The impact of cognitive style on the creative
performance of brainstorming participants
By
David van Londen
Master Thesis: MSc in BA, specialization: Strategic Innovation Management
University of Groningen Faculty of Economics and Business
First Supervisor: Dr. R. A. van der Eijk Second Supervisor: Dr. F. Noseleit
June 2014
ABSTRACT
Prevailing research indicates that brainstorming is inefficient due to factors related to the nature of brainstorming or its rules. However, personal characteristics of
participants are likely to influence creative performance as well, causing some groups to perform better than others. This study focuses on the relationship between
participant’s cognitive styles and creative performance in brainstorming. A distinction is made between three types of cognitive style: knowing, planning and creating
cognitive style. Data was collected by distributing questionnaires to participants of 66 brainstorming sessions. Individual scores were transformed to group-level by mean aggregation. Results indicate that planning and creating cognitive style are positively related to creative performance. The expected negative relation between knowing cognitive style and creative performance is not confirmed. Furthermore, no support is found for the interacting effects of motivation and creative experience on the
relationship between cognitive style and creative performance.
Keywords: idea generation, brainstorming, cognitive style, creativity, innovation, motivation, creative experience
EXECUTIVE SUMMARY
The purpose of this study is to investigate the impact of cognitive style on the creative performance of brainstorming participants. Brainstorming is an idea generation technique with the conviction that social interaction spurs individuals to greater
creativity. However, brainstorming is inefficient, due to the production-blocking nature of brainstorming, the occurrence of evaluation apprehension, and the possibility for participants to free ride, which leads to social loafing and downward performance matching.
This study adds to existing literature by investigating the role participants play in differing levels of creative performance by assessing their cognitive style on the group-level, in which a distinction is made between the knowing cognitive style, planning cognitive style and creating cognitive style. Individuals with a knowing cognitive style are associated with a search for facts; they like complex problems when they can find a rational solution. A planning cognitive style is characterized by a need for structure and control; these individuals like a well-structured environment. Lastly, people with a creating style tend to be creative and like experimentation. Problems are regarded as opportunities and are approached freely.
The relation between cognitive style and creative performance was measured through a hierarchical regression analysis. Furthermore, the interacting effects of creative
experience and motivation on the relationship between cognitive style and creative performance are examined. Creative performance was measured by the level of satisfaction participants had with the outcome of the brainstorming session. In regression analysis, first the control variables age, gender and work experience were added. Then the three variables measuring the cognitive style, as well as creative experience and motivation were added to the model. After that the first set of
interaction terms, regarding creative experience, were added; followed by the second set of interaction terms related to motivation in the fourth step.
TABLE OF CONTENTS
EXECUTIVE SUMMARY 2
1. INTRODUCTION 4
1.1. Research question 4
1.2. Research aim 5
1.3. Scope and domain 5
1.4. Overview of paper 6 2. THEORETICAL BACKGROUND 6 2.1.1. Brainstorming 7 2.1.2. Limitations of brainstorming 8 2.1.3. Benefits of brainstorming 9 2.2.1. Cognitive styles 10
2.2.2. Relevance of cognitive style 11
2.2.3. Types of cognitive style 12
2.2.4. Linking cognitive style and creativity 13
2.3. Interacting effects 13
2.3.1. Experience with brainstorming 14
2.3.2. Willingness to participate 14
2.4. Summary 15
3. HYPOTHESES AND CONCEPTUAL MODEL 15
3.1. Knowing cognitive style 16
3.2. Planning cognitive style 16
3.3. Creating cognitive style 17
3.4. Interacting effect of creative experience 17
3.5. Interacting effect of willingness to participate 18
4. METHODOLOGY 19
4.1. Sample and data collection 19
4.2. Measurement 20
4.3. Reliability 23
4.4. Analysis 24
4.5. Data aggregation 25
5. RESULTS 25
5.1. Descriptive statistics and Pearson correlations 25
5.2. Linear regression analysis 27
6. DISCUSSION 31
7. CONCLUSION 33
7.1. Research question 33
7.2. Theoretical implications 33
7.3. Managerial implications 34
7.4. Limitations and future research 34
8. REFERENCES 36
1. INTRODUCTION
Nowadays it is widely recognized that innovation is essential to the survival of an organization, and therefore creativity has become a vital managerial imperative for organizations. Creativity by individuals and teams is a starting point for innovation (Amabile, Conti, Coon, Lazenby & Herron, 1996) making idea generation an important phase in the creation of new innovative products and services. Idea
generation can be defined as “the retrieval of existing knowledge from memory and the combination of various aspects of existing knowledge into novel ideas” (Mumford, Mobley, Uhlman, Reiter-Palmos & Doares, 1991; Nijstad & Stroebe, 2006; Paulus & Brown, 2007). Creative ideas can be generated in various ways, from which
brainstorming is one of the most well-known and popular techniques. Brainstorming is an idea generation technique in a group setting with the conviction that social
interaction spurs individuals to greater creativity (Osborn, 1953). Despite its popularity, brainstorming is inefficient according to existing empirical research (Mullen, Johnson & Salas, 1991). Several explanations for brainstorming’s inefficiency that are addressed in prevailing research are free-riding, evaluation apprehension, production blocking (Diehl & Stroebe, 1987) and social loafing (Karau & Williams, 1993; Paulus & Brown, 2007). These explanations are limitations based on the nature of brainstorming, however more factors are likely to influence
performance in brainstorming. This paper aims to understand the limitations, or benefits, that are related to characteristics of selected participants partaking in the brainstorming session, arguing that some people perform better than others. It tries to achieve this by researching the influence of participant’s cognitive style, the consistent way in which he or she organizes and processes information and experience (Messick, 1984), on the creative output of a brainstorming session.
1.1. Research question
Consequently, the following sub questions have been defined: 1. What is brainstorming and why is it inefficient?
2. What is cognitive style and of what dimensions does it consist?
3. In what way do the cognitive styles of knowing, planning and creating relate to creative performance?
4. In what way does creative experience interact the relationship between cognitive style and creative performance in brainstorming?
5. In what way does willingness to participate interact the relationship between cognitive style and creative performance in brainstorming?
1.2. Research aim
This research aims to contribute to brainstorming literature in various ways. Firstly, cognitive styles have been researched in relation to creativity at the individual level (Scott & Bruce, 1995; Sagiv et al., 2010), however less is known about its effect on group-level creative performance. Cognitive styles have not been empirically
researched in relation to brainstorming yet and this research aims to address this void. Secondly, the interaction effect of creative experience with brainstorming and the willingness to participate in a specific session on the relation between cognitive style and creative performance will be researched in order to get a richer understanding of this relationship.
Moreover, the results of this research are of managerial value to organizations that use brainstorming as an idea generating technique. It can aid managers to select the most suitable participants for a brainstorming session based on their cognitive style and in this way they can compose the most productive brainstorming group possible.
1.3. Scope and domain
This research solely focuses on the influence of cognitive styles on the idea generating process in brainstorming. What is done with generated ideas afterwards is beyond the scope of this research and will not be discussed. Furthermore, this research does not take into account the inefficiencies of the nature of brainstorming identified by Diehl & Stroebe (1987), but will solely focus on the influence of cognitive style on
1.4. Overview of paper
In the upcoming section an overview of the literature concerning brainstorming and cognitive style will be elaborated. Furthermore, a table with definitions of key constructs used throughout this research will be given (table 1). After this the hypotheses are developed and depicted in a conceptual model; more precisely the expected relationship between the predominance of certain cognitive styles within a brainstorming group and the relation to creative output, as well as the anticipated moderating role of experience and a participant’s willingness to participate. After that the methodology of this research is provided, followed by the results of the analysis. Lastly, these results will be discussed, the theoretical and managerial implications will be pointed out and the limitations will be considered.
2. THEORETICAL BACKGROUND
Creativity is the starting point of innovation and can be defined as the ability to produce ideas that are novel and useful (Amabile, 1996). When confronted with a problem, organizations aim to solve these by employing creative problem solving. Creative problem solving is a process consisting of three steps: idea generation, idea evaluation and idea selection (Terwiesch & Ulrich, 2009). The focus of this research is on idea generation, the first stage of the creative problem solving process. Idea
Table 1. Definition of key constructs
Construct Definition
Creativity The ability to produce ideas that are novel and useful (Amabile, 1996).
Idea generation The retrieval of existing knowledge from memory and the combination of various aspects of existing knowledge into novel ideas (Mumford et al., 1991).
Brainstorming Idea generation technique in a group setting with the conviction that social interaction spurs individuals to greater creativity (Osborn, 1953).
Nominal group Idea generation technique in which people generate ideas individually, without interaction with other participants, and which outputs are combined in the end (Mullen et al, 1991). Cognitive style Consistent individual differences in ways of organizing and
processing information and experience (Messick, 1984). Knowing
cognitive style
People with a knowing style look for facts and data. They like complex problems if they can find a clear and rational solution (Cools, 2007).
Planning cognitive style
People with a planning style are characterized by a need for structure. They like to organize and control and prefer a well-structured environment (Cools, 2007).
Creating cognitive style
People with a creating style tend to be creative and like experimentation. They see problems as opportunities and
challenges, and they like uncertainty and freedom (Cools, 2007)
2.1.1. Brainstorming
Brainstorming is an idea generation technique popularized by Osborn (1953) reasoning that social interaction in a group setting spurs individuals to greater creativity. Osborn (1953) argued that it would generate twice as many ideas in comparison to established techniques. The logic behind this is that by interacting with people with different skills and knowledge individuals are able to recombine existing ideas in new context, which is expected to lead to a higher level of creativity (Kratzer, Leenders & Van Engelen, 2004). Osborn (1957) argues that the effectiveness of brainstorming could be
peers. Secondly, freewheeling is permitted, which means participants are encouraged to use their imagination and should strive for original ideas even if these seem
unrealistic. Moreover, brainstorming is based on the quantity of ideas believing that a higher quantity of ideas is more likely to contain useful ideas. Lastly, participants are expected to combine and improve generated ideas to form new ideas. Brainstorming can take on a variety of different formats, for example with or without an external facilitator, located internally or externally or with different time limits or types of problem statements.
2.1.2. Limitations of brainstorming
Group brainstorming is often compared with another idea generating technique by the use of nominal groups (Diehl & Stroebe, 1987; 1991; Mullen, Johnson & Salas, 1991; Paulus & Dzindolet, 1993; Rietzschel, Nijstad & Stroebe, 2005). Nominal groups are groups of individuals whose outputs are combined, but without interaction during the idea generation process (Mullen et al, 1991). Despite Osborn (1957) arguing that brainstorming would generate twice as many ideas compared to conventional
techniques, laboratory research has shown that process losses are more common than process gains in brainstorming and that the use of nominal groups is regarded as a more effective tool for idea generation (Mullen et al., 1991; Diehl & Stroebe, 1987; 1991). Given the same problem statement and time nominal groups produced a higher creative output.
There are several factors that negatively influence the productivity of brainstorming groups (Diehl & Stroebe, 1987) and which do not come into play when generating ideas individually. Firstly, brainstorming is production blocking (Lamm &
Trommsdorff, 1973; Diehl & Stroebe, 1987) since only one individual can speak at the same time and participants have to wait for their turn to express their ideas. A part from the obvious limitation of time, other participants might also forget their own ideas while listening to the speaker or get a feeling that their idea is somewhat similar and restrain from speaking up. This leads to a lower quantity of generated ideas. Secondly, participants might fear to speak up due to the presence of peers, which is called
evaluation apprehension (Diehl & Stroebe, 1987). They are afraid that their
task. Social loafing is one of the main causes of free riding (Karau & Williams, 1993; Paulus & Brown, 2007). Social loafing means that participants have a feeling of lower accountability for their individual performance when working in a group compared to when working individually. Free riding can ultimately can lead to downward
performance matching (Paulus & Dzindolet, 1993): other participants also lower their efforts because they notice people are putting in less effort compared to their own contribution. According to Diehl & Stroebe (1987) production blocking is the main issue, which can partly be overcome by assigning a trained facilitator to the brainstorm session (Offner, Kramer & Winter, 1996), who can be used to motivate participants and lead the process by reminding the group of the rules of brainstorming.
Additionally, using written notes or post-its (Paulus & Yang, 2000) to share ideas also decreases the influence of production blocking because participants can write down ideas simultaneously.
Table 2. Reasons why brainstorming is regarded as ineffective.
Reason Explanation
Production blocking Only one participant can speak at a time (Diehl & Stroebe, 1987).
Evaluation apprehension
Participants might fear to speak up due to the presence of peers and possible negative reactions (Diehl & Stroebe, 1987) Free-riding Individuals rely on effort of others when working on a task
(Diehl & Stroebe, 1987).
Social loafing Individuals have a feeling of reduced accountability for one’s individual performance when working in groups (Karau & Williams, 1993)
Downward performance matching
Tendency for group members to converge to a similar low level of performance (Paulus & Dzindolet, 1993).
2.1.3. Benefits of brainstorming
There are also factors that positively influence productivity of interactive groups. Downward performance matching (Paulus & Dzindolet, 1993) has been mentioned as one the disadvantages of brainstorming, however, contrariwise performance can also be explained positively. Exposure to productive brainstorming participants can push the performance of other participants to a higher level.
Additionally, brainstorming groups are likely to develop higher levels of cohesiveness, which has a positive effect on effectiveness (Henningsen & Hennigsen, 2013).
stimulating effects and spark a good idea from a participant’s less accessible area of knowledge (Nijstad & Stroebe, 2006). Additionally, this also leads to the recovery of more original ideas (Dugosh & Paulus, 2005). Furthermore, there is higher level of persistence in group brainstorming, which leads to longer periods of idea generation and therefore more ideas being generated (Nijstad & Stroebe, 2006). This is because a group generates a greater total number of ideas compared to individuals and as long as idea production is still reasonably high this is used by people as a signal to keep generating more ideas. A single individual would therefore stop earlier because at a certain point his production has declined to a low level and said individual is likely to use this as a signal to stop.
Despite the fact that brainstorming is seen as less effective, it remains one of the most popular and widely used techniques of idea generation. Due to the perseverance of organizations to use brainstorming it is interesting to look at factors that can positively influence its creativity outputs. The factors described in table 1 are not the only
possible explanations for the inefficiency of brainstorming.
Cognitive styles of participants are another explanation for possible differences in a brainstorming session’s creative output.
2.2.1. Cognitive styles
words such as analytic, deductive, rigorous, constrained, convergent, formal and critical (Nickerson, Perkins & Smith, 1985). Intuitive thinkers are more disposed toward novelty seeking without being restricted by rules and standards. On the other hand, individuals with an analytical-rational cognitive style translate new events in terms of existing knowledge, being more bound to rules in their problem-solving behaviour (Sagiv, Arieli, Goldenberg & Goldschmidt; 2010). All people use both modes of information processing, contingent to the type of situation and task (Smith & DeCoster, 2000), however one of these modes becomes recurrently prevailing for an individual and reflects his or her cognitive style (Sagiv et al., 2000).
Table 3. Definitions of cognitive style.
Witkin, Moore,
Goodenough & Cox (1977)
The individual way a person perceives, thinks, learns, solves problems, and relates to others.
Hunt, Krzystofiak, Meindl & Yousry (1989)
The way people process and organize information and arrive at judgements or conclusions on the basis of their observations.
Green & Schroeder (1990) Information-processing habits representing people’s dominant or preferred modes of perceiving, thinking, remembering, and problem solving.
Messick (1994) Consistent individual differences in ways of organizing and processing information and experience.
Cools (2007) The way people perceive stimuli and how they use this information to guide their behaviour (i.e., thinking, feelings, actions).
2.2.2. Relevance of cognitive style
guidance, task design, team composition, conflict management and training and development. The relation between cognitive style and team composition is especially interesting for this research, since the aim of this paper is to compose a brainstorming group of most suitable participants, based on cognitive style, to achieve creative performance. Based on Robertson’s (1985) work they argue cognitive style
characteristics provide a better basis for predicting performance on certain tasks than traditional measures of cognitive ability. Accordingly, Mintzberg (1976) states that tasks demanding logic and articulation require a rational cognitive style whereas jobs concerning ambiguity and complexity necessitate a more intuitive style.
2.2.3. Types of cognitive style
2.2.4. Linking cognitive style and creativity
Previous research indicates that cognitive styles at the individual level affect creative performance (Scott & Bruce, 1994; 1995; Sagiv et al., 2010). Cognitive styles have been differentiated in many different types, however these types are often argued to be different conceptions of the same dimension: analytical or intuitive (Allinson & Hayes, 1996). Scott & Bruce (1995) found that individuals who approach problems in a rational manner are less likely to be innovative because rationality limits the
boundaries of alternative problem formulation. In an earlier publication (1994) they come to similar results. They find creativity is related to the intuitive cognitive style because individuals examine information from various paradigms concurrently and are therefore more likely to come up with creative solutions. They use various areas of thought and use their imagination. On the other hand, systematic individuals come up with conservative explanations because they follow consistent methods and processes when problem solving. These studies discovered general innovative behaviour to be positively correlated with intuitive decision-making style and negatively correlated with analytical style. However, the effect of cognitive style on actual creative behaviour has rarely been examined empirically (Sagiv et al., 2010). Sagiv et al. (2010) address this gap by directly investigating the effect of analytical versus intuitive cognitive style on the creativity of a solution to a problem. They found empirical support for Scott & Bruce’s (1994; 1995) claims that intuitive participants generated more creative solutions. Furthermore, they found that under specific conditions
analytical individuals can be just as creative. This depends on the characteristics of the task at hand and the instructions participants received beforehand. Whereas intuitive individuals are more creative than analytical ones under free conditions, under highly structured conditions analytical people can reach the same level of creativity. Existing research focuses on the effect of cognitive styles on creativity on the individual level, however less is known about the effect of group-level prevalence of a specific
cognitive style on creative performance.
2.3. Interacting effects
2.3.1. Experience with brainstorming
The effect previous experience with brainstorming has on the relationship between the prevalence of a specific cognitive style and creative performance will be assessed by looking at the total number of brainstorming sessions participants have attended in their life. It is especially interesting to see if participants with an unfavourable cognitive style are able to overcome these disadvantages due to their experience. Several explanations for this are the fact that those participants who have brainstormed frequently before are accustomed to the rules of brainstorming (such as no criticism) and will feel more comfortable to share unusual ideas. Existing research examined the effect of experience on team creativity in a product-development setting (Gino, Argote, Spektor, Todoreva, 2009). They found that experience facilitated group creativity and also led to a higher level of component divergence. Products scoring high on
component divergence consist of new materials and technologies and therefore ideas need to be more imaginative to come up with this. Furthermore, prior research came to similar results. For example, previous experience leads to a faster execution of creative ideas (Taylor & Greve, 2006) and permits individuals to identify opportunities to be creative (Shane, 2000). By examining the role of participant’s experience with
brainstorming this will add value to literature by providing empirical evidence whether creative behaviour can be taught, or at least developed.
2.3.2. Willingness to participate
Idea generation is a cognitive process strongly moderated by social and motivational factors (Paulus & Brown, 2007). This indicates that it is likely willingness to
participate, which can be characterized as a motivational factor, will have an effect on the relationship between the prevalence of a specific cognitive style in a brainstorming session and creative performance. Many factors can influence a participants
contribute because of his or her enthusiasm and motivation. An individual’s intrinsic motivation is one of the most essential personal qualities for creative behaviour (Amabile, 1988; Tierney, Farmer & Graen; 1999) and will influence an individual’s willingness to participate.
2.4. Summary
Brainstorming is the most popular idea generation technique used by organizations, arguing that social interaction during the idea generation process spurs individuals to greater efficiency. It aims for the quantity, rather than the quality of ideas, based on the assumption that a larger set of generated ideas is likely to contain a larger set of useful ideas. However, empirical research shows that nominal groups are more effective than interactive groups. Factors that explain why nominal groups produce a higher level of creative compared to brainstorming are production blocking, evaluation apprehension, free-riding (Diehl & Stroebe; 1987), social loafing (Karau & Williams, 1993) and downward performance matching (Paulus & Dzindolet, 1993). This research investigates the role of personal characteristics of participants, by assessing their cognitive style, on performance in brainstorming. The cognitive style of participants influences the way in which he or she organizes and processes information and
experience and is therefore likely to lead to differences in performance. The distinction is made between a knowing, planning and creating cognitive style (Cools, 2007), from which the first two are correlated with analytical dimension of cognitive style and the latter with the intuitive style. Prevailing empirical research showed that individuals with an intuitive style are more creative on the individual-level compared to
individuals with an analytical cognitive style (Scott & Bruce, 1995). This research aims to examine whether this same effect holds on group-level.
3. HYPOTHESES AND CONCEPTUAL MODEL
intuitive cognitive style are more creative, that a prevalence of this style in a group setting leads to a higher level of creative performance as well.
3.1. Knowing cognitive style
Individuals with a knowing cognitive style rely on facts and data; they search for detailed explanations and information and look for rational solutions for complicated problems (Cools, 2007). During a brainstorm session not all required information is necessarily available at the start. Participants are encouraged to think outside the box and to leave existing structures when solving problems in order to achieve creative ideas. Furthermore, participants are required to contribute to idea generation immediately without a lot of time to integrate and assimilate outside information. Moreover, the knowing cognitive style is negatively correlated with the overall KAI scores (Kirton, 1976), which indicates that they are likely to stay within existing structures when solving problems. This leads to the following hypothesis:
H1A: A brainstorming group in which there is a prevalence of the knowing cognitive style negatively affects creative performance in brainstorming
3.2. Planning cognitive style
A planning cognitive style is characterized by a need for structure and control. Individuals with aforementioned cognitive style prefer to extensively prepare
themselves before solving a problem. Rules and regulations, step-by-step explanations and consistent procedures are warranted to reach objectives (Cools, 2007). Following a structured path is counterproductive according to existing literature on brainstorming. Furthermore, brainstorm sessions frequently do not require any preparation and often the purpose of the session is not even clearly articulated beforehand. This is likely to hinder the creative output of individuals with a planning cognitive style. Additionally, the knowing and planning cognitive style are negatively correlated with the overall KAI scores (Kirton, 1976), meaning that said individuals stay within the existing structure when solving problems, which makes it unlikely that they come up with original ideas. This gives further support for the expected negative relation between the planning cognitive style and creative performance. Consequently, the following
H1B: A brainstorming group in which there is a prevalence of the planning cognitive style negatively affects creative performance in brainstorming.
3.3. Creating cognitive style
Lastly, individuals with a creating cognitive style tend to be creative and like
experimentation. They approach problems as opportunities and challenges, and they like uncertainty and freedom. They prefer dynamic structures when problem solving and are constantly searching for hidden possibilities and new horizons (Cools, 2007). These characteristics are favourable for creative behaviour and therefore a positive effect is expected between a brainstorming group in which the creating cognitive style is dominant and its creative performance. This expected relation is further warranted by the positive correlation between the creating cognitive style and overall KAI scores (Kirton, 1976). Individuals with a positive KAI score tend to more innovative and restructure the situation when solving problems and making decisions. This leads to the following hypothesis:
H1C: A brainstorming group in which there is a prevalence of the creating cognitive style positively affects creative performance in brainstorming
3.4. Interacting effect of creative experience
Experience of participants with brainstorming is expected to be an interacting effect on the relationship between cognitive style and creative performance. When a
brainstorming group consists of participants with a lot of experience it is likely that they are familiar with all the rules related to it. Furthermore, they are used to trying to come up with creative solutions to problems. Even if this group prevalently possesses an unfavourable cognitive style on the group level, either the knowing or planning cognitive style, the aforementioned advantages are likely to overcome part of their disadvantage. One of the disadvantages an individual with an either knowing or
planning cognitive style has is that they require more time to assimilate knowledge and that they need more time to prepare themselves (Cools, 2007). However, Taylor & Greve (2006) discovered that previous experience with idea generation results in faster execution of creative ideas. Consequently, the following hypotheses have been
H2: Cognitive style and creative experience interact such that brainstorming groups consisting of a high level of average creative experience perform significantly better than brainstorming groups consisting of a low level of average creative experience, leading to a higher level of satisfaction with the outcome of the brainstorming session.
3.5. Interacting effect of willingness to participate
One of the factors that likely to affect an individual’s willingness to participate is differential ability (Stroebe & Frey, 1982; Stroebe et al., 2010). This means that participants feel that their contribution does not really add much to the group product and is therefore dispensable. For example, when all the other participants are experts concerning the specific problem at hand and one sees oneself as having a lesser degree of knowledge concerning that subject, said person might not try hard to contribute. Furthermore, when contribution is costly and it is hard to identify an individual’s contribution free riding might occur, due to social loafing. Other factors that influence an individual’s willingness to participate are his motivation and the degree to which he likes the subject at hand. All in all, it is expected that when all participants are willing to participate this will lead to a larger amount of ideas generated. Consequently, more ideas will be useful and therefore this is beneficial for creative performance. The following hypotheses have been constructed:
Figure 1 depicts the conceptual model.
4. METHODOLOGY
This research follows a quantitative approach since the research questions aim to investigate whether specific determinants predict behaviours at a statistically
significant level. Specifically, this research aims to investigate how different cognitive styles affect performance in brainstorming. The sample and data collection procedures, the measures used for analysis, reliability of used constructs, the models of analysis and the way in which data is aggregated to group-level are explained in the following sections.
4.1. Sample and data collection
collection was done with six students, all writing a thesis related to the theme of brainstorming, and therefore only part of the total questionnaire was useful for this specific research. A total of 450 companies were contacted, either by e-mail or by telephone, asking if they had a brainstorming session scheduled in the upcoming period that could be attended. A second, follow-up e-mail was sent to non-response companies in order to increase the number of sessions we could attend. In total, 68 meetings were attended, from which 66 meetings were useful for our analysis. This results in a response rate of 14,7%. These 66 meetings combined consisted of a total of 422 participants, from which 73 were initiators. From the final sample 58.4% are male and 41.6% female. Age ranged from 18 to 74 years old (M = 36.39, SD = 12.29). Work experience ranged from less than a year to 50 years (M = 14.56, SD = 11.97). The brainstorming sessions attended was heterogeneous in nature, meaning that they differed on various aspects such as time limits, structure, idea generating techniques and group sizes. The only thing they surely have in common is the fact that they are generating ideas in groups.
4.2. Measurement
Dependent variable:
Creative performance of a brainstorming session was measured by satisfaction. Both participants and initiators were asked to rate their level of satisfaction with the
outcome of the session on a 5-point Likert scale (1 = very low level of satisfaction, 5 = very high level of satisfaction).
At first, creative performance was measured by the total amount of useful ideas assessed by both the initiator and the participants. Existing literature on brainstorming has predominantly focused on productivity as a measurement of efficiency (Sutton & Hargadon, 1996), however productivity itself does not necessarily mean that this large pool of generated ideas also includes a larger collection of useful ideas. Therefore, only the ideas that were assessed as useful were taken into account. After all, useful ideas are the ones that are most likely to get developed further, and ultimately,
due to different perceptions of a useful idea. Additionally, due to the heterogeneous nature of this sample the number of useful ideas that indicate a brainstorming session is useful can vary greatly between sessions. Therefore this approach had to be abandoned and satisfaction was used as dependent variable.
Independent variables:
Knowing cognitive style was measured using the CoSI construct of Cools (2007). This
is a 4-item scale assessing the extent to which participants make analyses and like to have a full understanding of problems. A high score on this variable indicates that a person looks for facts and data and likes complex problems if they can find a rational solution. Knowing cognitive style is measured on a 5-Point Likert scale (1 = low degree of knowing cognitive style, 5 = high degree of knowing cognitive style).
Planning cognitive style was measured using the CoSI construct of Cools (2007). This
is a 7-item scale assessing the tendency of participants to make detailed plans and their preference of preparation. A high score on this variable indicates a need for structure; participants with a planning cognitive style like to organize and control. Planning cognitive style is measured on a 5-Point Likert scale (1 = low degree of planning cognitive style, 5 = high degree of planning cognitive style).
Creating cognitive style was measured using the CoSI construct of Cools (2007). This
is a 7-item scale assessing a participant’s preference for innovative solutions and their preference for avoiding routines and extending boundaries. A high score on this variable indicates that people see problems as opportunities, and they like uncertainty and freedom when solving them. Creating cognitive style is measured on a 5-Point Likert scale (1 = low degree of creating cognitive style, 5 = high degree of creating cognitive style).
The complete CoSi construct (18 items) will be included in Appendix 1.
Moderating variables:
Creative experience of a participant was measured by the total amount of
brainstorming sessions a participant has previously attended in his or her life. The questionnaire included an open question, asking the participant to fill in the approximate number of attended previously attended brainstorming sessions.
motivation for participating in that specific brainstorming session. Participants were asked to fill in the extent to which they were motivated for that specific brainstorming session on a 5-point Likert Scale (1 = very low level of motivation, 5 = very high level of motivation).
Control variables:
Control variables are incorporated to provide a stronger test of the hypotheses. The control variables used in this research are age, gender and tenure. These factors should be controlled for since individuals with a high age or tenure can limit said individual to a certain way of working, either increasing or decreasing, creative behaviour.
Similarly, gender is used as control variable in order to control for the fact that the outcomes might be gender-specific.
Table 4. Overview variables
Variable Type Measurement Scale Reference
Age Control variable Years Rational Gender Control variable 0 = Male 1 = Female Nominal Work Experience Control variable Years Rational Knowing Cognitive Style Independent variable 5-point Likert Scale Interval Cools (2007) Planning Cognitive Style Independent variable 5-point Likert Scale Interval Cools (2007) Creating Cognitive Style Independent variable 5-point Likert Scale Interval Cools (2007) Creative Experience Moderator Numbers of brainstorm sessions Rational
4.3. Reliability
Cronbach’s α is calculated to assess the reliability of the cognitive style construct. It calculates how closely related a set of items are as a group. Cronbach α is accepted when it is higher than universally accepted threshold of 0.70. Afterwards a factor analysis is conducted to look at the unidimensionality of the constucts. Table 5 shows the Cronbach alpha α for the knowing, planning and creating cognitive style. All three variables pass the threshold of 0.70 and are therefore acceptable. Furthermore, none of the variables’ Cronbach α becomes stronger after removing the weakest item, meaning all constructs can be reliably used.
Table 5. Cronbach α
Variable Cronbach’s α Knowing Cognitive Style 0.79
Planning Cognitive Style 0.88
Creating Cognitive Style 0.89
A principal components factor analysis was conducted to assess the unidimensionality of each construct. For knowing cognitive style, the total variance explained by the first item is 61,10%. Furthermore, the eigen value for the first factor is quite a bit larger than the eigen value of the next factor (2.44 versus 0.64). This suggests that the scale items are unidimensional. Likewise, planning cognitive style and creating cognitive style show similar results. The total variance explained for the first item of planning cognitive style is 57,51%, whereas the eigen value of the first factor in comparison to the second factor is 4.03 versus 0.85. Lastly, the total variance explained for the first item of creating cognitive style is 61,52%. The eigen value of the first factor in relation to the second factor is 4.31 versus 0.75. Table 6 shows the results of the factor
analysis.
Table 6a. Factor analysis knowing cognitive style
Table 6b. Factor analysis planning cognitive style
Component Initial Eigenvalue Total Initial Eigenvalue % of Variance 1 4.03 57.50 2 0.85 12.13 3 0.63 8.98 4 0.51 7.32 2 0.38 5.49 3 0.32 4.60 4 0.28 3.97
Table 6c. Factor analysis creating cognitive style
Component Initial Eigenvalue Total Initial Eigenvalue % of Variance 1 4.31 61.52 2 0.75 10.72 3 0.48 6.86 4 0.43 6.01 2 0.40 5.65 3 0.35 5.02 4 0.29 4.14 4.4. Analysis
The study uses SPSS software to conduct the statistical analysis. Descriptive analysis was conducted to explore the data set. After this, individual measurements were
gender, age and work experience were submitted into the model; afterwards in step 2 the independent variables and moderators were entered, followed by the first set of interaction terms regarding creative experience in step and the second set of interaction terms concerning motivation in step 4.
4.5. Data aggregation
Data was collected individually and aggregated to group-level data using the mean scores of the variables per meeting. Additionally, in order to check if similar results would be found, data was also aggregated using the standard deviations of the variables per meeting. Results of the bivariate analysis and regression analysis using this approach are found in Appendix 3. Not many striking correlations could be found and this was the main reason I decided to use aggregation by mean as data aggregation technique during my analysis, since it had more explanatory power. In the regression analysis the only significant relationship that was found was between the creating cognitive style and satisfaction. Furthermore, it is interested to note that in the regression analysis none of the values of R2 change or adjusted R2 are significant.
5. RESULTS
In the following section the results of the conducted analysis will be discussed. At first the correlations found in the bivariate analysis will be explained, followed by the results of the linear regression analysis.
5.1. Descriptive statistics and Pearson correlations
Furthermore, a significant correlation can be found between motivation and
satisfaction (r = .499, p < 0.01), indicating that participants who are more motivated to join the brainstorming session are also more satisfied with the results. Motivation is therefore positively related to creative performance. Motivation is also strongly correlated with the creating cognitive style (r = .534, p < 0.01), which indicates that brainstorming sessions with a prevalent creating cognitive style are more motivated to participate. People with a creating cognitive style are creative and see problem solving as a challenge and therefore brainstorming is an activity they enjoy and are motivated for. This is also confirmed by the positive correlation between creating cognitive style and creative experience (r = .270, p < 0.05). Since individuals with a creating cognitive style enjoy brainstorming and are creative, and they are likely to pursue jobs in which they frequently participate in brainstorming, their creative experience is higher.
There is also a positive correlation between creative experience and satisfaction, indicating that creativity experience enhances creative behaviour. This correlation is however not significant (r = .137, p = n.s.) so further research is warranted on this relationship. It is logical to assume however that by frequently participating in brainstorming, people become familiar with the rules associated with brainstorming and are trained to develop an information processing style that enhances performance.
Furthermore, a negative correlation can be found between the creating cognitive style and gender (r = -.211, p < 0.10). Gender scores were 0 for male and 1 for female, so a negative correlation means that men are more creative. When gender score moves from 0 to 1, or in other words from male to female, a negative relation to creating cognitive style is found, indicating that women are less inclined to have a creating cognitive style and are therefore less creative.
5.2. Linear regression analysis
Table 4 shows the results of the hierarchical linear regression analysis performed to test the hypotheses. In the first step, the control variables gender, age and work
experience were submitted into the model; then in step 2 the independent variables and moderators were entered, followed by the first set of interaction terms in step 3 and the second set of interaction terms in step 4. The interaction terms were separated so R2 change and adjusted R2 can be interpreted separately for the interaction terms. R2 measures the closeness of the data to the fitted regression line by looking at the percentage of the response variable variation. Adjusted R2 compares the explanatory power of regression models. The adjusted R2 increases when a new term in the regression analysis improves the model.
Hypothesis 1A states that knowing cognitive style is negatively related to creativity in brainstorming. The regression analysis does not confirm this hypothesis, however the relation between knowing cognitive style and satisfaction is negative (b = -.025, p = n.s.). This relationship is not significant.
Hypothesis 1C states that the creating cognitive style is positively related to creativity in brainstorming. This is confirmed by the results from the regression analysis, showing a positive relation (b = .595, p < 0.01) between the creating cognitive style and satisfaction. Individuals with a creating cognitive style are innovative and see problem solving as a challenge (Cools, 2007), which makes them suitable participants in brainstorming.
Hypothesis 2 looks at the interacting effect of creative experience with cognitive style, stating that brainstorming groups consisting of a high level of average creative
Table 6. Correlation table (N=66)
† p < .10, * p < .05, ** p < .01
Gender scores were 0 for male and 1 for female; age, work experience and creative experience were measured in years and the remaining variables on 5-point Likert scales. Data was aggregated using mean scores.
Variable Min Max Mean SD 1 2 3 4 5 6 7 8
Table 7. Regression analysis (N=66)
Satisfaction
Steps and variables entered 1 2 3 4
1. Age 0.48† -.012 -.012 -.010
Gender -.340 .037 .036 .44
Work Experience -.067* .001 .000 .002
2. Knowing cognitive style -.025 .002 .000
Planning cognitive style .300† .263 .209
Creating cognitive style .595** .596** .487**
Creative experience .000 .000 -.001 Motivation .161 .163 .182† 3. KnowCognXCreaExp .022 .014 PlanCognXCreaExp -.057 -.023 CreaCognXCreaExp -.010 -.034 4. KnowCognXMot -.127 PlanCognXMot .088 CreaCognXMot -0.68 R2 change .158* .404** .002 .036 Adjusted R2 .118* .501** .476** .491** a
N = 66. Unstandardized regression coefficient are reported for the respective regression steps. Data was aggregated using mean scores.
6. DISCUSSION
In this section the results of the hierarchical regression analysis will be discussed and related to relevant theory. Table 8 depicts an overview of the tested hypotheses.
Table 8. Hypotheses
H1A A brainstorming group in which there is a prevalence of the knowing cognitive style negatively affects creative performance in
brainstorming.
H1B A brainstorming group in which there is a prevalence of the planning cognitive style negatively affects creative performance in
brainstorming.
H1C A brainstorming group in which there is a prevalence of the creating cognitive style positively affects creative performance in
brainstorming.
H2 Cognitive style and creative experience interact such that
brainstorming groups consisting of a high level of average creative experience perform significantly better than brainstorming groups consisting of a low level of average creative experience, leading to a higher level of satisfaction with the outcome of the brainstorming session.
H3 Cognitive style and motivation interact such that brainstorming groups consisting of a high average level of motivation perform significantly better than brainstorming groups consisting of a low average level of motivation, leading to a higher level of satisfaction with the outcome of the brainstorming session.
Firstly, hypothesis H1A expects a negative relationship between the knowing
cognitive style and creative performance in brainstorming. During regression analysis a negative effect was found. However, this relation was not significant, and therefore hypothesis H1A is not confirmed. Future research should try to confirm this
relationship by repeating this research on a larger data set.
The most remarkable result is that the relationship between planning cognitive style and creative performance was positive. Hypothesis H1B predicts a negative
relationship between the planning cognitive style and creative performance in brainstorming. Prevailing literature on the planning cognitive style states that individuals with a planning style are characterized by a need for structure.
and emphasize routines (Allison & Hayes, 1996). Since individuals produce more creative work when they have freedom in how to accomplish their tasks (Amabile et al., 1996), a structured, routine-based approach was not expected to have a positive effect on creative output.
A possible explanation that addresses the differing results between the knowing and the planning style is that the knowing cognitive style is based on characteristics which concern the way in which a problem is tackled, whereas the planning cognitive style is more concerned with a preference for certain structural preferences (step-for-step explanations, preparation). If these structural preferences are met, individuals with a planning cognitive style can still display creative behaviour. This is, however, not likely for individuals with a knowing cognitive style since their characteristics are not based on structure, but on the problem statement itself. They want to have a full understanding of problems, understand the underlying logic and make detailed analyses. This prevents them from recombining generated ideas quickly and coming up with very explorative ideas. Furthermore, individuals with a knowing cognitive style are characterised as impersonal (Myers et al., 2003), which might hinder their interaction with other brainstorming participants. An additional explanation for the positive relation between planning cognitive style and creativity is that individuals with a planning cognitive style are highly efficient because of their superior planning. This efficiency allows them to generate ideas quickly and therefore results in a higher output of ideas causing a higher level of satisfaction. Furthermore, planning cognitive style was positively correlated (appendix 3) with the level of preparation done before the brainstorming session. It is a logical explanation that this made them perform well in the brainstorm session.
No support is found for the interacting effects of motivation and creative experience on the relationship between cognitive style and creative performance. These relations should be tested again on a bigger data set to find out whether or not these expected relationships hold. A possible explanation for the unconfirmed hypotheses are the limitations of the data collected, which will be further explained in the limitations section.
7. CONCLUSION 7.1. Research question
Brainstorming is ineffective (Diehl & Stroebe, 1987; Mullen et al., 1991), which is predominantly attributed to factors concerning the nature of brainstorming. This research finds support that the cognitive style of a participant is related to creative performance in brainstorming and can either hinder or benefit said individual during idea generation. When a brainstorming group consists predominantly of individuals with a creating or planning cognitive style this has a positive effect on creative performance, therefore finding support for the research question. No support was found for the expected negative relation between knowing cognitive style and
satisfaction. Furthermore, the interaction effects of motivation and creative experience were not confirmed in the analysis.
7.2. Theoretical implications
Prevailing research on cognitive style which linked cognitive style to creativity focused on the distinction between analytical and cognitive style (Sagiv et al. 2010; Scott & Bruce, 1995) and found that an analytical cognitive style was negatively related to creativity. This research indicates that the established division between an analytical and an intuitive dimension of cognitive style (Allison & Hayes, 1996) is too broad. Both planning and knowing cognitive style can be allocated into the analytical dimension, however they have different effects on creative performance. The
Furthermore, this research has implications for theory on brainstorming. While brainstorming has predominantly focused on reasons of inefficiency based on the brainstorming process of brainstorming (Diehl & Stroebe, 1987), selecting the right participants for the job should receive more attention. By composing a brainstorming group with the most suitable participants performance can be increased.
7.3. Managerial implications
Schweiger (1983) states that when a particular cognitive style is more appropriate in comparison to another style when conducting certain managerial activities, it affects the selection and composition of suitable individuals for these tasks. This study confirms that differences in cognitive style between participants results in different levels of creative behaviour in brainstorming and can therefore be used as a possible explanation for the inefficiency of brainstorming. Creating and planning were both positively related with high performance in brainstorming, measured by a
participant’s satisfaction with the outcome. Knowing cognitive style showed a negative relationship, although this relation was not significant. An important implication that can be derived from these results is that managers should keep in mind participant’s cognitive style when composing a brainstorming group. A manager should select individuals with the appropriate cognitive style, resulting in more
successful brainstorming sessions. Especially in an organizational context, managers can assess the cognitive styles of participants at some point in time and use these scores at a later point in time to invite the most useful employees to form a
brainstorming group. Furthermore, one might also adjust group composition based on the nature of the brainstorming session. If the session is highly structured more people with a planning cognitive style could be invited, whereas in a session with a problem statement regarding something intangible more people with a creating cognitive style could be asked to join. However, this is not empirically tested in this research and therefore an interesting field for future research, it is a logical assumption, however, that different natures of brainstorming session also require different group
compositions to come to the desired results.
7.4. Limitations and future research
This means that participants have to judge their own characteristics, which might cause them to fill in scores which are expected or preferred instead of actual scores. It is common knowledge creative behaviour is wanted in brainstorming, so participants might rate themselves too high on items regarding creativity in order to come across as a suitable person for the job. Secondly, a heterogeneous sample of brainstorming sessions is examined. This data set includes data from brainstorming sessions with different type of problem statements, problem-solving techniques, time limits and group sizes. Varying results of brainstorming outputs might be due to these
operational factors even though they are not included in this research. Additionally, the size of the sample used is relatively small. Future research should aim to conduct a similar research using a larger data set with similarly structured brainstorming
sessions to see if similar results are reached. Thirdly, due to practicality reasons I used the CoSI questionnaire (Cools, 2007), because it is a short and convenient assessment of cognitive style. It makes the distinction between knowing, planning and creating cognitive style. However, most research is conducted on the difference between analytical and intuitive cognitive style and part of my theory building is based on that too. The questionnaire measuring (Epstein, 1990) those two dimensions of cognitive style was however too elaborate and time-consuming, resulting in my choice for the CoSI questionnaire. Lastly, the dependent variable used in this research is not optimal. Being satisfied with outcome does not necessarily have to mean that many useful ideas are generated; it can for example also mean that participants are content with their contributions or with the realization other participants find the problem statement at hand important. Furthermore, individuals might fill in a relatively high score out of politeness. Future research should aim to find a reliable way of using the total amount of useful ideas, or the total amount of ideas which are eventually
8. REFERENCES
Allinson, C. W. & Hayes, J. (1996). The Cognitive Style Index: A measure of
intuition-analysis for organizational research. Journal of Management Studies, 33(1), 119-135
Amabile, T. M. (1988). A model of creativity and innovation in organizations.
Research in Organizational Behavior, 10, 123-167
Amabile, T. M. (1986). Creativity in context. Westview Press, Boulder, CO
Amabile, T. M., Conti, R., Coon, H., Lazenby, J., Herron, M. (1996). Assessing the work environment for creativity. Academy of Management Journal, 39(5), 1154-1184
Beyler & Schmeck (1992) – Assessment of individual differences in preferences for holistic-analytic strategies: evaluation of some commonly available instruments.
Educational and Psychological Measurement, 52, 709-719
Cools, E. & Van den Broeck, H. (2007) Development of the Cognitive Style Indicator. The Journal of Psychology, 141(4), 359-387
Diehl, M., Stroebe, W. (1987). Productivity loss in brainstorming groups: Toward the solution of a riddle. Journal of Personality and Social Psychology, 53(3), 497-509
Diehl, M., Stroebe, W. (1991). Productivity loss in idea-generating groups: Tracking down the blocking effect. Journal of Personality and Social Psychology, 61(3), 392-403
Dugosh, K. L., & Paulus, P. B. (2005). Cognitive and social comparison processes in brainstorming. Journal of Experimental Social Psychology, 41, 313-320
Epstein, S. (1990). Cognitive-experiential self theory. In L. Pervin, Handbook of
Epstein, S., Pacini, R., Denes-Raj, V., Heier, H. (1996). Individual differenes in intuitive-experiential and analytical-rational thinking styles. Journal of Personality
and Social Psychology, 71(2), 390-405
Gino, F., Argote, L.. Miron-Spektor, E., Todorova, G. (2010). First, get your feet wet: The effects of learning from direct and indirect experience on team creativity.
Organizational Behavior and Human Decision Processes, 111, 102-115
Green, K. E., & Schroeder, D. H. (1990). Psychometric quality of the verbalizer-visualizer questionnaire as a measure of cognitive style. Psychological Reports, 66, 939-945
Henningsen, D. D. & Henningsen, M. L. M. (2013). Generating ideas about the uses of brainstorming: Reconsidering the losses and gains of brainstorming groups relative to nominal groups. Southern Communication Journal, 78(1), 42-55
Karau, S. J. & Williams, K. S. (1993). Social loafing: A meta-analytic review and theoretical integration. Journal of Personality and Social Psychology, 65, 681-706
Kaufmann, G. (1979). The explorer and the assimilator: a cognitive style distinction and it’s potential for innovative problem solving. Scandinavian Journal of Education
Research, 23, 101-108
Kirton, M. (1976). Adaptors and innovators: A description and measure. Journal of Applied Psychology, 61, 622-629
Kratzer, J., Leenders, R, Th., A. J., Van Engelen, J. M. L. (2004). Stimulating the potential: creative performance and communication in innovation teams. Creativity &
Innovation Management, 13(1), 63-71
Lamm, H. & Trommsdorff , G. (1973). Group versus individual performance on tasks requiring ideational proficiency (brainstorming): A review. European Journal of
Madjar, N. (2008). Emotional and informational support from different sources and employee creativity. Journal of Occupational and Organizational Psychology, 81, 83-100
Martinsen, O (1997). The construct of cognitive style and its implications for creativity. High Ability Studies, 8(2), 135-158
Messick, S. (1984) The nature of cognitive styles: Problems and promises in educational research. Educational Psychologist, 19, 59-74
Mintzberg, H. (1976). Planning on the left side and managing on the right. Harvard
Business Review, 49-58
Mumford, M. D., Mobley, M. I., Uhlman, C. E., Reiter-Palmon, R. & Doares, L. M. (1991). Process analytic models of creative capacities. Creativity Research Journal, 4, 91-122
Mullen, B., Johnson, C., & Salas, E. (1991). Productivity loss in brainstorming groups: A meta-analytic integration. Basic and Applied Social Psychology, 12, 3-23
Nickerson, R., Perkins, D. and Smith E. (1985). The teaching of thinking. Erlbaum, Hillsdale, N.J.
Nijstad, B. & Stroebe, W. (2006). How the group affets the mind: A cognitive model of idea generation in groups. Personality and Social Psychology Review, 10, 186-213
Offner, A. K., Kramer, T. J., Winter, J. P. (1996). The effects of facilitation, recording and pauses upon group brainstorming. Small Group Research, 27(2), 283-29
Osborn, A. F. (1953). Applied imagination: Principles and procedures of creative problem solving. New York: Scribner
Paulus, P. B. & Brown, V. R. (2007). Toward more creative and innovative group idea generation: a cognitive-social-motivational perspective of brainstorming. Social
and Personality Compass, 248-265
Paulus, P. B. & Yang, H. C. (2000). Idea generation in groups: A basis for creativity in organizations. Organizational Behavior and Human Decision Processes, 82, 76-87
Paulus, P. B. & Dzindolet, M. T. (1993). Social influence processes in group brainstorming. Journal of Personality and Social Psychology, 64, 575-586
Riding, R. J. (1997). On the nature of cognitive style. Educational Psychology, 17, 29-49
Rietzschel, E. F., Nijstad, B. A., & Stroebe, W. (2006). Productivity is not enough: A comparison of interactive and nominal brainstorming groups on idea generation and selection. Journal of Experimental Social Psychology, 42, 244-251
Rietzschel, E. F., Nijstad, B. A., & Stroebe, W. (2014). Effects of Problem Scope and Creativity instructions on Idea Generation and Selection. Creativity Research Journal, 26(2), 185-191
Robertson, I. T. (1985). Human information – processing strategies and styles.
Behaviour and Information Technology, 4, 19-29
Sagiv, L., Arieli, S., Goldenberg, J. & Goldschmidt, A. (2010). Structure and freedom in creativity: The interplay between externally imposed structure and personal
cognitive style. Journal of Organizational Behavior, 31, 1086-1110
Schweiger, D. M. (1983). Measuring manager’s minds: a critical reply to Robey and Tagart. Academy of Management Review, 8, 143-151
Scott, S. G., & Bruce, R. A. (1995). Decision-making style: The development and assessment of a new measure. Educational and Psychological Measurement, 55, 818-831
Shane, S. (2000). Prior knowledge and the discovery of entrepreneurial opportunities.
Organization Science, 11(4), 448-469
Smith, E. R., & DeCoster, J. (2000). Dual process models in social psychology: Conceptual integration and links to underlying memory systems. Personality and
Social Psychology Review, 4, 108-131
Sutton, R. I., & Hargadon, A. (1996). Brainstorming groups in context: Effectiveness in a product design firm. Administrative Science Quarterly, 41, 685-718
Taylor, A. & Greve, H. R. (2006). Superman or the fantastic four? Knowledge combination and experience in innovation teams. Academy of Management Journal, 49(4), 723-740
Terwiesch, C., & Ulrich, K. T. (2009) Innovation Tournaments: Creating and Selecting Exceptional Opportunities. Harvard Business School Press.
9. APPENDIX
9.1. Appendix 1 – measures used in questionnaire
CoSI questionnaire (Cools, 2007)
Knowing style
K1. I want to have a full understanding of all problems. K2. I like to analyze problems.
K3. I make detailed analyses.
K4. I study each problem until I understand the underlying logic.
Planning style
P1. Developing a clear plan is very important to me. P2. I always want to know what should be done when. P3. I like detailed action plans.
P4. I prefer clear structures to do my job.
P5. I prefer well-prepared meeting with a clear agenda and strict time management. P6. I make definite engagements, and I follow up meticulously.
P7. A good task is a well-prepared task.
Creating style
C1. I like to contribute to innovative solutions. C2. I prefer to look for creative solutions. C3. I am motivated by on-going innovation. C4. I like much variety in my life.
C5. New ideas attract me more than existing solutions. C6. I like to extend my boundaries.
C7. I try to avoid routine.
Creative experience
Motivation
How would you characterize your motivation to join today’s brainstorm session?
Not motivated 1 2 3 4 5 Very motivated
Satisfaction
To what extent do you feel that today’s brainstorm session has been a success / are you happy with the output?
9.2. Appendix 2 – results of analysis of data aggregated by standard deviation Table 9. Correlation table (N=66)
Variable Min Max Mean SD 1 2 3 4 5 6 7 8
1. Age 0.58 19.86 7.33 4.48 2. Gender 0.00 0.58 0.40 0.22 .109 3. Work Experience 0.71 20.52 7.50 4.65 .942** .168 4. Creative Experience 0.00 264.60 27.80 45.36 -.090 .004 -.086 5. Motivation 0.00 1.82 0.66 0.42 -.062 -.195 -.145 .062 6. Knowing Cogn Style 0.14 1.57 0.70 0.28 .057 .071 .086 .119 -.189 7. Planning Cogn Style 0.18 1.41 0.69 0.23 -.253* -.106 -.228† -.026 -.099 .343** 8. Creative Cogn Style 0.14 1.38 0.61 0.27 -.182 .064 -.172 -.036 -.107 .441** .234 9. Satisfaction 0.00 1.41 0.67 0.28 -.051 -.132 -.006 -.095 .156 .128 .101 .263* † p < .10, * p < .05, ** p < .01
Table 10. Regression analysis (N=66)
Satisfaction
Steps and variables entered 1 2 3 4
1. Age -.017 -.022 -.025 -.024
Gender -.191 -.161 -.172 -.086
Work Experience .013 .022 .025 .019
2. Knowing cognitive style .031 -.026 -.102
Planning cognitive style .079 .054 .041
Creating cognitive style .277† .365* .444*
Creative experience -.001 .000 .000 Motivation .165 .181† .135 3. KnowCognXCreaExp -.067 -.095 PlanCognXCreaExp -.014 -.067 CreaCognXCreaExp .110 .156 4. KnowCognXMot -.006 PlanCognXMot .073 CreaCognXMot -.005 R2 change .029 .150 .017 .055 Adjusted R2 -.022 .052 .015 .023 a
N = 66. Unstandardized regression coefficient are reported for the respective regression steps. Data was aggregated using standard deviation.
9.3 Appendix 3 – correlations preparation and cognitive style
Correlations
KnowCogn_mea n
PlanCogn_mean CreaCogn_mean Prep_mean
KnowCogn_mean Pearson Correlation 1 ,612** ,412** ,221 Sig. (2-tailed) ,000 ,001 ,074 N 66 66 66 66 PlanCogn_mean Pearson Correlation ,612** 1 ,028 ,386** Sig. (2-tailed) ,000 ,825 ,001 N 66 66 66 66 CreaCogn_mean Pearson Correlation ,412** ,028 1 ,056 Sig. (2-tailed) ,001 ,825 ,655 N 66 66 66 66 Prep_mean Pearson Correlation ,221 ,386** ,056 1 Sig. (2-tailed) ,074 ,001 ,655 N 66 66 66 66
