Thesis
“The influence of status difference on the moderating effect of an
inclusive climate on innovative team behavior of functionally
diverse teams in higher education”
“Differences are not intended to separate, to alienate. We are different precisely in order to realize our need of one
another.” (Desmond Tutu)
Student : Justin Meerenburgh
Student number : 10684530
Date of submission : 31 March 2016
Version : Final version (1)
Course : Management studies – part time
Track : Leadership and Management
Institution : University of Amsterdam, ABS
Thesis supervisor : dr. C.K. Buengeler
2
Statement of originality
Statement of Originality
This document is written by Student Justin Meerenburgh who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Abstract
This team-level study focuses on the effect of functional diversity on innovative team behavior in the environment of higher education. Not only it is suggested in this study that the effect is positive, it is also hypothesized that the effect will even be stronger when team members experience a high level of climate inclusiveness. The effect is expected to decrease when the moderating effect of the inclusive climate is influenced by status difference in the team. In total 255 employees from 29 teams of Saxion University of Applied Science participated in this quantitative study. Saxion was selected as an organization since innovation is an important spear topic on their strategic agenda. Data was collected by requesting teams and their supervisors to fill out a questionnaire. Data analysis did not result in significant outcomes for the positive effect of functional diversity on innovative team behavior or for the expected influence of inclusive climate and status difference. However, the significant positive outcomes for the direct effect of an inclusive climate on innovative team behavior could help the supervisors of the university to create a working atmosphere that stimulates innovation. The significant outcomes on status difference provide us with an understanding that, within the context of the sample organization, perceived status difference by team members has a positive effect on the team’s innovative behavior. This finding is being supported by the absolute measure of status dispersion within the team.
Keywords: functional diversity; innovative team behavior; inclusive climate; status difference.
4
Acknowledgements
This thesis project is the final product which completes 2.5 years of studying at the
Amsterdam Business School, of the University of Amsterdam. A studyperiod that I
experience as an important milestone in my personal and professional development. It
was a major challenge to find the right balance betweenbeing a good student, a
dedicated employee and an excellent dad. The thesis period was an intense but
meaningful period. It gave me the opportunity to put the knowledge and skills I gained during the study period into practice, and I was also challenged to apply these to scenarios that had not been practiced. It was a challenge to get a deeper understanding of my own research topic, finding my way in managing my own research process and interpreting the results.
The successful completion of this study as well as my final thesis would not have been possible without the support of others. First, I would like to start by thanking the Management Team of the Hospitality Business School of Saxion for providing me with this opportunity. Thank you for believing in me. Second, I would like to thank dr. Buengeler of the University of Amsterdam. Thank you for your constructive feedback, support, and for sharing your knowledge. I truly admire the body of knowledge you possess. Third, I would like to thank my Saxion colleagues who I could ask for support and feedback when needed. Finally and most importantly, I would like to thank my wife, Ellen, and my son, Jonah, for their full support and love. I promise you, now it is family time!
5
Table of content
1. Introduction ... 6
2. Literature review ... 10
2.1 Functional diversity and innovative team behavior ... 10
2.2 Inclusive climate and status difference ... 16
3. Data and method ... 21
3.1 Sample and procedure ... 21
3.2 Measures ... 23
3.2.1 Functional Diversity ... 23
3.2.2 Innovative team behavior ... 24
3.2.3 Inclusive climate ... 26
3.2.4 Status difference ... 27
3.2.5 Control variables... 30
4. Results ... 33
5. Discussion ... 43
5.1 Limitations and future research ... 46
5.2 Managerial implications ... 49
6. Conclusion ... 51
7. References ... 52
6
1. Introduction
Over the years, groups in organizations have become more diverse and this trend still continues to evolve (Williams & O’Reilly, 1998; Knippenberg & Schippers, 2007). Differences between team members remain a challenge for organizations to manage and a struggle for researchers to conceptualize and study (Harrison & Klein, 2007), although educational or functional differences have become an important point of attention in the formation of teams within organizations (Van Knippenberg & Schippers, 2007). In this study the main effect of functional diversity on innovative team behavior is studied within the context of higher education. Higher education is currently being confronted with a constantly changing environment. In response to this constantly changing context The Onderwijsraad (2011) states in a press release that higher education plays a crucial role in laying the foundation for an innovative society. Furthermore, the Onderwijsraad (2011) states that it is the responsibility of higher education to create a climate in which students are ‘intellectually and professionally challenged’ and ‘develop creativity skills and innovative behavior’. However, if it is expected from the employees of these institutes to help develop these skills and behavior in students then it might be expected that the employees, and the teaching staff in particular, of these institutes are also challenged to exhibit such behavior? Since the innovative capacity of a team is of importance if it wants to survive in a constantly changing environment in an effective way (De Drue, 2002), universities should get an understanding of the elements that need to be in place, such as the composition of teams, in order to create an optimal climate stimulating innovative behavior. Therefore, this study is not only addressing the main effects of diversity but also focuses on models that consider potential
7 moderating variables explaining the main effects of diversity (Van Knippenberg & Schipper, 2007). One of the elements expected to influence the effect of functional diversity on innovative team behavior is the team’s inclusive climate. The literature on inclusive climate is currently growing but until recently was still experienced as being at its infancy (Shore et al., 2011). Although the positive effect of an inclusive climate for individuals in organizations is assumed, little is known on what causes this positive effect (Shore et al., 2011) or what might influence this positive effect. This study examines the effect of perceived status difference within the team on the potential moderating effect that an inclusive climate has on the relationship between functional diversity and innovative team behavior.
This study serves two purposes related to the understanding of what needs to be in place for teams to exhibit innovative team behavior. Firstly, it aims to find evidence for the positive effect of diversity on innovative behavior of teams in higher education, with a specific focus on the moderating effect of an inclusive climate on the relationship between functional team diversity and innovative team behavior. Secondly, it aims to find evidence for the negative effect of perceived status difference on the moderating effect of an inclusive climate, resulting in low degree of innovative team behavior by functionally diverse teams. These aims resulted in the development of the main research question. This study concentrates on the following main research question: “ To what extent does an inclusive
climate, influenced by within team status status difference, contribute to innovative team behavior of functionally diverse teams?”
8 Thus, in this study the relationship between functional diversity and innovative team behavior is studied, and the extent to which this relationship is influenced by, firstly, an inclusive climate and, secondly, the extent to which status difference has an influence on the effect of an inclusive climate (figure 1).
Saxion, University of Applied Sciences, is used in this case study as the sample organization. The organization is the biggest University of Applied Sciences in the east of the Netherlands, with locations in Enschede, Deventer and Apeldoorn. The board of directors of Saxion presents the university as having a growing interest in diversity and focus on innovative behavior. This increasing interest in diversity resulted in the establishment of the Diversity Workgroup. This study is conducted in close collaboration with the Chairman of the Diversity Workgroup, Mr. A. Dijkstra. Supervisors from all academies were approached by the chairman and the researcher and were requested to participate in this research with their team. These teams, academic teams or Knowledge Centre teams, are responsible for developing and providing education to students at Bachelor or Master level. In total twenty-nine teams and their supervisors took part, representing 8 out of 11 academies. The team
Functional team diversity Innovative team behavior Inclusive climate Figure 1.1: conceptual model.
Status difference
9 size varied from 2 to 22 team members. The supervisor of the teams and all team members were invited to participate and were requested to fill out an online questionnaire.
This study is structured as follows: chapter 2, which contains the literature review, follows this introduction. It will provide an overview of the most important findings from previously conducted studies related to the variables of this study and the proposed relationships. Chapter 3 provides a detailed description of the process of data collection within the organization of Saxion and the methods applied in this study. A quantitative strategy was most appropriate and questionnaires were developed as a data collection method to find hypothesis-related evidence. The results are briefly presented in chapter 4. Chapter 5 contains a discussion of the results, limitations and practical implications. Chapter 6 ends the thesis with a brief conclusion
10
2. Literature review
The literature review concentrates around studies conducted and literature exploring the following variables and the relationships between these variables: diversity, innovative team behavior, inclusive climate, and status difference. Work group diversity has raised the attention of different researchers from the field of behavioral science, however so far inconsistent results have been found on the effects of functionally diverse teams on team innovation (Hüttermann & Boerner, 2011). In diversity research the focus is often on demographic, societal or cultural diversity. Within this study the focus is on demographic diversity and in specific on the combination of two attributes of functional diversity: field of expertise and field of expertise related specialization. The phenomenon of inclusion in relation to diversity is currently still an underdeveloped area of research with limited agreement on the construct of inclusion (Shore et al., 2011) and its outcomes. Both workforce diversity and inclusive climate have been linked to status difference (e.g. DiTomaso, Post & Parks-Yancy, 2007; Ferdman & Deane, 2014) but as far as known not in combination with each other in one conceptual model, with an influence on innovative team behavior in the specific context of higher education.
2.1 Functional diversity and innovative team behavior
Diversity ‘is typically conceptualized as referring to differences between individuals on any attribute that may lead to the perception that an other person is different from the self’ (Williams & O’Reilly, 1998, Van Knippenberg & Schippers, 2007, p517). This means that the term ‘different’ can be interpreted in the broadest sense. It is about any dimension that differentiates individuals from each other (Noe, Hollenbeck, Gerhart & Wright, 2012).
11 Diversity dimensions that have been studied primarily are gender, race, ethnicity, tenure, educational background and functional background (Williams & O’Reilly, 1998; Van Knippenberg & Schippers, 2007). Within this study the focus is on the difference in job-related attributes between team members of the team. It concerns the functional diversity in the team based on field of expertise and the specialization within this field of expertise. Functional diversity is argued to be of importance since it ‘reflects a team member’s type of knowledge, as well as shaping a team member’s attitude and perspective (Bell, Villado, Lukasik, Belau & Briggs, 2011, p715). Within this study the aim is to research variety within teams based on functional diversity. Variety is the composition of differences, for example functional differences, between the members of the team (Harrison & Klein, 2007). A minimum state of variety is reached when there are no differences between team members in regards to particular attributes. All team members belong in this case to the same category. This is also referred to as a state of perfect homogeneity (Harrison & Klein, 2007). The maximum state of variety is reached when all team members belong to different categories in regards to the particular attribute (Harrison & Klein, 2007). The social categorization perspective and the information/decision making perspective are often used to explain the process and outcome difference between homogenous teams and heterogeneous teams. The social categorization perspective explains that ‘similarities and differences form the basis for categorization self and others into groups, distinguishing between similar in-group members and dissimilar out-group members’ (Van Knippenberg and Schippers, 2007, p517). The information/decision-making perspective is aimed at explaining the effects of diversity: ‘diversity may introduce differences in knowledge, expertise and perspectives that may help work groups reach higher quality and more
12 creative and innovative outcomes (Van Knippenberg and Schippers, 2007, p517). It is also proposed that the presence of a variety of perspectives and knowledge should make it possible for heterogeneous teams to outperform homogeneous teams (Bell et al., (2011). Outcomes of research indicate that, from the information/decision-making perspective, under ideal circumstances increased diversity might have a positive effect on the group’s performance but from the social categorization perspective diversity is likely to negatively influence group functioning (Williams and O’Reilly, 1998). Evidence is found for the positive performance oriented outcomes based on the information/decision-making perspective and the positive process outcomes based on the social categorization perspective. For example it has been proven that heterogeneous teams are more creative and are able to make more effective decisions in comparison to homogenous teams (Harrison & Klein, 2007). A possible explanation is that the uniqueness of the team members result in greater access to a more diverse pool of informational resources since the team members in a functionally diverse team differ in educational background, functional background and experience (Harrison & Klein, 2007). Homogenous teams are, in comparison to heterogeneous teams more cohesive, have a lower turnover and score higher on, which is expected to be the results of a more smoother group process and a higher satisfaction and attraction level to the group (Van Knippenberg & Schipper, 2007 P518).
In this study the effect of functional diversity on innovative team behavior is being researched. Team innovation is ‘the introduction and application in a team of ideas, processes, products, or procedures that are new to the team and are designed to be useful’ (Drach-Zahavy & Somech, 2013). Exposing innovative behavior, either as an individual or a team, is a continuous process. The output of the process does not necessarily need to be
13 innovative (Pieto & Pérez-Santana, 2014). This continuous process is multi-staged and goes beyond being creative. Creative behavior is aimed at the generation of new ideas. Innovative behavior extends this by also focusing on the implementation of these ideas (Woodman et al., 1993, Pieto & Pérez-Santana, 2014). The four stages of innovative behavior often used in studies are: 1) problem recognition, (2) idea generation, (3) idea promotion and (4) idea realization (Scott and Bruce, 1994; Dorenbosch, Van Engen and Verhagen, 2005). The stages of problem recognition and idea generation are both elements of the creativity-oriented behavior dimension. The stages of idea promotion and idea realization are elements of implementation-oriented behavior dimension (Dorenbosch, Van Engen and Verhagen, 2005).
In previous conducted studies various types of diversity, including functional diversity, are being linked to elements of innovative behavior. For example high demographic diversity, of which functional specialty was mentioned as one of the diversity variables, positively
influences problem solving abilities and creativity, where as low demographic diversity positively influences attraction to the group, commitment, ability to members’ needs, social integration, problem solving and implementation abilities (Williams and O’Reilly, 1998). Researchers agreed as well upon the positive effect of functional or background diversity on problem solving because diversity results in an increase of the range of knowledge, skills and contacts (e.g.; Pelled, Eisenhardt & Xin, 1997; Williams and O’Reilly, 1998). Functional
background diversity was also found to have a stronger positive effect on creativity and innovation than on efficiency (Bell et al. 2011). Finally a weak positive correlation was found between functional diversity and innovation (Hülsheger, Anderson & Salgado, 2009). And although the effect shows that functional diversity is conducive to innovative team behavior,
14 due to the weak effect it can be wondered which moderators are of influence on this
relationship (Hülsheger, Anderson & Salgado, 2009).
The management style of the supervisor is expected to have a positive effect on the relationship between functional diversity and innovative team behavior. Previous research confirmed already the positive effect of inclusive leadership on psychological safety and on employee involvement in creative work tasks (Carmeli, Reiter – Palmer & Ziv, 2010) as well as on employee creativity (Choi, Tran & Park, 2015). Inclusive leaders create an climate in which team members experience that the supervisor is ‘available, accessible and open in the interaction’ to them (Carmeli, Reiter – Palmer & Ziv, 2010). This could result in team members sharing their perspectives since they experience the safety of the climate and the support of their supervisor. The supervisor has the opportunity to collect and share the diversity of perspectives in collaboration with the team in order to improve the performance of the team. Previously conducted research already found support for the moderating effect of charismatic / transformational leaders on the relationship between education
specialization diversity and team creativity (Shin & Zou, 2007). Organizational citizenship behavior (OCB) was also found to be an influencing factor in the development of innovative behavior. A positive direct relationship was already found between these two variables (Harari, Reaves & Viswesvaran, 2016). It is expected that also in functionally diverse teams OCB has a positive influence on the team’s innovative behavior. It is being argued that OCB is an important asset for building social capital, which stimulates the development of sharing perspectives with the team (Harari, Reaves & Viswesvaran, 2016). A final potential
15 to cause lower turnover rates in demographically diverse team because of the high quality relationship between the team members and the supervisor and the focus on inclusion. (Nishii, Mayer & Kozlowski; 2009).
It is expected that diversity might not have a positive influence on innovative team behavior at all times. When the diversity in a team becomes too high it negatively influences group performance and group functioning (Williams and O’Reilly, 1998). Also the degree to which a team is diverse might influence the extent to which innovative behavior is shown by the team. A state of moderate variety in the team, opposed to maximum variety, increases the risk of coalition forming within the team (Harrison & Klein, 2007). Coalition formation might result in the formation of in-groups and out-groups and in addition to perceptions of high status and low status between groups. Status difference within teams could cause feelings of mistrust and group conflicts (Brewer, 1999; Nishii 2013) for example because information is not shared any more between the subgroups (Harrison and Klein, 2007). But also a state of high variety has its risks. Knowledge transfer might not be successful when experts use jargon and specialized language from their own field of expertise, which is not collectively shared within the team (Maznevski, 1994, Horwitz & Horwitz, 2007).
Although the effect of diversity has been researched in different studies, diversity researchers still struggle with the external validity of the found causalities of previous conducted laboratory experiments on the information/decision-making perspective and the social categorization perspective (Williams and O’Reilly, 1998). The main effect of functional diversity on innovative team behavior in the context of higher education will be researched first. It is expected that the organizational context will have an influence on this relationship.
16 Also because of the interest in moderating effects in the context of higher education it is of importance to first study the main effect. As a result this study will start by researching the following hypothesis:
Hypothesis 1: Functional team diversity positively influences innovative team behavior
2.2 Inclusive climate and status difference
In recent studies the difference between the concepts of ‘diversity management’ and ‘inclusion’ is at debate. Research has been conducted to determine the difference between both constructs. The outcome of this study suggest that ‘diversity may be managed through a variety of methods. More specifically, diversity in organizations may be supported by sets of practices to manage fair treatment issues, increase stakeholder diversity, and demonstrate leadership’s commitment to diversity, while inclusion may be supported by practices to integrate diversity into organizational systems and processes and encourage the full participation and contribution of employees’ (Roberson, 2004, p25). This distinction is being supported in this study. Inclusion has been defined as ‘the degree to which an employee perceives that he or she is an esteemed member of the workgroup through experiencing treatment that satisfies his or her needs for belongingness and uniqueness’ (Shore et al., 2011, p1265). A state of inclusion can be accomplished when team members experience high perceptions of belongingness and uniqueness (Shore et al., 2011). Team members then have the feeling of being treated as an insider in the team and that a team member’s unique features are accepted within the team since these are seen as learning opportunities resulting from the team’s diversity (Shore et al., 2011, Nishii, 2013). Recent studies resulted in a measure for inclusive climate based on three dimensions. The first
17 dimension is ‘fairly implemented employment practices’ referring to perceptions of fairness within in the team in regards to the implementation of HR and diversity practices. Biased practices in an organization or in teams result in perpetuating demographically based status differentials (Nishii, 2013). The second dimension is the ‘integration of differences’ of the diversity of members of the team that is related to the feeling of openness and the safety to be open without suffering unwanted consequences Nishii 2013). The third dimension is ‘inclusive decision making’ which refers to the extent to which within the team perspectives are being brought together regardless the effect of a perspective on the teams ‘status quo’. (Ely & Thomas, 2001; Mor Barak & Cherin, 1998; Nishii, 2013). The construct of an inclusive climate in this study is built upon these three dimensions. Previous research confirmed the interaction between the dimensions (Nishii, 2013). This means that a state of ‘high’ perceptions of inclusive climate is being reached when team members experience within the team an equal distribution of practices, the integration of differences, and inclusive decision-making. A state of ‘low’ perceptions of an inclusive climate is reached when the average score on the three dimensions can be interpreted as low. Since these dimensions interact it is expected that when one of the dimensions is perceived as low, also the other two dimensions are perceived as low.
It has been proposed that high perceptions of an inclusive climate positively influences creative behavior as an outcome (Shore et al., 2011). In a study on the indirect effect of cultural diversity on team creativity via team information exchange, an inclusive climate proved to have a positive moderating effect of cultural diversity on team information exchange (Li, Lin, Tien & Chen, 2015). An inclusive climate has the ability to foster interpersonal trust within the team, stimulating interpersonal communication between team
18 members (Ensari & Miller, 2006; Li, Lin, Tien & Chen, 2015). It is expected that team members in teams with a strong inclusive climate are more open to voice their opinions, share perspectives and feel save to speak up, which results, inline with the information-/decision making perspectives, in teams reaching higher quality levels and more creative and innovative outcomes (Van Knippenberg and Schippers, 2007) in comparison to teams with a weak inclusive climate. This assumption is tested through the following hypothesis:
Hypothesis 2: High perceptions of an inclusive climate positively influences the relationship
between functional diversity and innovative team behavior.
Functionally diverse teams exist out of a heterogeneous group of team members. Within the teams it might occur that team members are being experienced as more influential than others as a result of their gained status position. In different studies the term ‘status’ is conceptualized. Status is defined as group member’s relative standing in a group based on prestige, honor, and deference (Berger, Cohen, and Zelditch, 1972; Thye, 2000). But it is also being defined as a team member’s individual’s social standing or rank order among others within the social system, which is based on prestige, prominence, and respect (e.g., Berger, Cohen & Zelditch, 1972). According to the status characteristics theory a team member’s rank in the group (status) depends on the performance expectations that other team members have based on the observable traits of the particular member (Thye, 2000). These traits can be diffuse or specific. Diffuse traits create expectations about a task fulfillment in general based on for example a team member’s gender, age, education or occupation. Specific traits create expectations based on for example a team member’s specific or expertise (Thye, 2000). Status in a team based on subjective performance expectations
19 between team members can be defined as the informal hierarchy within the organization (Magee & Galinsky, 2008). The formal hierarchy in a team is based on the formalization of the organization of which the team is a part and is based for example on given job titles, position in the organizational chart or the communication structure (Magee & Galinsky, 2008). Both formal and informal hierarchy influence perceptions of status difference within
a team.As stated before, perceptions of status difference might be a source for mistrust and
group conflict (Brewer, 1999; Nishii 2013). It could also cause feelings of superiority versus inferiority resulting in low status members underestimating the value of their input, self-censor their contributions, and let higher status members make decisions (Bienefeld & Grote, 2014). Also a positive relationship between status difference and reluctance to speak up by low status members was found in previous conducted studies. Low-status members might not speak up because they fear negative social repercussions or personal consequences (Bienefeld & Grote, 2014). On the other hand it can be argued to what extent status difference is being perceived within a team. Status difference within a team is perceived as fair and legitimate as soon as the classification is based on ‘appropriate and agreed upon means and when higher ranked team members do not abuse their position’ (Magee & Galinsky, 2008). A team member’s status seems to play an inferior role when, based on the learning-and-integration perspective (Ely & Thomas, 2001), a team has a deeper commitment in regards to learning from each other. This perspective is initially aimed at seeing cultural difference as a valuable source because of the different insights, knowledge and experiences to be used for the team’s performance (Ely & Thomas, 2001). The bottom line of this perspective can be interpreted as that team members can expand their knowledge by being open to learn from those who are different from the self. It could
20 potentially be the case that in the context of higher education, a clear learning environment, status is inferior to the shared goal of creating knowledge. In heterogeneous teams in higher education, like in this study, it might be expected that some team members are seen as more knowledgeable or are perceived as an expert. Based on the status characteristics framework (Berger, Cohen and Zelditch, 1972, Bunderson, 2003) team members high in performance expectations are higher in status than other team members. The experts are being approached by the team members when certain decision have to be made are when collaborative action with the expert is required to accomplish the team goals (Bunderson, 2003). It is expected that in teams existing out of team members having their own expertise and specialization status difference does exist but is task related. The task-related expert is expected to be given more opportunities to participate and to have a higher influence in the team’s decision-making process and outcome (Bunderson, 2003). Researchers have suggested that there is a relationship between status difference and an inclusive climate. It is assumed that insider status associated practices as sharing information, participation in decision-making, and having voice are used in measures of inclusion (Shore et al., 2011). This suggestion is being supported in this study. It is expected that status difference does exist in teams in higher education, despite the learning environment, and that it has a negative effect on perceptions of the strength of an inclusive climate resulting in a less innovative team behavior of functionally diverse teams. The expectation is translated into the following hypothesis:
Hypothesis 3: High perceptions of status difference negatively influence the moderating
effect of an inclusive climate on the relationship between functional diversity and innovative team behavior.
21
3. Data and method
3.1 Sample and procedure
The case study can be defined as an explanatory study with the purpose of explaining the causal relationship between variables (Saunders, Lewis & Thornhill, 2012). The research strategy is quantitative of nature. The selected context of the case study is higher education and in specific Saxion, University of Applied Science. Saxion is situated in the east of the Netherlands in the cities of Apeldoorn, Deventer and Enschede. Saxion counts 11 academies across 3 locations. The institute offers a total of 85 courses on a bachelor or master level. Teams are responsible for the development and execution of these courses. Some teams are responsible for more than one course. Input for the development of these courses is also received from the research conducted by the academy’s Knowledge Centre(s). The KCs are
responsible for conducting research in collaboration with stakeholders, Saxion’s internal and
external environment and the integration of research in educational programs. Academic teams as well as Knowledge Centres are selected to participate in this study since both play an important role in the development of study courses for which innovative team behavior is required.
A team is expected to be a team when it exists out of at least two members. A total minimum of 30 teams was required (n>30). A total of 35 teams were approached and requested to participate. The supervisors of the teams were contacted by mail by the researcher and the chairman of the diversity workgroup. The email contained a request to participate and an introduction to this study (appendix 1). The researcher gave a more in-depth explanation on the research process once the supervisor confirmed the participation of the team. Furthermore, the supervisors were asked for their preference in regards to the
22 distribution of digital or printed versions of the questionnaires. One team and one supervisor requested for the printed version, while all other supervisors and teams participated online.
In addition, separate questionnaires were created for supervisors and team members (appendix 2, 3, 4 and 5). The questionnaires were created in English and translated to Dutch with the support of academic English lecturers. A test was performed amongst three individuals. This resulted in a few minor changes in regards to the interpretation of questions. Supervisors or secretary offices of the participating academies were requested to send an overview of the teams to the researcher. The overview stated the names of the teams, the particular supervisor related to the team, the first name, the family name and official position of all team members. Based on this list emails were sent out to the Saxion email addresses of the team supervisors and team members. The email contained a short introduction and the link to the Dutch and English version of the questionnaire. A response time of two weeks was given. During this two-week period the participating teams received two reminders with an aim to stimulate participation amongst the team members of the participating teams.
Some of the respondents knew the researcher in person, therefore it was emphasized in the introduction of the invitation to participate as well as in the questionnaire that the information provided by the participants is used only for the purpose of this study, that all information will be kept strictly confidential and that results will not allow any conclusions regarding the answers of individuals, singular teams or particular team supervisors. Anonymity of the respondents is guaranteed. Moreover, two incentives were offered for
23 filling out the questionnaire to stimulate participation. First, all participant had the possibility to participate in a raffle. Second, all participants had the possibility to receive a summary of the outcomes of the research.
In total 31 supervisors participated in this study. The data of two supervisors was removed from the dataset while one of the questionnaires was highly incomplete. The results of the other supervisor could not be related to a team. A total of 255 team members participated, of which 244 members filled out the online questionnaire and 6 members the hard copy version. These team members represented 29 teams. A total of five participants were removed from the data set as these respondents did not fill out the demographics and therefore could not be assigned to a team. Missing scores were indicated with the score ‘999’. Two outliers were detected in the question related to the respondent’s age. These scores where removed from the dataset and treated also as missing scores. The response rate per team ranged from 30 to 100 percent. The size of participating teams ranged from 2 to 22 team members.
3.2 Measures
3.2.1 Functional Diversity
Functional diversity is defined in this study as a team member’s expertise and specialization within this field of expertise. Team members were requested to answer the open questions ‘What is your expertise?’ and ‘What is your specialization within your field of expertise?’. An example was given in order to ensure right answering of the questions (e.g. When your field of expertise is 'Marketing', your specialization might be 'on line marketing'). Diversity in the teams is calculated by applying Blau’s index of heterogeneity (1977). The sample includes
24 teams of different sizes and therefore the adjusted formula of Blau’s index has been applied to calculate diversity levels within the participating teams (Biemann &Kearney (2010); Buengeler & Den Hartog (2015):
𝐵𝑙𝑎𝑢𝑁 = 1 − ∑ (𝑁𝑖
𝑘 𝑖=1
(𝑁𝑖 − 1)/𝑁(𝑁 − 1)
In the formula Ni represents the absolute number of team members in the ith diversity category and N represents the total number of members on a team (Buengeler & Den Hartog, 2015). The outcome values of Blau’s index can range from zero, indicating only team members from the same category, to one, indicating only team members from different categories (Harrison & Klein, 2007). Diversity categories on field of expertise and specialization were not created upfront because of the different industries represented by the academies. Diversity categories were determined per team while analyzing the data of the different teams. Categories were made by grouping team members based on similarity in specialization (e.g. languages, marketing, economics). If the similarity in specialization was ambiguous then the fields of expertise were being compared.
3.2.2 Innovative team behavior
Innovative team behavior was studied based on three different measures from both the perspective of the team members and the supervisor. The measures are used to study innovative team behavior from different perspectives. The first measure has a focus on the different dimensions of innovative behavior. The second measure is aimed at researching the team’s innovative behavior related to past work. With the third measure a comparison is made between the respondent’s team and other teams.
25 The first measure (ITB1) measured the team member’s creativity oriented work behavior and implementation oriented work behavior. These two dimension of innovative behavior dimensions are directly related to the four stages of innovative behavior: (1) problem recognition, (2) idea generation, (3) idea promotion and (4) idea realization (Dorenbosch, Van Engen & Verhagen, 2005). Problem recognition and idea generation are stages of creative oriented work behavior. Idea promotion and idea realization are stages of implementation oriented behavior. Questions on creativity-oriented work behavior were for example ‘To what extent do you try to detect impediments to collaboration and
coordination?’ and ‘To what extent do you generate new solutions to old problems?’. Questions on implementation-oriented behavior were for example ‘To what extent do you mobilize support from colleagues for your ideas and solutions?’ and ‘To what extent do you transform new ideas in a way that they become applicable in practice?’. The
creativity-oriented work behavior existed out of 10 items. The implementation-creativity-oriented work behavior existed out of 6 items. Some questions required being slightly adapted to the educational context of this study. Previous studies confirmed the high inter correlation between the creativity oriented dimension and the implementation oriented dimensions and therefore the two dimensions are not treated as two separate variables but as one (Scott & Bruce, 1994; Janssen, 2000; Dorenbosch, Van Engen & Verhagen (2005). The self-ratings measured innovative behavior on an individual level. The items from both dimensions were measured on a five-point Likert scale (from 1=to a very little extent to 5=to a very large extent). The individual scores on the dimensions of innovative behavior were used to calculate the mean score per team. Similar questions were asked to the supervisor (ITB4). In these questionnaires ‘To what extent do you…’ was replaced by ‘To what extent do team members
26
in this team…’. All items of the supervisor-rating questionnaire were measured on a
five-point Likert scale (from 1=to a very little extent to 5=to a very large extent) as well. Cronbach’s alfa for the self-rating was .871 and for the supervisor rating .899.
The second measure both the team member (ITB2) and the supervisor (ITB5) were requested to respond to 4 items of innovative team behavior while keeping in mind the teamwork of the past six months and trying to review performance data (Drach-Zahavy & Somech, 2001). Items of innovative team behavior were for example ‘The team initiated new
procedures and methods’ and ‘The team developed innovative ways of accomplishing work targets/objectives’. The items in both questionnaires were measured on a five-point Likert
scale (1=to a very little extent to 5=to a very large extent). Cronbach’s alfa of the self-rating was .875 and of the supervisor rating .912.
The third measure was aimed at comparing the innovative behavior of the own team to similar other teams (West, 1987). Both the team members (ITB3) and the supervisor (ITB6) were requested to compare the team to other teams on 5 items. Examples of these items are ‘Setting work targets and objectives’ and ‘Developing innovative ways of accomplishing
targets/objectives’. The items in both questionnaires were measured on a five-point Likert
scale (1=highly stable/few changes introduced to 5=highly innovative/many changes introduced). Cronbach’s alfa of the self-rating was .856 and of the supervisor rating .935.
3.2.3 Inclusive climate
The first moderator included in the conceptual model is ‘inclusive climate’. The construct of ‘inclusive climate’ exists out of three dimensions. The first dimension is ‘fairly implemented
27 employment practices’ referring to perceptions of fairness within the team in regards to the implementation of HR and diversity practices (Nishi, 2013). The second dimension is the ‘integration of differences’ of the diversity of members of the team, which is related to the feeling of openness and the safety to be open without suffering unwanted consequences (Nishii 2013). The third dimension is ‘inclusive decision making’ which refers to the extent to which perspectives in a team are being brought together regardless the effect of a perspective on the team’s ‘status quo’ (Ely & Thomas, 2001; Mor Barak & Cherin, 1998; Nishii, 2013). The 15-item shorted version of the existing inclusive climate measure of Nishii (2005) was used. Inclusive climate was measured as a supervisor rating and a self-rating. It measured the respondent’s perception on the three dimensions of the team’s inclusive climate. Items were related to the three dimensions: Foundation of equitable employment practices (e.g. ‘The team has a fair promotion process’), integration of differences (e.g. ‘This
team is characterized by a non-threatening environment in which people can reveal their “true” selves’), and inclusion in decision-making (e.g. ‘In this team, employee input is actively sought’). The measure of dimension one existed out of 5 items. The measure of dimension two existed out of 6 items. The measure of dimension three existed out of 4 items. All items were measured on a five-point Likert scale (from 1=strong disagree to 5=strongly agree). Cronbach’s alfa of the self-rating was .926 and of the supervisor rating .858.
3.2.4 Status difference
The second moderator in the conceptual model is ‘status difference’. Status difference is measured in two ways: the absolute and relative status difference. The absolute measure of status difference concerns the dispersion of power within the team and the team’s power level (Greer & Van Kleef, 2010). The dispersion of power within the team is the
28 concentration of power amongst team members (Harrison & Klein, 2007; Greer & Van Kleef, 2010). The team’s power level is the extent to which a group is able to modify the state of others in the organization (Greer & Van Kleef, 2010). The participating supervisors or the related secretary offices are requested to send an overview of the team members of the participating teams. This overview also contains the job title of every member. The different job titles are related to the ordinal scale (1 = student assistant, 2 = education and research officer 2, 3 = educational and research officer 1, 4 = lecturer/researcher 2, 5 = lecturer/researcher 1, 6 = Senior Lecturer, 7 = Professor). The average power level of the group will be calculated based on the ordinal scale. Power dispersion will be operationalized by calculating the standard deviation of the members’ levels of power within the particular group (Chan, 1998; Greer & Van Kleef, 2010). The two measures are going to be standardized and multiplied in order to create the overall interaction term (Aiken & West, 1991; Greer & Van Kleef, 2010).
The relative measure is based on the respondent’s perception of status dispersion. A quantitative measure on the perception did not exist. Therefore, a measure was developed. The social dominance literature served as the fundament and items proposed to form the SDO-dominance dimensions (Ho et al., 2012) were transformed into questions measuring the perception on status difference from both the supervisor and team member perspective. The reliability of the SDO-dominance measure can be interpreted as high. Across samples Cronbach’s alfa ranged from .810 to .920 (Ho et al., 2012). In total 10 items measured the respondent’s perception of status difference (SDI1). Items were, for example, ‘Some team
29
the top and other team members are at the bottom’. Some items required adaption to
measure the supervisor’s perception on status dispersion in the team. All items were measured on a seven-point Likert scale (from 1=strong disagree to 7=strongly agree). Cronbach’s alfa of the self-rating was .916 and of the supervisor rating .666.
A construct was developed for both the supervisor rating and the self-rating to measure perceptions of status difference in the team (SDI2) and to measure the extent to which the power difference is perceived as fair (SDI3). Respondents were requested to indicate on 7 items the extent to which a certain type of status difference was perceived to occur in the respondent’s team (e.g. ‘Team members who have been in this team for a longer period of
time are higher in status than team members who recently joined the team’) and if the
respondent perceived this type of status difference as fair (e.g. ‘It is fair that team members
who have been in this team for a longer period are higher in status than team members who recently joined the team’). All items of both measures were measured on a five-point Likert
scale (from 1=strong disagree to 5=strongly agree). Cronbach’s alfa for the perceived difference of the self-rating was .807 and of the supervisor rating .813. Cronbach’s alfa of the self-rating of the perceived fairness of status difference was .800 and of the supervisor rating .748. A factor analysis was conducted to test the factor loadings. KMO values of the self-rating measures can be interpreted as sufficient (status dispersion measure = .913, status dispersion in the team = .802, fairness of status difference in the team = .808). All items of all three self-rating measures load on one factor. KMO values of the supervisor rating measures can be interpreted as weak to just sufficient (status dispersion measure = .454, status dispersion in the team = .609, fairness of status difference in the team = .656).
30 The low(er) values are being accepted, as these are most likely the result of the small sample size.
3.2.5 Control variables
Additional variables for both employees and supervisors were added to the questionnaires. It is suggested that these variables might have an influence on relationships within the proposed model. Measures of charismatic leadership, inclusive leadership and task interdependence were specifically added to the team member questionnaire. Measures of organizational citizenship behavior (OCB) and the overall performance of the team were added to the supervisor questionnaire. Measures of leader-membership exchange (LMX) and diversity beliefs were added to both the supervisor and team member questionnaire.
Charismatic leadership was measured as a self-rating. It measures the perception to which a team member perceives his supervisor as a charismatic leader (De Hoogh, Den Hartog & Koopman, 2005). The measure existed out of 11 items. Respondents were requested to respond to statements such as ‘My supervisor has a vision and imagination of the future’ and
‘My supervisor mobilizes a collective sense of mission’. All items were measured on a
seven-point Likert scale (from 1=strong disagree to 7=strongly agree). Cronbach’s alfa was .950.
Inclusive leadership was measured as a self-rating, measuring the extent to which the team member perceives the supervisor applying this leadership style (Carmeli, Reiter-Palmon & Ziv, 2010). It measures the three domains of inclusive leadership: openness (e.g. ‘My
supervisor is open for hearing new ideas’), accessibility (e.g. ‘My supervisor is available for consultation on problems’), and availability (e.g. ‘My supervisor is accessible for discussing
31
emerging problems’). All items were measured on a five-point Likert scale (from 1=not at all
to 5=to a large extent). Cronbach’s alfa was .946.
Task interdependence was as well measured as a self-rating (Van der Vegt & Janssen, 2003). The measure existed out of 5 items. Respondents were requested to respond to statements such as ‘I have a one-person job’ and ‘I need to collaborate with fellow team members to
perform my job well. All items were measured on a five-point Likert scale (from 1=strong
disagree to 5=strongly agree). Cronbach’s alfa was .783.
The team’s organization citizenship behavior (OCB) was measured as a supervisor rating (Wong, Tjosvold & Liu, 2009). The OCB measure existed out of 21 items (e.g. ‘This team is
eager to tell outsiders good news about the organization and clarify their misunderstandings’) of which 7 items needed to be reversed coded (e.g. ‘This team often speaks ill of other teams behind their back’). All items were measured on a five-point Likert
scale (from 1=strongly disagree to 5=strongly agree). Cronbach’s alfa was .821.
The team’s overall performance was also measured as a supervisor rating (Van der Vegt & Bunderson, 2007). Supervisors were requested to rate the performance of the team they supervise on 6 items (e.g. the team’s ‘efficiency’ and ‘effectiveness’). All items were measured on a seven-point Likert scale (from 1= far below average to 7=far above average). Cronbach’s Alfa was .844.
32 Measures of the next variables were added to the both the supervisor and the team member questionnaire. A measure for Leader-Membership exchange (LMX) was added (Liden, Wayne & Stilwell, 1993). The questions of team members were aimed at the relationship between the respondent and the supervisor of his/her team (e.g. ‘Regardless of
how much power he/she has built into his/her position, my supervisor would be personally inclined to use his/her power to help me solve problems in my work’ and ‘I usually know how satisfied my supervisor is with me’). The same items were used in the supervisor
questionnaire but were adjusted to the supervisor perspective (e.g. ‘Regardless of how much
power I have built into my position, I would be personally inclined to use my power to help my team member solve problems in his or her work’ and ‘I think that I recognize my team member’s potential’). The questions answered by the supervisor were aimed at the
relationship of the respondent with individual employees of the team the particular respondent is supervising. All items were measured on a seven-point Likert scale (from 1=strongly disagree to 7=strongly agree). Cronbach’s alfa of the self-rating was .936 and of the supervisor rating .700.
In both questionnaires the respondent’s diversity beliefs are measured as a self-rating (Homan, 2004a; Homan et al., 2007). It existed out of 4 items measuring diversity beliefs on a five-point Likert scale (from 1=strongly disagree to 5=strongly agree). Respondents were requested to respond to statements such as ‘Diversity is an asset for teams’ and ‘I believe
that diversity is good‘. Cronbach’s alfa of the self-rating was .920 and of the supervisor rating
33
4. Results
Before interpreting the results normality was checked for all the variables and data was being analyzed on outliers and abnormal scores. Outliers were not detected. Two abnormal scores were detected in the age demographics’ of the respondents. The scores were removed from the dataset and the empty cells were treated as missing values. On a team level all variables showed a normal distribution, except for the control variable ‘diversity beliefs’. This variable was strongly negatively skewed. Removing items did not result in improvement of the normal distribution. Therefore it was decided to exclude this variable from this study. On a supervisor level not all scores were normally distributed. Signs of positive or negative skewness were accepted. The small sample size (n= <30) is expected to be the cause of the non-normal distributions. Despite the small sample size, the data was checked on linearity and homoscedasticity (appendix 5). The criteria for homoscedasticity and linearity are being met since the graphs do not show signs of curves or funneling out. These observed signs potentially indicate that a systematic relationship between the errors and the model’s predictions cannot be detected (Field, 2013). Finally the data was checked on signs of multicollinearity (appendix 6). The Variation Inflation Factor scores (VIF) of all variables of the conceptual model are well below five, meaning that multicollinearity is not an issue for these variables. This is confirmed by the tolerance statics. All tolerance scores are above 0,2. Next a correlation analyses was conducted to get a first understanding of the outcomes of this study. The outcome of the correlation analyses is presented from a team perspective (self rating) (table 4.1 and 4.2) and from the supervisors’ perspective (Table 4.3 and 4.4) The correlation matrix shows the variables’ mean, standard deviation, alpha coefficient, and the correlations including its significance. The team’s correlation matrix
34 based on the self-ratings reveals the following significant correlations in regards to the main variables: there is a significant positive correlation between inclusive climate and both ITB2
and ITB3. Both ITB5 and ITB6 have a significant positive correlation with SDI11. Finally, there
is a significant negative correlation between SDI1 and inclusive climate. The correlations based on the supervisors’ ratings reveals a significant positive correlation between inclusive climate and ITB4, ITB5 and ITB6. Both correlation matrixes did not provide any evidence for a relation between functional diversity and any of the ITB variables on a team level or on a supervisors’ level. None of the found correlations was positively or negatively significant.
The correlation matrix of the team based on the self-ratings reveals also other significant correlations between main variables and control variables. There is a significant positive correlation between charismatic leadership and ITB1 and charismatic leadership and ITB2. The correlation between performance (supervisor rating) and ITB1, ITB2, and ITB3 is significantly positive as well. There is also a significant positive correlation between LMX and inclusive climate. The correlation between gender diversity and SDI1 and SDI2 is significantly negative.
1 SDI1 = Status difference measure 1 rating), SDI2 = status difference measure 2
(self-rating), SDI3 = status difference measure 3 (self-(self-rating), SDI4 = status difference 1 (supervisor rating), SDI5 = status difference measure 2 (supervisor rating), SDI3 = status difference measure 3 (supervisor rating).
35 Team M SD α 1 2 3 4 5 6 7 8 9 10 1. Tenure 4,456 0,262 - - 2. Team size 8,52 6,145 - 0,288 - 3. Age div. 0,7489 0,198 - 0,144 -0,215 - 4. Gender div. 0,499 0,287 - 0,128 -0,184 0,34 - 5. LMX 5,302 0,604 0,936 0,106 -0,111 0,068 0,379* - 6. Charism. ls. 5,225 0,652 0,950 0,341 0,167 -0,032 0,097 0,710** - 7. Incl. ls. 3,963 0,427 0,946 0,223 0,142 -0,094 0,089 0,782 0,386** - 8. Task int. 3,965 0,291 0,783 0,067 0,081 -0,364 -0,200 -0,205 0,018 -0,104 - 9. Perform. 4,747 0,841 0,844 0,092 0,119 -0,159 -0,100 0,091 0,261 0,203 0,034 - 10. Funct. div. 0,768 0,297 - 0,401* 0,168 0,470* 0,399* 0,203 0,047 0,185 -0,449* -0,133 - 11. ITB (1) 3,326 0,222 0,871 -0,140 0,038 -0,303 0,272 0,247 0,073 0,06 0,313 -0,234 -0,128 12. ITB (2) 3,287 0,385 0,875 0,353 0,252 -0,031 0,271 0,295 0,438* 0,185 0,08 0,186 0,046 13. ITB (3) 3,077 0,367 0,935 0,384* 0,060 0,140 0,293 0,476** 0,641** 0,389 0,061 0,214 -0,05 14. ITB (4) 3,532 0,508 0,899 0,275 0,018 -0,081 0,034 0,114 0,316 0,092 0,107 0,603** -0,185 15. ITB (5) 3,44 0,875 0,912 0,121 0,313 -0,111 -0,219 -0,06 0,304 0,090 0,298 0,520** -0,203 16. ITB (6) 3,145 0,935 0,935 0,136 0,090 -0,099 0,006 0,063 0,251 0,121 0,091 0,678** 0,044 17. Incl. climate 3,55 0,338 0,926 -0,020 -0,215 0,014 0,365 0,721** 0,512 0,44 -0,028 0,231 -0,116 18. SDI1 4,058 0,892 0,916 -0,001 0,414* -0,143 -0,439* -0,324 0,167 0,046 0,236 0,191 -0,119 19. SDI2 2,714 0,531 0,807 -0,236 0,239 -0,319 -0,499* -0,322 0,041 -0,139 0,17 -0,016 -0,326 20. SDI3 2,585 0,418 0,800 -0,280 0,152 -0,309 -0,357 -0,045 0,135 0,044 0,087 0,179 -0,194 21. Status dispersion 4,8599 0,79305 - -0,078 0,317 -0,276 -0,201 0,008 0,343 0,102 0,161 0,394* -0,256
36 Team M SD α 11 12 13 14 15 16 17 18 19 20 21 11. ITB (1) 3,326 0,222 0,871 - 12. ITB (2) 3,287 0,385 0,875 0,366 - 13. ITB (3) 3,077 0,367 0,935 0,210 0,670** - 14. ITB (4) 3,532 0,508 0,899 -0,024 0,267 0,496** - 15. ITB (5) 3,44 0,875 0,912 0,082 0,395* 0,356 0,643** - 16. ITB (6) 3,145 0,935 0,935 -0,021 0,214 0,279 0,758** 0,777** - 17. Incl. climate 3,55 0,338 0,926 0,295 0,542** 0,620** 0,310 0,106 0,090 - 18. SDI1 4,058 0,892 0,916 -0,056 -0,128 -0,031 0,277 0,536** 0,384* -0,435* - 19. SDI2 2,714 0,531 0,807 0,017 -0,141 -0,199 0,185 0,192 0,113 -0,370* 0,724** - 20. SDI3 2,585 0,418 0,800 0,085 -0,096 -0,144 0,274 0,362 0,280 -0,051 0,571** 0,636** - 21. Status dispersion 4,8599 0,79305 - 0,045 0,271 0,066 0,410* 0,482* 0,448* 0,003 0,458* 0,632** 0,636** - Table 4.2: Correlation matrix Teams part 2. N = 29 teams. Means, standard deviations, alpha coefficients and correlations among variables. *p<0,05, **p<0,01
37 Supervisor M SD α 1 2 3 4 5 6 7 8 1. Tenure 4,456 0,262 - - 2. Team size 8,52 6,145 - 0,288 - 3. Age div. 0,7489 0,198 - 0,144 -0,215 - 4. Gender div. 0,499 0,287 - 0,128 -0,184 0,340 - 5. LMX 5,823 , 575 0,700 0,107 -0,074 -0,084 0,222 - 6. OCB 3,842 0,384 0,821 0,324 0,061 0,099 -0,134 0,337 - 7. Overall performance 4,747 0,841 0,844 0,435* 0,119 -0,159 -0,100 0,488** 0,541** - 8. Functional diversity 0,768 0,297 - -0,188 0,168 0,470* 0,399* -0,085 -0,109 -0,133 - 9. ITB (4) 3,532 0,508 0,899 0,157 0,018 -0,081 0,034 0,560** 0,364 0,603** -0,185 10. ITB (5) 3,44 0,875 0,912 0,399* 0,313 -0,111 -0,219 0,313 0,527** 0,520** -0,203 11. ITB (6) 3,145 0,935 0,935 0,324 0,090 -0,099 0,006 0,555** 0,540** 0,678** 0,044 12. Inclusive climate 3,763 0,487 0,858 0,220 -0,077 -0,039 0,036 0,592** 0,615* 0,514** -0,310 13. SDI4 4,441 0,627 0,666 0,137 0,136 0,219 0,025 -0,142 0,044 -0,205 0,008 14. SDI5 2,8 0,707 0,813 -0,164 0 -0,029 0,283 0,030 -0,104 -0,086 -0,165 15. SDI6 2,71 0,647 0,748 -0,262 -0,141 0,140 0,235 -0,105 -0,330 -0,343 -0,211 16. Status dispersion 4,8599 0,79305 - -0,078 0,317 -0,276 -0,201 0,457* 0,315 0,394* -0,256
38 Supervisor M SD α 9 10 11 12 13 14 15 16 9. ITB (4) 3,532 0,508 0,899 - 10. ITB (5) 3,44 0,875 0,912 0,643** - 11. ITB (6) 3,145 0,935 0,935 0,758** 0,777** - 12. Inclusive climate 3,763 0,487 0,858 0,636** 0,445* 0,554** - 13. SDI4 4,441 0,627 0,666 -0,266 0,083 -0,186 -0,196 - 14. SDI5 2,8 0,707 0,813 0,176 -0,095 -0,151 -0,073 0,284 - 15. SDI6 2,71 0,647 0,748 0,086 -0,237 -0,396* -0,112 0,309 0,755** - 16. Status dispersion 4,8599 0,79305 - 0,410* 0,482** 0,448* 0,545 0,247 -0,047 -0,084 -
39 Hierarchical regressions were performed based on both the self-rating and the supervisor rating measures. The regressions were created for all the ITB measures. The first hierarchical regressions included all the control variables (appendix 7). Two-way and three-way interactions had to be created in order to test the model. The interactions were computed by using SPSS and based on standardized values. All the control variables were entered in step 1. The main variables (functional diversity, inclusive climate and SDI) were added in step 2. This did not result in any significant findings in regards to the main effect of functional diversity on any of the ITB measures. The two-way and three-way interactions were included in step 3 and 4. The two-way interactions of the main variables (functional diversity x inclusive climate, functional diversity x SDI, and inclusive climate x SDI) were included in step 3. The three-way interaction (functional diversity x inclusive climate x SDI) was entered in step 4. Entering the three-way interactions tested the full model. All two-way and three-way interactions were non-significant.
A second hierarchical regression was performed but without including the control variables. At first the main effect of functional diversity on the innovative team behavior variables was tested. None of the main effects were significant. Next, the moderating effect of the inclusive climate was tested by including the main effects in the first step, followed by including the two-way interactions between the main variables in the second step (functional diversity x inclusive climate, functional diversity x SDI, and inclusive climate x SDI). This did also not result in any significant results in regards to the two-way interactions. Finally, also the three-way interactions (functional diversity x inclusive climate x SDI) were included in the third step. The three-way interactions were also not significant.
40 A third hierarchical regression was performed in which the SDI measures were replaced by the absolute measure of status difference: the interaction of power level x power dispersion. Also here the outcomes in regards to the main effect, the two-way interaction, and the three-way interaction did not result in significant findings.
Hierarchical regression 1, 2 and 3, based on both self-rating and the supervisor-rating measures, resulted all in non-significant findings in regards to the direct effect of functional diversity on innovative team behavior, the moderating role of an inclusive climate, and the moderated moderation effect of absolute and relative status difference. These findings result in the rejection of hypothesis 1, 2, and 3.
Since evidence was not found in regards to the hypothesis, next the dataset was analyzed on significant overall model fits of single main variables and two way interaction of main variables on the ITB measures. For this a simple regression was performed. The found significant relationships are equal to the outcomes as stated on the correlation matrix. The B-values resulting from the simple regression are given as unstandardized coefficients. The simple regression exposed the strong significant positive direct effect on inclusive climate (self-rating) on ITB2 (B= .616, p<0,01) and ITB3 (B= .677, p<0,01.). A similar direct effect was found for the effect of inclusive climate (supervisor rating) on ITB3 (B= .300, p<0,05), ITB4 (B= .664, p<0,01), ITB5 (B= .800, p<0,05), and ITB6 (B= .677, p<0,01). Some of the relative status difference measures also exposed significant direct effect. A significant positive effect was revealed for the direct effect of SDI1 on ITB5 (B= .525, p<0,01) and ITB6 (B= .403,
