The effect of team composition changes
on innovation capabilities
Leon van Vliet (10354751) Date: 27 July 2017
Words: 9400
Abstract: This thesis studies the relationship between team composition changes (TCCs) and innovation in the video-game industry. Previous research is inconclusive about this
relationship, and shows contradictory findings. We introduce an integrated theoretical framework showing the relationship between TCCs and innovation, where social effects disrupt informational effects. Using data on 400 video-games reviews we test the hypothesis that the relationship between TCCs and innovation is inverted U-shaped. We operationalize the construct innovation by Computer Assisted Text Analysis, and introduce a model on the relation between TCCs and innovation. Contrary to predictions we do not find a significant relationship between TCCs and innovation.
Statement of originality
This document is written by student Leon van Vliet, 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.
Introduction
In 2017 the game True Story was released. True Story involved a range of innovations involving new ideas, and creativity to challenge the buyer of the game. Another innovative application can be found in the medical industry; an app has been made by a team of people who haven’t met before, to let technicians communicate with hospitals. Also in the consulting industry market participants are eager to innovate; a web-tool was made to help run
workshops for clients.
These examples show innovative applications. Indeed, in today’s quickly changing, and challenging environments, innovations are crucial for firms to grow, and maintain a sustainable and competitive advantage (Leifer, 2000; Tellis, Prabhu, & Chandy, 2009; West & Anderson, 1996). Tellis et al. (2009) note that organizations failing to innovate can be brought down because of this. Firms recognize this and prioritize it more, trying to find ways in which innovative capabilities may be improved (Anderson, Potočnik, & Zhou, 2014).
The examples also show how the innovations that the organizations introduced, were made by teams. One of the ways in which firms might improve innovativeness is by forming teams inside the organization (Brown & Eisenhardt, 1995). Zaccaro and Bader (2003) note that teams are more adaptive, because they have a larger array of capacities, and networks to draw on. This adaptivity results in innovation. Organizations have picked up on this fact, and increasingly rely on teams for its innovations (Ilgen, Hollenbeck, Johnson, & Jundt, 2005).
Like in the case of True Story, These teams often change as to the needs of the project the team works on (Huckman, Staats, & Upton, 2009). In fact, the one thing that stood out for the game and the other applications was that the teams that had created them solely consisted of freelancers; it thus consisted of employees who have never worked together, as The New York Times describes it: “The workers were all freelancers who typically had never met and, perhaps more striking, the entire organization existed solely to create the game and then disbanded” (Scheiber, 2017).
Teams don’t stay the same. Changing team composition has become easier, partly due to advances in technology: “It’s way easier to search for people, bargain and contract with them.”(Scheiber, 2017). Sometimes new members enter, and others leave. This is also the case for Top Management Teams (TMTs) who, according to upper echelons theory, have a large influence on the performance of the organization (Carpenter, Geletkancz, & Sanders, 2004).
In this paper, we study the effect of membership change, or team composition changes (TCCs), in top management teams on team innovation. This is of both practical and
theoretical importance. Practically, innovation is crucially important for the performance of the firm. Thereby, knowing the effect of changing a team’s composition on innovation may guide a manager’s decision on whether it is necessary, or not, to shuffle teams.
Theoretically, team innovation has traditionally received little research attention (Kurtzberg & Amabile, 2001; West, 2002). Kurtzberg and Amabile (2001) argue that
Guilford’s (1950) pioneering work on individual creativity has left little room for research on the effects of interpersonal interactions and teams on creativity and innovation. Team
innovation thus has traditionally received little research. Besides the lack of research on team innovation, the role of team longevity has also been largely ignored in past research on team cognition (Mohammed, Ferzandi, & Hamilton, 2010), and the research that has been done has generated an ongoing debate on the up-and-down sides of team longevity in general, and on innovation in particular (Humphrey & Aime, 2014).
On the upside, research has shown that when keeping teams stable and not introducing TCCs, teams become more innovative. One reason for this could be that, for example, the team members come to trust each other as they spend more time together, increasing
communication, and thereby increasing innovation (Mcevily, Perrone, & Zaheer, 2003). On the downside, increasing TCCs would increase innovation, for example simply because of the introduction of new ideas of the newcomer into the team (Choi & Thompson, 2005).
The present study looks at the effect of team composition changes in top management teams on innovation in the video-game industry. Studying this effect in the video-game industry is especially revealing, as the case for the need of innovation may be even more explicit in cultural industries like the video-game industry, as Lampel, Lant and Shamsie (2000) put it: “competition in cultural industries is driven by a search for novelty.”
Nowadays the video games are of such scope – the size of the video game-industry in 2015 was 92 billion dollars (Takashi, 2016) - that they get developed by teams consisting of specialists that get reshuffled with every new release (Huckman et al., 2009). Thus, there is both a need for innovation and TCCs in the top management in the video-game industry.
A TMT may have worked together as a team in the past, or it may have been assembled from individuals who have not worked together before. These TMTs thus have different degrees of experience working together. Because of the possibility, necessity, and widespread use of team composition changes, and the importance of innovation in the video
game industry, the question that arises, and that will be dealt with in this paper is: does having top management team composition changes have effects on the innovative capabilities of a team. In effect, would changing the team composition harm or help the video-game the team is creating in terms of its innovativeness?
This paper contributes to organizational theory in two ways. Firstly it proposes an integrated theoretical framework, adapted from Knippenberg, De Dreu and Homan (2004), that shows the relationship between TCCs and innovation. In the framework, we depart from the view that social, and informational or cognitive factors can be seen as separate; we
theorize that social and informational factors, both leading from TCCs, interact, to predict that the effects of TCCs on innovation evolves from negative, to positive, to negative, to form an inverted U-shape. Secondly, this paper provides empirical evidence of the relationship between Team Composition Changes and innovation, by testing the proposed theoretical model.
The outline of this thesis is as follows: firstly, we define and conceptualize innovation and team composition changes. We then review literature on the relationship between these, and propose a model based on this literature. Next the methods section describes the setting, data and statistical methods, after which follow the results. Lastly, we discuss these results, including the implications for practice, limitations of the current research and suggestions for future research.
Theoretical framework
Some research findings suggest that changing the team composition would harm innovation capabilities, other research suggests otherwise. To get a clear understanding of what might happen when an organization changes a top management team, and its effect on the
innovative capabilities of the team, we first look at what innovation is, and what the determinants for innovation are. We then look at what TCCs are, and what happens when TCCs occur; what constructs change when we introduce TCCs? Lastly, we look at how the constructs affected by TCCs, affect the determinants of innovation. Knowing this we propose a model on the relationship between TCCs and innovation.
Innovation
As noted, team innovation is very important to a firm. However, it is not always clear how researchers and practitioners define team innovation (West, 2002). Researching the
relationship between team composition changes and innovation requires understanding of the construct. West and Farr (1990) define team innovation as “the introduction or application within a team of ideas, processes, products, or procedures that are new to that team and that are designed to be useful”. The introduction of an idea signifies that innovation relates to, and needs, creativity. This creativity can be defined as the generation of original and useful ideas (Amabile, 1983), and can be seen as the first step towards innovation.
However, for innovation to happen, the team needs more than just creativity, as noted in the definition of innovation by the word “application”. Innovation starts with creativity, but it also depends on other factors (Amabile, Conti, Coon, Lazenby, & Herron, 1996).
Sometimes creativity stimulation may have adverse effects on the application of the creative idea (West, 2002), and sometimes innovation may occur, even when no new creative ideas in the team are created by the team itself, but rather by the introduction of outside creative ideas.
In their paper on innovation Scott and Bruce (1994) have defined three behavioral tasks that are involved in innovation: idea generation, idea promotion, and idea realization. These tasks are stages in a multistage process, and for each stage the team needs different behaviors. The authors argue that innovation begins with the recognition of a problem, and the subsequent generation of ideas or solutions. These ideas can be either novel or adopted. Next, the individual with the idea must communicate this idea, and get support from the team. Lastly, the team must complete innovative idea by first producing a prototype, and finally implementing the innovation by mass-production.
These behavioral tasks show what the team needs for innovation to take place: not only should team-members be creative, and come up with ideas but they should also communicate their ideas, and information, and finally implement the innovation. Knowing these requirements for innovation we do a study on whether and how team composition changes might affect these determinants of innovation.
Team composition changes
Team composition changes in organizations have many different effects within an
organization. Research on these effect can roughly be categorized into two different areas: informational, or knowledge, and social factors (van Knippenberg, 2017). informational factors involve the effects on teams on the amount of information created, shared, the amount of shared knowledge, and the knowing of who knows what (Humphrey & Aime, 2014; Ren, Carley, & Argote, 2006). Social factors involve the effect on teams that are interpersonal and social of nature. These are for example: team climate, social categorization, conflict,
cooperation and trust (van Knippenberg, 2017).
The knowledge perspective posits that teams are more innovative than individuals alone because the teams are the natural place where knowledge, information and ideas get shared, thus giving the teams access to a wider array of information which the team can combine to induce innovation. Central to the social perspective is the notion that people in teams tend to seek similarities and differences between work group members, and other workgroups, by which they can categorize themselves, and distinguish between in-group and out-group members.
Gersick (1991) has proposed that change can be conceptualized as a punctuated equilibrium. Normally everything in the organization is in equilibrium. In this equilibrium, only incremental adaptations are possible. Only when the equilibrium is punctuated, for example by the change of a member, can brief periods of revolutionary upheaval occur. In these periods, radical innovation can occur. Thus, in this paper, we conceptualize the team composition changes as equilibrium punctuations that influence certain constructs. In effect, these constructs have an impact on one, or more, of the three behavioral tasks that are
involved in innovation, thereby improving or worsening the innovative capabilities of a team. Next, we elaborate on the informational and social constructs that may have an impact on the innovative capabilities of a team. After which, we model this relationship between TCCs and innovation.
Effects of team composition changes on innovation capabilities
In general, literature on teams has made a clear distinction between the informational and social factors; For example, Ilgen et al. (2005) find that psychological safety (a social factor) leads to more innovation. This occurs through an increase in communication (or information sharing), which is a knowledge factor. Also, Ren et al. (2006) argue that knowing who knows what (a knowledge factor) improves innovation. However, most of the time the informational and social effects seem to be interconnected. For example, having more trust in yourcoworkers may influence the amount of information that you share, and creating conflict in your team may increase the amount of information created, if someone takes the role of a ‘devil’s advocate’, and rejects the first best idea.
A reconciliation is offered by Van Knippenberg, De Dreu and Homan (2004), who have studied the relationship between work group diversity and performance. They found that the relationship between work group diversity and performance yielded inconsistent results, and argued that this is because of three reasons: a lack of attention to group information processing; an oversimplification of social processes, and the fact that researchers have seen social and informational processes as separate. The authors then propose a new model: the Categorization-Elaboration Model (CEM), in which the informational and social processes in teams interact, in such a way that intergroup biases flowing from social categorization disrupt the elaboration (in depth-processing) of task relevant information and perspectives. Social factors thus interact with informational factors
In this paper, we assume that the effects of team composition changes on innovation are like the effect of diversity on group performance in the sense that informational, and social factors interact, and cannot be seen as being separate; social processes disrupt the elaboration of task relevant information and perspectives. Like the relationship between diversity and performance, the relationship between TCCs and innovation has yielded inconsistent results. Social factors and information factors are also often seen as separate. Also, TCCs and diversity are linked; changing membership can diversify the team by the introduction of new ideas, and insights. Generally, increasing TCCs increases diversity, showing a link between the two research areas.
Informational effects
We propose, much like van Knippenberg et al. (2004), that the elaboration of the task-relevant information is one of the primary process underlying the positive effects of TCCs on innovation. The elaboration of the task-relevant information is not only related to creativity, increasing the amount of ideas generated, but also to the implementation of innovations, improving innovation capabilities (Scott & Bruce, 1994). When more ideas get
communicated, there is more information available to each individual team member, and more knowledge integration, which improves innovation (Jin & Sun, 2010). similar effects were proposed and found by Edmondson (1999) and Jin and Sun (2010), who found that a higher level of information sharing, and knowledge integration leads to innovation. Besides, Richter, Hirst, van Knippenberg and Baer (2012) find that the exposure, and integration of diverse knowledge and perspectives stimulates creativity for individual team members.
Van Knippenberg et al. (2004) define the elaboration of information as “the exchange of information and perspectives, individual-level processing of the information and
perspectives, the process of feeding back the results of this individual-level processing into the group, and discussion and integration of its implications.” The most important
determinant of innovation thus is not so much the availability, quantity and diversity of the information, but rather the use, and elaboration of this information that is used in the task in which the team is involved. For example, Ren et al. (2006) researched the effect of transactive memory systems on innovation and found that with better memory systems there was a better knowledge retrieval process, which was linked to innovation.
The information of the individuals and the team can be conceptualized in two different ways: as transactive memory systems, and as shared mental models. The transactive memory system concept describes the team-member’s system for encoding, storing and retrieving information. This includes each team member’s knowledge of who knows what; where the expertise, knowledge and information is, that is required for the team to perform (Kozlowski & Ilgen, 2006). The other concept is the shared mental model, which is defined as the shared understanding of the task and the involved team work. (Espevik, 2011). These are two different theoretical perspectives; the first emphasizes specialization, and the distinction between team-members. The latter emphasizes that team members work together and have shared knowledge.
Both the transactive memory systems and the shared mental model acknowledge that the information is both about the work content (what needs to be done? how do we solve this problem), and about the work practices (how do we do things? What are the routines?) The
latter might be implicit, and not often talked about, as they are just how things are. This means that work practice information normally doesn’t get much elaborated upon, meaning that social effects mostly influence work content information. We will first elaborate on the work content information, where the social effects have an impact on.
Work content information
The work content information is the information needed on the job. This could be for example information on how a product could be improved. Katila and Ahuja (2002) found that the search for innovations occurs in two dimensions; the search depth, or how often the team uses existing information, and search scope, or the extent to which the team explores new knowledge. With the introduction of a new member in the team the amount of existing information might increase.
Some research has found that groups think more like each other the more they spend time with each other, thus the breadth of information available, and the diversity of
information decreases (Van Dijk, Van Engen, & Van Knippenberg, 2012). However, other research has found, contrary to predictions, that the team’s mental models of the group’s work and each other’s expertise didn’t become more similar as the team worked more together, because of an increase in specialization (Levesque, Wilson, & Wholey, 2001).
Introducing TCCs thus might improve the availability, and diversity of information. As noted increasing the availability, quantity and diversity of information has the potential of increasing innovation capabilities, simply because there is a larger pool of information with which the team can work, and which later can be used to make inferences, or gain new insights, improving innovation.
Introducing a new team member, who has spent little or no time with other team members means that the total information available might increase, dependent on whether the team member’s mental models converge or not. If they do not converge the new member has information which the other team members don’t know about. Another factor that might limit the available information is the fact that after some time team members become increasingly isolated from other information sources (Katz, 1982). This could hamper the availability of information even further. As mentioned earlier, this does not mean that innovation
automatically improves with the introduction of TCCs, because this new information first must be elaborated upon. It does, however, introduce the potential for more innovation.
The only direct informational effect of TCCs come from the fact that the availability of information might increase, and thereby increase the potential for innovation. This
potential doesn’t bring any innovation however, if it is never realized through other processes; the elaboration of this information. Social processes may improve or harm this in certain ways.
Work practice information
Unlike work content, work practice information concerns information about the strategy, how things are done, routines, and procedures. As teams spend more time together they tend to get better at performing the existing routines (Huckman et al., 2009). These organizational routines are the rules and customs in the behavior of the members of an organization (A. C. Edmondson, Bohmer, & Pisano, 2001). These existing routines may however be suboptimal, even if these routines, strategies and procedures have been optimal in the past (Leonard-Barton, 1993).
Choi and Levine (2004) found that with the introduction of newcomers team members were induced to rethink the work practices, changing the work atmosphere, and helping the creation of innovative ideas. Thus, with more TCCs team reflexivity, or the process in which team members evaluate and reflect on the objectives, routines, and the work practices in general, increased. Likewise, Schippers, West and Dawson (2015) found that highly reflexive teams were more innovative in demanding situations, than teams that were less reflexive; with an increase in reflexivity the suboptimal routines got revised, which led to an increase in innovation.
Correspondingly, Katz (1982) argued that when team members are doing their job for a long time they tend to rely on the customary ways of doing things. When new members are introduced to the team these customary ways of doing things will be broken, and new, more innovative ways of doing things might get introduced. This introduction of new innovative ways may be intentional, or unintentional (Levine, Moreland, & Choi, 2001). The
introduction of new members may however be inefficient in this way because the old members don’t change; they pose their way of doing things onto the newcomers, without reflection (Lewis, Belliveau, Herndon, & Keller, 2007).
A shared vision in the team is another part of the work practice, which the doesn’t talk about often. Increasing TCCs may hamper the creation of a shared vision, as with the
introduction of new members to a team, they will have to adjust to each other, and create this shared vision. Pearce and Ensley (2004) found that innovation and a shared vision are
reciprocally and longitudinally related. Hülsheger, Anderson and Salgado (2009) confirm this finding in their meta-analysis; they found that a shared vision displays one of the strongest
relationships with creativity and innovation. Thus, increasing TCCs decreases the notion of a shared vision within the team, hampering innovation.
Social effects
We argue that social factors mediate the relationship between information elaboration and innovation; increasing certain social factors may have a positive or negative effect on the extent of information elaboration, and in-depth thinking. Thus, everything that might increase elaboration, communication, in-depth thinking stimulates innovation, and vice versa: anything disrupting elaboration, communication, in-depth thinking will block innovation. We will now elaborate on the social constructs that may have an effect on innovation, through the interplay of information elaboration, and in-depth information processing.
Psychological safety and trust
With the introduction of TCCs new members will have to start a new relationship with other team members. As is often the case, team members will become familiar with each other and build trust over the lifetime of this relationship; starting from little, and gaining more as they build experiences together. Thus, increasing TCCs will likely decrease the extent to which team members trust each other, as they are less familiar with each other. A lack of trust can impede team members sharing information, managing conflicts between members, and negotiating effectively (Mcevily et al., 2003). Thus, with the introduction of team
composition changes, team-members will not use their informational resources, because they may not trust their team members enough. This lack of information use may impede
innovation capabilities.
Related to the concept of trust is the concept of psychological safety. Psychological safety is “the shared belief held by members of a team that the team is safe for interpersonal risk raking” (A. Edmondson, 1999). Hofmann and Stetzer (1996) found that psychological safety leads to increased communication. This view is supported by research by Edmondson, Bohmer and Pisano (2001), and Nijstad, Berger-Selman and De Dreu (2012), who have shown that higher levels of psychological safety is correlated with innovation.
Huckman et al. (2009) found that coordination and a “willingness to engage in a relationship” increased with more familiarity with the team. Both of these are related to more information communication and elaboration. This means that with more TCCs there will be less coordination, less engagement in relationships, and thus less communication of ideas. Likelwise, in a computational study Singh, Dong and Gero (2012) found that a minimum threshold of team familiarity is needed for team members to learn from each other, and thus
increase information elaboration. This threshold suggests that introducing TCCs will harm information elaboration, and thereby hamper innovation capabilities.
Minority dissent
Team composition changes can in some cases lead to minorities. In a team of four, changing one member may lead to a minority. Changing all members however, will not likely
immediately lead to minorities, since all members have the same amount of interpersonal experience. De Dreu & West (2001) found that more innovations are introduced when there is a high level of minority dissent, as compared to a low level of minority dissent. This was only the case however if the minorities in the group felt free to participate. The reason behind the increase in innovation, according to the researchers, is that minority dissent improves creativity and divergent thought, ultimately resulting in innovation. Similarly, Nemeth and Chiles (1988) argue that minority dissent increases the chance that team members resist to conformity pressures, which may lead to premature decision making, decreasing innovation capabilities.
Likewise, Nijstad, Berger-Selman and De Dreu (2012) found in their research that minority dissent was positively related to the number of innovations implemented by TMTs. One criterion however, for minority dissent to influence innovation is that they experience participative safety. Also, Smith (1970) found that membership in a new group creates intellectual tension, and conflict of ideas, creating a ‘condition of novelty’. The author proposes that the groups that have been working together for a longer time could benefit if they could retain this intellectual tension. Increasing TCCs to some extent, where minorities are possible, will thus increase the information elaboration, and thereby help the
innovativeness of the team.
Intellectual and personal conflict
Related, but not equal to minority dissent is intellectual and personal conflict. Whereas minority dissent is some form of conflict involving minorities, intellectual and personal conflict can arise regardless of whether there are minorities in the team. With the introduction of TCCs team members have to get to know each other, and stability within the team
decreases, creating possibilities for conflict, both task-related or intellectual and personal. Intellectual, or task conflict is conflict that is centred on the task at hand. Task conflict may have positive effects on innovation; it may facilitate superior decision making by preventing premature consensus (Jehn, 1995), and it may stimulate team members to think more critically
(de Wit, Greer, & Jehn, 2012). At the same time, task conflict may also be detrimental to innovation capabilities because cognitive resources are required for the conflict, that cannot be directed towards creating innovative ideas (De Dreu, 2008)
Whereas task conflict has been shown to have possible positive effects on innovation, research has shown that personal conflict rarely leads to innovation (Jehn, 1995). Farh, Lee and Farh (2010) have shown that relationship conflicts harm group creativity. Personal conflicts heighten member anxiety, and because of the personal nature of the conflict they represent ego threats (de Wit et al., 2012). At the same time, personal conflict also has the potential to decrease trust among team members, reducing communication.
Several meta-studies have been done on conflict in general: Jehn, Rispen and Thatcher (2010) found that task conflict and relationship were respectively marginally positively, and negatively correlated with innovation. De Dreu (2006) found that task conflict is positively related with innovation when the level of task conflict was moderate, and negatively related when it was low or high, showing an inverse U-shaped relationship. This was not the case for relationship conflict, which has been shown to lead to negative outcomes with regards to innovation.
A similar finding was done by Farh (2010), who has shown that the relationship between task conflict and team creativity is curvilinear in such a way that moderate levels of task conflict lead to the best creative outcomes. There thus doesn’t seem to be a
straightforward relation between task conflict and innovation, as it sometimes leads to increased information elaboration, but also captures the cognitive resources of the team members.
Social categorization processes
Social categorization processes are the processes in which a group of people create subgroups. These subgroups can then exhibit in-group-out-group behaviour. Moreland (1985) has shown that these affective, cognitive and behavioural biases towards the outgroup are largest in the beginning, or at the creation of the group, and tend to decrease as the distinction between the new and the old members become less visible. Thus, increasing TCCs generally increases social categorization processes.
Kane, Argote, Levine and Stern (2004) found that long-time members of groups will adjust to, and accept more of ingroup members, than outgroup members. These biases towards the outgroup, in this case the new member, or members, can result in less
communication, and cooperation with this team member, impairing innovation within the group (Levine et al., 2001).
Choi and Levine (2004) found that newcomers are more influential when teams had been assigned rather than chosen, and that the newcomers that had been chosen were very susceptible to the influences of the people already in the team. When people can choose their own coworkers they tend to choose people with whom they already have a working
relationship with (Hinds, Carley, Krackhardt, & Wholey, 2000). Similarity, affect,
competence and familiarity increase the probability that a member of a team wants to work with another team member. This potentially increases team similarity and familiarity, and decreases social categorization processes.
Modelling the relationship between team composition changes and innovation
How do informational and social factors, induced by team composition changes affect the innovative capabilities of a team? As suggested earlier, social factors interact with
informational factors and can disrupt them. The model we propose shows that there are a lot of factors that determine whether TCCs lead to innovation. We find that there are three major ways in which innovation capabilities increase with TCCs; firstly, the team increases its access to new information. This factor does not necessarily lead to increased innovation, but creates the potential to do so. Secondly, with TCCs certain social factors influence the extent to which team members elaborate on information, and engage in in-depth thinking. This increases the amount and quality of ideas, and improves decision making capabilities. Lastly, with the introduction of TCC, the team has the potential to reflect on, and change work practices to better innovation capabilities.
The model is shown in figure 1. Relationships between constructs are depicted by plus or minus signs on the lines. Relationships between TCCs and the constructs are depicted by plus or minus signs in the construct-box. For example: increasing TCCs has a detrimental effect on psychological safety and trust, and since psychological safety and trust are positively related to the elaboration of information, increasing TCCs results in less information that is elaborated upon, finally resulting in less innovation. Increasing TCCs increases task conflict. This may, or may not increase information elaboration. Thus, the effect of task conflict on innovation is unclear, and may be non-linear.
There are many contradictory effects of TCCs on innovation. Not only are there constructs that are influenced by TCCs, like task conflict, that have an unclear, possibly non-linear, effect on information elaboration, but there are multiple constructs that simultaneously influence the innovation-capabilities of the team. This finding can be reconciled by accepting the Too-Much-Of-A-Good-Thing effect, as proposed by Pierce and Aguinis (2013). When there is too much of a good thing, like intellectual challenge, or psychological safety, this harms the outcome. This effect can clearly be seen in the task conflict factor; as task conflict increases, team members will first be more likely to reject a first, inferior idea to look for a better one. But when task conflict goes beyond an inflection point it becomes detrimental to innovation, because the team members spend too much cognitive effort on the conflict. As team composition changes increase, and team tenure decreases, innovation capabilities will first rise sharply, then flatten, and eventually decrease.
Several research findings suggest an inverted U-shaped relationship between TCCs and innovation: Zheng and Yang (2015) found that in interorganizational routines, more and
longer cooperation leads to an inverted U-shaped relationship with innovation; with few collaborations breakthrough innovations increase as collaborations increase, but after too many collaborations the introduction of innovations decline. The authors argue that this is because inertia and rigidity forming in the relationship between firms. This finding could be extrapolated to intrafirm relations.
Also Uzzi (1997) found an inverse U-curved relationship between similar constructs; embeddedness and complex adaptation. The author argues that economies of time, and efficiencies can be achieved as embeddedness increases, but that after a certain threshold this embeddedness can derail the performance of the firm by insulating the firm from information beyond the network of the firm. Hereby complex adaptations, which are related to
innovations, decrease.
Other research that supports the U-curved relationship hypothesis is from Berman, Down and Hill (2002). They found that as shared team experience increases, initially tacit group knowledge increases, improving the innovativeness and performance of the team. This relationship however flattens and even becomes negative as teams have too much shared experience. Equally, both Farh et al. (2010) and De Dreu (2006) found curvilinear relationships between task conflict and creativity and innovation.
Another, complementary, way of explaining the contradictory findings is by viewing the innovation capabilities as a logical effect of many factors; when one or a few is missing, innovation-capabilities suffer. For example, with no TCCs psychological safety is highest, suggesting higher information elaboration. At the same time, however, no TCCs reduce task conflict, and minority dissent, leading to less information elaboration. This theoretical view of the effect of TCCs on innovation can be compared to a bottle-neck; there is always a factor limiting innovation capabilities. Increasing other factors will henceforth have no effect on the innovation capabilities. Increasing TCC might for example improve task conflict, but it decreases psychological safety at the same. If psychological safety is at that point the bottle-neck restricting innovation, then increasing TCCs will not result in innovation capabilities increasing, but decreasing.
This relationship is illustrated in figure 2. Figure 2 only illustrates two exemplary factors of the model: minority dissent and trust. The dotted line here represents innovation capabilities, as a function of the two factors. As there are more team composition changes, trust and comfort will decrease, and minority dissent will increase, but for innovation to
occur both are needed; with no TCCs there will be no innovation because of a lack of
intellectual challenge, and with too many TCCs there will be no innovation because of a lack of social comfort. The result is that only moderately changed teams innovate.
Adopting this view results in the same inverse U-shape as suggested by Pierce and Aguinis (2013), because initially increasing TCCs (from no TCCs to a moderate amount) leads to an improvement of the bottle neck that is limiting innovation capabilities. Even if it decreases other factors which were already large, innovativeness will increase. Increasing TCCs even further will have detrimental effects on innovation, because this will make the bottle neck of factors that were already small, smaller. The optimal amount of TCCs is thus when the smallest bottleneck is biggest; when TCC are moderate. Our hypothesis is thus the following:
Hypothesis: There is an inverted U-shaped relation between team composition changes and innovation.
Methods
operationalize these constructs as follows: TCCs are operationalized by two measures: team familiarity and team intactness. Innovation is operationalized by measuring the relative count of words that indicate innovativeness. These measures are further described after laying out the setting and the data.
Setting and data
To research the relationship between team composition changes and innovation we analyzed the innovativeness of video games. The setting of this research is the thus video game
industry. This industry generated a revenue of 92 billion dollar in 2015 (Takashi, 2016). In the video game industry, there is an abundance of communication about the quality of its products in the form of reviews. In this research we focus on console video games (Playstation, Xbox, Wii), since this is the most prevalent way of playing video games (Essential facts about the computer and video games industry, 2014), and it provides the most useful data on team composition changes and innovation.
The data for the research has been obtained in earlier, yet unpublished research by Situmeang (2017). This data contains information on management teams, their positions, tenures, and who everyone has worked with. The data also contains consumer and expert reviews of video games, release dates, and information about the sales performance. In his research, data was obtained by searching in the databases of Metacritic, Mobygames, and VGChartz, using web crawling software. Data on TCCs has been obtained from
mobygames.com, which provides information on video games from 145 video game
platforms. Data on innovation has been obtained from Metacritic.com, which shows expert and consumer reviews from 5,689 titles. The information has been integrated using a fuzzy matching technique.
The total time-period of the data collected on video games by the research from Situmeang (2017) ranges is from 1984 to 2014. Text review data is available from 2006 to 2014. TCC data is available from 1984 to 2012. Data from which both innovation and TCC data are available ranges from 2006 to 2013.
Sample
The final datasets from Situmeang (2017) consisted of data from reviews and teams. Review data has been collected on 2,451 video games for which a total of 41,874 customer reviews, and 37,436 expert reviews have been written. Team data has been collected on 2147 video games. After filtering out data from which there is either no TCC or innovation data available, the data provides both innovation and team information on 392 video games.
On the video-games for which there is both innovation and team data the average management teams consisted of seven members, ranging from one to fifty-three (SD = 8.4). These management teams are generally divided in three functional departments: game designers, responsible for creative development, game programmers, responsible for the technical aspects of the game, and general project managers. The total number of managers is 3,341: 1,051 designers, 1,604 programmers, and 686 general managers.
Video game reviews have either been written by an expert or a consumer. On average, an expert review consists of 36 words, ranging from 1 to 167 (SD = 19.1). The average review count per game is 15.2. In the final dataset, where both innovation and team data is available, the average amount of words written in expert reviews per video game is 1098, ranging from 9 to 4,333. A consumer review, in general, consists of 135 words, ranging from 1 to 3,633 words (SD = 150.8). On average, for one game 18 customer reviews have been written. Each video game had on average 3,789 words written about in reviews.
Measures
Independent variables: Management Team Intactness and Familiarity
Team composition change is a complex concept and there is little agreement on how to measure this. For example: changing every member of a team does not necessarily mean that the team members are entirely unfamiliar, as perhaps certain team members have worked together before on previous projects. In line with Situmeang (2017) this research measures team composition changes by individual familiarity, and management team intactness. This is in line with previous research studying the interpersonal experience of temporary teams (Huckman et al., 2009; Reagans, Argote, & Brooks, 2005; Situmeang, 2017)
The first variable, team intactness, is a measure on the difference in team composition between previous projects; has the management team remained unchanged from the earlier project? This measure thus expresses whether, and to what extent, the team has worked consecutively in projects. When the team has worked together as one unit in the past, but with interruptions (different projects in the meantime) the score on team intactness will be low. The variable is calculated using a fuzzy matching algorithm obtained from Situmeang (2017).
The latter variable, team familiarity, is a measure of the extent to which team members have experience working with each other in the past. This variable is calculated by taking the average of the total number of times each individual member of the team has worked together in the past with another individual member of the team. Team members may thus have
worked together in the past, but not all together in a team; TCCs will thus be lower for a team like this, than for a team composed of members that have never worked together in the past, as pairs or in teams in general.
Dependent variable: Innovative word count in reviews
West & Farr (1990) defined innovation as “the introduction or application within a team of ideas, processes, products, or procedures that are new to that team and that are designed to be useful.” Innovative products then, are products that introduce new and improved ways of doing things. Even though the authors describe innovation as products that are new to the team, in this research innovation is operationalized by measuring to what extent customers and experts view the products as innovative. This is done for two reasons. First,
self-judgement of innovation could be positively biased. Second, there is no readily available data on the extent to which a team judges its products to be innovative.
Newness and usefulness are thus best judged by the customer. These judgements from the customer can readily be found on the internet; for every game that gets released reviews are written by final consumers (gamers), and by journalists (experts). The latter does this for others to decide on whether to buy a certain game or not. Since these reviews are an unbiased opinion of the consumer, they can be used to measure innovativeness of a product. This is done by counting the amount of words that show innovativeness in reviews written about the products that are made by the teams, and averaging them by dividing by the total amount of words.
The construct “innovativeness” is validated by using Computer-Aided Text Analysis (CATA), as advocated by Short, Broberg, Cogliser & Brigham (2010). Words that indicate innovativeness are collected in two steps: first a list of words is collected from the paper by Short et al.(2010). These include words like: “inventive”, “imaginative”, and “change”. We tried to find additional innovative words from the reviews, by sampling 100 of these reviews, and judging whether there are words that are not in the list of Short et al. (2010). No
additional words could be found. The words that indicate innovativeness are summed up in appendix 1.
The score of innovativeness is constructed using a software program called
shown that construct validity, and reliability is high, and scores on sentiment deduced by the program has shown high correlations with sentiment scores as judged by humans. In this research using SentiStrength involves replacing the sentiment words, that are originally included in the program, by the words from the list of Short et al (2010).
The program is set to count the occurrences of innovative words in the review, giving each a score. The program handles negation in front of an innovative word (e.g. not
innovative) by subtracting one point from the innovation score, making it possible for a game to receive a negative score. However, no words are included in the vocabulary of the program that may indicate the opposite of innovation. These may be words like “unimaginative”. Negative scores thus are only possible by negation.
The final score for each video-game indicating innovativeness is given by dividing the total amount of innovative words by the total amount of words written about a game. This ensures that games for which a lot of words have been written, do not artificially receive a higher score.
Control variables
To minimize bias in the analysis we include control variables. Ratings from users and experts are used as a control variable; users and experts may use words that indicate innovation, when they like a game, thus there might be a relationship between the amount of words used in reviews that indicate innovativeness, and the review-score. We also include the number of sales of the game; highly innovative games might be more or less likely to get sold a lot. Similar to other studies (cf. Situmeang, 2017) we include as a control variable the genre of the game; action and action-adventure game are coded as 1, other genres as 0. It might be the case that action and action adventure games are rated as more or less innovative. Lastly, we include the variable sequel, indicating whether or not the game is a follow-up to a different game; the sequel may be seen as less innovative, as it has to adopt things from the game that came before it, and cannot be entirely new.
Statistical approach
Hierarchical regression is used to test whether the relationship between TCCs and innovation is non-linear, and inverted U-shaped. All variables are standardized, except for dummy variables. We first added the control variables before adding the main independent variables familiarity and intactness. In a third step, we add the squared terms of familiarity and
intactness, to test whether the relationship between the independent and dependent variables is non-linear, and inverted U-shaped.
Results
Table 2 presents the mean, standard deviations and correlations on the control, dependent and independent variables. Management team intactness is not significantly correlated with expert and user innovation scores. Management team familiarity is also not significantly correlated with expert and user innovation scores. We do find significant correlations between expert, and consumer innovation scores, and between team familiarity and team intactness.
Table 2: Means, standard deviations, correlations
Variable Mean S.D. 1 2 3 4 5 6 7 8 1 Expert innovation score .578 .577 2 Consumer innovation score .400 .545 .277** 3 Sales 2.012 3.242 .056 .008 4 Sequel .744 .437 .053 .036 .120* 5 Genre .224 .418 .001 -.078 .022 -.050 6 Expert score 6.077 2.933 .279** .122* .249** .015 -.065 7 Consumer score 6.813 1.972 .042 .035 .050 .059 .119* .362** 8 Team familiarity .539 1.335 -.095 .037 -.003 -.144** .059 -.056 .049 9 Team intactness .44 .484 .081 .020 -.091 .047 -.118* .112* -.034 -.301** N = 406. * p < 0.05. ** p < 0.01
We hypothesized an inverted U-shaped relationship between TMT intactness and familiarity, and the innovativeness of the product, as judged by customers and experts. As can be seen in table 3 to 4, partly contrary to our predictions, only team familiarity, and not team intactness, was non-linearly and significantly related to innovativeness as judged by experts. The other hypothesized non-linear relationships between team intactness and innovation were non-significant, both for innovation as judged by experts and consumers.
Table 3: Results of hierarchical regression for expert innovativeness score
Variable Step 1 Step 2 Step 3
Step 1: control variables
Expert score .315 *** .305 *** .301 ***
Consumer score -.078 -.071 .068
Sequel .058 .047 .033
Sales -.027 -.021 -.014
Genre .034 .039 .040
Step 2: Main effects
Team familiarity -.062 -.260 *
Team intactness .026 .626
Step 3: Squared term
Team familiarity2 .196 *
Team intactness2 -.631
R2 .086 .091 .106
Δ R2 .000 .005 .015
F 6.768 5.133 4.678
Table 4: Results of hierarchical regression for customer innovativeness score
Variable Step 1 Step 2 Step 3
Step 1: control variables
Expert score .124 * .128 * .127 *
Consumer score -.002 -.007 -.007
Sequel .034 .042 .044
Sales -.026 -.026 -.027
Genre -.067 -.068 .-.067
Step 2: Main effects
Team familiarity .057 .076
Team intactness .010 -.219
Step 3: Squared term
Team familiarity2 -.009
Team intactness2 .235
R2 .021 .024 .025
Δ R2 .000 .003 .001
F 1.565 1.269 1.013
N = 406. Coefficients are Beta coefficients. * p < 0.05. ** p < 0.01. *** p < 0.001
Discussion
This paper studied the relationship between team composition changes in top management teams and innovativeness. A theoretical framework has been proposed that predicts that, through the interaction of social and informational factors each influenced by TCCs, TCCs and innovation are non-linearly related. This means that with either a few, or a lot of team composition changes innovation capabilities suffer, and that only with a moderate amount of member changes in the team we find a positive effect on innovativeness. We hypothesized an inverted U-shaped relationship between team composition changes and innovativeness. In an empirical study of over 400 video games we could only find support for a non-linear
relationship between team familiarity and innovation as judged by experts. There are several possible explanations for these findings, which we will discuss here.
The first explanations for the findings concern the measurement of the constructs innovation, and team composition changes. This research thus included a few limitations, which should be addressed in future research. Innovation in this research has been
conceptualized as the relative number of words that indicate innovativeness in reviews of video games. The innovation score that has been determined for both the expert and user reviews is based on a Computer Assisted Text Analysis (CATA) using the computer software program SentiStrength. Using CATA poses a few limitations: firstly, even though research has shown that for the purpose of sentiment analysis accuracies are high (Thelwall et al., 2010), we could not provide a score for accuracy in this research where different words are used to determine a score of innovativeness, instead of sentiment. This limits our capability to say that the innovation-scores provided are indeed indications of innovativeness of the video game.
The SentiStrength analysis software has been originally built for sentiment analysis of short texts like messages on Twitter; the software can detect negation (“this game is not innovative”), but is rather crude in the detection of the intentions of the writer of the review; the list of words from Short et al (2010) contain words like “clever”, which may be used for other purposes than indicating innovativeness. Words that could indicate the opposite of innovativeness, like “old-fashioned”, or “uncreative” are not included. Future research could thus focus on the improvement of the innovation-analysis of the reviews, by improving the innovation-word-list, and doing tests on the accuracy of the CATA-findings. Future research could also combine the results of CATA, with text analysis by humans. Other options for future research, to improve the accuracy of the construct innovativeness, are to create a panel of experts who independently judge the innovativeness of the video game.
Another point of debate is the measurement of the construct team composition
changes; we constructed this by means of two measures: team familiarity and team intactness. Due to the intricacies of teamwork and human relationships it is often hard to tell how
familiar team members are with each other; they may have worked together closely, or they may have been in the same team, but working mostly on separate terms. One possibility for future research would be to include an additional measure of team familiarity: a
self-judgement on how much the team has changed. A highly-respected manager leaving the team increases team composition changes substantially, in contrast with a minor team change of a contested member. This would be reflected in the added self-judgement measure.
Other explanations for the results that were found concern the proposed model. There are several arguments for why no non-linear effect would be found. For example, with a
moderate amount of TCCs you could get less innovation, because of subgroup formation. Thus, merging two groups who have worked together in the past, but not with the other group, may not result in innovation because there is a higher chance these two groups will stick together and not share information with each other. This effect has been found by Gibson and Vermeulen (2003). Per this argument, the solution is to either not change the group, or
completely overhaul the group. No presence of subgroups gives the group some degree of psychological safety.
Another possibility is that having partial membership change, for example changing two members in a group of four, would risk lead the old team members to impose their
routines on the new team members, not changing anything within the team, and not increasing innovation (Lewis et al., 2007). If this were the case, we would only see innovation in the case of complete team overhaul.
Implications for research
An important contribution of this research is that it proposes an integrated theoretical framework which integrates research from both social, cognitive and informational
viewpoints. In this thesis, we theorize that the social effects interact with the informational effects, such that they can improve or harm information elaboration. This theoretical
framework gives a possible explanation for the contradictory findings in previous research, by posing that the limiting factor is determining the total amount of innovation; improving other factors will thus not improve innovation capabilities. Viewing the relationship between team composition changes and innovation this way can explain why some findings suggest a positive relationship, and others a negative. In the case of a positive relationship we suggest that the limiting factor has improved. In the case of a negative relationship we suggest that the limiting factor has worsened. The effects of other constructs are then irrelevant.
Another contribution of this thesis is that the research has been done on fluidly
composed teams. Research by Situmeang (2017) already noted that most of the TMT research focuses on stable, ongoing teams. For these teams, team tenure or team longevity is an
appropriate measure to study team composition changes. However, as the research from Situmeang (2017) has shown, fluid teams in dynamic settings can offer benefits. This research contributes to this by looking at the effects of working experience of team members in fluid management teams on the innovation of the products they make.
Also, the data from research by Situmeang (2017) includes all games from a large timeframe. This study thus enjoys the benefits of using large amounts of data, which can offer
advantages for both research and practice (van Knippenberg, Dahlander, Haas, & George, 2015). The longer time period of the data allows us to see the effects of very few or very many TCCs; with shorter time periods, there will be no opportunity for research to study teams that stay together for a very long or short time.
Future research could focus on the testing of the proposed model, where social factors interrupt, or improve the elaboration of information in the team in such a way that the limiting factor dictates the amount of innovation. It would be useful to incorporate measures of the social factors, besides the measures of team composition changes, firstly to see whether the TCCs influence the social factors, and secondly whether the social factors influence
innovation in that setting.
Implications for practice
This thesis has a few implications for practice. Innovation is for a lot of companies a priority. Knowing when to change the team composition to maximize innovation may be useful to know. Little results in this study could be found suggesting that changing the team has effects on innovativeness. This means that changing team members for the sole purpose of improving the innovativeness the product they are making is futile.
There is an ongoing tension between forces for creativity, spurring innovation and those for rational interests, generally driving profitability (Tschang, 2007). Lampel et al. (2000) note that the consumers want, and need, familiarity to grasp the game, but at the same time they will need novelty and innovation to enjoy it; they don’t want to play the same game over and over again. Innovation however is not the same as success. Teams have to balance exploration, involved in the innovation of products, versus exploitation, using known concepts that work. We can assume that more innovative products increase variability in returns, and thereby risk, but that it is necessary to survive in the creative industry (Taylor & Greve, 2006).
The game True Story and the applications described in the introduction serve as an example where the entire team has been changed, and where innovativeness was high. These however only serve as examples. Besides, it is hard to tell what is the optimal level of team composition changes is, especially with many circumstantial factors. Thus, when an optimal level of exploration is determined, it is still very much dependent on circumstances what level of TCC is required.
Conclusion
This study researched the effect of team composition changes on the innovativeness of teams in the video game industry. The study reveals that there are only marginal effects to be found; team familiarity, as measured by the extent to which team members have experience working with each other, is only related to innovativeness as judged by experts. This relationship is non-linear. The other constructs, team intactness and innovativeness as judged by consumers, are not related. This suggests that, contrary to the hypothesis, that there is no relationship between team composition changes and innovation.
Bibliography
Amabile, T. M. (1983). The social psychology of creativity: A componential
conceptualization. Journal of Personality and Social Psychology, 45(2), 357–376. https://doi.org/10.1037/0022-3514.45.2.357
Amabile, T. M., Conti, R., Coon, H., Lazenby, J., & Herron, M. (1996). Assessing the Work Environment for Creativity Assessing the Work Environment for Creativity. The Academy of Management Journal, 39(5), 1154–1184. https://doi.org/10.2307/256995 Anderson, N., Potočnik, K., & Zhou, J. (2014). Innovation and Creativity in Organizations: A
State-of-the-Science Review, Prospective Commentary, and Guiding Framework. Journal of Management, 40(5), 1297–1333. https://doi.org/10.1177/0149206314527128 Berman, S. L., Down, J., & Hill, C. (2002). Tacit knowledge as a source of competitive
advantage. Academy of Management Journal. https://doi.org/10.2307/3069282 Brown, S. L., & Eisenhardt, K. M. (1995). Product development: past research , present
findings, and future directions. Academy of Management Review, 20(2), 343–378. Retrieved from http://www.jstor.org/stable/pdf/258850.pdf
Carpenter, M. A., Geletkancz, M. A., & Sanders, W. G. (2004). Upper echelons research revisited: Antecedents, elements, and consequences of top management team composition. Journal of Management. https://doi.org/10.1016/j.jm.2004.06.001 Choi, H. S., & Levine, J. M. (2004). Minority influence in work teams: The impact of
newcomers. Journal of Experimental Social Psychology, 40(2), 273–280. https://doi.org/10.1016/S0022-1031(03)00101-X
Choi, H. S., & Thompson, L. (2005). Old wine in a new bottle: Impact of membership change on group creativity. Organizational Behavior and Human Decision Processes, 98(2), 121–132. https://doi.org/10.1016/j.obhdp.2005.06.003
De Dreu, C. K. W. (2006). When Too Little or Too Much Hurts: Evidence for a Curvilinear Relationship Between Task Conflict and Innovation in Teams. Journal of Management, 32(1), 83–107. https://doi.org/10.1177/0149206305277795
De Dreu, C. K. W. (2008). The virtue and vice of workplace conflict: food for (pessimistic) thought. Journal of Organizational Behavior, 29(1), 5–18.
https://doi.org/10.1002/job.474
De Dreu, C. K. W., & West, M. A. (2001). Minority dissen and Team Innovation: The Importance of Participation in Decision Making. Journal of Applied Psychology, 86(6), 1191–1201.
de Wit, F. R. C., Greer, L. L., & Jehn, K. A. (2012). The paradox of intragroup conflict: A meta-analysis. Journal of Applied Psychology, 97(2), 360–390.
https://doi.org/10.1037/a0024844
Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383. https://doi.org/10.2307/2666999 Edmondson, A. C., Bohmer, R. M., & Pisano, G. P. (2001). Disrupted Routines : Team
Learning and New Technology Implementation in Hospitals. Administrative Science Quarterly, 46(4), 685–716. Retrieved from
http://journals.sagepub.com/doi/pdf/10.2307/3094828
Espevik, R. (2011). Expert Teams: Do Shared Mental Models of Team Members make a Difference?, 1–36.
Essential facts about the computer and video games industry. (2014). Retrieved from http://www.theesa.com/wp-content/uploads/2014/10/ESA_EF_2014.pdf
Farh, J.-L., Lee, C., & Farh, C. I. C. (2010). Task conflict and team creativity: A question of how much and when. Journal of Applied Psychology, 95(6), 1173–1180.
https://doi.org/10.1037/a0020015
Gersick, C. J. G. (1991). Revolutionary change theories: a multilevel exploration of the punctuated equilibrium paradigm. Academy of Management Review, 16(1), 10–36. https://doi.org/10.5465/AMR.1991.4278988
Gibson, C., & Vermeulen, F. (2003). A Healthy Divide : Subgroups as a Stimulus for Team Learning Behavior. Administrative Science Quarterly, 48(2), 202–239.
https://doi.org/10.2307/3556657
Guilford. (1950). Creativity. American Psychologist, 3, 444–454.
Hinds, P. J., Carley, K. M., Krackhardt, D., & Wholey, D. (2000). Choosing Work Group Members: Balancing Similarity, Competence, and Familiarity. Organizational Behavior
and Human Decision Processes, 81(2), 226–251. https://doi.org/10.1006/obhd.1999.2875
Hofmann, D. A., & Stetzer, A. (1996). a Cross-Level Investigation of Factors Influencing Unsafe Behaviors and Accidents. Personnel Psychology, 49(2), 307–339.
https://doi.org/10.1111/j.1744-6570.1996.tb01802.x
Huckman, R. S., Staats, B. R., & Upton, D. M. (2009). Team Familiarity , Role Experience , and Performance : Evidence from Indian Software Services. Management Science, 55(1), 85–100. https://doi.org/10.1287/mnsc.1080.0921
Hülsheger, U. R., Anderson, N., & Salgado, J. F. (2009). Team-level predictors of innovation at work: a comprehensive meta-analysis spanning three decades of research. The Journal of Applied Psychology, 94(5), 1128–1145. https://doi.org/10.1037/a0015978
Humphrey, S. E., & Aime, F. (2014). Team Microdynamics: Toward an Organizing Approach to Teamwork. The Academy of Management Annals, 8(1), 443–503.
https://doi.org/0.1080/19416520.2014.904140
Ilgen, D. R., Hollenbeck, J. R., Johnson, M., & Jundt, D. (2005). Teams in organizations: from input-process-output models to IMOI models. Annual Review of Psychology, 56, 517–43. https://doi.org/10.1146/annurev.psych.56.091103.070250
Jehn, K. A. (1995). A Multimethod Examination of the Benefits and Detriments of Intragroup Conflict. Administrative Science Quarterly, 40(2), 256. https://doi.org/10.2307/2393638 Jehn, K. A., Rispens, S., & Thatcher, S. M. B. (2010). The Effects of Conflict Assymmetry on
Work Group and Individual Outcomes. Academy of Management Journal, 53(3), 596. https://doi.org/10.5465/amj.2010.51468978
Jin, L., & Sun, H. (2010). The effect of researchers’ interdisciplinary characteristics on team innovation performance: evidence from university R&D teams in China. The
International Journal of Human Resource Management, 21(13), 2488–2502. https://doi.org/10.1080/09585192.2010.516599
Kane, A. A., Argote, L., Levine, J. M., & Stern, L. N. (2004). Knowledge transfer between groups via personnel rotation: Effects of social identity and knowledge quality. https://doi.org/10.1016/j.obhdp.2004.09.002
Katila, R., & Gautam, A. (2002). Something Old , Something New : A Longitudinal Study of Search Behavior and New Product Introduction. The Academy of Management Journal, 45(6), 1183–1194. https://doi.org/10.2307/3069433
Katz, R. (1982). The effects of group longevity on project communication and performance. Administrative Science Quarterly, 27(1), 81–104. https://doi.org/10.2307/2392547
Kozlowski, S. W. J. J., & Ilgen, D. R. (2006). Enhancing the Effectiveness of Work Groups and Teams. Psychological Science in the Public Interest, 7(3), 77–124.
https://doi.org/10.1111/j.1529-1006.2006.00030.x
Kurtzberg, T. R., & Amabile, T. M. (2001). From Guilford to Creative Synergy: Opening the Black Box of Team-Level Creativity. Creativity Research Journal, 13(3–4), 285–294. https://doi.org/10.1207/S15326934CRJ1334_06
Lampel, J., Lant, T., & Shamsie, J. (2000). Balancing Act: Learning from Organizing Practices in Cultural Industries. Organization Science, 11(3), 263–269.
https://doi.org/10.1287/orsc.11.3.263.12503
Leifer, R. (2000). Radical innovation: How mature companies can outsmart upstarts. Boston: Harvard Business Press.
Leonard-Barton, D. (1993). Core capabilities and core rigidities: A paradox in managing new product development: Strategic Management Journal , 13 , 111–125 (Summer 1992). Long Range Planning. https://doi.org/10.1016/0024-6301(93)90313-5
Levesque, L. L., Wilson, J. M., & Wholey, D. R. (2001). Cognitive divergence and shared mental models in software development project teams. Journal of Organizational Behavior J. Organiz. Behav, 22, 135–144. https://doi.org/10.1002/job.87
Levine, J. M., Moreland, R. L., & Choi, H. S. (2001). Group Socialization and Newcomer Innovation. In Blackwell Handbook of Social Psychology: Group Processes (pp. 86– 106).
Lewis, K., Belliveau, M., Herndon, B., & Keller, J. (2007). Group cognition, membership change, and performance: Investigating the benefits and detriments of collective
knowledge. Organizational Behavior and Human Decision Processes, 103(2), 159–178. https://doi.org/10.1016/j.obhdp.2007.01.005
Mcevily, B., Perrone, V., & Zaheer, A. (2003). Trust as an Organizing Principle. Organization Science, 14(1), 91–103. https://doi.org/10.1287/orsc.14.1.91.12814
Mohammed, S., Ferzandi, L., & Hamilton, K. (2010). Metaphor No More: A 15-Year Review of the Team Mental Model Construct. Journal of Management, 36(4), 876–910.
https://doi.org/10.1177/0149206309356804
Moreland, R. L. (1985). Social categorization and the assimilation of “new” group members. Journal of Personality and Social Psychology, 48(5), 1173–1190.
https://doi.org/10.1037/0022-3514.48.5.1173
Nemeth, C., & Chiles, C. (1988). Modelling courage: the role of dissent in fostering independence. European Journal of Social Psychology, 18(3), 275–280.
