Transformational
Leadership Dimensions
and IT Innovativeness:
Knowledge Sharing Effects in Virtual Teams
ANNE DIJK2
Transformational Leadership Dimensions
and IT Innovativeness:
Knowledge Sharing Effects in Virtual Teams
Author Anne Dijk Barmaheerd 41 9737 MG Groningen The Netherlands +31(0) 630 25 49 93 University of Groningen
Faculty of Economics and Business
Master Strategy and Innovation Management Master’s Thesis
Supervisors
3
Abstract
The aim of this study is to provide several insights in the effect of transformational leadership on IT innovativeness in the context of virtual teams. By constructing IT innovativeness based on the concept of self leadership, two different types of innovative behavior are formulated: The IT innovative behavior of a virtual team member to support his or her own tasks; and IT innovative behavior of the individual to support the collective team tasks. With an online questionnaire data was collected from respondents working in virtual teams. Based on the literature, transformational leadership was expected to predict both types of IT innovative behavior, but this was not supported. An even more in-depth look was taken by testing the six dimensions of transformational leadership which are distinguished in literature individually, but again none was supported. To provide a different perspective, the moderator knowledge sharing was taken into account. This was expected to lead to more IT innovative behavior for team use, as it emphasizes the virtual team’s priority, and indeed it was supported. By additional analyses, the six transformational leadership dimensions showed several significant interaction effects with knowledge sharing, in predicting IT innovative behavior to support virtual team’s tasks.
Key words: TRANSFORMATIONAL LEADERSHIP; SELF LEADERSHIP; IT
4
Table of Content
Abstract... 3 1. Introduction ... 5 2. Literature review ... 7 2.2.1 Self leadership ... 82.2.2 Self leadership towards IT Innovativeness ... 9
2.3.1 Transformational leadership ... 10
2.2.2 Transformational leadership and IT innovative behavior ... 11
2.4 Knowledge sharing ... 13
3. Methodology ... 16
3.2 Sample ... 16
3.3 Measures ... 17
3.3.1 Dependent Variables ... 18
IT Innovativeness for Own Use (ITI) ... 19
IT Innovativeness for Team Use (ITI Team) ... 19
3.3.2 Independent Variables ... 20
Transformational Leadership (TFL) ... 20
Knowledge Sharing (KS) ... 21
4. Results ... 22
4.1 Factor Analysis Independent Variables ... 22
4.2 Factor Analysis Dependent Variables ... 24
4.2.1 Multiple Regression Analyses ... 26
Multiple Regression analysis: TFL on IT Innovativeness for Own Use ... 26
Multiple Regression analysis: TFL on IT Innovativeness for Team Use ... 27
4.2.2 Moderating Regression Analysis... 28
MRA: KS moderating TFL on ITI Innovativeness for Team Use ... 28
MRA: KS moderating the six transformational dimensions ... 29
5. Discussion ... 33
Theoretical contributions ... 36
Limitations and Future Research Directions ... 36
Acknowledgement ... 36
6. References ... 37
7. Appendix ... 42
Appendix I – Factor Analysis Transformational Leadership ... 42
Appendix II – Questions ITI ... 44
Appendix III – Questions ITI Team ... 45
Appendix IV – Questions Transformational Leadership (6 dimensions) ... 46
5
1. Introduction
In the globalizing economies of today, how do companies remain effective if their relevant expertise is scattered around the globe? The increasing inter-organizational activities require new organizational forms, since the organizations may be located all over the world making face to face meetings between the team members more difficult (Chang, 2011). These difficulties suggest that firms need to deal with increased challenges to coordinate tasks across time zones, physical boundaries, and organizational contexts (Kayworth and Leidner, 2002). An answer for managing this development relies in the advances in information and communication technology (also referred to as IT), which created new opportunities for organizations to establish and manage virtual teams (Kirkman et al., 2004).
Virtual teams are functioning teams that rely on technology-mediated communication while crossing several different boundaries, which are commonly boundaries geography, time and organization (Martins, 2004). The context of virtual teams is interesting for this research, as several studies have pointed out virtual teams are becoming commonplace in organizations today (Martins et al., 2004). Kanawattanachai and Yoo (2002) already found that more than 60% of professional employees work in virtual teams. Next to this, virtual teams have the potential to deliver unique strategic flexibility by enabling the rapid formation and disbanding of teams comprised of the best talent available all over the world (Lipnack and Stamps, 1997).
The communication and coordination activities of virtual team members are facilitated by information technologies that can be characterized by three continua, being time, space and level of group support (Warkentin, Sayeed and Hightower, 2007). The virtual teams can communicate synchronously or asynchronously; they may be located together or remotely; and the technology can provide task support primarily for the individual team member or for the group’s activities (Warkentin, Sayeed and Hightower, 2007).
So, for virtual teams it is crucial to be effective in their IT use as it has great impact on the total team performance given their dependence on IT to facilitate knowledge exchange, transfer and sharing (Zakaria, Amelinckx and Wilemon, 2004). An innovative and exploratory approach of the team members is needed to benefit from new information technologies (Agarwal and Prasad, 1998). This innovative behavior is the production or adoption of useful new ideas and the implementation of it, to increase activity effectiveness (Scott and Bruce, 1994). This is connected to the concept of IT self leadership by Wang, Li and Hsieh (2011), which regards to team members motivating themselves to use IT in an innovative and creative way. But how can this IT innovative behavior of virtual team members be realized?
6 transformational leadership style is known for supporting this innovative behavior of team members (Den Hartog, Van Muijen and Koopman (1996). Transformational leaders stimulate the group members to realize higher performances as is expected of them (Bass, 1985). This type of leader accomplishes this by motivating team members to work for transcendental goals that go beyond their immediate self-interests (Bass, 1997).
Therefore, this study will provide virtual teams as novel context for the influence of transformational leadership on team member IT self leadership. However, as Warkentin et al. (2007) stated, virtual team’s technology can provide task support primarily for the individual team member or for the group’s activities. Based on these two initial purposes of the innovative behavior, a team member is in this study considered to be innovative with IT for own individual use or be innovative for team use. This leaves out the question whether or not the team members are likely to innovate for team purposes: What’s in it for the individual? (Husted and Michailova, 2002)
According to Hu, Horng and Sun (2009), organizations have to create a positive team culture for sharing knowledge to stimulate this innovative behavior for team uses. The level of IT related knowledge sharing within a virtual team is based on the processes of sharing one’s experiences with another, thereby creating tacit knowledge in the form of mental models and technical skills like IT innovations (Nonaka and Takeuchi, 1995). The importance of knowledge sharing to support innovative behavior has been recognized by the literature (Hallin and Marnburg, 2008). To point out this effect of knowledge sharing on the relationship of transformational leadership and IT self leadership, for individual use and team use, the perception of knowledge sharing among the team members will be analyzed.
So, taking the knowledge sharing into account, the following research questions are formulated:
Does transformational leadership, in a virtual team, lead to IT innovativeness for individual and team use?;
And does knowledge sharing has a moderating effect on transformational leadership resulting into more IT innovative behavior for team use?
7
2. Literature review
2.1 Virtual teams
Virtual teams are functioning teams that rely on technology-mediated communication while crossing several different boundaries, which are commonly boundaries geography, time and organization (Martins, Gilson and Maynard, 2004). The context of virtual teams is interesting for this research, as several studies have pointed out virtual teams are becoming commonplace in organizations today (Martins et al., 2004). The rise of the virtual teams is being attributed to a confluence of technological and organizational developments along with a range of business benefits associated with using these types of teams (Solomon, 2001).
The virtual teams have the potential to deliver unique strategic flexibility compared to traditional face-to-face teams by enabling the rapid formation and disbanding of teams comprised of the best talent available all over the world (Lipnack and Stamps, 1999). This flexibility is realized by virtual team members to cooperate when they are geographically dispersed within a single firm, business unit or even between several firms to support the organizational objectives (Berry, 2011; Gibson and Cohen, 2003). Not only does the dynamic structure of the virtual teams enables creative and flexible responses to challenging development needs, it also allows access to expertise by composing the best individuals for the task on an as-needed basis (Brown and Eisenhardt, 1995; Lipnack and Stamps, 1999). For example, virtual teams can be temporary and focused on the completion of a specific project, or they can be long lasting, with stable membership over several months or years (Duarte and Snyder, 1999).
The communication and coordination activities of virtual team members are facilitated by information technologies that can be characterized by three continua, being time, space and level of group support (Warkentin, Sayeed and Hightower, 1997). These computer-based communication technologies are utilized to overcome the space and time burdens of face-to-face meetings and to improve group task performance effectiveness (McGrath and Hollingshead, 1994). Berry (2011) acknowledges the importance of IT by stating, the use of information technologies does not imply a team being virtual, the degree of virtualness increases if the group members rely more on the ICT compared to face-to-face interactions when working together to complete the tasks.
8 technologies (Agarwal and Prasad, 1998).Virtual teams therefore rely on the IT related innovative behavior of their team members which is called IT self leadership by Wang et al. (2011). This IT self leadership concept was based on the self leadership ideas of Manz and Sims (1980).
2.2.1 Self leadership
According to Manz (1992), self leadership involves the influence people exert over themselves, to achieve the self-motivation and self-direction needed to behave in desirable ways. Self leadership consists of specific behavioral and cognitive strategies designed to positively influence personal effectiveness (Neck and Houghton, 2006). These three strategies are: Behavior-focused strategies; natural reward strategies; and constructive thought pattern strategies (Prussia, Anderson and Manz, 1998; Houghton and Neck, 2002; Neck and Houghton, 2006).
Behavior-focused strategies strive to heighten an individual’s self-awareness in order to facilitate behavioral management, especially the management of behaviors related to necessary but unpleasant tasks (Manz and Neck, 2004). In other words, this type of strategy is designed to encourage positive, desirable behaviors that lead to successful outcomes, while suppressing negative, undesirable behaviors that lead to unsuccessful outcomes (Neck and Houghton, 2006).
Natural reward strategies are intended to create situations in which a person is motivated or rewarded by inherently enjoyable aspects of the task or activity (Manz and Neck, 2004). This type of strategy is designed to help create feelings of competence and self-determination, which in turn energize performance-enhancing task-related behaviors (Neck and Houghton, 2006).
Constructive thought pattern strategies are designed to facilitate the formation of constructive thought patterns and habitual ways of thinking that can positively impact performance (Manz and Neck, 2004). So, individuals who envision successful performance of an activity in advance of actual performance are more likely to perform successfully when faced with the actual task (Manz and Neck, 2004).
9 behavior in their tasks (Carmeli et al., 2006). This experimental behavior is the production or adoption of useful new ideas and the implementation of it, to increase activity effectiveness (Scott and Bruce, 1994). Our focus is the innovative behavior of virtual team members regarding IT, as it is introduced by the concept of IT self leadership by Wang et al. (2011).
2.2.2 Self leadership towards IT Innovativeness
Literature is not extensive about the linkage between self leadership and IT innovativeness. The link between innovativeness and self leadership is found more often. Carmeli, Meitar and Weisberg (2006) found a positive relationship between self leadership and innovative individual behavior. Innovative behavior is found to be a complex process, which refers to an activity whose aim is to develop, carry, react to, and modify ideas (Van de Ven, 1986; Janssen, Van de Vliert and West, 2004). It is not only the intentional act of generating new ideas, which is creativity, but also the introduction and application of new ideas, all aimed at improving organizational performance (Janssen et al., 2004).
Regarding to our focus on IT innovativeness, Agarwal and Prasad (1997) define the personal innovativeness with IT as a trait which contains the willingness of an individual to try out any new information technology. In this study, the IT Innovativeness will not be perceived as a trait, like Agarwal and Prasad (1997) used, but as the behavioral outcome of the team members. Wang et al. (2011) did use the behavioral outcome to measure the personal IT innovativeness and found that the mediators perceived usefulness and satisfaction are positively related to personal innovativeness with IT. Based on their findings, we will we use the definition of Wang et al. (2011) to describe the IT innovativeness, which is the generation and implementation of individual users’ creative ideas in the form of IT usage behaviors. Specifically, IT innovativeness describes a user’s applying IT in novel ways to support his or her task performance (Wang et al, 2011).
However, a distinction in the behavioral outcomes of IT innovativeness can be made. As Warkentin et al. (2007) state that the communication and coordination activities of virtual team members are facilitated by information technologies, they also state that these IT can be characterized by three continua, being time, space and level of group support. The virtual teams can communicate synchronously or asynchronously; they may be located together or remotely; and the technology can provide task support primarily for the individual team member or for the group’s activities (Warkentin et al., 2007). Of which third continuum could lead to two opposite types of IT innovation outcomes.
10 support the virtual team’s collective tasks, which will be called IT innovativeness for team
use.
For the IT innovativeness for own use, the behavior will be defined as experimental or playful use with the available or discovered IT by a team member, to find solutions for his or her own task supporting benefit and use. Whereas the IT self leadership for team use will be defined as the innovative IT behavior of a team member with the available or discovered IT, to find novel IT support and solutions for team purposes.
A way of stimulating the virtual team members to perform these two types of IT innovative behaviors is appointed to the team leader. According to Bryant (2003), leaders have direct control on team members’ behavior by encouraging by rewarding and value the desired innovative behavior. The transformational leadership style is known for supporting this innovative behavior of team members (Bryant, 2003).
2.3.1 Transformational leadership
According to Bass (1999), a team leader’s task is to align the team member's interests with the organizational interest. This can be done in a participative or directive way, so called transformational or transactional leadership respectively. Transformational leaders stimulate the group members to realize higher performances as is expected of them (Bass, 1985). These types of leaders accomplish this by motivating team members to work for transcendental goals that go beyond their immediate self-interests (Bass, 1999), whereas transactional leadership refers to the exchange relationship between leader and follower to meet their own self-interests.
Literature suggests that some contexts are better suited for transformational leaders where others are better for transactional leaders. For virtual teams, an empirical study found that as a communication medium becomes more anonymous a transformational leader may be a more effective leadership style compared to transactional leadership (Huang et al., 2010). Therefore, the main interest of this research is to examine how virtual teams could benefit of this transformational leadership style.
11
2.2.2 Transformational leadership and IT innovative behavior
Literature has pointed out only several different relationships between transformational leadership and IT innovativeness. According to Borekci (2009), IT use of leaders influences the perception of the followers of their leaders. Another study of Sosik et al. (1997) showed that transformational leadership was associated with higher levels of group potency, being the group's belief that it can be effective.
To get a more in-depth view of the influence of transformational leadership on IT innovativeness, we will take a look at the six dimensions of transformational leadership as they were found by Podsakoff et al. (1990). These dimensions of transformational leadership are based on the different characteristics or abilities appointed to this style of leadership. These dimensions are the ability of the leader to: Foster the acceptance of group goals; show individual support; provide intellectual stimulation; provide a relevant model for the team members; articulate a vision; and to have high performance expectations. A summary of the six dimensions is listed by Podsakoff et al. (1990):
Fostering the Acceptance of Group Goals – Behavior on the part of the leader aimed at promoting cooperation among employees and getting them to work together toward a common goal.
Providing Individualized Support – Behavior on the part of the leader that indicates that he/she respects followers and is concerned about their personal feelings and needs.
Intellectual Stimulation – Behavior on the part of the leader that challenges followers to re-examine some of their assumptions about their work and rethink how it can be performed.
Providing an Appropriate Model – Behavior on the part of the leader that sets an example for employees to follow that is consistent with the values the leader espouses.
Identifying and Articulating a Vision – Behavior on the part of the leader aimed at identifying new opportunities for his or her team, and developing, articulating, and inspiring others with his or her vision of the future.
High Performance Expectations – Behavior that demonstrates the leader’s expectations for excellence, quality, and/or high performance on the part of followers.
Each of these behaviors has been identified as an important element in the literature on transformational leadership (Podsakoff et al., 1990).
12 the individual influences of the six dimensions of transformational leadership on IT
innovativeness. However, as the IT innovative behavior could have two different outcomes, remember IT innovativeness for own use and IT innovativeness for team use, the influence of transformational leadership and its sub constructs will be tested on both outcomes:
H1. Transformational leadership has a positive influence on IT innovative behavior of virtual team members, for own tasks.
H1a. Fostering the Acceptance of Group Goals has a positive influence on IT innovative behavior of virtual team members, for own tasks.
H1b. Providing Individualized Support has a positive influence on IT innovative behavior of virtual team members, for own tasks.
H1c. Intellectual Stimulation has a positive influence on IT innovative behavior of virtual team members, for own tasks.
H1d. Providing an Appropriate Model
has a positive influence on IT innovative behavior of virtual team members, for own tasks.
H1e. Identifying and Articulating a Vision
has a positive influence on IT innovative behavior of virtual team members, for own tasks.
H1f. High Performance Expectations has a positive influence on IT innovative behavior of virtual team members, for own tasks.
H2. Transformational leadership has a positive influence on IT innovative behavior of virtual team members, for collective team tasks.
H2a. Fostering the Acceptance of Group Goals has a positive influence on IT innovative behavior of virtual team members, for team tasks.
H2b. Providing Individualized Support has a positive influence on IT innovative behavior of virtual team members, for team tasks.
H2c. Intellectual Stimulation has a positive influence on IT innovative behavior of virtual team members, for team tasks.
H2d. Providing an Appropriate Model
has a positive influence on IT innovative behavior of virtual team members, for team tasks.
H2e. Identifying and Articulating a Vision
has a positive influence on IT innovative behavior of virtual team members, for team tasks.
H2f. High Performance Expectations has a positive influence on IT innovative behavior of virtual team members, for team tasks.
13 However, another perspective could be taken to analyze the relationship between
transformational leadership and IT innovativeness. According to Wang et al. (2011) moderators can provide a more comprehensive picture of the connections between variables than just linear relationships.
A potential moderator for this relationship was found as Warkentin et al. (2007) stated that virtual team’s technology can provide task support primarily for the individual team member or for the group’s activities. As said, based on these two initial purposes of the innovative behavior, a team member is in this study considered to be innovative with IT for own individual use or be innovative for team use. This leaves out the question whether or not the team members are likely to innovate for team purposes: What’s in it for the individual? (Husted and Michailova, 2002) According to Hu, Horng and Sun (2009), if organizations are able to create a positive team culture for sharing knowledge, this will stimulate innovative behavior for team uses.
2.4 Knowledge sharing
Many studies have recognized the importance of knowledge sharing to support innovative behavior in an organization (Finnegan and Willcocks, 2006; Hallin and Marnburg, 2008; and Mohamed et al., 2004). The level of IT related knowledge sharing within a virtual team is defined by the processes of sharing one’s experiences with another, thereby creating tacit knowledge in the form of mental models and technical skills like IT innovations (Nonaka and Takeuchi, 1995). The importance of knowledge sharing to support innovative behavior has been recognized many times by the literature (Hallin and Marnburg, 2008).
However, sharing such IT innovations is not always a logical consequence of finding useful innovations: it might be good for the team, but what is in it for the individual (Husted and Michailova, 2002)?
14 According to Bryant (2003), leaders have direct control on team members’ behavior
by encouraging by rewarding and value the desired behavior. He states that leaders encourage workers to share their ideas by creating a climate that is receptive to new ideas.
The transformational leaders create a constructive climate by creating an atmosphere conducive to knowledge creation and sharing (Bryant, 2003). By encouraging intellectual development and paying individual attention to workers, transformational leaders motivate their workers to create and share knowledge. Therefore, transformational leadership should lead to positive perception of knowledge sharing among the virtual team members, which leads to sharing and more IT innovative outcomes at the benefit of the collective team tasks compared to IT innovative outcomes for the individual’s tasks. The following hypotheses will be used to test the moderating effect of knowledge sharing on the IT innovative outcomes:
H3. The perceived knowledge sharing within a virtual team moderates the relationship between transformational leadership and innovative IT behavior for collective team use.
H3a. Fostering the Acceptance of Group Goals within a virtual team moderates the relationship between transformational leadership and innovative IT behavior for collective team use.
H3b. Providing Individualized Support within a virtual team moderates the relationship between transformational leadership and innovative IT behavior for collective team use
H3c. Intellectual Stimulation within a virtual team moderates the relationship between transformational leadership and innovative IT behavior for collective team use
H3d. Providing an Appropriate Model within a virtual team moderates the relationship between transformational leadership and innovative IT behavior for collective team use
H3e. Identifying and Articulating a Vision within a virtual team moderates the relationship between transformational leadership and innovative IT behavior for collective team use
H3f. High Performance Expectations has a positive influence on IT innovative behavior of virtual team members, for own tasks.
15
VIRTUAL TEAM CONTEXT
H2.
IT
Innovativeness
Transformation
al Leadership
(TFL)
a. FAG b. ISU c. IST d. PAM e. AV f. HPEKnowledge
Sharing
IT Innovativenessfor Own Use
IT Innovativeness
for Team Use
H1.
16
3. Methodology
3.1 Research Design
Data was collected from individual virtual team members of random chosen companies. The secure a randomized data collection we chose 30 companies for each of the 10 general industry sectors in the Netherlands, which are distinguished by the Chamber of Commerce and the Central Bureau of Statistics of the Netherlands. However, the companies were selected based on the likelihood of making use of virtual teams.
Of each company, the employees who had experience with working in virtual teams were asked to fill out a structured digital questionnaire, which they were able to access by a website link provided in an email. Over 300 respondents were approached by a telephone call and asked to participate in the online survey. The respondents that did not finished or started the questionnaire were reminded of their intended participation one week after the initial phone call by follow up telephone call or email.
The survey was done online as it has the advantages of low research costs, flexibility and the ease of use (Lumsden and Morgan, 2005). All the questions used in the survey were taken from original scales found in the literature and translated from English to Dutch by four native Dutch speakers. Later on, the English questionnaire was also made available on the website which provided some extra opportunities to gain respondents.
3.2 Sample
Of the approximately 300 respondents approached for this research, 166 participated in the online survey leading to a initial response rate of 55.33%. The sample used for the analysis consisted of 166 respondents, of which 46 respondents started the questionnaire but did completed the questionnaire and 6 respondents filled it out after the deadline of June 9, 2013. This led to 114 respondents remaining of which 6 respondents had too much missing values and were therefore also left out of the final sample, resulting in 108 useful respondents for the analyses.
17 The random selection of the companies has turned out quite well, as every branch
had at least one respondent and not the gross of the respondents are still distributed over 6 of the branches.
The average size of the virtual teams participated by the respondent consisted of 12 team members, with a minimum score found of 2 team members and a maximum of 80 team members.
Table 1: Sectors of Industry Table 2: Sample characteristics
Variable Frequency (%) Sex Male 61 (56.5%) Female 47 (45.5%) Total 108 (100%) Age 20-29 35 (32.5%) 30-39 22 (20.4%) 40-49 25 (23.1%) 50-59 22 (20.4%) 60+ 2 (1.9%) Unknown 2 (1.9%) Total 108 (100%)
3.3 Measures
In this section we will discuss the measurements used from the literature and assess them on their construct validity and internal consistency. In Table 3, the literature sources of the questions per variable are provided. Tested constructs from the literature were used to support the validity of the questions and scales. An overview of all the questions we used in the questionnaire to measure our variables can be found in the Appendix.
Sectors of Industry Frequency (%)
Health Care 16 (14.8%)
Other 16 (14.8%)
Education 15 (13.9%)
IT Industry 14 (13.0%)
General Business Services 10 (9.3%)
Fast Moving Consumer Goods 10 (9.3%)
Other Services 5 (4.7%)
Transport 3 (2.8%)
Bank and Insurance 3 (2.8%)
Governmental or Law Organizations 3 (2.8%)
Retail and Wholesale 3 (2.8%)
Development or NGOs 2 (1,9%)
Paper and Print 1 (1.0%)
Culture, Sports and Leisure 1 (1.0%)
Fuel, Plastics, Chemical Industry 1 (1.0%)
Construction 1 (1.0%)
Agriculture, Forestry and Fishing 1 (1.0%)
Hospitality 1 (1.0%)
Unknown 2 (1.9%)
18 Before we tested the hypotheses, we executed a Principal Component Analyses on
the multi-item scales and retained the measures for each construct according to the following three criteria: First, each measure should have a loading of greater than .50; second, a measure cannot have a loading of .40 to more than one factor; and third, each factor must load into the correct factor. The Principal Component Analysis was used as it is considered to be superior to Exploratory Factor Analysis in testing the scales drawn from existing literature (Velicer and Jackson, 1990).
After the constructs were determined by the factor analysis, the Cronbach’s alpha was measured on each multi-item scale to assess the internal consistency of the scales, which has to be higher than the 0.7 norm of DeVilles (2003). The Cronbach’s alpha of the constructs can be find on the diagonal axe of Table 8.
Table 3: Variables and Measurements
Variables Measures No. of Questions
Dependent Variable IT Innovativeness for Own Use (ITI)
Agarwald & Prasad (1998); Ahuja &Thatcher (2005)
4 + 2
IT Innovativeness for Team Use (ITI Team)
Agarwald & Prasad (1998) *modified; Ahuja & Thatcher (2005) *modified
4 + 2 Independent Variable Transformational Leadership (TFL) Podsakoff et al. (1996) 24 Moderating Variable
Knowledge Sharing (KS) Faraj & Sproull (2000); Durham (1997)
4 + 3
3.3.1 Dependent Variables
19 This include adapting the initial IT innovative behavior questions and transform them
to clearly measure the IT innovative behavior which has the purpose of supporting the virtual team tasks. This was done by emphasizing the difference purposes of the two question sets by underscored and bold letter type in the description of the questions. We will now describe the development of the constructs for these two dependent variables and also discuss the reliability and validity of these variables.
IT Innovativeness for Own Use (ITI)
In the literature no complete set of scales or questions is provided to measure the construct of IT Innovativeness for Own Use. We found Innovative with IT (IwIT) of Wang et al. (2011) and Personal Innovativeness with IT (PIIT) of Agarwald and Prasad (1998) as comparable constructs. To compose a scale for this construct, three steps had to be taken. First, the validated questions from the literature about the alike constructs (Agarwald and Prasad, 1998; Wang et al., 2011) were combined to provide a six-item scale. The reliability of these separate scales in the literature was good, both IwIT and PIIT showed a Cronbach’s alpha of 0.86 in the study of Wang et al. (2011). Second, a Principal Component Analysis was executed to confirm that these two scales about IT innovativeness extracted from the literature (Agarwald and Prasad, 1999; Wang, 2011) could form one 6-item scale. The open factor analysis was done with the Oblimin Rotation method and the loadings showed that it was appropriate to form one scale. This was also supported in our third step by the Cronbach’s alpha, which indicated this 6-item scale to have a relative high internal consistency with α = .871 compared to the 0.7 norm of DeVilles (2003).
IT Innovativeness for Team Use (ITI Team)
For the second dependent variable IT Innovativeness for Team Use, the same questions were derived from the literature as the questions implicated a strong focus on the IT innovativeness for the individual own goals.
The interest of this research is to compare this individual IT innovating purpose with the team supporting IT innovating behavior of the individuals. Therefore, the questions were modified by focusing on the innovative behavior to serve the virtual group goals. Same as with the ITI, the literature offered no IT Innovations for Team Use related questions. Therefore, the questions for this construct were based on the former dependent variable ITI. The questions were modified to emphasize the different purposes of the innovative behaviors which can be performed by the respondents.
20 in to one scale. It also showed the clear different loadings between the constructs of the two
dependent variables, as can be seen in Table 7.
For combining the two scales into the construct of IT innovativeness for team uses, the factor analysis showed that is was preferred to remove the third (reversed) question ‘In
general I am reluctant to introduce new ICTs, that would better support team tasks, to the team.’ from the scale. Still, the Cronbach’s alpha of α = .910 for this 5-item scale for ITI for Team Use was also relative high, which is concluded a scale with high internal consistency.
3.3.2 Independent Variables
Transformational Leadership (TFL)
The initial questions of for measuring the transformational leadership in the virtual team were taken from the study of Podsakoff et al. (1990), which identified 6 key transformational leadership dimensions. These dimensions all showed a good internal consistency in the literature (FAG α = .89; ISU α = .90; IST α = .82; PAM α = .84; AV α = .87; HPE α = .80)
The factor analysis showed whether or not the six dimensions of Podsakoff et al. (1990) could be extracted from the data of our survey. By executing 12 Principal Component analyses with the Oblimin rotation method, the weak or cross loading questions were removed one measure by one analysis. The factor analyses showed that was most appropriate to remove 11 items from the initial scales and form the six dimensions with the 13 remaining questions as they are presented in the Appendix I, Table C and D.
To test our proposed relationships more in-depth, the transformational leadership variable was used as a general construct as well as the six different dimensions of Podsakoff et al. (1990). Therefore, two steps were taken to assess the internal consistency of this construct, first the consistency of the total transformational leadership construct was measured and second the multi-item scales of the six dimensions were examined individually to give a more detailed view about the internal consistency.
Of the first dimension ‘fostering acceptance of group goals’ (FAG), 1 item was removed, leaving a three-item scale which displayed a Cronbach’s alpha of α = .939, suggesting a very high internal consistency.
Of the second dimension ‘providing individual support’ (ISU), two items of the three-item scale were removed, leaving a single-three-item scale.
21 From the fourth dimension ‘providing a relevant model’ (PAM) three questions were
removed by the factor analysis, leaving a single-item scale.
The fifth dimension ‘articulating a vision’ (AV), four items of the five-item scale were removed, leaving a single-item scale.
The sixth dimension ‘high performance expectations’ (HPE), a three-item scale, displayed a relative very high internal consistency with a Cronbach’s alpha of α = .902.
Finally, the internal consistency of the transformational leadership construct in general was assessed and was considered to be very high, remaining 13 item-scale showed α = .930 for the Cronbach’s alpha.
Knowledge Sharing (KS)
The questions for the construct knowledge sharing were composed by Srivastava, Bartol and Locke (2006). They found support for combining the four-item scale of Faraj and Sproull (2000) with a three-item scale of Durham (1997), with α = .94. Faraj and Sproull (2000) used the scale to measure the perception of expertise sharing in teams among team members, whereas Durham’s (1997) scale is measuring the actual information sharing in teams. With combining these scales the team members express their feelings about the perceived knowledge sharing in their virtual team.
22
4. Results
In this section, the significant results of the several statistic tests that are conducted are presented. This chapter starts with assessment of correlation to measure the associations between constructs of the study which will be important for the tests later on. First a factor analysis is presented to show that there are no underlying structures among the different independent variables, next the same will presented for the dependent variables.
This assessment will be followed by the results of several multiple Regression analyses to provide answers to Hypothesis 1 and 2 and their sub hypotheses. The final tests performed are the Moderating Regression Analysis, this type of test is used to measure the moderating effect of Knowledge Sharing (KS) on the relationship of Transformational Leadership (TFL), including the dimensions, and ITI for Team Use (ITI Team).
4.1 Factor Analysis Independent Variables
By performing a Factor Analysis for the independent variables all together, we get a clear view of potential underlying structures of the variables, reducing the data into a few usable constructs for the analysis.
Table 4: Factor analysis
KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy. ,875
Bartlett's Test of Sphericity
Approx. Chi-Square 1399,114 136 df
Sig. ,000
23 Table 5: Factor analysis Independent Variables
Factor Loadings from Factor Analysis
Component
TFL FAG KS TFL ISU TFL IST TFL PAM TFL HPE TFL AV TFL_FAG4 ,947 TFL_FAG3 ,832 TFL_FAG1 ,806 TFL_FAG2 ,710 KS 3 ,943 KS 4 ,936 KS 1 ,930 KS 2 ,825 TFL_ISU1r ,971 TFL_IST3 ,952 TFL_IST4 ,793 TFL_IST5 ,753 TFL_PAM3 -,969 TFL_HPE2 ,856 TFL_HPE1 ,714 TFL_HPE3 ,707 TFL_AV1 -,879
Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization.a a. Rotation converged in 12 iterations.
24
4.2 Factor Analysis Dependent Variables
The open Principal Component Factor analysis of the dependent variables was justified to run according to the KMO score, having .825>.050. The results of the factor analysis, done with the Oblimin Rotation method, are presented in Table 6. IT innovativeness for own use and IT innovativeness for team use showed two individual constructs and both having good loading questions on each construct.
Table 6: KMO and Bartlett’s Test Table 7: Factor analysis Dependent
Variables
KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy. ,825
Bartlett's Test of Sphericity
Approx. Chi-Square 686,206
df 45
Sig. ,000
After the constructs of the variables were determined a correlation matrix is presented in Table 6 to provide a clear overview of the correlating variables in this study. The table shows no significant correlations between variables of our main focus. Neither transformational leadership, nor one of its six sub constructs, is significantly correlated to IT innovativeness for own or IT innovativeness for team use. The transformational leadership constructs are also not significantly correlated to our moderator knowledge sharing. However, the moderator is significantly correlated to both dependent variables, IT innovativeness for own use (p < 0.01) and IT innovativeness for team use (p < 0.05). The dependent variables also correlated significant with each other (p < 0.01).
Factor Loadings from Factor Analysis
Component ITI Team ITI IT_TEAM_IWIT2 ,903 IT_TEAM_PIIT4 IT_TEAM_IWIT1 ,896 ,873 IT_TEAM_PIIT2 IT_TEAM_PIIT1 ,856 ,731 IWIT2 ,865 IWIT1 ,832 PIIT3(r) ,746 ,659 PIIT1 PIIT4 ,606
Extraction Method: Principal Component Analysis.
Rotation Method: Oblimin with Kaiser Normalization.
Table 8: Correlation matrix
Cronbach’s Alpha Mean (SD)
1 2 3 4 5 6 7 8 9 10
1. ITI Pearson Correlation
Sig. (2-tailed)
4.08 (1.313)
.864
2. ITI Team Pearson Correlation Sig. (2-tailed) 3.82 ,645** .910 (1.381) ,000 3. TFL Pearson Correlation Sig. (2-tailed) 4.60 ,134 ,110 .930 (1.119) ,172 ,261
4. TFL Foster Acceptance of Group Goals Pearson Correlation Sig. (2-tailed)
4.67 ,153 ,131 ,910** .939 (1.432) ,117 ,182 ,000
5. TFL Intellectual Stimulation Pearson Correlation Sig. (2-tailed)
4.53 ,115 ,182 ,801** ,591** .893 (1.307) ,241 ,061 ,000 ,000
6. TFL High Performance Expectations Pearson Correlation Sig. (2-tailed)
4.79 ,095 ,027 ,833** ,669** ,614** .902 (1.357) ,333 ,782 ,000 ,000 ,000
7. TFL Individual Support Pearson Correlation Sig. (2-tailed)
4.47 -,043 -,122 ,445** ,439** ,216* ,158 1 (1.777) ,664 ,212 ,000 ,000 ,025 ,104
8. TFL Articulating a Vision Pearson Correlation Sig. (2-tailed)
4.53 ,145 ,026 ,755** ,620** ,564** ,622** ,289** 1 (1.532) ,138 ,789 ,000 ,000 ,000 ,000 ,003
9. TFL Providing a Model Pearson Correlation Sig. (2-tailed) 4.17 ,056 ,144 ,598** ,525** ,415** ,404** ,174 ,439** 1 (1.450) ,567 ,140 ,000 ,000 ,000 ,000 ,073 ,000 10. KS Pearson Correlation Sig. (2-tailed) 4.67 ,428** ,230* ,015 ,075 -,015 ,022 -,090 -,034 -,028 .927 (1.349 ,000 ,017 ,881 ,442 ,882 ,824 ,355 ,729 ,778 N 107 107 107 107 107 107 107 107 107 108
4.2.1 Multiple Regression Analyses
Multiple Regression analysis: TFL on IT Innovativeness for Own Use
To test hypothesis 1 we look at Model 1 of Table 9, only regressing the independent variable transformational leadership (TFL) on IT innovativeness for own use (ITI), which has an Adjusted R-square of only .008 so can only be accounted for 8% of the variance of IT innovativeness for own use. We use Adjusted R-square (Adj.R²) because of the little respondents involved in this survey, which is now taken into consideration in the test. Next to the low Adj.R², the F statistic (1.889) is not considered significant (p > .05). With these results be considered not significant our hypothesis 1a: ‘Transformational leadership has a
positive influence on IT innovative behavior of team members, for own individual tasks.’ is not supported.
Next to this general construct of transformational leadership, none of the six sub constructs provided significant support for their sub hypothesis (H1a – H1f) in predicting IT Innovativeness for own use.
Table 9: Results Multiple Regression Analysis IT Innovativeness for Own Use
Note: *p<.05, **p<.01
27
Multiple Regression analysis: TFL on IT Innovativeness for Team Use
To test hypothesis 2 we look at Model 1 of Table 10, regressing the independent variable transformational leadership (TFL) on the second dependent variable IT innovativeness for team use (ITI Team). This model has an even lower adjusted Adj.R² of .003. Again, the F
statistic (1.276) is not considered significant (p > .05). Therefore, our hypothesis 1b: ‘Transformational leadership has a positive influence on IT innovative behavior of team members, for collective team tasks’ is not supported.
Again, next to this general construct of transformational leadership, none of the six sub constructs provided significant support for their sub hypothesis (H2a – H2f) in predicting IT Innovativeness for team use.
We will continue with this dependent variable, IT innovativeness for Team Use, as it is also used for testing the third hypothesis, which will include the moderating effect of knowledge sharing.
Table 10: Results Multiple Regression Analysis IT Innovativeness for Team Use
Note: *p<.05, **p<.01
28
4.2.2 Moderating Regression Analysis
MRA: KS moderating TFL on ITI Innovativeness for Team Use
To test the hypothesis 3: ‘Knowledge sharing within a virtual team moderates the IT innovative behavior for collective team use.’ we used three regression models. In Table 10
the results are presented of the models. We mean-centered all variables, as was recommended by Aiken and West (1991) to test the test interaction effects. No serious multicollinearity problems were found in the mean-centered regression analyses.
Model 1, which only contained the independent variable transformational leadership (TFL), with Adj.R² =.011 though the F statistic (1.276) is not considered significant (p >.05). Model 2 included the independent variable knowledge sharing which is considered to be the moderating variable, show Adj.R²=.048 and the F statistic (3.635) to be significant (p < .05). Model 3 adds the interaction variable which can be seen as the product of the moderating effect of knowledge sharing on transformational leadership in predicting IT innovativeness for team use. This model has Adj.R²=.101, which is therefore only accounted for approximately 10,1% of the variance of IT Innovativeness for team use, and the F statistic (4.918) is found to be significant (p < 0.05) and β=.188 (SE = .071). Based on this significant
interaction effect, as predicted by hypothesis 3: Knowledge sharing within a virtual team moderates the IT innovative behavior for collective team use.’ this moderating effect is
supported.
29 To examine these results in greater depth, we conducted a spotlight analyses. In a
spotlight analysis, the researcher can focus the ‘spotlight’ on a particular level of a continuous independent variable by shifting the mean of this variable up or down (Fitzsimons 2008). This was done by creating a new variable which categorized the data into three levels of the moderating variable knowledge sharing, being high, moderate and low levels of knowledge sharing. A spotlight analysis at one standard deviation above and one standard deviation below the mean of knowledge sharing illustrate (Figure 2) the interaction effect between transformational leadership on IT innovativeness for team use.
The spotlight analysis shows that individuals that perceive low levels of knowledge sharing in their virtual team, only correlate with β=.184 (0.034 square rooted). For the moderate level this correlation is square root 0.048 = β=.219. But for the high level of KS, the correlation is square rooted .066, which is β=.257. That is what the interaction effect of knowledge sharing is. The nature or strength of the relationship between transformational leadership and IT innovativeness for team use changes as a function of another variable. In this case, the higher the level of knowledge sharing, the more stronger the relations gets between transformational leadership and IT innovativeness for team use. In other words, the effect of TFL leading to ITI Team is increased because there is a feeling of knowledge sharing among the virtual team members.
An additional analysis is done to examine whether or not knowledge sharing would have the same significant interaction effect along with transformational leadership on IT innovativeness for own use. As is shown by Model 3 of Table 10 the Adj.R² =.190 and the F
statistic (9.222) is found to be significant (p < 0.05). However, the interaction on this dependent variable is not found significant (p > 0.05). Therefore, a different effect of knowledge sharing on the two variables is found.
MRA: KS moderating the six transformational dimensions
30 Remember the six dimensions being the leader’s ability to foster acceptance of group
goals (TFL FAG); show individual support (TFL ISU); provide intellectual stimulation (TFL IST); providing a relevant model for its team members (TFL PAM); articulating a vision (TFL AV); and have high performance expectations (TFL HPE).
So, with six multiple regression analyses we found out which of the six dimensions did have a significant interaction effect with knowledge sharing on IT Innovativeness for Team Use. Of the six performed multiple regression analyses, four dimensions showed significant interaction effect with knowledge sharing. The significant moderating effects were found with the sub constructs ‘foster acceptance of group goals’ (FAG, Table 11); ‘providing a relevant model’ (PAM, Table 12); ‘articulating a vision’ (AV, Table 13); and ‘individualized support’ (ISU, Table 14).
The first tested dimension of transformational leadership was the ability of the transformational team leader to foster acceptance of group goals (TFL FAG). This third model of Table 11 showed Adj.R²=.111 and the F statistic (1.276) of this model was significant (p < .01). Also, the interaction effect of knowledge sharing on this dimension of transformational leadership was considered significant (p < 0.01), with a β=.151 (Standard Error = 0.053) in predicting the IT Innovativeness for Team Use.
Table 11: Results Multiple Regression Analysis IT Innovativeness for Team Use
Note: *p<.05, **p<.01
The second tested dimension was the ability to show individual support (TFL ISU) to the team members. Model 3 of Table 14 showed that the Adj.R² = .104 and the F statistic
(5.068) of this model were significant (p < .01). The interaction effect of these independent variables was also significant (p < .01) with a β=.144 (SE= .052).
31 The third dimension, the transformational leader’s ability to provide intellectual
stimulation (TFL IST) for its team members. The performed regression analysis showed that the Adj.R² = .071 and the F statistic (3.658) of this model were significant (p < .05). Though,
the interaction effect of knowledge sharing with intellectual stimulation was not considered significant (p > 0.05).
Table 14: Results Multiple Regression Analysis IT Innovativeness for Team Use
Note: *p<.05, **p<.01
Dimension four, providing a relevant model for its team members (TFL PAM) is shown in Model 3 of Table 12. The Adj.R² = .130 and the F statistic (6.223) of this model
were significant (p < .01). The interaction effect of this dimension was found significant (p < 0.01) with β=.175 (SE= .057).
The fifth dimension is the ability to articulate a vision (TFL AV). Model 3 of Table 13 show Adj.R² = .133 and the F statistic (8.817) of this model were significant (p < .01). The
interaction effect of this dimension with knowledge sharing was found significant (p < 0.01) with a coefficient β= .166 (SE= .055).
32 Table 12: Results Multiple Regression Analysis IT Innovativeness for Team Use
Note: *p<.05, **p<.01
The sixth and last dimension, the transformational leader to create high performance expectations (TFL HPE) showed Adj.R² = .071 and the F statistic (3.031) of this model were
significant (p < .05). Though, the interaction effect of knowledge sharing with these high performance expectations was not considered significant (p > 0.05).
Table 13: Results Multiple Regression Analysis IT Innovativeness for Team Use
Note: *p<.05, **p<.01
Variables Dependent variable: IT Innovativeness for Team Use Model 1 Coefficient Estimate (Standard Error) Model 2 Coefficient Estimate (Standard Error) Model 3 Coefficient Estimate (Standard Error) Direct effect TFL PAM .137 (.092) .140 (.090) .139 (.087) KS .241* (.096) .252** (.093) Interaction effect TFL PAM x KS .175** (.057) F value 2.214 4.283 6.223 Adj.R² .011 .059 .130
33
5. Discussion
Table 14 summarizes the findings of our hypotheses.
To provide answers to our two research questions: ‘Does transformational
leadership, in a virtual team, lead to IT innovativeness for individual and team use?’ and
‘Does knowledge sharing has a moderating effect on transformational leadership resulting into more IT innovative behavior for team use?’, we will first take a look at the relationships of transformational leadership and the two dependent variables found in this study.
Comparing the two dependent variables, both were not significant correlated with transformational leadership or were significant predicted by transformational leadership in the regression analyses. Therefore, the first research question, which was examined by
hypotheses 1 and 2 (both found no empirical support), is answered negatively.
So, where literature (Den Hartog, Van Muijen and Koopman, 1996; Bryant, 2003; Bass, 1985) assumed that transformational leadership would lead to IT innovative behavior from the virtual team members, either to support their own individual tasks or the collective team tasks, this study showed that this is not always the case.
Does this mean transformational leadership is not useful for virtual teams? Next to our limited research, which will be discussed later on, the virtual team could still provide a good context to work effective by transformational leadership (Huang et al., 2010). However, a direct relationship with IT innovativeness for own use or team use was not found. This was even further supported by looking at the transformational leadership on the level of the six dimensions of Podsakoff (1996). Still, even with these sublevels of transformational leadership, no direct relationship was found with either IT innovative behavior to support individual task or team tasks.
Contrary to the first research question, the second research question did found a variety of support to positively answer this question. Regarding to the moderating effect of knowledge sharing on the transformational leadership in virtual teams, the hypotheses 3 did provide significant support (Figure 2).
The hypothesis was strengthen by the spotlight analysis, concluding that high levels of knowledge sharing combined with transformational leadership would lead to more IT innovative behavior in virtual team for the collective teams uses. This effect is even checked for a possible similar effect on IT innovative behavior for individual uses, but on this dependent variable no significant interaction effect was found between knowledge sharing and transformational leadership.
34 sharing climate (Bryant, 2003) would stimulate the team member to balance the concern for
personal and group outcomes, cooperation, creativity, and free information exchange (Cooke and Szumal, 1994), leading to more innovations for team tasks but not necessarily more for individual tasks. Based on our findings, to ensure transformational leadership to stimulate IT innovativeness for team uses, knowledge sharing is needed to strengthen the collective priority.
To get a more in-depth look in the effect of knowledge sharing on IT innovativeness for team uses, we did some additional analyses. Next to the perspective of knowledge sharing in the spotlight analysis, we took another perspective by splitting up transformational leadership into the six dimensions of Podsakoff (1996).
The interaction effect of these dimensions showed that 4 out of 6 dimension caused a significant interaction effect, being ‘foster the acceptance of group goals’ (FAG), ‘providing individual support’ (ISU), ‘providing a relevant model’ (PAM) and ‘articulating a vision’ (AV).
Table 14: Summary of the hypotheses findings
Factors Hypotheses Findings
TFL H1: TFL ITI Not supported
1a: FAG ITI Not supported
1b: ISU ITI Not supported
1c: IST ITI Not supported
1d: PAM ITI Not supported
1e: AV ITI Not supported
1f: HPE ITI Not supported
TFL H2: TFL ITI Team Not supported
2a: FAG ITI Team Not supported
2b: ISU ITI Team Not supported
2c: IST ITI Team Not supported
2d: PAM ITI Team Not supported
2e: AV ITI Team Not supported
2f: HPE ITI Team Not supported
KS H3: Moderate TFL ITI Team Supported
3a: Moderate FAG ITI Team Supported
3b: Moderate ISU ITI Team Supported
3c: Moderate IST ITI Team Not supported
3d: Moderate PAM ITI Team Supported
3e: Moderate AV ITI Team Supported
35 The ability of transformational leaders to ‘foster acceptance of group goals’ and
‘providing a relevant model’ are both likely to interact with knowledge sharing as they emphasize the importance or priority of the team. Knowledge sharing assists in the formation of shared mental models that enable people to be ‘on the same page’ during task execution and achieve higher team performance (Srivastava, Bartol and Locke, 2006). Therefore, the team members need to learn from each other and provide others access to their own knowledge to accomplish those collective goals, which could lead to more willingness to share knowledge.
The significant effect of ‘providing individual support’ could be explained by the findings of Podsakoff et al. (1990), which found that team members that experienced individualized support from their leaders are more altruistic compared to others. This altruistic behavior is likely to strengthen the supportive culture in the virtual team leading consequently to more knowledge sharing.
The last significant dimension of transformational leadership, ‘providing a vision’, is explained by stimulating sportsmanship (Podsakoff et al., 1990). By providing a clear and general vision for the virtual team, the virtual team members are more likely to show sportsmanship which results into more fellowship among the team members. This increased cohesiveness of a virtual team will also secure for more knowledge sharing behavior as the team members are aware of the collective goals.
To provide possible explanations for the insignificant moderating effect of knowledge sharing with the two remaining dimensions ‘intellectual stimulation’ and ‘high performance expectations’, we consider the more individualized focus of these two dimensions (Podsakoff et al., 1990). This is comparable with the significant ‘individualized support’ dimension, though these two insignificant dimensions did not show altruistic behavior in the study of Podsakoff et al. (1990). Intellectual stimulation would differ more per individual as it does not provide guidelines for the team outcome, but it is more individual centered.
The high performance expectation is also lacking empirical support for an interaction effect with knowledge sharing, which could be explained by the team members are more focused on their own performance as the transformational leader raises the bar for each individual to take even a step further (Podsakoff et al., 1990).
36
Theoretical contributions
This study indicates that transformational leadership is on itself not always a significant predictor of IT innovativeness within a virtual team. For the IT innovation literature this study did provided empirical support for IT innovativeness to be separated in two different interpretations of IT innovation purposes, being for individual use and team use, based on the supporting goal of IT (Warkentin et al., 2007). This distinction in the definition of IT Innovativeness is not only supported by the factor analysis, but also by the final results showing significant differences in the moderating effect of knowledge sharing leading to only more IT innovative behavior for team use.
More insights were provided on the six dimensions of transformational leadership, as they were not all significantly moderated by knowledge sharing. Based on comparisons and findings of existing literature (Podsakoff et al., 1990; Cooke and Szumal, 1994; Bryant, 2003; Srivastava, Bartol and Locke, 2006), the individual or collective nature of the dimension influenced the interaction with knowledge sharing, as knowledge sharing did interact with the more collective focused dimensions. Knowledge sharing on itself was also found to be correlated to both interpretations of IT innovativeness and also to transformational leadership.
Limitations and Future Research Directions
This study had several limitations. The relatively small sample does not allow us to make any proper generalizations. This external validity is weak due to fact that this study only focused on virtual teams which operate in the Netherlands or are operating for a Dutch company abroad. Therefore these research findings are limited to this sample, because it does not take into account for instance the cultural differences or IT facilities of the virtual teams.
As we have seen the influences of the transformational leadership behaviors on the knowledge sharing within a virtual team, more research can be done upon the influences of different cultural aspects or availability of IT facilities within the virtual teams. Do these teams really provide a constructive climate; and which aspects of this climate do support the knowledge sharing and IT innovative behaviors of team members? With this, more specific managerial implications can be made based on the characteristics of the team to more effective apply certain transformational leadership behaviors.
Next to this, a more in-depth look at the crucial role of knowledge sharing could lead to more useful insights and managerial implications for virtual team leaders. This study only focused on the perceived knowledge sharing among team members, but does not include specific aspects or characteristics of this knowledge sharing.
Acknowledgement
37
6. References
Agarwal, R., and Prasad, J. (1998). A conceptual and operational definition of personal innovativeness in the domain of information technology. Information systems research. Vol. 9(2): 204-215.
Aiken, L.S., & West, S.G. (1991). Multiple regression: Testing and interpreting interactions.
Sage Publications, Incorporated.
Amabile, T.M. (1988). A Model of Creativity and Innovation in Organizations. Research in Organizational Behavior. Vol. 10: 123-167.
Amabile, T.M. (1997). Motivating Creativity in Organizations: On Doing What You Love and Loving What You Do. California Management Review. Vol. 40(1): 39-58.
Bass, B.M. (1985). Leadership: Good, Better, Best. Organizational Dynamics. Vol.13(3): 26-40.
Bass, B.M. (1999). Two Decades of Research and Development in Transformational Leadership. European Journal of Work and Organizational Psychology. Vol. 8(1): 9-32.
Berry, G.R. (2011). Enhancing Effectiveness on Virtual Teams Understanding Why Traditional Team Skills Are Insufficient. Journal of Business Communication, Vol.48(2): 186-206.
Börekçi, D.Y. (2009). Leader’s ICT usage’s influence on follower’s positive work attitudes via perceived leader-follower relations. Journal of Leadership & Organizational Studies. Vol.16(2): 141-158.
Brown, S.L., and Eisenhardt, K.M. (1995). Product development: past research, present findings, and future directions. Academy of management review, 343-378.
