Shared Leadership and Diversity
The effects on knowledge sharing and team creativity in an IT environment.
Student:
Mart Kerkvliet
Student number:
10203885
Thesis supervisor:
Pernill van der Rijt
Word count:
8038
2 Introduction
In response to rapid change and fierce competition, creativity is an important factor to develop and implement innovation. Hence, most firms have pursued diverse strategies to promote individual and team creativity in the workplace (Lee et al., 2014). This call for creativity and innovation leads to the question of how. How can a business improve its creativity, specifically how can it improve the level of creativity within a team? Measuring at the team level is especially interesting because many organizations nowadays have adopted team- based work systems in order to increase their responsiveness and in order to better foster innovation (Lee at al. 2014).
The answer to this may lie in the degree of knowledge sharing within an organization. Knowledge is the foundation of a firm’s competitive advantage and, ultimately, the primary driver of a firm’s value (Bock et al., 2005). Knowledge sharing is thought of as a process through which employees exchange knowledge and jointly create new knowledge (Lin, 2007). It is not hard to imagine that the more knowledge that is being shared and exploited the higher the degree of creativity within a team. This because knowledge sharing creates opportunities to maximize organization ability to meet those needs and generates solutions and efficiencies that provide a business with a competitive advantage (Reid, 2003). Many scholars therefore stress the importance of knowledge sharing to enhance creativity (Lin and Lee, 2006). The question therefore is, how exactly can knowledge sharing be stimulated?
In this study it is expected that shared leadership and diversity may lead to increase levels of knowledge sharing. This expectation is based on an earlier study by Lee et al. (2014) who found significant relationship between both shared leadership and knowledge sharing, and diversity and knowledge sharing. When we think of leadership we tend to think about the traditional leadership styles, however since work is becoming increasingly team- based we need to ask ourselves whether the traditional models and approaches to leadership are still reliable (Pearce & Barkus, 2004). It is possible that more horizontal types of leadership could be more enticing for teams. One of these types is known as shared leadership. Shared leadership emerges when all members of a team are not afraid to influence and guide their team members so that they can maximize their output as a team. Pearce & Barkus (2004) identified three characteristics of knowledge work that are related to the need for shared leadership, these include interdependence, complexity and creativity. This indicates that a social network with a high density of shared leadership might lead to better knowledge sharing as a team tries to maximize its potential output which may in turn lead to amongst other things, a higher level of creativity.
3
In addition to shared leadership, diversity might also play a decisive role in the advancement of knowledge sharing and consequent improvement of team creativity.
This study will focus on the effects of shared leadership, diversity and knowledge sharing on team creativity in which the main question is: what are the effects of shared leadership, diversity and knowledge sharing on team creativity? The goal of this study is to replicate, for the most part, a previous study by Lee et al. (2014) into the effects of shared leadership, diversity and knowledge sharing on team creativity. However, in their study they focused solely on students and therefore mentioned the results might differ from theirs in an actual workplace. Therefore, in this study the focus will be on full time employees. In addition, in this study diversity will be split into the two dimensions of informational and demographic diversity, this because different effects are expected from both types of diversity. Managers might know that knowledge sharing and team creativity are important, it is however also important to know how knowledge sharing and team creativity can be improved. By breaking down diversity into its two dimensions it is possible to give a more accurate image of how these processes work thus providing them with better and sharper tools to better utilize the potential of knowledge sharing and team creativity.
In addition, it is always valuable to conduct similar surveys in different contexts. With this study parts of the study of Lee et al. (2014) will be re-examined however this time in a working environment, as opposed to a student environment. Specifically it will be attempted to replicate their findings with regards to the effects of shared leadership and diversity onto knowledge sharing and that of knowledge sharing onto team creativity. The results might yield insight into the workings of shared leadership, diversity, knowledge sharing, and team creativity in a working (IT) environment as well as adding to the already existing literature on shared leadership, diversity, knowledge sharing, and team creativity.
To examine whether shared leadership and diversity leads to higher degree of knowledge sharing and whether this in turn will lead to a higher degree of team creativity a survey will be held.
Theoretical framework
Shared leadership and knowledge sharing
It was Gibb (1954) who first came up with the notion of two forms of leadership, distributed and focused. Focused leadership will occur when leadership rests on the shoulders of a single individual. By contrast distributed leadership will occur when two or more individuals share
4
the role of leadership, along with its’ responsibilities and functions. These two forms are however not to be mistaken as rigid either- or categories, but as two endpoints on a continuum. In addition, leadership can be thought of in two ways. Either through the strength of influence, for example the quality of leadership or its effectiveness. Or the source of influence, meaning single versus multiple team members. In this study the focus will be on source of influence. Shared leadership is a phenomenon in which team members exert mutual influence on each other as they work toward team objectives. It involves dynamic, interactive influence processes among and between individuals in teams (Pearce and Conger, 2003). Therefore shared leadership offers something that traditional vertical leadership cannot offer. The practice of leadership on a team-level wherein behaviors are enacted by multiple individuals rather than one single individual. In addition, shared leadership isn’t merely about leading it is more than that, it focusses on leadership as a social process, a dynamic, multidirectional and collective activity (Fletcher and Kaufer, 2003).
Shared leadership involves multiple individuals collaborating in a group’s leadership toward the attainment of their common goals (Pearce & Conger, 2003).This vision of leadership contrasts with the traditional forms of leadership. These traditional types of leadership are based on a leader role which rests on a single individual. In contrast, shared leadership is regarded as an emergent team property resulting from the distribution of leadership influence across multiple team members (Carson et al., 2007). It therefore differs greatly from the earlier leadership paradigm known as “vertical leadership” as defined by Pearce and Sims (2002). Due to the nature of shared leadership it might especially be suited for knowledge work. Knowledge work is the sort of work which leans heavily on a persons’ knowledge or professional skills. This type of work is however becoming more and more dependent on the successful integration of the knowledge and skills of these professionals in order to create true innovation (Cox et al., 2003). Cox et al. (2003) reckon that it is more and more difficult for any one person to have all the required knowledge, skills and abilities necessary to be able to show true leadership in all aspects of knowledge work. It is for this very reason that teams are highly dependent on the ability to coordinate and integrate their ideas and abilities of all individuals with different backgrounds, experiences and approaches in order to successfully complete any project (Bligh, Pearce & Kohles, 2006). It is therefore reasoned that only when team members feel empowered to lead themselves and share influence with their fellow teammates in matters such as defining problems, making decisions and solving problems that true knowledge sharing is likely to result.
5
Davenport and Prusak (1998) defined knowledge as “a fluid mix of framed experience, values, contextual information, and expert insights that provides a framework for evaluating and incorporating new experiences and information”. According to Davenport and Prusak (1998) knowledge is personal. If an organization is seeking to effectively manage knowledge resources, they need to ensure that employees are willing to cooperate with their fellow colleagues in order to contribute knowledge to the firm. Knowledge sharing has two dimensions, that of knowledge donating and knowledge collecting. In the former, the emphasis is to make individual knowledge become group and in a later stage even organizational knowledge. This leads to improvements in the degree of knowledge available within a firm. Second is knowledge collecting, in which several processes and mechanisms are used for gathering information and knowledge from internal and external sources. In contrast to knowledge donating, the process of knowledge collecting means taking organizational knowledge and turning this into group and/or individual knowledge through the process of internalization and socialization of knowledge (Lin, 2007). In addition, when talking about ‘’sharing’’ this may be any type of sharing, face to face, over long distance with the help of electronic technology or over time with access to information which was archived by others (Armbrecht et al., 2001).
An organization has two options for promoting a knowledge sharing culture. First off, the organization can attempt to incorporate knowledge into its’ business strategy. In addition, it can focus on changing employee attitudes and behaviors to promote willing and consistent knowledge sharing (Connelly and Kelloway, 2003). One way of potentially altering employees’ attitudes and behaviors might be through the emergence of shared leadership.
Therefore the following hypothesis are formulated with regards to shared leadership and knowledge sharing:
H1: Higher levels of shared leadership will be significantly and positively related to higher levels of knowledge sharing within a team.
Diversity and knowledge sharing
In the previous section it was explained what knowledge sharing entails and how shared leadership is expected to be positively related to higher levels of knowledge sharing. It is however highly unlikely that shared leadership alone can boost levels of knowledge sharing. In addition, previous studies (Lee et al. 2014; van Knippenberg, de Dreu & Homan, 2004) have
6
already demonstrated the impact of diversity on knowledge sharing. Therefore this section will explain how diversity may also play a role in knowledge sharing.
Work-group diversity is a fact of organizational life. Groups in organizations have become more and more diverse in terms of their demographic composition over the years and will continue to become more diverse in the years to come (van Knippenberg, de Dreu & Homan, 2004). Diversity refers to the ways in which people are similar of different from each other. It may be defined by any characteristic that varies within a particular network unit such as gender, race, age, education or functional background (Bauer & Ergodan, 2005). Diversity itself can be divided into two dimensions, that of social category diversity which entails differences in easily detectable attributes such as sex, age and ethnicity and informational diversity which are those differences which are less visible like somebody’s function or educational background (Jackson, 1992). Diversity in the workplace can be regarded as a double edged sword. It has been related to both positive and negative outcomes in work groups (Hofhuis, van der Zee & Otten, 2012). By using the different perspectives, ideas and knowledge diverse groups have the potential to be more productive and more efficient. Moreover, it may lead to an enhanced level of creativity and knowledge.
In the social categorization perspective similarities and differences can be seen as the basis for categorizing yourself and others into subgroups. With these subgroups one can distinguish between one’s own in- group and one or more out- groups (van Knippenberg, de Dreu & Homan, 2004). This categorization of people in sub- groups stems forth from the fact that people tend to like and trust in- group members more than out- group members and thus generally tend to favor in- groups over out- groups (Tajfel & Turner, 1986). This means that categorization processes may lead to the formation of subgroups within work groups, creating an atmosphere of “us” versus “them” which consequently may result in problematic inter- subgroup relations.
The idea is based on the similarity/attraction paradigm (Byrne, 1971). Pfeffer (1983) wrote about how the distribution of demographic differences in both teams and organizations could affect both process and performance. Pfeffer (1983) based his idea on the fact that the demographic composition of a group could lead to differences in the way a group communicates, but also in the cohesion and integration of said group. Underlying this fact is the degree in which members of a team, organization or any form of group thought of themselves of being either similar or different from the rest of the group. Therefore, individuals who are more similar may share common life experiences and values and may find interacting
7
with one another more easy. It is for this reason that similarity/attraction deems heterogeneous groups to be less effective (Williams & O’Reilly, 1998). Empirical findings show dissimilarity often leads to losses in performance and distortions in processes, this includes outcomes such as less positive attitudes towards one another, less frequent communication between one another and most harmful for an organization, a higher likelihood of turnover from the group (Williams & O’Reilly, 1998).
The opposite is believed to be true for informational diversity, also known as functional or knowledge diversity. Where the social categorization has a strong focus on the relational aspect, the focus of informational diversity is far more concentrated on task- related aspects of the group processes (Van Knippenberg, De Dreu & Homan, 2004). Informational diversity refers to difference in knowledge bases and perspectives that members of a team have. Therefore Jehn et al. (1999) defined informational diversity as “differences in knowledge bases and perspectives that members bring to the group”. These differences in educational background, training, and work experience increase the likelihood that diverse perspectives and opinions exist in a workgroup (Stasser, 1982). As said before, informational diversity is more concentrated on the task- related aspects of a group process. Research by Jehn, Chadwick and Thatcher (1997) indeed confirms this. They concluded amongst other things that informational diversity led to an increase in task- related debates in work teams. With task- related debates the authors meant debates about either the content or the process of the task. Herein content refers to the what question and process refers to the how question. Therefore, since more debates arise between team members, one might assume that the amount of knowledge being shared in the process of debating is higher too.
Additionally, van Knippenberg, De Dreu & Homan (2004) contend that informational diversity has the ability to enhance group performance by stimulating the elaboration of task- relevant information and perspectives. In light of this Van Knippenberg, De Dreu & Homan (2004) defined a concept known as group information elaboration as the transfer of information, and perspectives, individual level processing of the information and perspectives, feeding back the results of this individual- level processing into the group, and discussion and integration of their implications. The authors argue that this deeper and more comprehensive processing of task- relevant information may lead to diverse teams performing better than more homogeneous teams.
Not only are teams able to use to more different perspectives within a team, individuals in diverse groups may also have greater access to informational access outside their own group.
8
The availability of this extra information may therefore enhance group performance even though social categorical diversity has negative impacts on group processes (Williams & O’Reilly, 1998).
Based on the above it is expected that informational diversity will have a positive influence on knowledge sharing. Conversely social categorical diversity is expected to have a reversed effect and hamper knowledge sharing. From this, the following two hypotheses were formulated:
H2a: Higher levels of social categorical diversity will be significantly and negatively related to knowledge sharing.
H2b: Higher levels of informational making diversity will be significantly and positively related to knowledge sharing.
Moderating effects of shared leadership on the relation between diversity and knowledge sharing
Though both shared leadership and diversity may have a direct effect on the amount of knowledge sharing within a team, it isn’t entirely unthinkable that shared leadership might also act as a moderator on the relationship between shared leadership and diversity. In the previous section it was explained how diversity can lead to the formation of subgroups which result in a person’s own in- and out- groups, this due to the process of social categorization. On the contrary, shared leadership was defined as ‘multiple individuals collaborating in a group’s leadership toward the attainment of their common goals’ (Pearce & Conger, 2003). Therefore, the assumption can be made that shared leadership can have a moderating effect on the relationships between social categorical and informational diversity and knowledge sharing. Simply, because when a team is high in shared leadership the effects of diversity should in the case of social categorization be mitigated and possibly even strengthened in the case of informational diversity.
With shared leadership any member of the group can influence what the group does, how it is done, and the way individuals in the group relate to each other (Yukl, 1994). Therefore the assumption can be made that the subgroups cease to exist in a team with high shared leadership due to the fact that each and every member has influence in the team processes.
9
Diversity- oriented shared leadership may be useful in the prevention of intergroup biases that stem forth from the process of social categorization and the consequent negative affective evaluations that team members experience from the feeling of dissimilarity among group members. It is argued by Van Knippenberg, de Dreu & Homan (2004) that intergroup biases are less likely to occur when team members are encouraged to see the value in difference and also try to protect the difference in identities that team members have. Therefore, shared leadership with an orientation towards diversity can reduce tension and hostility due to individual differences and foster mutual acceptance and respect among group members. Based on the arguments presented above it is expected that shared leadership will positively moderate the relationship between both social categorical and informational diversity on the degree of knowledge sharing. From this the following two hypotheses were formulated:
H3a: Shared leadership will have a positive moderating effect on the relationship between social categorical diversity and knowledge sharing.
H3b: Shared leadership will have a positive moderating effect on the relationship between informational diversity and knowledge sharing.
Knowledge sharing and team creativity
In the previous sections it was discussed how shared leadership and diversity might affect each other and also how they might hold a relationship with knowledge sharing. In this chapter it will be explained how knowledge sharing might lead to higher levels of team creativity. Team creativity is defined as the degree to which a project team’s processes are novel in the context of the project’s objectives. Team creativity is a process in which individual team members together build towards interrelating their ideas with the perspectives and unique skills of other individuals in the team (Drazin, Glynn & Kazanjian, 1999).
Knowledge sharing creates opportunities to maximize organization ability to meet those needs and generates solutions and efficiencies that provide a business with a competitive advantage (Reid, 2003). There is research supporting the notion that a firm’s ability to transform and exploit knowledge may determine its level of organizational innovation, such as faster problem- solving capability and enhanced rapid reaction to new information (Lin, 2007).
According to Armbrecht et al. (2001) knowledge sharing is the catalyst for creativity and subsequently for innovation. This because knowledge sharing provides the means by which
10
innovative ideas can be captured, shared or tested. In turn this balances the communal knowledge of for example a team and therefore leads to new and improved ideas. Also Nonaka and Takeuchi (1995) suggest that the process of creativity starts when team members meet to share knowledge in a given area, much of which is tacit. Tacit knowledge may include insights into customer needs, hunches about what might fix an intractable problem, lessons learned from previous experience, how other have approach similar problems and information about new technologies. Sharing such tacit knowledge creates a flow of novel ideas that contribute to successful outcomes, such as new products, processes and patents.
A study performed by Drazin, Glynn & Kazanjian (1999) also indicates that knowledge sharing leads to higher levels of team creativity. They explain how team creativity results from finding novel associations and linkages among the diverse ideas, perspectives, and domain expertise that individual team members hold. Now team members, especially those in diverse teams often bring different ideas, perspectives and expertise to a team. Therefore, when knowledge sharing within a team is high Drazin, Glynn & Kazanjian (1999) suggest that the team has access to a variety of alternatives, example solutions, and ideas which in turn have the ability to potentially lead to higher creativity. They claim that once individually held expertise is integrated at the team level, a mechanism is provided for enhancing team creativity. Due to the fact that team members are now able to access, explore and moreover use diverse information from related knowledge domains associated with the project. If the assumption were to be made that integration of expertise equals knowledge sharing then this most certainly means that knowledge sharing is expected to lead to higher levels of team creativity.
This is also predicted by the Categorization- Elaboration Model or CEM for short. The model states that the elaboration of task- relevant information and perspectives leads to better performance, which consist of both creativity and innovation but also decision quality (Van Knippenberg, De Dreu & Homan, 2004). Herein elaboration is defined as the exchange of 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. Van Knippenberg, De Dreu & Homan (2004) go on and explain how high- quality performance is dependent on members informing each other on the basis of their own expertise. These members must then process the different perspectives offered by their team members in order for them to understand the implications these perspectives have on their of area of expertise. Once the implications have been become clear they must feed these implications back to the team and consequently design the optimal product through integrating all the different viewpoints.
11
Based on these arguments it is expected that knowledge sharing will positively impact team creativity which led to the formulation of the following hypothesis:
H4: Higher levels of knowledge sharing will be significantly and positively related to team creativity
Mediating effects of knowledge sharing
Based on the earlier presented hypothesis, more hypothesis can be formulated. Once more, it is expected that shared leadership will have an influence on the degree of knowledge sharing. In addition, it is expected that knowledge sharing will have an influence on the degree of team creativity. Therefore the assumption can be made that shared leadership might also have a direct influence on the degree of team creativity, which in turn will be mediated by the knowledge sharing. Therefore the following hypothesis is formulated:
H5: The relationship between shared leadership and team creativity will be mediated by the degree of knowledge sharing.
Similar mediating effects may be expected for both social categorical and informational diversity. Therefore it is expected that both social categorical and informational diversity will have a relationship with team creativity, which is mediated by the degree of knowledge sharing. Therefore the final hypotheses in this study are:
H6A: The relationship between social categorical diversity and team creativity will be mediated by the degree of knowledge sharing.
H6B: The relationship between informational diversity and team creativity will be mediated by the degree of knowledge sharing.
In the sections above it has been explained how each of the variables might hold a
relationship with each other and from this the hypotheses were formulated. Based on these hypothesis the following conceptual model is constructed.
12
Figure 1: Conceptual model
Shared Leadership Knowledge Sharing Team Creativity
Informational/Functio nal Diversity Social Categorical Diversity H1 H2B H2A H3B H3A H4 H5 H6A H6B Methodology
Participants and procedure
The method for data collection in this study was by means of a survey. The survey was held amongst teams of an IT department of a large organization in the Netherlands. Each of the employees are part of a team within this IT department. The teams themselves were e-mailed separately using the respective teams’ e-mail addresses. They were asked whether they wanted to participate in a study with regards to shared leadership, diversity, knowledge sharing and team creativity. In total 160 employees were approached which the request to fill in the survey. In total 109 employees agreed to participate in the study, of which 100 finished the survey which gives a response rate of 62,5 percent.
Measurements
Diversity. Participants were asked to provide categorical answers about age, gender, demography with regards to social categorical diversity and questions about major (in what field they and their colleagues studied), role (e.g. developer or tester) and years of working experience within the organization with regards to informational diversity. Blau’s diversity index was used to calculate the varying diversity indexes. For example, when calculating the gender diversity in a team with five females and ten males, the following calculation was used: B = 1 - ((.33)2 + (.66)2). In order to measure the social categorical diversity, the indexes from age, gender and demography were added together and divided by three to give an mean index about social categorical diversity. The same was done for informational diversity with regards to major, role and work experience diversity.
13
Shared leadership. In this study a social network approach (Carson et al., 2007) was used to measure the density of shared leadership. The density is a measure of the total amount of leadership displayed by team members as perceived by others on the team. The equation used for calculating the density for shared leadership is shown below.
𝐷𝑒𝑛𝑠𝑖𝑡𝑦 = 𝑆
7𝑁(𝑁 − 1).
In this equation, S is the sum of all values that team members would rate each other for leadership. N equals the number of team members; N (N – 1) is the total number of possible ties in a team. The number 7 represents the maximum value rated by a peer in a team. The number that stems forth from the equation shows the density of shared leadership, in which a low number would indicate a low amount of shared leadership whereas a high number would represent a high amount of shared leadership
Knowledge sharing. In order to measure the amount of knowledge sharing, items were used from Lee et al. (2014). An example of one of the items is ‘people in this team are willing to share their knowledge/ideas with others’. In total this scale had three of such items (appendix B), which after the reliability analysis proofed to be very reliable, α = 0.83.
Team creativity. In order to measure the amount of team creativity, again items were used from Lee et al. (2014). An example of one of the items is ‘our work unit comes up with many new ideas about how work should be done’. In total this scale was composed of two items (Appendix B), which after the reliability analysis proofed to be very reliable too, α = 0.81.
Results
The analysis of the data will be reported in four different sections. Linear regression analyses were used to test hypotheses 1, 2A, 2B and 4. These are the hypotheses for testing the influence of shared leadership and both types of diversity on knowledge sharing and the hypotheses with regards to knowledge sharing and team creativity. To test hypotheses 3A and 3B a multiple regression analysis was used, in order to test the moderation effect as predicted by these hypotheses. Lastly, to test hypothesis 5, 6A, and 6B multiple regression analyses were used, in order to test the mediation effect as predicted by the hypothesis.
Effects of shared leadership on knowledge sharing
The expectation was that higher levels of shared leadership would lead to higher levels of knowledge sharing (hypotheses 1). The regression model with the amount of shared leadership as perceived by each individual in a team as an independent variable and the degree of
14
knowledge sharing within a team as the dependent variable is significant, F (1,85) = 9,557, p = .003. The regression model can therefore be used to predict the level of knowledge sharing within a team, however the strength of the prediction is moderate at best. Just over ten percent of the variation of knowledge sharing is predicted by the amount of shared leadership within a team (R² = 0.102). Shared leadership, b, = 1.405, t = 3.09, p = .003 has a significant and positive relationship with the degree of knowledge sharing within a team. Hypothesis 1 is therefore supported.
Effects of diversity on knowledge sharing
The expectation was that social categorical diversity would lead to lower levels of knowledge sharing (hypotheses 2B). The regression model with social categorical diversity as the independent variable and the degree of knowledge sharing as the independent variable is insignificant, F (1,84) = 0.644, p = 0.424. The regression model can therefore not be used to predict the level of knowledge sharing within a team. Social categorical diversity, b = .549, t = 0.803, p = 0.424 has an insignificant relationship with the degree of knowledge sharing within a team. Hypotheses 2A is therefore rejected.
Next it was expected that informational diversity would lead to higher levels of knowledge sharing (hypotheses 2B). The regression model with informational diversity as the independent variable and the degree of knowledge sharing as the dependent variable is also insignificant F, (1,71) = 1.040, p = 0.311. This means that this regression model cannot be used to predict the level of knowledge sharing within a team. Informational diversity, b = .522, t = 1.020, p = 0.121 has an insignificant relationship with the degree of knowledge sharing. Hypotheses 2B is therefore rejected.
Moderating effects of diversity on the relationship between shared leadership and knowledge sharing.
The expectation was that shared leadership would negatively affect the relationship between social categorical diversity and knowledge sharing (hypothesis H3A). The multiple regression model with shared leadership and social categorical diversity as the independent variables and knowledge sharing as the dependent variable is significant, F (3,77) = 5.11, p = 0.003. Therefore the model can be used to predict the level of knowledge sharing within a team under the influence of both shared leadership and social categorical diversity, however the strength of the prediction is moderate at best. Nearly seventeen percent of the variation of knowledge sharing is predicted by shared leadership and social categorical diversity (R² = 0.166). Shared
15
leadership, b = 0.25, t = 3.22, p = 0.002 has a significant and positive relationship with the degree of knowledge sharing within a team. Social categorical diversity, b = 0.06, t = .73, p = 0.46 does not have a significant relationship with the degree of knowledge sharing. However, the analysis did reveal a significant interaction effect between shared leadership and social categorical diversity, b = 0.15, t = 2, p = 0.05. Hypothesis 3A therefore supported.
The graph below displays the multiple regression of shared leadership on the relationship between social categorical diversity and knowledge sharing. Knowledge sharing is the highest when there is high social categorical diversity and high shared leadership. Here we can see that H3A is supported since knowledge sharing is slightly higher when there is high shared leadership and high social categorical diversity, but slightly lower when there is high social categorical diversity and low shared leadership. Therefore the effects of social categorical diversity are slightly moderated by the effects of shared leadership.
Figure 1: Multiple regression analysis of social categorical diversity on the relationship between shared leadership and knowledge sharing.
With hypothesis 3B the expectation was that shared leadership would positively affect the relationship between that informational diversity and knowledge sharing. The multiple regression model with shared leadership and informational diversity as the independent variables and knowledge sharing as the dependent variable is significant, F (3, 64) = 2.84, p = 0.04. The model can therefore be used to predict the degree of knowledge sharing within a team,
1 2 3 4 5 6 7
Low Informational Diversity High Informational Diversity
KN OW L ED GE S HA R ING Low Shared Leadership High Shared Leadership
16
however the strength of the prediction is fairly weak, with only twelve percent of the variation of knowledge sharing being predicted by shared leadership and informational diversity (R² = 0.118). Shared leadership, b = 0.21, t = 2.5, p = 0.01 has a significant and positive relationship with the degree of knowledge sharing. Informational diversity however, b = 0.053, t = 0.55, p = 0.59 is insignificant. Moreover, the analysis showed that the interaction effect between shared leadership and informational diversity too was insignificant, b = 0.078, t = 1.05, p = 0.3. Therefore hypothesis 3B was rejected.
Effects of knowledge sharing on team creativity
Another expectation in this study was that higher levels of knowledge sharing would lead to higher levels of team creativity (hypotheses 4). The regression model with knowledge sharing as the independent variable and team creativity as the dependent variable is significant F, (1,89) = 25.296, p <.001. Therefore the model can be used to predict the degree of team creativity. In addition, the strength of the prediction is moderate as well. Twenty two percent of the variation of team creativity is predicted by the degree of knowledge sharing (R² = 0,221). Knowledge sharing, b = 0.693, t – 5.030, p <.001 has a significant and positive relationship with the degree of team creativity. Hypotheses 4 is therefore supported.
Mediating effects of knowledge sharing on the relationship between shared leadership and team creativity
With hypothesis 5 it was expected that knowledge sharing would mediate the relationship between shared leadership and team creativity. It was first tested whether shared leadership held a significant relationship with team creativity. The regression model with shared leadership as an independent variable and team creativity as a dependent variable is significant F (1,85) = 5.84, p = 0.018. However, the strength of the prediction is fairly weak. Only 6,5 percent of the variation in team creativity is predicted by shared leadership (R² = 0,065). The next step was to see whether controlling for the degree of knowledge sharing would mediate this relationship. The regression model with shared leadership and the degree of knowledge sharing as the independent variables and team creativity as the dependent variable is significant, F (2,85) = 13,38, p <.001. Therefore the model can be used to predict the degree of team creativity. The strength of the prediction is moderate, with twenty five percent of the variation of team creativity being predicted by the degree of shared leadership and the degree of knowledge sharing. Knowledge sharing, b = .694, t = 4.61, p <.001 has a positive and significant relationship with the degree of team creativity. Shared leadership, b = .713, t = 1.08, p = 0.28, is insignificant when controlled for knowledge sharing. This indicates a full mediation as the
17
difference is such that shared leadership no longer has a significant effect, meaning that controlling for knowledge sharing removed all the effects from shared leadership. Furthermore, Sobel’s Z = 4.67, p <.001 which indicates that this full mediation is significant too. Therefore, hypothesis 5 is supported.
Mediating effects of knowledge sharing on the relationship between diversity and team creativity.
Next it was expected that the effects social categorical diversity on team creativity would be mediated by the degree of knowledge sharing (hypothesis 6A). It was first tested whether social categorical diversity held a significant relationship with team creativity. The regression model with social categorical diversity as the independent variable and team creativity as the dependent variable is insignificant F (1, 84) = 0.18, p = 0.68. Therefore the model cannot be used to predict the degree of team creativity. Furthermore, a mediation analysis cannot done when the effects are insignificant to begin with, as the chances of finding mediation effects are next to zero. Therefore, hypothesis 6A is rejected.
Lastly it was expected that knowledge sharing would mediate the relationship between informational diversity and team creativity (hypothesis 6B). It was first tested whether informational diversity held a significant relationship with team creativity. The regression model with informational diversity as the independent variable and team creativity as the dependent variable is insignificant F (1, 71) = 0.264, p = 0.61. Therefore the regression model cannot be used to predict the degree of team creativity. And once more, a mediation analysis cannot be performed when the effects are insignificant to begin with, since it is highly unlikely that a mediation effect will be found. Therefore, hypothesis 6B is rejected.
Conclusion & Discussion
The aim of this study was in part to replicate a previous study into shared leadership, diversity, knowledge sharing and team creativity by Lee et al. (2014). In addition, the aim was to expand on the aforementioned study by examining respondents within a professional (IT) setting. In this study a significant relationship was found between the degree of shared leadership and the degree of knowledge sharing within a team. Therefore this study was successful in replicating these results in a different context. The findings suggest that one way of boosting the amount of knowledge sharing is by encouraging teams to engage in shared leadership as opposed to only vertical leadership.
18
In contrast, no significant relationships were found for both social categorical as informational diversity and the degree of knowledge sharing. This certainly doesn’t mean that social categorical diversity won’t have any negative effects on the degree of knowledge nor does it mean that informational diversity won’t have a positive effect on the degree of knowledge sharing. There have been a variety of studies that did find these effects for both types of diversity (e.g. van Knippenberg, de Dreu and Homan, 2004). Not finding any significant results is in this instance probably caused by flawed measurements rather than flaws in the theory. Why this might be the case will be explained at a later point.
Additionally, it was attempted to find a moderating effect for both social categorical diversity as informational diversity on the relationship between shared leadership and knowledge sharing. The results showed a significant and positive moderating effect for shared leadership on the relationship between social categorical diversity and knowledge sharing. Meaning that when both shared leadership and social categorical diversity are high, that knowledge sharing would also be slightly higher than when shared leadership was low. A moderating effect wasn’t found for informational diversity. It was expected that informational diversity together with shared leadership would lead to even higher levels of knowledge sharing. A cause for not finding any significant effects might be that the degree of knowledge sharing was already extremely high (M = 6.06) which means that the degree of knowledge sharing could hardly be improved already and therefore the difference between low and high shared leadership was probably too small to be significant.
With hypothesis 4 it was again tried to replicate results from earlier studies (e.g. Lee et al. 2014). A significant relationship was found between the degree of knowledge sharing and the degree of team creativity. This indicates that when trying to boost levels of team creativity, it is also necessary to take into account the degree of knowledge sharing within a team. Therefore a first step in creating more creativity might be by encouraging team members to share what they know.
Lastly it was attempted to find significant mediation effects with regards to knowledge sharing. With hypothesis 5 it was expected that knowledge sharing would mediate the relationship between shared leadership and knowledge sharing. The analyses showed at first that shared leadership had a significant relationship with the degree of team creativity. However, when controlled for knowledge sharing this relationship completely disappeared and became insignificant. Therefore knowledge sharing fully mediated the relationship of shared leadership and moreover this mediation was significant too. This indicates, as said in the previous section,
19
that to when trying to boost levels of team creativity, it is important to focus on the degree of knowledge sharing when trying to enhance the degree of team creativity. By contrast, no such effects were found for both types of diversity. Mostly because the effects of diversity weren’t significant to begin with, and therefore it was not attempted to do a mediation analysis as chances of finding a mediation are virtually zero. Once again, not finding any significant results with regards to diversity may be found in the way it was measured.
With diversity the respondents noted that they found it difficult to think of the age of their team mates. This is reflected in the results, as the results showed very inconsistent data within teams (e.g. people from the same team rating people at different age groups). But the biggest problem seemed to lie in measuring the informational diversity. Not surprising since these sorts of diversity aren’t directly observable, which was obvious from the data. Major diversity (i.e. what each of team member studied) was almost impossible to make any sense of as it was missing a lot of data and if there was some data, this was highly inconsistent within teams. Since it was therefore almost impossible to construct any meaningful data from it, it was decided not to use major diversity for analysis. Informational diversity in this study therefore consisted of only role and experience diversity. But the problem didn’t restrain itself to just major diversity. Respondents remarked that they often had to guess as to what somebody had studied, and the same applied to the work experience within the organization. Therefore the informational diversity probably yielded very inconsistent data. Though asking for somebody’s perceived diversity had the benefit of being able to run team analyses even though only one person from that team had filled out the survey, this was completely downplayed by the hugely inconsistent data. Therefore future research might yield better results using strictly objective data (e.g. a person’s age, work experience, etc.). However, answers are needed from an entire team to make any analyses. However, using only objective data would rule out the perception people have about diversity. For instance, a team might be very diverse with regards to age, but a person might not perceive this diversity and therefore no effects might be found. Ideally, future research would thus measure diversity both subjectively as well as objectively. In doing so a comparison can be made as well between the diversity that somebody experiences and the diversity as it is exists within the team. But, if future research were to again use perceived measurements then they would have to make absolutely sure that the respondents will be able to fill out the questions properly, in order to avoid guessing as was done in this study.
In this study an attempt was made to replicate previous results with regards to shared leadership, diversity, knowledge sharing and creativity. In part this succeeded, by finding significant results
20
for shared leadership on knowledge sharing, and for knowledge sharing on team creativity. Next to replication, additions were made with findings with regards to the moderation analysis (H3A) and the mediation analysis (H5). The rest of the hypothesis however were rejected. In addition, no analysis were done at the team level, simply because the amount of teams in the survey was too low (n = 19), which simply wasn’t enough to make any proper analysis. Therefore, future research looking to find effects of shared leadership and diversity at the team level ought to make sure that they have a sufficient amount of teams in their sample.
Furthermore, the way shared leadership was measured could be improved upon as well. In this study the data that was collected reflects the overall quantity of leadership, but not the distribution (i.e. does everybody have an equal leadership role or do some have more than others?). According to Carson et al. (2007) network density calculated using valued data is simply another version of “group-as-a-whole” aggregation. It indicates nothing about the distribution of leadership among members and therefore is a deficient operationalization of shared leadership. Therefore future research should focus on network centralization. In this, centrality value are calculated for each individual and represents the number of connections each individual has with other. This described the extent to which connections are concentrated around one individual (Hanneman & Riddle, 2005). Therefore this approach is a measure of distribution and is an appropriate operationalization of shared leadership (Small and Rentsch, 2010).
The non- significant results can however also be blamed on the limited number of respondents. Even though the response rate with 62,5 percent was pretty decent, given the fact that this is a large organization which usually yields far lower response rates, the amount of respondents still only reached the bare minimum. Therefore, in future research the aim should be on more respondents in order to ensure that proper analyses can be done.
In addition, this study doesn’t have any data with regards to the composition of the respondents. Due to the nature of how diversity was measured, any questions with regards to the person were left out as it was only interesting to know the perceived diversity and not the objective diversity. This means that no claims can be made with regards to the composition of the respondents and therefore it is uncertain whether this sample is representative for the rest of the Netherlands. This also means that it cannot be said for sure if this sample is made up of only full- time employees. Having only full- time employees in your data set would be beneficial, especially when measuring shared leadership, this because it can be expected that shared leadership in a team will be lower when a few of the team members only work part- time as
21
opposed to full time. Therefore, future research should make sure that their target group contains only fulltime employees.
However, despite the flaws and failure to replicate most of the results from previous studies, this study did offer some insights. It was once again found that knowledge sharing does indeed lead to higher levels of team creativity. This supports all the previously described theories and studies into this area, which all stated that knowledge sharing indeed leads to better performance/creativity. In addition, the multiple regression analysis showed a significant interaction effect between shared leadership and social categorical diversity. Meaning that the effects of shared leadership as perceived by the individual were weakened by the effects of social categorical diversity. This too is in line with research into social categorical diversity. Unfortunately any other interaction effects weren’t found with regards to diversity. This is most likely due to the earlier mentioned difficulties respondents had with filling out their team members major, role and years of work experience within the organization. Therefore, future research is needed into diversity with more reliable measurements for (informational) diversity. With the findings of this study a bit more insight is gained into areas of interest within the domain of corporate communication. This includes insights into the workings of shared leadership, diversity, knowledge sharing and creativity in an environment which is highly dependent on knowledge sharing and creativity, namely that of the IT sector. IT work is very heavily knowledge based and nowadays a lot of IT development projects work with the Scrum development methodology. The Scrum methodology pretty much revolves around shared leadership and diversity. This because it revolves for the most part around multi- disciplinary roles and working together to reach their goals. Therefore knowing how shared leadership and diversity impact the degree of knowledge sharing and creativity provides a basis on which future research can build. In addition, this knowledge is also beneficial for companies seeking to boost knowledge sharing levels and enhance creativity. With the findings from this study it has been established that raising the degree of knowledge sharing may help enhance the degree of team creativity. In addition, organizations may consider encouraging shared leadership within teams in addition to the traditional vertical leadership as this may in turn boost knowledge sharing .
22 References
Armbrecht, F.M., Chapas, R.B., Chappelow, C.C., Farris, G.F., Friga, P.N., Hartz, C.A., Mcilvaine, M.E., Postle, S.R., & Whitwell, G.E. (2001). Knowledge Management in Research and Development. Research- Technology Management, Vol 44, No.4, pp 28-48.
Bauer, T.N., & Ergodan, B. (2005). Enhancing Career Benefits of employee proactive personality: The role of fit with jobs and organizations. Personnel Psychology, Vol. 58, No. 4, pp 859-891.
Bligh, M.C., Pearce, C.L., & Kohles, J.C. (2006). The importance of self- and shared leadership in team based knowledge work. Journal of Managerial Psychology, Vol. 21 No.4, pp. 296-318. Bock, G.W., Zmud, R.W., & Kim, Y.G. (2005). Behavioral intention formation in knowledge sharing: examining the roles of extrinsic motivators, social-psychological forces, and organizational climate. MIS Quarterly, Vol. 29 No. 1, pp. 87-111.
Byrne, D. (1971). The attraction paradigm. New York: Academic Press.
Carson, J. B., Tesluk, P. E., & Marrone, J. A. (2007). Shared leadership in teams: An investigation of antecedent conditions and performance. Academy of Management Journal, 50, 1217–1234.
Connelly, C.E. and Kelloway, E.K. (2003). Predictors of employees’ perceptions of knowledge sharing culture. Leadership & Organization Development Journal, Vol. 24 No. 5, pp. 294-301. Cox J.F., Pearce L., & Sims.H.P. (2003). Toward a broader agenda for leadership development: Extending the traditional transactional–transformational duality by developing directive, empowering and shared leadership skills; in: S. E. Murphy and R. E. Riggio, editors, The Future of Leadership Development, Lawrence Earlbaum, Mahwah, NJ ,pp. 161–180.
23
Davenport, T. H., & Prusak, L. (1998). Working knowledge: How organizations manage what they know. Boston: Harvard Business School Press.
Drazin, R., Glynn, M.A., & Kazanjian, R. (1999). Multilevel theorizing about creativity in organizations: A sensemaking perspective. Academy of Management Review Vol 24, No. 2, pp 86-307.
Fletcher, J.K., & Kaufer, K. (2003). Shared Leadership: Reframing the hows and whys of Leadership. California: Sage Publications.
Gibb, C. A. 1954. Leadership. In G. Lindzey (Ed.), Hand book of social psychology, 2, 877-917. Reading, MA: Addison-Wesley.
Hanneman, R. A., & Riddle, M. (2005). Introduction to social network methods. Riverside, CA: University of California, Riverside.
Hofhuis, J., van der Zee, K, I. & Otten, S. (2012) Social identity patterns in cultural diverse organizations: The role of diversity climate. Journal of Applied Social Psychology. doi: 10.1111/j.1559-1816.2011.00848.x
Jackson, S. E. (1992). Team composition in organizational settings: Issues in managing an increasingly diverse workforce. In S. Worchel, W.Wood, & J. A. Simpson (Eds.), Group processes and productivity (pp.136–180). Newbury Park, CA: Sage.
Jehn K.A., Chadwick C., Thatcher S. (1997). To agree or not to agree: the effects of value congruence, member diversity, and conflict on workgroup outcomes. Int. J. Confl. Manage. 8:287–305
Jehn K.A., Northcraft G.B., & Neale M.A. (1999). Why differences make a difference: a field study of diversity, conflict, and performance in workgroups. Admin. Sci. Q. 44:741–63
Lee, D,S., Lee, K.C., Seo, Y.W., Choi D.Y. (2014). An analysis of shared leadership, diversity, and team creativity in an e-learning environment. Computer in Human Behavior, 42, pp. 47-56.
24
Lin, H.F. and Lee, G.G. (2006). Effects of socio-technical factors on organizational intention to encourage knowledge sharing. Management Decision, Vol. 44, No. 1, pp. 74-88.
Lin, H.F. (2007). Knowlegde Sharing and firm innovation capability: an empirical study. International Journal of Manpower, Vol 28, pp. 315-332.
Nonaka. I., and Takeuchi, H. (1995). The knowledge- creating company: How Japanese companies create the dynamics of innovation. New York: Oxford University Press.
Pearce, C. L., & Sims, H. P. (2002). The relative influence of vertical vs. shared leadership on the longitudinal effectiveness of change management teams. Group Dynamics: Theory, Research, and Practice, 6(2): 172-197. Pearce, C. L., Yoo,
Pearce, C. L., & Conger, J. A. (2003). Shared leadership: Reframing the hows and whys of leadership. Thousand Oaks, CA: Sage.
Pearce, C.L., & Conger, J.A. (2003). All those years ago. The historical underpinnings of shared leadership, in Pearce, C.L. and Conger, J.A. (Eds), Shared Leadership. Reframing the Hows and Whys of Leadership, Sage, Thousand Oaks, CA.
Pearce, C. L., & Barkus, B. (2004). The future of leadership: Combining vertical and shared leadership to transform knowledge work. The Academy of Management Executive, 18(1), 47– 59
Pfeffer, J., (1983). Organizational demography. In Larry L. Cummings and Barry M. Staw (eds.), Research in Organizational Behavior, 5: 299- 357. Greenwich, CT: JAI Press.
Reid, F. (2003). Creating a knowledge sharing culture among diverse business units. Employment Relations Today, Vol. 30, No. 3, pp. 43-9.
Small, E.E., & Rentsch J.R. (2010). Shared Leadership in Teams: A Matter of Distribution. Journal of Personnel Psychology, 9, pp. 203-211.
25
Stasser, G. (1982). Information salience and the discovery of hidden profiles by decision making groups: A “thought experiment”. Organizational Behavior and Human Decision Processes, Vol. 52, No. 1, pp 156-181.
Tajfel, H. & Turner, J. C. (1986). The social identity theory of intergroup behavior. Psychology of intergroup relations, 2, 7-24.
Van Knippenberg, D., De Dreu, C. K. W. & Homan, A. C. (2004). Work group diversity and group performance: An integrative model and research agenda. Journal of Applied Psychology, 89, 1008–1022
Williams, K. Y., & O’Reilly, C. A. (1998). Demography and diversity in organizations: A review of 40 years of research. Research in Organizational Behavior, 20, 77–140.
26 Appendices
Appendix A – Tables of significant regression analyses.
Table 1: Linear Regression analysis of informational diversity at the team level on the degree of knowledge sharing
B Std. Error T Sig.
Constant 5.56 0.25 22.59 <.000
Informational Diversity Team 0.94 0.45 2.1 0.04
Table 2: Multiple regression analysis of social categorical diversity at the individual level on the relationship between shared leadership at the individual level and knowledge sharing.
B Std. Error T Sig.
Constant 6.03 .076 78.931 <.000
(Zscore) Shared Leadership Individual 0.25 0.08 3.22 .002
(Zscore) Social Categorical Diversity Individual
-0.48 .27 -1.75 .084
(Zscore) SL x SCD Individual 0.91 0.45 2 0.05
Table 3: Linear regression analysis of knowledge sharing on the degree of team creativity.
B Std. Error T Sig.
Constant 1.086 .841 1,291 ,200
27
Appendix B – Questionnaire
Q1 Which team do you primarily belong to? (names removed)
Teams were asked to select their team from a dropdown list, which would activate a display logic and forward them to questions with regards to shared leadership for that specific team. Answer If Which team do you primarily belong to? Team Is Selected
Q37 Please rate each of your team members on the question below. Please answer "N/A" for yourself To what degree does your team rely on this individual for leadership?
1 (1) 2 (2) 3 (3) 4 (4) 5 (5) 6 (6) 7 (7) N/A (8) Team member (1) Team member (2) Team member (3) Team member (4) Team member (5) Team member (6)
28
Q14 Knowledge Sharing.The following questions are about the level of knowledge sharing within your team.
1 (1) 2 (2) 3 (3) 4 (4) 5 (5) 6 (6) 7 (7)
People in this team are willing to share
knowledge/ideas with others (1)
People in this team share their ideas openly (2) People in this team with expert knowledge are willing to help others in this team (3) This team is good at using the knowledge/ideas of employees (4)
29
Q15 Team creativity The following questions are about the level of team creativity within your team.
1 (1) 2 (2) 3 (3) 4 (4) 5 (5) 6 (6) 7 (7)
Our team comes up with many new ideas about how work should be done (1) If a new way of doing work is introduced, it often comes from within the team (2)
Our team is frequently the source of ideas that are copied by other teams (3)
Q16 Could you indicate your own age and those of your team members? (I.e. if two of your team members have an age of between 29-39, then note a "2").
18-28 (1) 29-39 (2) 40-50 (3) 51-61 (4) 62+ (5)
Q17 How many males and females are there in your team? (Including yourself) Male (1)
30
Q31 Please indicate how many representatives of a given world region are working in your team, including yourself? (i.e. West Europe - 2, North America - 1)
Africa (1)
Australia & South Pacific (2) Caribbean (3)
Central America (4)
East Europe & Former Soviet Union (5) East Asia (6) Middle East (7) North America (8) South America (9) South Asia (10) Southeast Asia (11) West Europe (12)
31
Q21 Please indicate in which field you and your colleagues primarily studied. (I.e. if both you and colleague studied Business & Finance, then note a ''2'').
Business & Finance (1) Computer & Technology (2) Construction (3)
Engineering (4) Health & Medical (5)
Hospitality, Travel & Tourism (6) Law (7)
Management (8) Public Service (9)
Media & Communications (10) Office Administration (11) Production & Manufacturing (12) Psychology & Counseling (13) Sales (14)
Science (15)
Other, namely (+ amount)... (16)
Q24 Please indicate which and how many roles are being represented within your team? (Business Analyst - 2, Developer - 2, Tester - 1)
Application Architect (1) BI Consultant (2) Business Analyst (3) Business Architect (4) Developer (5) Functional Maintenance (6) IT/Ops (7) Manager (8) Product Owner (9) Program Supporter (10) Tester (11)
Other, namely (+ amount)... (12)
Q32 Please indicate how many team members (including yourself) have the following amount of work experience within the Rabobank:
(1) 1 - 2 years (2) 3-5 years (3) 6-10 years (4) 11 - 15 years (5) 15+ years (6)
