2013 Rijksuniversiteit Groningen

(1)

2013

Rijksuniversiteit Groningen

EFFECT OF THINKING STYLES AND CHARACTERISTICS OF NPD TEAMS

ON THE DIMENSIONS OF PROJECT PERFORMANCE IN TERMS OF

QUALITY, ADHERENCE TO TIME AND ADHERENCE TO BUDGET.

Author:

Supervisors:

Silvia Woudberg

M. de Visser

1936530

G. Balau

(2)

2 TABLE OF CONTENTS Abstract 3 Introduction 4 Theoretical Background 7 Cognitive Styles 7 Project Performance 10

Cognitive Styles & Project Performance 11

Hypotheses 13

Group Work 13

Effectiveness 13

Rational thinking style 13

Experiential thinking style 14

Efficiency 14

Adherence to schedule 14

Adherence to budget 17

Methodology 19

Sample & Data Collection 19

Dependent variable 20

Independent variable 20

Control variables 23

Analysis 23

Results 25

Discussion & Conclusion 30

Managerial implications 32

Limitations & Future Research 33

References 35

Appendix 39

Appendix A 39

(3)

3

ABSTRACT

(4)

4

INTRODUCTION

Companies develop many types of new products ranging from incremental projects to radically new projects (Benner & Tushman, 2002). This new product development (NPD) process is critical,

because new products are becoming the nexus of competition of many firms (e.g., Clark & Fuijimoto, 1991). Project groups have become a favorite means by which organizations that do such R&D conduct their work (Keller, 1986). The major advantages of project groups are their ability to bring together scientists and engineers from several disciplines and the flexibility of their structure and duration (Katz & Allen, 1985). Although many studies have reported considerable improvement in organizational life since the introduction of teamwork, some organization theorist and organization managers have experienced difficulties in getting groups to work as effectively as theorists and organizations would expect (Sinclair, 1992; Schrage, 1995).This increasing reliance on teams in organizations raises the question of how these teams should be formed (Ancona & Caldwell,1992). Most studies that have tested the effect of team configuration on team creativity or innovation have focused on overt demographic variables, such as education and functional background, age, and organizational tenure (e.g., Hulsheger, Anderson, & Salgado, 2009; Lovelance et al., 2001). Although previous research focused more on how demographic differences (e.g., age, tenure) influence team performance, studies ignored a focus on how deep level characteristics impact team performance. Focusing on cognitive styles and their impact on team performance as underlying psychological characteristics such as personality attributes have been found to be better predictors of team performance over time (Bell, 2007; Harrison et al., 2002). Therefore it is important to investigate more on deep-level variables (i.e., cognitive styles) that are less readily apparent (Bell,2007). Deep-level variables refers to team member’s psychological characteristics, including personalities, values, and attitudes (Harrison et al., 1998).

(5)

5

within target costs with both the project and the finished product and adherence to schedules means starting the manufacturing and/or marketing on the target date. (Hoegl & Gemuenden, 2001).

It is interesting to know how different team cognitive styles influence the different task-related components, because project teams can have different focus to reach different outcomes. A project team can per example be formed to develop a new product within 3 months where the costs of this development are less important. Where another project team can be formed to develop a high quality new product where there is no deadline for the release of this product. Therefore it is valuable for firms to know how different team cognitive styles influence the different components of team performance (i.e., quality, cost and time objectives) of overall project performance. Members of NPD project teams are highly interdependent in that team members must work together to complete their assignment and that they also must work extensively with nonmembers (Ancona & Caldwell, 1992). Hence cooperation in NPD project teams is necessary to complete their assignments (e.g., in terms of quality, adherence to schedule and budget), it is important to study the cognitive styles of these NPD project team members. Cognitive styles of team members are, as we already mentioned, crucial for information exchange (Harrison et al., 2002), and therefore important for the cooperation in NPD project teams.

The goal of this paper is a contribution to existing literature about cognitive psychology and team performance. In summary, the main gap in the existing literature is that there is less known about the impact of deep-level characteristics (i.e. cognitive styles) on team performance. To dive even deeper in this concept, little research has been done about the influence on the different aspects of this team performance (i.e. in terms of quality, adherence to schedule and budget). This research addresses these different literature gaps which lead to the following questions:

To what extent do different team cognitive styles have an effect on the components of team performance?

- To what extent do different team cognitive styles have an effect on team performance in terms of quality?

- To what extent do different team cognitive styles have an effect on team performance in terms of the adherence to budget?

- To what extent do different team cognitive styles have an effect on team performance in terms of adherence to schedule?

(6)

6

(7)

7

THEORETICAL BACKGROUND

Cognitive Styles

Cognitive styles are crucial for information exchange between team members (Harrison et al., 2002), and because members of NPD project teams are highly interdependent, it is important to do research about cognitive styles in NPD project teams. Cognitive styles are also important to do research about, because these underlying psychological characteristics have been found to be better predictors of team performance over time than overt demographic variables (Bell, 2007; Harrison et al., 2002).

In the literature there are a lot of definitions of cognitive style, in this paper the definitions of Ausburn & Ausburn (1978) and Messick (1976) are used. Cognitive style historically has referred to a psychological dimension representing consistencies in an individual’s manner of cognitive functioning, particularly with respect to acquiring and processing information (Ausburn & Ausburn, 1978). Messick (1976) defined cognitive styles as stable attitudes, preferences or habitual strategies that determine individuals’ modes of perceiving, remembering, thinking and problem solving. Both definition focus on the fact that a cognitive style is stable and consistent for an individual. Therefore it can be valuable for team composition to know the cognitive styles of all these individuals, because it can been seen as part of an individual.

The first studies about cognitive styles attempt to organize the array of cognitive styles revolved around the idea that there is a unified structure based on an analytic-holistic (or analytic-intuitive) style (e.g., Allinson & Hayes, 1996; Entwistle, 1981). Most of these approaches related the analytical-holistic dimension to the hemispheric lateralization of the brain based on the assumption that the left and right hemispheres have different cognitive functions during information processing (e.g., the left hemisphere processes information analytically, whereas the right hemisphere processes information holistically) (Kozhevnikov, 2007).

(8)

8 characterize as right brain and left brain”.

Many authors have appealed to dual processes in so many different ways, where authors have proposed a number of names for the two kinds of thinking they contrast (Evans, 2008). Evans (2008) came up with a table (table 1) where he listed the labels attachted to this dual-processing in the literature.

Table 1, Labels (Evans, 2008).

Through time the theory about the cognitive styles has changed a lot. Previous studies described both processing modes along a continuum (e.g. Hammond et al., 1987), where the basis of this theory is that individual will have a preference for a paricular cognitive mode (Hammond et al., 1987). This means that the more a person is analytic in information processing, the less the person is in intuitive information processing. As a conclusion you may say that it conceptualize the two modes of information processing as a balance between the two (e.g. figure 1).

(9)

9

In this paper the cognitive-experiential self-theory (CEST; Epstein, 1994) has been used for the dual-processing systems. The labels “experiential” and “rational” derived from this theory are therefore used in this paper. This more recent theory states that the two modes of processing do not work as a continuum, but can be conceputalized as an orthogonal relationship (Epstein & Pacini, 1999). This means that the process modes operate as two parallel systems, so that an individual can be experiential and rational at the same time. In table 2 the differences between the Experiential system and the Rational system have been listed.

(10)

10 Project Performance

It is important to do research about project performance because there is an increasing reliance on teams in organizations (Ancona & Caldwell, 1992). Project groups have become a favourite means by which R&D in organizations do their work

(Keller, 1986).

The general term project success covers a broad area and because it is not easy to define this concept there is no single uniform measure for NPD project success (Morteza & Kanyar, 2009).

There has been done a lot of different suggestions about what are the criteria for success, there are 3 success measurement criteria (i.e., quality, cost, time)

Figure 2, Iron Triangle

that has not really changed over almost 50 years (Atkinson, 1999). These success measurement criteria are often referred to the “iron triangle” in figure 2.

Per example the definition for project management given by Oisen (1971), where he also mentioned the three angles of the Iron Triangle:

Project management is the application of a collection of tools and techniques to direct the use of diverse resources toward the accomplishment of a unique complex, one-time task within time, cost and quality constraints. Each task requires a particular mix of these tools and techniques structured to fit the task environment and life cycle of the task.

Hoegl and Gemuenden (2001) also relate team performance to the elements of the iron triangle, divided into two subcategories.

(11)

11

In this research we use the following definitions of team performance in terms of projects: “Team performance is the extent to which a team is able to meet established quality, cost and time objectives” (Schrader & Goepfert, 1996; Gemuenden & Lechler, 1997).

In this research we therefore divide project performance to the most common known dimensions:

- Quality

- Adherence to schedule - Adherence to budget

Cognitive styles & Project Performance

In this paper we focus on the impact of cognitive styles on project performance. This indication of impact of cognitive style is based on previous research. Personality attributes as cognitive styles have been found to be better predictors of team performance over time than demographic attributes (Bell, 2007; Harrison et al., 2002). Previous research focused especially on how demographic differences (e.g., age, tenure, education) might impact team performance, but ignored how deep level characteristics as cognitive styles might impact team performance.

Ancona & Caldwell (1992) per example found evidence that high levels of functional diversity are directly associated with lower levels of performance. They also found evidence that diversity of tenure shows a similar, however less strong, negative relationship with performance. Less research is conducted on deep-lever variables, for instance Scott and Bruce (1994) stressed the relationship between cognitive style and innovative behavior of individuals. They found that individuals with an intuitive problem-solving style, who have a propensity to process information from different paradigms, are more likely to induce innovative behavior than members that have a systematic problem-solving style.

Cognitive styles are crucial for information exchange between team members (Harrison et al., 2002), and therefore might individual differences in cognitive styles among NPD project team members have a significant impact on performance of these NPD projects. Identifying and understanding each member’s cognitive style can allow managers to improve individual and team performance (Volkema & Gorman, 1998). The focus in this research lies on NPD project teams, because in these teams it is nescessary to work extensively together to complete their assignment (Ancona & Caldwell, 1992).

(12)

12

(13)

13

HYPOTHESES

Group Work

The individual style in which members approach their work and the type of tasks that groups have to perform may be strongly linked to the overall group functioning (Priola et al., 2004). Therefore we can analyze the effect of the cognitive styles of the individual members in NPD project teams and relate this to the overall functioning of these NPD project teams.

Effectiveness

From the perspective of quality this means adherence to predefined properties of the product, service or process to be developed, per example functionality and performance (Hoegl & Gemuenden, 2001). In this research quality will be measured by adherence to this predefined specifications. To meet this specifications, thoroughness and addressing all the little details, is required (Cole, 1999; Kirton, 1976).

Rational thinking style

Miron et al. (2011) found that conformists and attentive-to-detail team members have a positive impact on the adherence to standards.

Conformist team members reflects a person’s tendency to perform within given constraints when solving a problem (Kirton, 1976; Miron et al., 2004). Whereas attentive-to-detail members are systematic, reliable, precise, and carefully attentive to the implementation of their ideas (Goldsmith, 1989). Karwowski (2008) found that rationalists were significantly more conformist and less oriented towards a creative style of behavior than individuals with an experiential thinking style. This indicates that an individual with a rational thinking style is able to perform within given constraints as predefined properties in terms of quality of a NPD project performance. Allinson and Hayes (1996) argue that analytic decision makers (i.e. rational thinking style) may prefer to pay attention to detail and focus on solid data. This is also an indication that an individual who score high on rational thinking style is able to perform in an environment where attention to detail is necessary, which is the case for a NPD project with predefined properties in terms of quality (Miron et al., 2011).

(14)

14

Based on the above findings we can state that the degree of rational thinking style in a NPD product team has a positive impact on team performance in terms of quality.

H1a: The degree of rational thinking style in new product development teams has a positive impact on team performance in terms of quality.

Experiential thinking style

Miron et al. (2011) state that creative members tent to initiate task conflicts and to not adherence to rules. They found that creative team members have a negative impact on this adherence to specifications. Creative people are less likely to perform well when the task requires accuracy and adherence to rules, what points at the dark side of creativity (Miron et al., 2011).

Experiential thinking style process information from different paradigms and are therefore more oriented to the creative style (Scott & Bruce, 1994). Scott & Bruce (1994) found that individuals with a intuitive problem-solving style (i.e. experiential thinking style) are more likely to induce innovative behavior than members that have a systematic problem-solving style (i.e. rational thinking style). Also Cools et al. (2010) state that the creating style (i.e. experiential thinking style) have a preference for a creative and unconventional way of decision making. These creative and unconventional way of decision making can lead to deviation of the predefined properties in terms of quality. The experiential system can also be characterized as an holistic type of information processing, due to this lack of this attentive-to-detail, the predefined properties (i.e. quality properties) can be neglect.

These articles indicate that an individual with an experiential thinking style is not able to perform well in an environment where adherence to rules is required. Adherence to predefined properties (i.e. quality properties) can be seen as rules, and therefore a hard environment to perform well for an individual with an experiential thinking style.

Based on above findings we can state that the degree of experiential thinking style in NPD teams has a negative impact on team performance in terms of quality.

H1b: The degree of experiential thinking style in new product development teams has a negative impact on team performance in terms of quality.

Efficiency

(15)

15 Adherence to schedule

As mentioned above team performance in terms of time means adherence to schedules. Adherence to schedule with regard to NPD project team performance can be related to meeting the deadline for finishing the project.

Armstrong and Priola (2001) found that rational individuals tended to be more task oriented, more impersonal, and more self-controlling in their emotional behavior. Based on these findings Cools et al. (2010) found that teams with a dominant planning style (which can be related to the rational thinking style) tend to be more task oriented. They found that these task-related processes are more concerned with goal attainment. Adherence to time, in case of a NPD project team, meeting the deadline for finishing the project can be seen as a clear goal. Therefore this is a strong argument that the degree of rational thinking style has a positive impact on team performance in terms of adherence to schedule.

From the characteristics of table 2 we can also conclude that the rational thinking styles are reacting from what is rational and “are in control of their thoughts” (Epstein, 1991). The degree of rational thinking style in NPD teams will therefore have a positive effect on team performance in terms of adherence to schedule. When you are in control of your thoughts you will spend less time on irrelevant thoughts.

Scott and Bruce (1994) found that the analytic processing (i.e. rational processing) to be essential as it is based on a following set of routines, adherence to rules and disciplinary boundaries. Adherence to schedule/time can be seen as a rule of disciplinary boundary, so that is an essential situation for an individual with a rational thinking style. Allinson & Hayes (1996) also found in another paper that the rational thinking style can be related to a compliant system, what means that if a NPD project team has a deadline the individual with a rational thinking style is compliant to meet that deadline.

The arguments above from different articles provide enough evidence to come up with the following hypothesis:

(16)

16 Experiential thinking style

From the characteristics of the table, listed by Epstein (1991), we can cite that the experiential thinking style has a more rapid processing and is more oriented toward immediate action. Epstein (1991) also found that the experiential thinking style has an automatic and effortless processing system. Based on these findings it is obvious to state that the degree of experiential thinking style in NPD teams a positive impact will have on team performance in terms of time. In this research we doubt about this linear relation between these variables. We think that there will be another effect in the extremes (i.e. relatively low/relatively high) of the degree of experiential thinking style in a NPD team.

When the degree of experiential thinking style in a NPD team is relatively high, there will occur more effects with regard to the relation with performance in terms of adherence to schedule. Experiential individuals tended to be more interpersonally oriented and expressive (Armstrong & Priola, 2001). Based on these findings Cools et al. (2010) found that the dominant creating style (which can be related to experiential) were more relational oriented. These relational oriented processes refers to group solidarity, integration and destruction of harmony. Besides that the experiential thinking style takes a more broad perspective on a problem before they reach conclusions (Armstrong and Priola, 2001). Based on these findings we can state that when the degree of experiential thinking style in NPD teams is relatively high, this will have a negative effect on performance in terms of time. Experiential thinking style will spend more time on the interpersonal activities and will need more time before they reach conclusions.

Levitt (2002), confirms these suggestions and state that experiential thinking styles have plenty of ideas but little businesslike follow-through. They have a lot of creativity without action-oriented follow-through. Creativity disturbs the order to get the intended job done efficiently and on time. Scott and Bruce (1994) found that creativity is related to the experiential thinking style and therefore we can state that when the degree of experiential thinking style is relatively high, this will have a negative effect to team performance in terms of adherence to schedule.

H2b: The degree of experiential thinking style in new product development teams will have an inverted U-shape relation with team performance in terms of adherence to schedule/time.

(17)

17

homogenous. He gave both groups the same assignments and the result was that the ‘rational’ dominated group was immediately task oriented and solved the assignment within 13 minutes. The ‘experiential’ dominated group solved the assignments after 20 minutes and was very delayed by the different perspectives on the assignments and therefore cannot made a decision quickly.

This experiment supported our theoretical arguments about the degree of experiential thinking style or degree of rational thinking style within a NPD project team on the team performance in terms of adherence to schedule/time.

Adherence to budget

Budget means per example staying within target costs with both the project and the finished product and adherence to schedules means starting the manufacturing and/or marketing on the target date. (Hoegl & Gemuenden, 2001).

A team consisting of mostly rational thinking styles will react form the logical reasoning and they will first require enough evidence and logic to take action (Epstein 1991). This provides evidence that a team with a high degree of rational thinking style will make relatively less mistakes, because they will take action based on enough evidence. Therefore we can state that a team with a high degree of rational thinking style compared to a team with a low degree of rational thinking style will spend less money on solving mistakes and errors.

The reasoning behind this perspective to take action can be proven and can be explained by logical connections. Due to this logical reasoning there will be a low level of disagreement with a higher degree of rational thinking style. Moreover, the rational thinking style can understand the operation of the experiential thinking style, whereas the reverse is not true (Epstein, 2003). This lower level of disagreement is related to a less conflicts, which are costly to solve. The higher the level of conflicts in a team the higher the coordination costs of that team (Holahan & Mooney, 2004). The coordination costs for a team with high level of rational thinking style will be low.

(18)

18

compliant (Scott & Bruce, 1992). This are arguments that the degree of rational thinking style has a positive impact on performance in terms of adherence to budget.

H3a: The degree of rational thinking style in new product development teams has a positive impact on team performance in terms of adherence to budget.

Experiential thinking style

Experiential thinking styles react from emotion and use self-evidently valid: “experiencing is believing” (Epstein et al., 1996). They will use trial and error to come to the most excellent idea in their eyes. Due to the trial and error/experiencing is believing there will occur more mistakes than “think before doing” style which will be more costly.

As we already mentioned the experiential thinking style will take a broader perspective on a problem. The higher the degree of experiential thinking style in a NPD project team the more perspectives there will be a on a problem, which will lead to higher coordination costs. Therefore we can state that the degree of experiential thinking style will have a negative impact on team performance in terms of adherence to budget.

Due to these more perspectives and ‘experiencing is believing’ characteristics of the experiential system this will provide a high level of conflict. Conflict can improve decision making, but can also degrade decision making as it distracts team members from the essential purposes of the project (Holahan & Mooney, 2004). These distractions of the actual accomplishments of a team occur often through emotions (Holahan & Mooney, 2004). This distraction and the fact that the experiential thinking style is characterized as less performing on attentive-to-detail (Miron et al., 2011) can be seen as arguments that degree of experiential thinking style in a NPD project team has a negative impact on project performance in terms of adherence to budget. Also the fact that the experiential system is characterized by experiences seized by emotions, the opportunity of emotional conflict is in a team with a high degree of experiential thinking style, high.

(19)

19

METHODOLOGY

All analyses were at the individual level and conducted using a sample of product development teams. The data consist of questionnaires from team members, using the scale of Hoegl (i.e. team performance) and the REI scale of Epstein (i.e. cognitive styles).

Sample & Data Collection

In order to empirically test the hypotheses stated in the theoretical section it is order to collect data. The purpose of the data collection was to create a shared database with two other master students. The first step was to select all Dutch manufacturing companies which are appropriate for this research. All the companies were called to try to get in touch with the R&D manager, or other representatives for the NPD processes within the companies. After explaining the purpose of this research, an email has been send where we explained the research more in detail. A lot of companies were not able to participate into this research, due to a lot of different reasons. The most common reason mentioned by companies was that they were not able to schedule time for this research due to the economic crisis. Another reason mentioned a lot, was that they were very interested in this research but that they cannot arrange a contribution to this research within our time path and deadlines. Even though there were more companies which are not willing to cooperate we found four companies which were willing to schedule a meeting with. Batavus, Ihc Merwerde, BASF and Philips Eindhoven scheduled an appointment with us. Batavus and Ihc Merwerde were very

enthusiastic to contribute to our research, but unfortunately we decided after consideration with our supervisors, that both of the companies were too small for our research. The problems we then will occur were per example that there were too much employees which contributed to all most every NPD project, due to the relative small size of the R&D department. Philips Eindhoven was also very interested in our research and we are now organizing a cooperation with different departments of Philips Eindhoven. Unfortunately it was for Philips not possible to contribute to our research before our deadline to hand in our master thesis. Philips Eindhoven is an important innovator and has a long history in product development. It is even the company which is the number one at the list of top 30 R&D-companies in the Netherlands for a couple of years now, concluded by the Technical Journal. Therefore we found it that interesting that they are willing to contribute to our research that we continued our appointments and data collection at Philips Eindhoven. Also BASF was very interested in our research, and also we continue our contact with them to collect data there in the future.

(20)

20

our master thesis. This is a large database, but due to the fact that we could not use our own scales, we were all limited to the data which was appropriate for our individual thesis’s. This resulted for me in 76 project groups from 4 different companies. Due to some missing values I decided to remove 2 projects (i.e. Project Number 5 & Project Number 42) from the sample, what resulted at the end in a sample of 232 individuals distributed over 74 projects.

Dependent Variable

Because there is a growing awareness that the reliance on teams in organizations increases (e.g. Ancona & Caldwell, 1992), there has been a lot of research about how team performance can be measured. In this research the scale of Hoegl and Gemuenden (2001) will be used to measure the team performance of the NPD teams of our sample. This scale measures the team performance with regard to the different success criteria of the iron triangle. Team performance is in this scale

described in terms of the variables effectiveness and efficiency.

These performance measurements make use of a 7-points Likert scale. Ranging from strong disagree to strong agree.

Quality item

 Going by the status of the project, the project meets quality specifications. Budget item

 Going by the status of the project, the project expenditures are on budget. Schedule item

 Going by the status of the project, the project duration is on schedule.

For every individual we calculated the mean score for every item, and expanded this to the group level. By doing this we calculated the mean score per project for every single item.

Independent Variable

There has been a lot of research about how to measure cognitive styles of individuals. Cognitive styles was previously measured alongside a continuum (e.g., Allinson & Hayes, 1996), but recently there is an increased attention towards looking at the cognitive styles as being orthogonal constructs. Epstein came up with the CEST in 1994 and broadened this theory. From the perspective of cognitive-experiential self-theory (CEST; Epstein, 1994), what people are experiencing are the outcomes of two different information-processing systems, rational and experiential.

(21)

21

versions as the REI-59 which is the original measure. The REI-40 was designed to assess preferences for information processing, where there is a distinguishing between 2 cognitive styles (Epstein & Pacini, 1999). To measure the degree to which a person processes information intuitively 5 items (EP1-EP5) from the “Faith in Intuition” scale (i.e. FI) of the REI have been used. This was also the case to measure the degree to which a person processes information analytically, therefore 5 items (EP6-EP10) from the “Need for Cognition” scale (i.e. NfC) have been used. The choice of using these 10 7-point Likert scale items was based on the highest factor loadings (Epstein & Pacini, 1999). A

confirmatory factor analysis of the 10 items on the sample of 232 individuals has been made to measure the factor loadings. First the assumptions, if it is appropriate to run a factor analysis, has been analyzed. One of the assumptions is that it is important that the answer patterns of the items which can been seen as one factor do not have much overlap. This is important because every single item needs to contribute uniquely to the underlying construct. To analyze the existence of this effect of multicollinearity between the items it is case to create a correlation matrix. The database is not appropriate for a factor analysis if the inter-item correlation is higher than 0.9. This is for all the items not the fact, so based on this assumption we can run a factor analysis.

EP1 EP2 EP3 EP4 EP5

EP1 1.000 .511 .435 .341 .513 EP2 .511 1.000 .458 .384 .511 EP3 .435 .458 1.000 .571 .491 EP4 .341 .384 .571 1.000 .549 EP5 .513 .511 .491 .549 1.000 Determinant = .190

Table 3, Correlation Matrix EP1-EP5

EP6 EP7 EP8 EP9

EP6 1.000 .542 .588 .543

EP7 .542 1.000 .534 .665

EP8 .588 .534 1.000 .722

EP9 .543 .665 .722 1.000

Determinant = .153

Table 4, Correlation Matrix EP6-EP9

(22)

22

Kaiser-Meyer-Olkin Measure of Sampling Adequacy. Approx. Chi-Square

.807 379.838

Bartlett’s Test of Sphericity Df Sig.

10 .000

Table 5, KMO and Bartlett’s Test EP1-EP5

Kaiser-Meyer-Olkin Measure of Sampling Adequacy. Approx. Chi-Square

.769 429.358

Bartlett’s Test of Sphericity Df Sig.

6 .000

Table 6, KMO and Bartlett’s Test EP6-EP9

Therefore we can conclude that it is appropriate to run a factor analysis for these items. We found out that one item (EP10) had such a low factor loading (0.572) compared to the other items that we decided to delete that single item. This resulted in 4 items that measure the degree of rational thinking style and 5 items to measure the degree of experiential thinking style. In table beneath shows the 2-factor model with the remaining factor loadings for each item sorted by scale.

Item Rational (NfC) Experiential (FI)

I like to rely on my intuitive impressions. (EP1) 0.728

Using my gut feeling usually works well for me in figuring out problems in my life. (EP2)

0.746

I believe in trusting my hunches. (EP3) 0.749

Intuition can be a very useful way to solve problems. (EP4) 0.777

I often go by my instincts when deciding on a course of action. (EP5)

0.811 I try to avoid situations that require thinking in depth about

something. (EP6)

0.791

I enjoy solving problems that require hard thinking. (EP7) 0.817

I am much better at figuring things out logically than most people. (EP8)

0.854

I have a logical mind. (EP9) 0.882

Table 7, Confirmatory Factor Analysis

(23)

23

Cronbach’s Alpha N of Items

.816 5

Table 8, Reliability Statistics EP1-EP5 Table 9, Reliability Statistics EP6-EP9

For both the factors the Cronbach’s Alpha is higher than 0.7, so we can state that the internal consistency is reliable enough.

All above findings provided strong support for the validity and reliability of data to measure the degree of rational thinking style and experiential thinking style. For every individual we calculated the mean score on both scales and from these mean scores we expanded this to the group level, by calculating the mean scores for every group on both scales.

Control Variables

There are factors that can affect team cognitive styles or team performance, of course. We assessed two factors in an attempt to eliminate alternative interpretations.

The first control variable we take into account in our research is organizational tenure. Zenger and Lawrence (1989) found that organization tenure is related to frequent communication and

performance (O’Reilly and Flatt, 1989). To prevent that the performance will be higher due to a high level of organizational tenure in a group this variable will be included to control this effect. Hence a variable measuring the company experience of an individual has been included in the questionnaire. We calculated the mean company experience for each team in years.

The second control variable we assesed in this research is team size of the NPD project teams. There are a lot of articles which state that large teams are superior to small teams, due to the fact that larger teams have more capabilities and resources to solve the group tasks (e.g. Jackson, 1992). Per example an increase in team size, increases the number of ideas/solution for a problem statement or increases the number of perspectives on the new product development process. Therefore we included a variable measuring the team size of the NPD project team to control this effect on the performance of that team. We calculated the mean teamsize for each team.

Analysis

Before starting the analysis, it is necessary to look at ouliers of the data. Grubs (1969) state that an outlier is one that appears to deviate noticeably from other members of the sample in which it occurs. To analyze whether there are outliers in this database the outlier labeling rule introduced by Turkey (1977) has been used. This is based on multiyplying the Interquartile range (IQR) by a factor of 1.5 (Tukery, 1977). In this database this resulted in the project teams 1, 2, 9, 47 en 59 to be

considered as an outlier. However there has been a lot of resistance to this factor 1.5. Hoaglin &

Cronbach’s Alpha N of Items

(24)

24

Iglewicz (1987) argued that almost 50 % of the outliers observerd by this factor 1.5 are not really outliers. They suggested that it is beter to use a multiplier of 2.2, when you have a small sample size for your analysis. The sample size in this research is quite small (N=74) and therefore the outliers labeling rule based on multiplying the IQR by the factor of 2.2 has been used. This resulted that there are no outliers in this database.

To run a multiple regression it is important that the date is appropriate with regard to a few

assumptions. We check our data with regard to the four assumptions of parametic test (Field, 2005). The first assumption is that the dependent and independent variables are measured at the interval level. The dependent variables and independent variables are all measured with a 7-point Likert scale, which can be interpreted as an interval level. The second assumption is that the relation between the variables is lineair, which is the case for every hypothesis except H2b. The next assumption is that the residuals are independent of each other, normally distributed and that they have a constant variation (homoscedecity) (Field, 2005). The respondents in our research have filled in their questionnaire fully individual, so therefore we can state that these are independent of each other. To check the normal distribution it is necessary to look at the values of Skweness en Kurtosis (Field, 2005). The Skewness measures the assymetry of a distribution, where a Skewness value of zero indicates that the normal distribution is symmetric (Field, 2005). The Kurtosis measures in his place the extent to which the values cluster around a central point, where a Kurtosis value of zero indicates for a normal distribution (Field, 2005).

NfC FI BUDGET TIME QUALTIY TeamSize OrgTenure

Skewness -.690 -1.213 .058 .386 -.463 1.406 .435

Std. Error of Skewness .279 .279 .279 .279 .279 .279 .279

Kurtosis .045 3.328 .177 -.709 .389 2.109 -.950

Std. Error of Kurtosis .552 .552 .552 .552 .552 .552 .552

Table 10, Skewness and Kurtsosis Test

(25)

25

The last assumption we need to check before we can do a regression analysis, is to check whether the sample has a constant variance (homoscedecity). The ‘Homogeneity of variance’, more known as Levene’s Test, tests if the variances in groups are equals (Field, 2005). In our case we measure if the variances across the different companies are equal for the dependent variables. If de test shows that the significant is p<.05, then we can conclude that the variances are not equal (heteroscedecity) (Field, 2005). We found that the variances were not equal across groups, which means that heteroscedecity occurs (see Appendix B). After logarithmic transformations of the dependent variables the significant values improved with regard to homoscedecity (see Appendix B). Therefore we found evidence that by using the logarithmic values of the dependent variables the variances across groups near the significant score of p<0.05 or even p>0.05 (see Appendix B). In our analysis of our hypotheses we therefore work with the logarithmic transformations of the dependent variables. We can assume that all the assumptions to do a parametic test (Field, 2005) are roughly (due to the limited database) met.

Results

To check whether there is a lack of multicollinearity among the independent variables, we standardized the variables and try to find support by the obtained variance inflation factor (VIF) values. The VIF values of the independent variables ranged from 1.011 to 1.035, which were all below the cut-off value of 10 (Field, 2005). This indicates that multicollinearity among these independent variables is not a major problem.

Variable VIF

ZNfC (Rational) 1.028

ZFI (Experiential) 1.011

ZTeamSize 1.027

ZOrgTenure 1.035

Table 11, Variance Inflation Factor

(26)

26

Variable Mean Standard

Deviation

Budget Time Quality NfC FI TeamSize OrgTenure

Budget 3.441 .767 1 Time 3.197 .906 .411*** 1 Quality 3.697 .707 .140 .470*** 1 NfC 5.302 .745 .046 -.059 .056 1 FI 4.584 .622 -.199* .020 .011 .135 1 TeamSize 3.135 1.328 -.257** -.123 .018 .030 .054 1 OrgTenure 14.747 8.127 -.279** -.022 .225* -.081 .099 -.083 1 *p<0.1 **p<0.05 ***p<0.01

Table 12, Descriptive Statistics and correlations (two-tailed)

Table 13-15 summarizes the findings of the General Linear Modeling analysis whereby in every table another dimension of project performance acts as the dependent variable. In table 13 this is the dependent variable adherence to budget, in table 14 the dimension quality and in table 15 the variable adherence to time acts as the dependent variable. It can be observed that there are no significant direct effects between the two processing styles and the dependent variables. The only significant effect we found, is that the results show that company 4 significantly over performs compared to the other companies. Furthermore we did not found any significant relations between our control variables, team size and organizational tenure, and the dependent variables.

Model 1 Model 2 Model 3

Intercept .515*** .517*** .508***

Company 1 .024 .021 .025

Company 2 -.015 -.017 -.002

Company 3 .037 .035 .047

Company 4 Reference Reference Reference

ZTeamSize -.018 -.018 -.017

ZOrgTenure -.024 -.024 -.027

ZNfC (Rational) .003

ZFI (Experiential) -.014

R Squared / Adjusted R Squared .154 / .092 .154 / .079. .170 / .096

*p<0.1 **p<0.05 ***p<0.01

Table 13, Results of General Linear Modeling analyses – dependent variable: Budget (logarithmic) (N=74)

Intercept .553*** .561*** .552***

Company 1 -.019 -.031 -.019

Company 2 .024 -.014 .026

Company 3 .008 .001 .009

ZTeamSize .006 .006 .006

ZOrgTenure .009 .011 .009

R Squared / Adjusted R Squared .065 / -.004 .077 / -.005 .065 / -.018

*p<0.1 **p<0.05 ***p<0.01

(27)

27

Intercept .522*** .526*** .527***

Company 1 -.050 -.056 -.051

Company 2 -.081 -.087 -.092

Company 3 .004 .000 -.004

ZTeamSize -.003 -.003 -.003

ZOrgTenure .008 .009 .010

ZFI (Experiential) .012

R Squared / Adjusted R Squared .089 / .002 .091 / .009 .096 / .016

*p<0.1 **p<0.05 ***p<0.01

Table 15, Results of General Linear Modeling analyses – dependent variable: Time(logarithmic) (N=74) Due to the above findings, we found it interesting to observe if there is a relation between the two thinking styles and the operational performance, where we took the three single items as one. We did this by computing the mean of the three single items per project team. Before doing this we conducted a factor analysis where we found that it is appropriate to take these three single items as one. Also here we a General Linear Modeling analysis, observed in table 16. We also did not find here any significant effect on the dependent variable, which was in this case Operational

Performance.

Intercept .530*** .534*** .529***

Company 1 -.015 -.022 -.015

Company 2 -.024 -.030 -.083

Company 3 .016 .012 .017

ZTeamSize -.005 -.005 -.005

ZOrgTenure -.150 -.001 -.003

R Squared / Adjusted R Squared .051 / -.019 .056 / -.028 .051 / -.034

*p<0.1 **p<0.05 ***p<0.01

(28)

28 Figure 3, Quadratic curve estimation

Due to the fact that we did not found any significant relationships between the two processing systems and the dimensions of project performance, and neither for operational performance, we will observe the other correlations mentioned in table 12.

The dependent variable ‘Budget’ seems to correlate with the independent variables ‘FI’, ‘TeamSize’ and ‘OrgTenure’. To see if there is a linear or quadratic relationship between these variables and the dependent variable budget, a plot of both the relationships has been made. Based on the significant level, one quadratic relationship and two linear relationship can be observed. The independent variable ‘FI’ has a significant (p<.1) quadratic relationship with the dependent variable budget (figure 4). Team size seems to have a significant (p<.1) negative relationship with budget (figure 5) and organizational tenure also seems to have a significant (p<.05) negative relationship with budget (figure 6).Organizational tenure seems to have a significant (p<.1) positive relationship with quality (figure 7).

(29)

29

Figure 6, Linear curve estimation Figure 7, Linear curve estimation

Table 12 also indicates that time correlates with quality (p<.01) and budget (p<.01), but that there is no correlation between quality and budget. Therefore the correlation between Operational

performance and the three dimension has also been observed. This resulted in a correlation (table 17) between Operational performance and time and quality, but not with budget.

Variable Mean Standard

Deviation

Budget Time Quality Operational

Performance Budget 3.441 .767 1 Time 3.197 .906 .411*** 1 Quality 3.697 .707 .140 .470*** 1 Operational Performance 3.632 .622 .098 .583*** .790*** 1 *p<0.1 **p<0.05 ***p<0.01

(30)

30

DISCUSSION & CONCLUSION

This section is structures as follows, first the findings of this research will be shown. After this the managerial implications will be given and in the end the main limitations of this research and suggestions for future research will be given. These findings can contribute to existing literature about team composition and project performance. It can also provide managers insights in the way they should form a NPD project team.

This research was conducted to found the empirical evidence to support the hypotheses stated in theoretical section of this paper. There were not found any significant relations between the degree of thinking styles in NPD project teams and performance in terms of budget. Therefore the results from the analysis conducted between these variables can only be seen as an indication of the relations between them. The degree of rational thinking style has a positive impact on performance in terms of quality, adherence to time and adherence to budget. These findings are not significant, but can be seen as an indication for the relationship between these variables. These results are consistent with the hypotheses stated in the theoretical section of this paper. The degree of experiential thinking has a negative impact on performance in terms of quality which can also be seen as an indication, but is nevertheless in line with the hypothesis stated. The degree of

experiential thinking style has a positive impact on performance in terms of time. This indication is not consistent with the hypothesis stated, because a inverted U-shape relation was the expectation. An explanation for this positive relation can be that the experiential thinking style has a rapid processing system and is oriented toward immediate action (Epstein & Pacini, 1991). However these indication is not consistent with the experiment of Hermann (1981), where he found that the ‘experiential’ dominated group performed less in terms of adherence to time compared to the ‘rational’ dominated group. It is hard to conclude out of these findings what the relation between these variables is, or if there is a relation, because the findings of the research conducted in this paper can only be seen as an indication.

(31)

31

two types of innovation projects have different characteristics. Explorative activities can be related to search, risk-taking, experimentation, radical, flexible, while exploitative can be characterized as efficiency, selection, execution, incremental and refinement (March, 1991). These different activities, would require different degree of cognitive styles and that can be an explanation for the variance in impact of the degree of experiential thinking style and performance in terms of budget.

This research showed that the degree of organizational tenure has a positive impact on quality, but negative impact on budget. There is thus evidence that the degree of organizational tenure has not the same impact on all the different dimensions of project performance. For these findings there are some explanation in the literature about organization tenure. According to per example Meyer and Allen (1997), more tenured workers have a higher opportunity to have positive work experience, what can be a explanation for higher levels of commitment. Whereas commitment is related to productivity and performance (Meyer et al., 2002). Zenger and Lawrence (1989) also found that organization tenure is correlated with the frequency of communication. Whereas members of NPD project teams are highly interdependent in that team members must work together with each other and with nonmembers (Ancona & Caldwell, 1992), communication is very important to perform well as a team. Therefore the higher the organization tenure in a NPD project team, the better they perform in terms of quality. There is also evidence in the literature that degree of organization tenure has negative impact on project performance in terms of budget. The following definition of tenure has been given by Brimeyer et al. (2010): “Tenure provides an indication of accumulation of work skills, greater job security, and economic security”. Brimeyer et al. (2010) also state that longer tenure gives workers the experience that produces autonomy. Due to the fact that the higher the organization tenure, the higher the experience, security and autonomy, the more it will take guidelines in to stride (Brimeyer et al., 2010). This can be an explanation that the higher the organization tenure in a project team, the lower the team will perform in terms of adherence to budget. A more tenured employee, per example, will be less concerned with consequences when he will not met the targeted date of the project, because he is more secure of his job.

(32)

32

more constructive conflicts and therefore require more coordination costs. An increase in team size also increases the number of opinions. These arguments can be an explanation that the larger the team, the lower the team will perform in terms of adherence to budget.

The most significant (p<.01) results found in this research were the correlations between the different dimension of project performance (i.e. operational performance) and the construct as one with the dimensions as single items (figure 8). It seems obvious that operational performance (the mean score of quality, time and budget) would correlate with the different dimensions. This was for quality and adherence to time the case, but with adherence to budget there was no correlation found. An explanation for this phenomenon could be that budget should be seen as an independent construct. In this research the correlation between the dimensions has been observed. This resulted in correlations between time and quality; and time and budget, but not in a correlation between quality and budget. This finding plus the result that project performance (i.e. operational

performance) did not correlate with budget, seems to be a strong argument that adherence to budget is acting in a different way than the other dimensions.

Figure 8, conceptual design of correlations

Managerial Implications

(33)

33 Limitations and Future Research

The first limitation of our study is that the value of the dependent variables in this research were only measured with one single item. The disadvantages of using one single item is that you cannot test on reliability (e.g. consistency and homogeneity) of your item. A suggestion for further research is that it is better to use multiple-item scales to make your findings more reliable.

Another limitation of this research is that the findings are based on subjective perceptions of project performance. This can be one of the causes that the values given by individuals to the different dimensions of project performance did not deviate much from each other. Per example the value given for adherence to schedule was often the same for the other to dimensions of project performance. We measured finished projects so it could be hard for individuals to have it clear in mind what exactly was performed well in a project. People often only remind if a project was well performed or bad performed, and not exactly what kind of dimensions, so that they will fill in for all the dimensions almost the same value. 22,4 % of the individuals filled in the same values for the three single items of operational performance. For future research it would be more evident to use objective data about project performance, and to prevent this problem mentioned above.

The last, but not least limitations is that our study is based on a limited database. We provided our findings based on only 74 projects, which can be cause that we did not provide a lot significant effects. For future research it is valuable to do research about the correlated effects shown in the results part of our study. It could be interesting to investigate what the effect of team size is on the different dimensions of project performance. It can also be valuable to do research how the organizational tenure effects the different dimensions of project performance in the different innovation setting (i.e. explorative vs. exploitative). Furthermore it could be interesting to do more research about the experiential thinking style and the adherence to budget. We found correlation between these dimensions and a significant quadratic relationship between both. Probably with a larger database and more items related to this dimension of project performance it could deliver even more valuable findings about this relationship.

(34)

34

Furthermore it is interesting to do more research about the scale of Hoegl and Gemuenden (2001), because in this research the item which measures adherence to budget seems to be an independent construct.

(35)

35

REFERENCES

Allen, J., Darrel, K., Rigby, K. Gruver, (2009). Innovation in turbulent times. Harvard Business Review. Allinson, J., & Hayes, C. (1996). The cognitive style index, a measure of intuition-analysis for

organizational research. Journal of Management studies, 33, 119-135.

Ancona, D., Caldwell, D.F. (1992). Demography and design: predictors of new product team performance. Organization Science 3(3): 321-341

Atkinson, R. W. (1999) Project Management: cost, time and quality, two best guesses and a

phenomenon, it’s time to accept other success criteria. International Journal of Project Management, 17 (6) 337-342.

Armstrong S.J. and Priola V. (2001) Individual Differences in Cognitive Style and their effects on task and social orientations of Self-Managed Work Teams, Small Group Research, 32(3), 283-312

Ausburn, L., J., & Ausburn, F. B. (1978). Cognitive styles: Some information and implications for instructional design. Educational Communication and Technology, 26, 337-354.

Bell, S. T. (2007). Deep-level composition variables and predictors of team performance a meta-analysis. Journal of Applied Psychology, 92: 595-615.

Benner, M. J., &Tushman, M., (2002), Process management and the technological innovation: A longitudinal study of the photography and pain industries. Administrative Science Quarterly, 47: 676-706.

Brimeyer, T. M., R. Perrucci, and S. M. Wadsworth (2010). “Age, Tenure, Resourses for Control, and Organizational Commitment”. Social Science Quarterly, Vol. 91, No. 2, 511-530.

Bunderson, J. S. (2003). Team member functional background and involvement in management teams: Direct effects and the moderating role of power centralization. Academy of

Management Journal ,48 (4), 458-474.

Clark, K. B., & Fuijimoto, T. (1991). Product development performance. Boston: Harvard Business School Press.

Cohen, S. G., & Bailey, D. E. (1997). What makes teams work: Group effectiveness research from the shop floor to the executive suite. Journal of Management ,2(3), 230-290.

Cole, E. R. (1999). Managing quality fads. New York: Oxford University Press.

Cools, E., Van der Heyden, K. & Lommelen, B., (2010), Cognitive styles and teamwork: examining the impact of team composition on team processes and outcomes. Vlerick Leuven Gent Working Paper Series 2010/10

Entwistle, N. J. (1981). Styles of teaching and learning: An integrated outline of educational psychology for students, teachers and lecturers. Chichester, United Kingdom: Wiley

(36)

36

Epstein, S., Pacini, R., Denes-Raj, V., &Heier, H. (1996). Individual differences in intuitive-esperiential and analytical-rational thinking styles. Journal of Personality and Social Psychology, 71, 390-405. Epstein, S., Pacini, R., (1999). The relation of Rational and Experiential information processing-styles to personality, basic beliefs, and the Ratio-Bias phenomenon. Journal of Personality and Social Psychology, 76, 972-987.

Epstein, S. (2003). Cognitive-experiential self-theory of personality. In Millon, T., & Lerner, M. J. (Eds), Comprehensive Handbook of Psychology, Volume 5: Personality and Social Psychology ( pp. 159-184). Hoboken, NJ: Wiley & Sons.

Evans, J.S.B.T., (2008). Dual processing accounts of reasoning, judgment and social cognition. Annual Review of Psychology, 59, 255-278.

Field, A. (2005). Discovering Statistics Using SPSS. Londen, Sage publications.

Gemuenden, H. G., Lechler, T. (1997), Success factors of project management: The critical few. Reviewed paper, Portland Internat. Conf. Management of Eng. Tech. Portland, Oregon July 27-31. Goldshmith, R. (1989). Creative style and personality theory. In M. Kirton (Ed.), Adaptors and innovators: 37–55. London: Routledge.

Grubbs, F.E., (1969). Procedure for identifying outlying observations in samples. Technometrics, 11: 1 21

Hammond, K.R., Hamm, R.M., Grassia, J., Pearson, T., (1987). Direct comparison of the efficacy of intuitive and analytic cognition in expert judgment. IEEE Transactions on Systems. Man and Cybernetics, 17(5), 753-770.

Harrison, D., Price, K., & Bell, M. (1998). Beyond relational demography: Time and the effects of surface- and deep-level diversity on work group cohesion. Academy of Management Journal, 41, 96-107

Jackson, S., May, E., & Whitney, K. (1995). Understanding the dynamics of diversity in decision-making teams. In Guzzo, R & E. Salas (Eds.), Team decision decision-making effectiveness in organizations (pp. 204–261). San Francisco, C: Jossey-Bass.

Harrison, D. A., Price, K. H., Gavin, J. H., & Florey, A. T. (2002). Time, teams, and task performance: Changing effects of surface- and deep-level diversity on group functioning. Academy of Management Journal, 45, 1029-1045

Hayes, J., &Allinson, C. W. (1998). Cognitive style and the theory and practice of individual and collective learning in organizations. Human relations, 51, 847-871.

Hermann, N. 1981. The creative brain, Training and development journal, vol. 11.

Hoaglin, D. C., and Iglewicz, B. (1987), “Fine Tuning Some Resistant Rules for Outlier Labeling”, Journal of American Statistical Association., 82, 1147-1149

(37)

37

Holahan, P.J., & Mooney, A.C (2004) "Conflict, Decision Making, and Project Team Performance", Academy of Management Annual Meetings, New Orleans, LA.

Horwitz, S. K., (2005), The compositional impact of team diversity on performance: Theoretical considerations. Human Resource Development Review, 4;219.

Hulsheger, U. R., Anderson, N., & Salgado, J. F. (2009). Team-level predictors of innovation at work: A comprehensive meta-analysis three decades of research. Journal of Applied Psychology, 94: 1128-1145

Jackson, S. E. (1992). Consequences of group composition for the interpersonal dynamics of strategic issue processing. Advances in Strategic management, vol. 8: 345-382.

Karwowski, M. (2008). Giftedness and Intuition. Gifted and Talented International . Vol 23. No. 1 Katz, R., & Allen, T. J. (1985). Project performance and the locus of influence in the R&D matrix, Academy of Management Journal, 28: 67-87.

Keller, R. T. (1986), Predictors of the performance of project groups in R&D organizations, Academy of Management Journal 29: 715-726.

Kirton, M. (1976). Adaptors and innovators: a description and measure. Journal of Applied Psychology, 61(5), 622–629.

Kozhevnikov, M., (2007). Cognitive styles in the context of modern psychology: Toward an integrated framework of cognitive style. Psychological Bulletin. Vol. 133, No. 3, 464-481.

Levitt, T. (2002). "Creativity Is Not Enough." Harvard Business Review 80(8) August:137-145. Lovelace, K., Shapiro, D. L., & Weingart, L., R. (2001). Maximizing cross-functional new products teams’ innovativeness and constraints adherence: A conflict communication perspective. Academy of Management Journal, 44: 779-793.

March, J.G., (1991). Exploration and exploitation in organizational learning. Organization Science, 2, 71-87.

Messick, S. (1976). Personality consistencies in cognition and creativity. In S. Messick (Ed.), Individually in learning (pp. 4-23). San Francisco: Jossey-Bass.

Meyer, John P., and Natalie J. Allen. (1997). Commitment in the Workplace: Theory, Research, and Application. Thousand Oaks, CA: Sage Publications.

Meyer, John P., David J. Stanley, Lynne Herscovitch, and Laryssa Topolnstsky. (2002). ‘‘Affective, Continuance, and Normative Commitment to the Organization: A Meta- Analysis of Antecedents, Correlates, and Consequences.’’ Journal of Vocational Behavior 61:20–52.

Miron, E., Erez, M., & Naveh, E. (2004). Do personal characteristics and cultural values that promote innovation, quality, and efficiency compete or complement each other? Journal of Organizational Behavior, 25: 175–199.

(38)

38 Journal 54(4): 740-760

Morteza S. K., Kanyar, K. C. (2009), Generic project success and project management success criteria and factors: literature review and survey, transaction on business and economics, 8(6), pp. 456-468. Oisen, R. P. (1971), Can project management be defined?, Project management quarterly, 2(1), 12-14.

Priola, V., Smith, J. L., Armstrong, S. J., (2004), Group work and cognitive style: a discursive integration, Small group Research, Vol. 35. 565-595.

O’Reilly, C., Flatt, S., (1989). Executive team demography, organizational innovation, and firm performance. Paper presented at the 49th annual meeting of the Academy of Management, Washington,DC.

Schrader, S., J. Goepfert, (1996), Structureing manufacturer-supplier interaction in new product development teams: An empirical analysis. H. G. Gemuenden, T. Ritter, A. Walter, eds. Proc.

12thinternat. Conf. Industrial Marketing and Purchasing (1) Universitaet Karlsruhe, Karlsruhe,

Germany, 557-598.

Schrage, (1995). No more Teams! New York: Doubleday/Currency.

Scott. S, G., & Bruce, R. A. (1994). Determinants of innovative behavior: A path model of individual innovation in the workplace. Academy of Management Journal 137. 580-607.

Sinclair, A. (1992). The tyranny of a team ideology. Organization Studies. 13, 611-626.

Volkema, R.J., & Gorman, R. (1998). The influence of cognitive-based group composition on decision-making process and outcome. Journal of Management Studies, 35: 105-121.

(39)

39

Appendix

Appendix A

The normal distribution of FI

(40)

40 Appendix B

Levene Statistic Sig.

Based on Mean 3.342 .024

Based on Median 1.979 .125

Based on Median and with adjusted df 1.979 .129

Based on trimmed mean 3.472 .021

Test of Homogeneity of Variance – Dependent Variable Budget

Based on Mean 2.886 .042

Test of Homogeneity of Variance – Dependent Variable Logarithmic transformation of Budget

Based on Mean 3.347 .024

Test of Homogeneity of Variance – Dependent Variable Time

Based on Mean 1.730 .169

Test of Homogeneity of Variance – Dependent Variable Logarithmic transformation of Time

Based on Mean 3.275 .026

Test of Homogeneity of Variance – Dependent Variable Quality

Based on Mean 2.734 .050