of Process is used (written by Hayes for SPSS). This model can test for two mediators at the same time. Table 4 shows the results of the mediations. There are no significant relations between diversity and task and relationship conflict, diversity and creativity, and both types of conflict and creativity. Additionally, the result of the indirect effects (see Table 5) shows that both effects are not statistically different from zero. In the case of task conflict, a1b1 = .0033, which means that teams that differ by one
unit regarding their diversity index score .0033 units higher on their creativity as a result of their task conflict. In the case of relationship conflict, a1b1 = -.149, which means that teams
that differ by one unit on their diversity index score .149 units lower on their creativity as a result their relationship conflict. Both effects are not statistically different from zero, as revealed by a 95% BC bootstrap confidence that is below and above zero (task conflict: [-.755, .725] and relationship conflict: [-1.521, .266]). Therefore, no support was found for the first and the second hypotheses, that relationship and task conflict mediate the relation between gender diversity and creativity.
Moderations
To test hypotheses 3a, 3b, 4, 5, and 6, a linear regression analysis was used. To test the effect of the moderator variables, interaction terms were used by computing the product of the moderator and the independent variables after standardizing them to overcome a potential multicollinearity problem. The hypotheses were tested in two steps: first, the main effect was tested and then the interaction terms and moderator were added.
Hypothesis 3a states that when conflict occurs in the beginning of the project, task-related conflict will have a stronger positive effect on creativity than if conflict occurred just before the deadline. Model 1 shows that task conflict has a negative, but insignificant effect, on creativity (b=-.466, se=.661). This is the opposite direction than hypothesized. Model 2 shows that when the task conflict takes place positively relates to creativity (b=1.423*, se=.581). This means that when the difference in task conflict between time point 1 and time point 2 was higher such that it was higher at the first meeting, this positively relates to creativity. However, no positive moderation was found between task conflict and time of conflict. Therefore, no support was found for hypothesis 3a.
Hypothesis 3b states that if relationship conflict occurs in the beginning of the project, it will have a weaker negative effect on creativity than if it occurs just before the deadline. No support was found for this hypothesis. No support was found for hypothesis 4 either. Relationship conflict does not influence the relation between task conflict and creativity.
Table 7 shows the results of testing hypotheses 5 and 6. Model 6 shows that diversity slightly positively relates to task conflict, but this result is insignificant (b=120, se =.424). After adding the moderator and the interaction terms in model 7, a negative (significant at a 10% level) relation was found between debate and task conflict. But again, the interaction term is insignificant, so hypothesis 5 was not supported. Therefore, debate does not moderate the relation between diversity and task conflict. Model 8 shows that diversity positively, but insignificantly, relates to relationship conflict. Additionally, inclusive climate does not influence this relationship. Therefore, hypothesis 6 was not supported.
Table 3: Descriptive statistics and correlation coefficients p†<0.10; *p<0.05; **p<0.01; ***p<0.001. Variable Mean S.D. 1 2 3 4 5 6 7 8 9 10 1. Creativity 1.290 1.553 - 2. Gender diversity .376 .245 -.011 - 3. Task conflict 1.985 .423 -.122 .078 - 4. Relationship conflict 1.580 .404 -.212 .214 .822*** - 5. Debate 47.839 29.983 .035 -.149 -.119 -.174 - 6. Inclusive climate 170.161 55.917 .016 -.075 .056 .004 -.189 -
7. Time of conflict task -.194 .448 .261 -.004 -.208 -.088 .552** -.249 -
8. Time of conflict relationship -.192 .394 .028 -.076 -.321† -.244 .449* -.181 .554** -
9. Course (pre-master = 0,
bachelor = 1)
.58 .502 -.352 .215 -.199 -.016 .004 -.145 .298 .224 -
Table 4: Mediation - Regression coefficients, Standard errors, proportion explained variance, and F statistic
Consequent Antecedent
Task Conflict (M1) Relationship Conflict (M2) Creativity (Y)
Coeff. SE p Coeff. SE p Coeff. SE p
Gender diversity (X) a1 .032 .350 .929 a1 .201 .335 .554 c1’ 1.566 1.225 .213
Task Conflict (M1) - - - b1 .104 1.173 .930
Relationship Conflict (M2) - - - b1 -.738 1.222 .551
Course (Control 1) -.308 .180 .098 -.161 .172 .358 -.601 .660 .372
Team size (Control 2) -.201 .151 .195 -.187 .145 .209 .948 .539 .091
Constant i1 3.153 .859 .001 i1 2.514 .825 .005 i2 2.109 1.436 .153 R2 = .113 R2 = .105 R2 =.275 F(3,27) = 1.143 p = .350 F(3,27) = 1.052 p = .386 F (5,25)= 1.895 p = .131 N=31, Bootstrap resamples: 5000
Table 5: Mediation – Direct, total, and indirect effects
Effect SE p LLCI ULCI
Direct effect c1’ 1.566 1.224 .213 -2.177 2.872
Total effect c1 1.421 1.180 .239 -2.470 2.337
Boot SE Boot LLCI Boot ULCI
Indirect effect Task a1b1 .0033 .369 -.755 .725
Indirect effect Relationship
a1b1 -.149 .382 -1.521 .266
Table 6: Moderation – Creativity, task conflict, relationship conflict, and time of conflict
*p<0.05; **p<0.01; ***p<0.001.
Hypothesis 3a Hypothesis 3b Hypothesis 4
Dependent variable: Creativity Model 1 Model 2 Model 3 Model 4 Model 5
b se b se b se b se b se
Independent variable
Task conflict -.466 .661 -.168 .623 -.116 1.197
Relationship conflict -.548 .682 -.385 .758
Moderator variable
Time of conflict task 1.423* .581
Time of conflict relationship .377 .739
Relationship conflict -.213 1.424
Interaction terms
Task conflict X Time of conflict task 1.622 1.837 Relationship conflict X Time of
conflict relationship
.728 2.652
Relationship conflict X Task conflict -.424 1.188
Control variables Course -.677 .650 -1.131 .646 -.623 .626 -.690 .657 -.638 .679 Team size .721 .495 .659 .486 .696 .498 .711 .528 .701 .517 Constant -1.848 2.667 -1.219 2.654 -1.755 2.667 -1.764 2.838 -1.713 2.774 R2 .223 .384 .227 .238 .231 F(3.27)=2.585 p=.074 F(5,25)=3.111 p=.026 F(3.27)=2.647 p=.069 F(5,25)=1.565 p=.206 F(5,25)=1.505 p=.224
Table 7: Moderation – Task conflict, relationship conflict, debate, and inclusive climate
p†<0.10 *p<0.05; **p<0.01; ***p<0.001.
Hypothesis 5 Hypothesis 6
Dependent variable: Task conflict Model 6 Model 7 Dependent variable: Rel. conflict Model 8 Model 9
b se b se b se b se
Independent variable Independent variable
Gender diversity .120 .424 -.014 .420 Gender Diversity .346 .412 .378 .433
Moderator variable Moderator variable
Debate -.006† .003 Inclusive Climate .001 .002
Interaction terms Interaction terms
Gender diversity X Debate -.002 .011 Gender diversity X Inclusive Climate
-.004 .011
Control variables Control variables
Course -.433 .218 -.435 .212 Course -.233 .212 -.247 .235
Team size -.176 .183 -.205 .180 Team size -.141 .178 -.129 .189
Constant 3.169** .971 3.338** .952 Constant 2.501* .943 2.449* 1.000 R2 .132 .245 R2 .086 .092 F(3,27)=1.364 p=.275 F(5.25)=1.624 p=.190 F(3,27)=.847 p=.480 F(5,25)=.509 p=.767
DISCUSSION AND CONCLUSION
The purpose of this study was to examine the role conflict in the diversity-creativity link. This study added to the current research by using of a more objective performance rating than previous studies—rating creative ideas and solution without the involvement of the team members. Additionally, the interpersonal interactions during team meetings were observed to access how diversity can lead to a task or relationship conflict. No self-reporting questionnaires were used, rather, the meetings were videotaped and the interactions between team members were coded to overcome biases that could occur in teams where gender minorities are less appreciated (van Dijk et al., 2012). Additionally, this study tried to provide clarity in how task and relationship conflict relate to each other and what role time of conflict plays in the effectiveness or damage of conflict. The academic relevance, managerial implications, limitations and suggestions for future research are discussed below.
Although no support was found for the hypotheses, this study has some academic relevance, so some results are discussed. First, I hypothesized that task conflict positively mediates the relation between diversity and creativity, while relationship conflict negatively mediates this relation. As in previously research, this study found a high correlation between task and relationship conflict (Simons & Peterson, 2000; Chen, 2006; de Dreu & Weingart, 2003; de Wit et al., 2012; Pelled et al., 1999). A meta-analysis shows that a high correlation between task and relationship conflict influences the relation between task conflict and performance such that studies with a higher correlation between task and relationship conflict found a more negative relation between task conflict and performance ratings (de Dreu & Weingart, 2003).
Although task conflict negatively correlates with creativity in this study, it shows a very small positive indirect (but insignificant) effect as mediator between gender diversity and creativity, in contrast to relationship conflict, which only shows a negative relation with creativity. This suggests that although there is a strong positive correlation between task and relationship conflict, the types of conflict differ in explaining the potential link between diversity and creativity. However, this conclusion needs to be taken with a degree of caution, as no significant support was found for both mediations.
Early research shows that task conflict was damaging for creativity in the end stage of a project (Farh et al., 2010). But the current literature was unclear about the effect of task conflict in the earliest phase of a project (Jehn & Mannix, 2001; Farh et al., 2010). This study did not find a moderation of the timing of conflict with conflict on creativity. However, when
task conflict occurs relates to creativity. This means that when more task conflict is perceived in the first meeting than in the second meeting, this positively relates to creativity. The amount of conflict does not play a role, such that a high amount of task conflict does not more positively affect creativity when it occurs in the early stage of the project than when it occurs at the end.
The present study tried to find another way of measuring the concepts debate and inclusive climate. The main purpose was to research the actual interpersonal interaction instead of what team members perceive. This study was a first attempt in translating the concepts debate and inclusive climate in observable interactions. Future research is needed to develop and validate these measurements.
Remarkably, debate and task conflict show a negative (significant at a 10% level) relation. This result indicates that if team members state a problem with a solution, show interest in change, and do not end a discussion early, they perceive less task conflict. However, no support was found for the hypothesis that a debate strengthens the relationship between diversity and task conflict.
In this study, the videotaped meetings are used to require a more objective way of performance measurement. Previous studies have suggested that rater biases might influence the results in diversity research (van Dijk et al., 2012). This could explain finding no support for the hypotheses in this study. The perception of performance might be linked with the perception of diversity and team processes such that an objective creativity is not related to these concepts. Further research is necessary to study this.
Despite the lack of statistical evidence for the hypotheses, this study provides some managerial implications. A possible explanation for finding no relation between gender diversity and relationship conflict is that the students in the sample do not make categorizations in the team based on gender (Randel, 2002). Therefore, it could be helpful for managers to find out if inaccurate categorizations based on gender are made in their teams, for example, task related stereotypes, like men are better in this kind of task than women (van Dijk, 2013). Only if that is the case, it is necessary to change the stereotypes.
Although relationship conflict does not show significant effects in this study, it seems to have only negative results. Therefore, avoiding or managing relationship conflict in a team is desirable. Research shows that setting cooperative goals instead of competitive goals influences the outcomes of conflict (Tjosvold, 1998). Therefore, if a team clearly communicates a common goal, team members see conflict as a mutual problem that needs a common approach (Alper, Tjosvold, & Law, 2000). When the goals of different team
members are negatively linked, such that if one team member reaches a goal another does not, conflict resolution will not work. Therefore, managers have to ensure that their team members have a common goal or positively linked goals by developing a team-based incentive structure. When relationship conflict occurs, managers could reframe this conflict as a mutual problem. This lets the team members have a common goal to resolve the conflict (Deutsch, 2006).
This study has some limitations. First, the power of this study is not high, as only 31 teams were examined. The power of a study is the ability to find an effect when it exists. When the power is low, it is hard to find significant results, especially when it is about small effects. Future research should increase the sample size to find out whether there really are no significant results or that the insignificant results of the present study were caused by a lack of power.
When it comes to the operationalization of creativity, debate, and inclusive climate, this study attempted a new way of measuring. Although the operationalization is based on validated concepts, coding, and techniques, it requires more research to validate this way of measuring.
Additionally, the meetings were videotaped. Research shows that this could lead to productivity losses and that respondents do not act as they normally would (Kurtzberg & Amabile, 2001). Another possible factor that plays a role in the results is that the students were free to choose to participate in the course related study. Therefore, the students in the study might not be representative of the other students in the course. For example, it is likely that the students who participate in the study are more motivated regarding their studies.
Based on the results, some other opportunities for future research arise. The connection between debate and task conflict should be further explored. The perception of task conflict is not in line with the behaviors that reflect the positive side of a task conflict in the theory. The respondents do not seem to perceive an open discussion as a conflict.
Additionally, the mediating role of task conflict requires more research. The role of task conflict in explaining the relation between diversity and creativity is still unclear. Other aspects likely play an important role in the relation between task conflict and creativity, for example, how teams perceive their conflict and their diversity. Their conflict resolution strategy or the trust within the team could also have an effect (Simons & Peterson, 2000; de Dreu & Weingart, 2003). Future studies could take these factors into account.
This study explored the role of conflict in the relation between diversity and creativity. Previous studies show inconsistent findings. By adding contingency factors and
using another way of measuring creativity, the present study tried to explain these inconsistent results. Unfortunately, no support was found for the hypotheses. Nevertheless, this study has raised the issue of a gap between perceptions and actual behavior. Videotaped meetings for data collection allowed observation of actual interactions between team members during their meetings. The results suggest that the perception team members have about conflict in the team is not always in line with the actual interpersonal interactions. Therefore, more objective ways of measuring should be used in research to determine the reality instead of the perceptions.
REFERENCES
Alper, S., Tjosvold, D., & Law, K. S. (2000). Conflict management, efficacy, and performance in organizational teams. Personnel Psychology, 53(3), 625-642.
Amabile, T. M. (1982). Social psychology of creativity: A consensual assessment technique. Journal of Personality and Social Psychology, 43(5), 997-1013.
Amason, A. C., & Sapienza, H. J. (1997). The effects of top management team size and interaction norms on cognitive and affective conflict. Journal of Management, 23(4), 495-516.
Barsade, S. G., Ward, A. J., Turner, J. D., & Sonnenfeld, J. A. (2000). To your heart's content: A model of affective diversity in top management teams. Administrative Science Quarterly, 45(4), 802-836.
Biemann, T., & Kearney, E. (2010). Size does matter: How varying group sizes in a sample affect the most common measures of group diversity. Organizational Research Methods, 13(3), 582-599.
Byrne, D. (1997). An overview (and underview) of research and theory within the attraction paradigm. Journal of Social and Personal Relationships, 14(3), 417-431.
Chen, M. H. (2006). Understanding the benefits and detriments of conflict on team creativity process. Creativity and Innovation Management, 15(1), 105-116.
Dahlin, K. B., Weingart, L. R., & Hinds, P. J. (2005). Team diversity and information use. Academy of Management Journal, 48(6), 1107-1123.
De Dreu, C. K., & Weingart, L. R. (2003). Task versus relationship conflict, team performance, and team member satisfaction: A meta-analysis. Journal of Applied Psychology, 88(4), 741-749.
Deutsch, M. (2006). Cooperation and competition. In M. Deutsch, P. T. Coleman, & E. C. Marcus (Eds.), The handbook of Conflict Resolution: Theory and practice (pp. 23-40). San Francisco, CA: Jossey-Bass.
De Wit, F. R., Greer, L. L., & Jehn, K. A. (2012). The paradox of intragroup conflict: A meta-analysis. Journal of Applied Psychology, 97(2), 360-390.
Ely, R. J., & Thomas, D. A. (2001). Cultural diversity at work: The effects of diversity perspectives on work group processes and outcomes. Administrative Science Quarterly, 46(2), 229-273.
Ensari, N. K., & Miller, N. (2006). The application of the personalization model in diversity management. Group Processes & Intergroup Relations, 9(4), 589-607.
Farh, J. L., Lee, C., & Farh, C. I. (2010). Task conflict and team creativity: A question of how much and when. Journal of Applied Psychology, 95(6), 1173-1180.
Field, A. (2013). Discovering statistics using IBM SPSS statistics. London, England: Sage.
Gersick, C. J. (1988). Time and transition in work teams: Toward a new model of group development. Academy of Management Journal, 31(1), 9-41.
Greer, L. L., Saygi, O., Aaldering, H., & de Dreu, C. K. (2012). Conflict in medical teams: Opportunity or danger?. Medical Education, 46(10), 935-942.
Gilson, L. L. (2015). Creativity in teams: Processes and outcomes in creative industries. In C. Jones, M. Lorenzen, & J. Sapsed (Eds.), Oxford Handbook of Creative Industries (pp.50-74). Oxford, England: Oxford University Press.
Harrison, D. A., & Klein, K. J. (2007). What's the difference? Diversity constructs as separation, variety, or disparity in organizations. Academy of Management Review, 32(4), 1199-1228.
Harrison, D. A., Price, K. H., Gavin, J. H., & Florey, A. T. (2002). Time, teams, and task performance: Changing effects of surface-and deep-level diversity on group functioning. Academy of Management Journal, 45(5), 1029-1045.
Hoever, I. J., van Knippenberg, D., van Ginkel, W. P., & Barkema, H. G. (2012). Fostering team creativity: Perspective taking as key to unlocking diversity's potential. Journal of Applied Psychology, 97(5), 982-996.
