Laughing Out Loud: The Effects of a Humorous Stimulus on Creative Performance in the Front-End of the Innovation Process through Humorous Mood Induction

(1)

Laughing Out Loud: The Effects of a Humorous Stimulus on

Creative Performance in the Front-End of the Innovation

Process through Humorous Mood Induction

Rémon Lemmens

S1772678

MSc Strategic Innovation Management

Dr. W.G. Biemans

Dr. K.R.E. Huizingh

Faculty of Economics & Business

University of Groningen

30-03-2015

(2)

2

Laughing Out Loud: The Effects of a Humorous Stimulus on

Creative Performance in the Front-End of the Innovation

Process through Humorous Mood Induction

Rémon Lemmens

Abstract. Many studies have studied the effects of humour in diverse research fields. While some studies

examined the effects of humour on physical capabilities, other studies focused on cognitive capabilities, such as creativity. This study aims to connect the fields of humour, creativity, and innovation and examines the effects of being exposed to a humorous stimulus on creative performance through the induction of a humorous mood. Creative performance is measured in terms of an idea generation task, which takes into account the number and the quality of the ideas generated, since idea generation is an important aspect in the front-end innovation process. In total, 151 subjects participated in this research. Results show that there was no direct effect of the humorous stimulus on creative performance. The results did show that participants who were exposed to a humorous stimulus were in a more humorous mood, regardless of their sense of humour. Additionally, there was a small significant positive relationship between being in a humorous mood and the amount of ideas generated. There were, however, no significant relationships between being in a humorous mood and the quality of the ideas.

Introduction

lmost everyone has been in a situation in which they laughed about something they found amusing. Laughter is a common response to something humorous such as a funny fragment on television, a funny sound, a joke, a funny situation or a funny person (Ziv, 1976; Bennet & Lengacher, 2006) and can positively influence someone’s emotional state (Ruch, 1993; Martin, 2001). In recent literature, humour is widely discussed in different fields. Various studies focused on the effects of humour on health, both physically (Salovey, Rothman, Detweiler & Steward, 2000; Martin, 2001; Bennet & Lengacher, 2006) and mentally (Howrigan & MacDonald, 2008; Tice, Baumeister, Shmueli & Muraven, 2007). In

(5)

5 followed by implementation and adoption (Van de Ven, 1986). According to Mumford, Scott, Gaddis, and Strange (2002), the innovation process comprises of two stages. In the first stage, novel and useful ideas are generated and this process is often referred to as creativity (Amabile, Conti, Coon, Lazenby, & Herron, 1996; Mumford & Gustafson, 1988; Scott & Bruce, 1994). Hence, creativity, and especially idea generation, is regarded as an important stage of the multistage process what is called innovation (Kanter, 1988).

There is, however, little research conducted on the link between being in a humorous mood, triggered by exposure to a humorous stimulus, and creative performance in terms of idea generation. Whereas previous studies mainly focused on the creativity of the ideas this study also addresses two important aspects of the innovation process: implementation and adoption. This study aims to address this gap in the literature by making a link to the innovation research field by also taking into account how much an idea benefits the customer and if it is feasible to implement a certain idea and if being in a humorous mood contributes to generating ideas which score higher on these aspects. This study aims to do so by first analyzing the effect of being exposed to a stimulus on someone’s humorous mood. Hereafter, it is examined whether this humorous mood has an influence on creative performance in terms of idea generation.

This study contributes to the existing literature by connecting the research fields of humour, innovation and creativity in a new way, namely by examining the link between being exposed to a humorous stimulus and creative performance, in terms of idea generation, through the possible mediating effect of humorous mood induction and also taking into account factors that are important for the implementation and adoption phases in the innovation process.

The results showed that exposure to a humorous stimulus has no significant direct effect on creative performance. Exposure to a humorous stimulus did positively influence the participants’ humorous mood, which in turn positively influenced the amount of ideas generated. No significant effects between being in a humorous mood and the quality of the ideas were found.

(6)

6

1. Literature review

This section defines and discusses this study’s key topics of humour and creativity and its underlying concepts

.

1.1. Humour

Although there is an abundance of theories and studies on the topic of humour, it remains a complex human phenomenon which makes it hard to clarify and analyze (Ziv, 1976). Many different studies focus on different aspects of humour and therefore use different definitions. Whereas some studies focus on humour as a defence mechanism (Freud, 1928), other studies focus more on the interaction effect between a sender and receiver of humour (Crawford, 1994; Cooper, 2005). Humour entails behavioural, emotional, psycho-physiological, cognitive, and social aspects (Martin, 2001). Humour can refer to a mental process and how humour is perceived by someone. It can refer to a stimulus and to a response. Additionally, it can refer to someone’s emotional state or a personal trait (Ruch & Hofmann, 2012). This study aims to analyze the effects of a humorous stimulus on a person’s creative performance through the induction of a humorous mood. In this study, the term “humour” is described as “a process initiated by a humour stimulus . . . and terminating with some response indicative of experienced pleasure, such as laughter” (Godkewitsch, 2007).

1.2. Sense of Humour

The term sense of humour does not have one specific meaning. Having a sense of humour may mean three things. Firstly, it can have a conformist meaning which means that someone who has a sense of humour laughs at the same things as we do (Eysenck, 1972). Secondly, it may have a productive meaning which indicates

that someone amuses other people by, for example, telling stories (Eysenck, 1972). Thirdly, it may have a quantitative meaning which means that a person may laugh a lot and has a low threshold for laughter (Eysenck, 1972).

This paper focuses on the quantitative meaning about how people react to humour and how susceptible they are for humorous stimuli. There may be individual differences in sense of humour (Hehl and Ruch, 1985), which relate to the degree to which someone creates humorous comments and other stimuli, how someone comprehends these stimuli, and if someone appreciates these types of humorous stimuli. Sense of humour is a personality trait and describes how someone enjoys and perceives certain forms of humour (Martin, 1998). Someone who has a high sense of humour may focus more on the positive elements and less on the negative elements of a certain stimulus which may result in a humorous response (Cann, Holt & Calhoun, 1999).

(7)

7

1.3. Humorous Mood

Everyone has good days and bad days and fluctuations in their mood states. In this paper, the approach of Ruch et al (1997) is used to describe humorous mood as a state. Humorous mood is described by the same underlying concepts as sense of humour, namely by the cheerfulness state, bad mood state, and seriousness state. As opposed to the underlying concepts of sense of humour, which describe personality traits, these concepts describe one’s current mood state. When being in a cheerful state people are feeling happy and portray themselves as “being in good spirits”. When being in a bad mood state, on the contrary, one’s state is described as being sad or “gloomy”. Moreover, someone can be in a serious state, which may be described as a mental attitude (Ruch et al, 1997). When someone is in a state of high seriousness, this person may feel that he or she is engaged in something important. This attitude inhibits this person from being amused and to reach a cheerful state. Ruch et al (1997) found that state cheerfulness is negatively correlated with both state seriousness and state bad mood, being more negatively correlated with the latter. Thus, being in a humorous mood is being in a combination of a high cheerfulness state and low bad mood and seriousness states.

1.4. Creativity

The definition of creativity depends on how it is studied (Batey, 2012). The majority of the researchers concur that creativity can be defined as something that is new and useful (Mumford, 2003). Creativity is “a process that results in a novel product or idea which is accepted as useful, tenable, or satisfying by a significant group of others at some point in time” (Stein, 1974). Amabile (1982) builds on this definition

of Stein, by adding that something is seen as being creative when independent experts in the field agree that this particular idea is creative. In this paper, the consensual definition of Amabile (1982) combined with Mumford’s (2003) definition is used to describe and measure creativity. Creativity is “the production of outcomes that reliably can be assessed as novel or valuable by expert observers”.

Creativity in the Front-End of the Innovation Process

In many research studies, the terms innovation and creativity are frequently used to describe the same phenomenon. However, there is some conformity about how both terms should be defined. While creativity deals with producing useful and novel ideas (Mumford & Gustafson, 1998), innovation deals with producing novel and useful ideas followed by implementation and adoption (Van de Ven, 1986). Hence, creativity, and especially idea generation, is regarded as an important stage of the innovation process (Kanter, 1988). Ideas are the foundation of innovation and “it is people who develop, carry, react to, and modify ideas” (Van de Ven, 1986).

The generation of ideas is the first stage in the NPD process and therefore vital to an organization’s design, marketing, and launch strategies. It is the main element of the so-called “fuzzy front end” and one of the firm’s utmost important leverage points (Dahan & Hauser, 2001).

Idea Quality versus Quantity

(8)

(9)

9

2. Hypotheses & Conceptual Model

Previous studies have focused on the influence of humour on creative performance and found that humour leads to better creative performance and that a humorous environment may lead to an increase in creativity due to the created contagious “fun mood” (Ziv, 1983). Whereas previous studies focused on the creativity of the ideas, this study makes the link to the innovation process by focusing on the quality of the ideas since it are the best ideas that are most relevant for companies in the idea selection process. Additionally, this study takes into account that the influence of being exposed to a humorous stimulus on creative performance is mediated by the creation of a humorous mood.

The following section discusses all relationships between the various variables. Firstly, the possible direct influence of the humorous stimulus on creative performance is described. Secondly, the relationship between being exposed to a humorous stimulus and its effect on humorous mood is explained. Thirdly, the possible moderating effect of one’s sense of humour on the susceptibility of being induced in a humorous mood by a stimulus is examined. Finally, the influence of being in a humorous mood on creative performance is analyzed. By analyzing each relationship separately, the effect of being exposed to a humorous stimulus on creative performance through the mediating effect of humorous mood induction, can be determined. Finally, a conceptual model is presented which illustrates all relationships.

2.1. Inducing Creative Performance

Previous studies have shown that stimuli may increase creative performance (Ziv, 1976), for example by auditory stimuli (Mehta, Zhu & Cheema, 2012). In this paper, the focus lies on visual humorous stimuli since these produce the desired response, laughter, best to reach a humorous mood (Martin, Sadler, Barret & Beaven, 2008).

Exposure to humorous stimuli increases creative thinking (Ziv, 1976). This does not mean that the stimulus makes people more creative individuals. However, it allows people to think in different ways which are more unconventional and thus may generate unconventional answers (Ziv, 1976). This may increase the likelihood of making more remote and creative associations (Isen, Daubman & Nowicki, 1987; Isen, 2000). Therefore, being exposed to humour might generate different, more creative, ideas. Hence:

H1: Being exposed to a humorous stimulus has a positive effect on one’s creative performance.

2.2. Humorous Mood Induction

(10)

10 indicates that one is high in cheerful state and both low in bad mood and serious states, it is hypothesized that:

H2: Being exposed to a humorous stimulus has a positive effect on one’s humorous mood.

2.3. Susceptibility to Humorous Stimuli

Having a sense of humour influences one’s susceptibility to humorous stimuli and may therefore have an effect on one’s mood (Ruch, Köhler & van Thriel, 1997). Having a high cheerfulness trait has been given a prominent role in determining sense of humour and is an important factor for determining someone’s humorous mood. Besides the fact that people having a high cheerfulness trait are more susceptible to being induced in a positive or humorous mood by a humorous stimulus (Ruch, Köhler & van Thriel, 1997), they tend to stay in this mood longer (Ruch & Hoffman, 2012).

The reverse is true for the bad mood trait. Having a high bad mood trait is one of the signs of having a low sense of humour (McGhee, 1996). A bad mood trait raises the threshold for reaching the humorous mood since the participants are less prone to react positively to the presented stimulus (Ruch & Hoffman, 2012). Another personality trait that affects the effect of a stimulus on one’s humorous mood is one’s “frame of mind”, which is described by the seriousness trait. People with a more serious personality have a higher threshold for being induced into a humorous mood than people with a less serious personality (Ruch & Hoffman, 2012). Since seriousness indicates a certain frame of mind, it is possible that combinations occur. It is possible to be not cheerful and serious simultaneously indicating having a low sense of humour. The opposite is also possible,

which indicates a highly cheerful and non-serious person as having a high sense of humour. Additionally, someone might have a serious personality trait, while also having a cheerful personality. This might be the fact with introverts (Ruch, 1994). Moreover, one might be both low in cheerfulness and seriousness (e.g. nihilists). Therefore, cheerfulness and seriousness are slightly negatively correlated. Additionally, previous studies show that seriousness and bad mood are positively correlated (Ruch, Köhler & van Thriel, 1997). Since having a high or low sense of humour may influence the threshold for one’s susceptibility to a humorous stimulus and thus influence the effect of the stimulus on one’s humorous mood, it is hypothesized that:

H3: The effect of being exposed to a humorous stimulus on one’s humorous mood is moderated by one’s sense of humour.

2.4. Humorous Mood on Creative Performance

(11)

11 Idea generation

Previous studies show that being in a humorous mood may have a positive influence on divergent thinking, because it creates a climate in which people have the liberty to “play” with ideas in original ways (Ziv, 1983; Vosburg, 1998). Being in a bad mood, however, may inhibit this divergent thinking performance (Vosburg, 1998).

Idea Quantity

Both Vosburg (1998) and Ziv (1976) found a relationship between being in a humorous mood and the amount of ideas generated. A humorous mood facilitates the flow of idea production, while a bad mood inhibits this flow. Being in a positive humorous mood state may lead to more creative fluency and may increase the amount of ideas generated than being in a negative bad mood state (De Dreu, Baas, and Nijstad, 2008). Therefore:

H4: Being in a humorous mood has a positive effect on the amount of ideas generated.

Idea Quality

Several studies examined the effect of being in a humorous mood on the quality of ideas and showed contradictory results. The study of Grawitch et al (2003)found that the originality of ideas was increased when participants were in a humorous mood. The studies of Ziv (1983) and De Dreu et al (2008) confirm this finding. Vosburg (1998), on the contrary, did not find any relationship between a positive mood and a quality indicator. Although the study of Vosburg (1998) did not find any relationship between a humorous mood and idea quality, the studies of Grawitch et al (2003) and Ziv (1983) give reason

to believe that there might be a positive relationship. Since being in a humorous mood indicates that one is high in cheerful state and both low in bad mood and serious states, it is hypothesized that:

H5: Being in a humorous mood has a positive effect on average idea quality.

This research makes a distinction between average idea quality and best idea quality with an emphasis on the best idea quality since the best ideas are most relevant for companies (Poetz & Schreier, 2012). Consequently:

H6: Being in a humorous mood has a positive effect on the quality of the best ideas.

2.5. Control Variables

Gender

(12)

12 Creative Personality

Several studies have revealed that certain personality characteristics relate consistently and positively to creative performance in multiple domains (Gough, 1979; Barron & Harrington, 1981; Martindale, 1989). Additionally, Zhou and Oldham (2001) showed in their study that employees who scored high on the Creative Personality Scale (Gough, 1979), produced the most creative suggestions. Someone’s creative personality traits may be the reason for someone’s creative performance instead of being in a humorous mood and therefore we have to control for this possible influence.

Time of Day

Time of day is included in this study, since this may have an influence on someone’s humorous mood. Egloff, Tausch, Kohlmann, & Krohne (1995) found that the maximum activation of a positive mood is reached in the afternoon.

Additionally, tasks involving creativity may be influenced by the time-of-day effect (Wieth & Zacks, 2011). Since three of the sessions were conducted in the morning and three in the afternoon, time of day might be of influence.

Number of Canteen Visits

Finally, number of canteen visits is included as a control variable since someone who visits the canteen facilities more frequently may have better insights in how to improve the current facilities.

Age

Age was also included for obtaining a more complete picture on the possible influences of exposure to a humorous stimulus on humorous mood and creative performance. However, no major effects are expected since all participants were drawn from an international department of a university in which variance in age is low.

(13)

13

3. Research Method

This section describes the research design, all variables and the sample from which data was collected, how this data was collected during experimental sessions, which measurements were used, and how this data was analyzed.

3.1. Appropriateness of Research Design

This research used an experimental design to analyze the influence of exposure to a humorous stimulus on creative performance by inducing a humorous mood. Participants first filled in several general questions, which measured the control variables. After the general questions, the participants filled in a question form which measured their sense of humour. Sense of humour was measured before being exposed to the stimulus to predict the participants’ susceptibility to the stimulus. Hereafter, the participants’ creative personality was measured. The creative personality was also measured before exposure to the stimulus to prevent a possible influence of one’s mood state on the results since this is a self-assessment test. After completing these forms, the participants were exposed to a video stimulus. The participants were exposed simultaneously to the stimulus in a group context since previous studies have shown that people laugh more when they are in groups than when they are exposed individually (Leventhal & Cupchik, 1975; Malpass & Fitzpatrick, 1959; Young & Frye, 1966). Humorous mood was measured after being exposed to the stimulus to capture the participants’ mood state best. Subsequent to measuring the humorous mood, an idea generation exercise had to be fulfilled. This exercise was executed after the exposure to the video stimulus to examine the effects of the participants’ mood on their creative performance. Finally, the participants were given the opportunity to leave a comment about

how they experienced the experimental session and points for improvement.

By conducting the experiment in this order, the influence of exposure to a humorous stimulus on creative performance, by inducing a humorous mood, was measured. In order to induce participants into a humorous mood without being affected by external factors, it was required to expose the participants to exactly the same stimuli under exactly the same experimental conditions. To examine these relationships, the following variables were measured:

Independent Variable

Humorous stimulus. Participants were exposed to a video stimulus. Video stimuli are most effective in inducing moods and therefore most relevant to our research (Westermann, Spies, Stahl & Hesse, 1996). Three of the groups were exposed to a humorous stimulus, while the other three groups were control groups and exposed to a neutral stimulus. Since the aim of this study is to induce a humorous mood and analyze the positive effects of a humorous mood on creative performance, a negative stimulus was not included in this study.

Mediating Variable

(14)

14 Humorous mood is measured using three separate underlying factors instead of combining all factors into one mood factor. Ruch, Köhler and Von Thriel (1997) considered combining the underlying factors into one state variable by reflecting both the bad mood and seriousness subscales and summing up the values of each subscale. However, “this gave impetus to discard state exhilarability as a scale” (Ruch, Köhler & von Thriel, 1997). Humorous mood is measured by the cheerfulness state, bad mood state, and seriousness state. Being in a humorous mood is having a high cheerfulness state and low bad mood and seriousness states.

Moderating Variable

Sense of Humour. The participants’ sense of humour determines how susceptible they are to humorous stimuli. It may lower or raise the threshold for inducing a humorous mood and thus may affect the influence of the humorous stimulus on someone’s humorous mood. Having a sense of humour is measured in this paper by the cheerfulness trait, bad mood trait, and seriousness trait. Having a sense of humour means that one has a high cheerfulness trait and low bad mood and seriousness traits.

Dependent Variables

Creative Performance consists of three underlying dependent variables: Amount of Ideas, Average Idea Quality, and Best Idea Quality. Creativity is frequently seen in the literature as generating novel and useful ideas (Amabile, 1996; Mumford et al, 2002). To measure the participants’ creative performance, the participants had to generate ideas about the canteen facilities on a university campus. All participants were required to have visited the

canteen at least one time to be able to participate in this study. Since the participants were familiar with the current facilities, and thus familiar with potential points for improvement, they were better able to generate novel and useful ideas.

Control Variables

Finally, several control variables were included in this study since these variables might influence certain relationships between the variables. The included control variables are Creative Personality, Gender, Time of Day, Age, and Number of Canteen Visits.

3.2. Pilot Study

(15)

15 were included in the final humorous stimulus. (Appendix A).

The second pilot study identified the neutral stimulus. To identify the final neutral stimulus, 10 participants were exposed to 4 nature documentaries (Vohs & Heatherton, 2000; Volkow et al, 2006; Tice et al, 2007) with a total duration of 25 minutes. Between each fragment, there was a short break of 10 seconds during which the participants had to rate the previous fragment on a 5-point Likert scale based on three criteria: funniness, informative, and interesting. The criterion informative was merely used to mislead the participants. Interesting was used as a criterion, because it is important that all participants focus on the stimulus while being exposed to create the desired mood. Eventually, 3 fragments which scored low on funniness and high on interesting were included in the final neutral stimulus (Appendix A).

In addition, all instructions and scales were tested for any ambiguities and errors. This led to the adjustment of the general instructions and one of the general questions. (Appendix A). The duration of each section was set beforehand. During the pilot study, the researchers observed if each section could be completed within the set time frame. No altercations to the time frame were made.

3.3. Research Participants

In total, 151 participants, who have visited the canteen facilities at least one time, completed the test. Due to a technical malfunction, a surplus of three blank entries was registered. These blank entries were deleted, but did not influence the sample in any way. All participants were Dutch and part of an international department of a Dutch university. This way, language would not become a barrier for filling in the English

question forms which measured one’s sense of humour, creative personality, and humorous mood and for fulfilling the Dutch idea generation task. In total, 93 male (62%) and 58 (38%) female participants, with an average age of 22.7 years, participated in the final experiment (Appendix B). The distribution of men and women is approximately equivalent to the gender distribution of the department (male= 66%, female=34%) (Appendix B).

3.4. Instruments and Measures

Independent Variable

Humorous Stimulus. For the humorous group, a stimulus consisting of 7 funny fragments with a total duration of 10 minutes and 27 seconds was used. The control group was exposed to a neutral stimulus consisting of nature documentaries, which is in line with previous literature (Vohs & Heatherton, 2000; Volkow et al, 2006; Tice et al, 2007). The neutral stimulus consisted of 3 fragments with a total duration of 10 minutes and 3 seconds.

Mediating Variable

(16)

16 words to prevent a possible language barrier (Appendix C).

Moderating Variable

Sense of Humour was measured by the short Trait-form of the State-Trait Cheerfulness Inventory (Ruch, Kohler & van Thriel, 1996) and consisted of 30 statements on one’s mood and mentality in general. Translations in Dutch were added for several words to prevent a possible language barrier. Participants rated each statement based on a 4-point Likert scale (1=strongly disagree and 4= strongly agree). The results were summed up using the Trait-scoring key to form a score on the underlying dimensions (CH, SE, and BM) (Appendix C).

Dependent Variables

Creative Performance was measured by the expert ratings on the ideas, which were generated during the idea generation task. Besides merely determining creative performance by analyzing the amount of ideas, the ideas were rated by two experts on canteen facilities after the experiment. Both experts were team leaders of the canteen facilities at the university campus and work in the canteen on a daily basis (Appendix E). The experts rated the ideas using the Consensual Assessment Technique (Amabile, 1982). This technique assumes that a creative idea about a particular subject is best measured by the combined assessment of experts in that particular field. In this paper, the average quality of the ideas and the quality of the best idea is measured in terms of the evaluation score of the experts based on a three-way interaction term of the following criteria: novelty, customer benefit, and feasibility.

Control Variables

Creative Personality was calculated by the Creative Personality Scale (CPS) (Gough, 1979). The scale consists of 30 adjectives, which describe personality traits. Out of the 30 adjectives, 18 adjectives are associated with a creative personality and 12 with a less creative personality. The 18 adjectives which were associated with a creative personality were scored by a +1, while the other adjectives were scored by a -1. Following Oldham and Cummings (1996), all scores were summed to form a CPS index which could range from -12 to +18. A higher score on this CPS index indicates a more creative personality (Appendix C).

Gender. The participants’ gender was established by one of the general questions in the beginning of the experiment and measured with a binary scale in which female=0 and male=1.

Time of Day. Out of the six experimental sessions, three were conducted in the morning and three were conducted in the afternoon. The sessions that were conducted in the morning at 10 am received a value of 0, while the other three sessions that were conducted at 1pm received a value of 1 to measure the effect of Time of Day in the analysis.

Age. Information about the participants’ age is retrieved from the answers on the general question and is measured in years.

(17)

17 week, 6=2-3 times a week, 7=daily). (Appendix B).

3.5. Procedure

Data was collected during six experimental sessions in a time period of two weeks. During three of the sessions, the participants were exposed to the humorous stimulus and during the other three sessions, the participants were exposed to the neutral stimulus. The sessions in the first week took place at 10 AM and the sessions in the second week at 1PM in a secluded computer lab. All sessions had a duration of approximately 40 minutes. Access to each session closed 5 minutes after the designated time. Participants who wanted to enter the room after this time, were denied access since the instructions were already given and the experiment was conducted within a set, collective timeframe.

The experiment was carried out with a fixed script for every session (Appendix D). When participants arrived, they were registered by one of the researchers, directed to a seat and requested not to use the computer before the researchers indicated to do so. Every session started with the same, pre-written introduction text, read out by the same researcher during all sessions. This way, each participant received exactly the same instructions. Data was entered by making use of computers and recorded with Qualtrics survey software. Each session contained, in chronological order, the following elements: general questions, a humorous trait test, a creative personality test, a video stimulus (neutral or humorous), a humorous state test, an idea generation exercise, and a comments section. The instructions after each section stated if the participants were allowed to proceed to the next section or had to wait for the

other participants to finish that particular section. This way, it was possible to expose all participants simultaneously to the stimulus. The experiment ended with several closing remarks and the handing out of the lunch voucher that was provided to all participants.

3.6. Idea Generation

The canteen facilities of a university campus were used as the subject to generate ideas about. The participants had to generate ideas within a 5 minute time frame, which is based on the results of the pilot test (Appendix A). Each idea had to contain a short description of 10-15 words to allow for serious evaluation by the experts (Poetz & Schreier, 2012). The best idea was rewarded with a prize of €100 to motivate the participants to be as creative as possible. The participants were allowed to execute the idea generation task in both English and Dutch to prevent the possible language skills to be a barrier for the participants’ creativity (Appendix E).

3.7. Idea Evaluation

(18)

18 The experts rated the ideas using the Consensual Assessment Technique (Amabile, 1982). Based on previous literature, the ideas were rated on novelty, customer benefit, and feasibility (Piller & Walcher, 2006; Wittel et al, 2010; Poetz & Schreier, 2012) on a 7-point Likert scale (Appendix E). All ideas were rated separately by the experts on one dimension before continuing to the second and third dimension. Eventually, all ideas received a total score using a three-way interaction term (novelty x customer benefit x feasibility). Ideas were rated based on the interaction term since the ideas that score high on all three dimensions are most relevant for companies. Very novel ideas, which cannot be implemented or do not have any benefit for the customer, are not as valuable as ideas that score high on all three dimensions. The ideas were all presented in Dutch to avoid a possible language barrier for the experts (Appendix E).

3.8. Data Analysis

Factor and Reliability Analysis

The State and Trait scales were purified by performing a factor analysis using principal component analysis with varimax rotation and a cut-off value of .50 in SPSS. Number of factors to be extracted was 3 based on existing literature for both the State and Trait form. The State-form showed minor problematic results. The CH subscale had a Cronbach’s alpha of .85, which is above the threshold of .70 (Cortina, 1993). Although items 3 and 17 on the CH subscale had fairly high cross loadings, this was expected since the states are related (Ruch & Hofmann, 2012). Removing these items would worsen the scale and were therefore retained. The BM and SE subscales had alphas of .89 and .85 respectively (Appendix F).

The Trait-form showed major problems. Many factors loaded weakly or had strong cross loadings. Based on a cut off value of 0.50 (Matsunaga, 2010) items 1, 6, 8, and 29 were deleted from the CH subscale. This led to a Cronbach’s alpha of .78 for this particular subscale. For the BM subscale items 3, 11 and 23 were deleted, which generated a Cronbach’s alpha of .82. The SE subscale was most problematic with a Cronbach’s alpha of .63. Deleting items 2, 17, 20, and 24 generated a Cronbach’s alpha of .69. After deleting these items, most items loaded strongly on the correct factor. Although several items had to be deleted, further analysis using the Trait-scale was conducted since the remaining items captured the essence of each subscale (Appendix G).

(19)

19 creative personality better. The total alpha of the CPS index was .68 (Appendix H).

Inter-rater Reliability

Inter-rater reliability was established by the following steps to ensure that there was consensus among both experts on how to rate the ideas. Firstly, a copy of the general instructions was distributed to both experts. One of the researchers read the instructions out loud. Hereafter, the experts got the possibility to mention any ambiguities or ask questions regarding the general instructions. Secondly, one of the researchers explained what the first dimension, novelty, meant and how it should be rated. The experts got 10 example ideas that were generated during the pilot test and had to rate each example idea separately (Appendix E). The researchers emphasized that each expert should take as much time as needed to rate the ideas. The scores of both experts were compared and both experts got the possibility to discuss each other’s ratings. This way, the rating behaviour will be aligned as much as possible. If the ratings were far apart, the previous steps were repeated until consensus was reached. All ideas were rated on one dimension, before continuing to the next one. Thirdly, the previous steps were applied to the second and third dimension; customer benefit and feasibility respectively.

Inter-rater reliability was calculated using the intraclass correlation method. Since the raters were not drawn randomly from a larger population of raters, but were established beforehand, a two-way mixed measure is used. Consistency had an average measure of .79 with a lower bound of .78 and an upper bound of .81.

The Cronbach’s alpha for the inter-rater reliability is .80 (Appendix I).

Further Analysis

(20)

20

4. Results

This section describes the results by analyzing the data. First, the descriptive statistics and bivariate correlation matrix are presented. Second, the relationships between the variables are analyzed using ANCOVA and regression analysis. Finally, the mediation analysis is described.

4.1. Descriptive Statistics

Descriptive statistics are summarized in table 1. In this research, the standard deviations of the dependent, independent, and mediating variables are rather high. Especially the standard deviations with regard to the quality of the generated ideas are high, which suggests large differences in idea quality. The high fluctuations in humorous state are expected, since the control groups were exposed to a neutral stimulus instead of a humorous stimulus.

The dependent variables show anomaly. Best Idea Quality has Z-skewness and Z-kurtosis values within the range of +1.96 and -1.96.

Average Idea Quality and Amount of Ideas, however, show values that are outside this range. Additionally, the Shapiro-Wilk-test shows that all dependent variables are non-normally distributed (Appendix J). Only the CPS seems to be normally distributed. Transforming Best Idea Quality did not increase normality and is therefore used in the regression without any transformation. Transforming Average Idea Quality and Amount of Ideas using a log10 transformation, however, did reduce the variables’ skewness and increases normality (Appendix K).

Table 1. Descriptive Statistics

N=151 Minimum Maximum Mean Std. Deviation Cronbach’s α ZSkewness ZKurtosis

Best Idea Quality 30.00 295.75 187.42 56.84 .80 -1.47 -.41

Average Idea Quality 24.85 186.38 83.30 28.75 .80 2.77 1.98

Log10 Average Quality 1.40 2.27 1.89 .16 - -2.49 .24

Amount of Ideas 1 36 8.12 4.26 - 14.56 35.40

Log10 Amount of Ideas .00 1.56 .86 .20 - -1.81 7.48

(21)

21

4.2. Bivariate Correlation Analysis

Table 2 shows all bivariate correlations among the variables. The correlations show how the variables are related to one another.

Average Idea Quality and Best Idea Quality are significantly related to each other (.70**), which is not surprising since having a higher score on one’s best idea increases the average idea score. The Amount of Ideas significantly correlates with Best Idea Quality (.23**), which may suggest that generating more ideas increases the chance of scoring high on at least one idea. Moreover, both Trait Bad Mood (.17*) and Trait Seriousness (.17*) weakly correlate with Best Idea Quality and Average Idea Quality respectively. This may indicate that one’s Sense of Humour affects the quality of the ideas generated to some extent. Additionally, Humorous Mood has a weak significant correlation with the Amount of Ideas generated in which State Cheerfulness (.20*) is positively correlated to Amount of Ideas and both State Bad Mood .20*) and State Seriousness (-.19*) are negatively correlated. Furthermore, it is expected that all states significantly correlate with each other, which was also the case in previous studies. Interesting to see is that the

Type of Group significantly influences all states. Although this was expected since one group was exposed to a humorous stimulus and the other to a neutral stimulus, it shows that the humorous stimulus probably positively influences one’s Humorous Mood. Time of Day significantly correlates with Type of Group (-.33**), State Cheerfulness (-.18*) and State Seriousness (.24**). The former was expected since one of the three experimental sessions which were conducted in the afternoon was the humour group, while two of the three were neutral groups. The correlations with State Cheerfulness and Seriousness, however, may suggest that the time at which the experiment was conducted may have an influence on someone’s mood. Gender correlated with Creative Personality and Number of Visits, which indicates that the male participants had a higher creative personality, but do not visit the canteen as often as the female participants. Finally, Age significantly correlates with the Number of Visits. This indicates that older people visit the canteen more frequently.

Table 2. Correlation Matrix

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1. Best Idea Quality 1

2. Average Idea Quality .70** ₁

3. Amount of Ideas .23** _-.08 ₁ 4. State CH .04 -.07 .20* ₁ 5.State SE -.09 .01 -.19* _-.57**₁ 6. State BM -.01 .09 -.20* _-.72**_.45** ₁ 7. Trait CH .01 -.09 .06 .32** _-.05 _-.30** ₁ 8. Trait SE .08 .17* _-.10 _-.31**_.24** _.52** _-.51** ₁ 9. Trait BM .17* _.10 _.12 _-.09 _.13 _.10 _-.14 _.13 ₁ 10. CPS .04 .00 .15 .06 .04 -.09 .11 -.09 .01 1 11. Type Group 1=Humour .05 -.01 .10 .64** -.60** -.44** -.03 -.06 .03 -.00 1 12. Time of Day 1=afternoon .11 .05 .07 -.18* .24** .07 -.00 .01 .06 .00 -.33** 1 13. Nr Visits .08 .05 .06 .02 .03 -.05 .07 -.18* _.18* _-.11 _.09 _-.04 ₁ 14. Gender 1=male -.05 .09 -.06 -.02 .18* _.05 _-.12 _.05 _.03 _.31** _.01 _-.12 _-.17* ₁ 15. Age .01 .08 .03 .00 -.04 -.02 .07 -.07 .13 -.11 .10 .07 .27** _-.02 ₁

(22)

22

4.3. ANCOVA & Regression Analysis Results

The Influence of the Humorous Stimulus on Creative Performance

The effect of the stimulus on the dependent variables Amount of Ideas, Average Idea Quality, and Best Idea Quality was tested using ANCOVA analysis. This way, it was possible to include several control variables in the analysis. To be able to conduct an ANCOVA analysis the dependent variables had to approximate normality and were therefore transformed using a log10 transformation.

First, the influence of the stimulus on the amount of ideas generated was analyzed. Levene’s test showed that the variances between the groups (neutral and humour) are homogeneous (.09). Additionally, the ANCOVA results in table 3 show that there is no significant difference in the amount of ideas generated between the humour and neutral groups after controlling for the other factors using a 95% confidence interval, F(1, 144)=1.53, p=.22).

The same applies to the average quality of the ideas. Levene’s test of equality of error variance is not significant (.32). Additionally, the ANCOVA results in table 4 show that there are no significant differences with regard to the average quality of the ideas between the two groups after controlling for the other factors, F(1, 144)=.06, p=.82).

Finally, the effect of the stimulus on the quality of the best ideas generated was analyzed. Levene’s test of equality of error variance is not significant (.80) indicating that the variances between the groups (neutral and humour) are homogeneous. The ANCOVA analysis in table 5 indicates that there are no significant differences between the humour and neutral group with regard to the quality of the best idea, F(1, 144)=1.06, p=.30.

Since the humorous stimulus did not have a significant direct positive effect on the amount of ideas generated, the average quality of the ideas, and the quality of the best ideas, hypothesis 1 is fully rejected.

Table 3. Tests of Between-Subjects Effects. Dependent Variable: log10Amount of Ideas Source Type III Sum of

(23)

23

Table 4. Tests of Between-Subjects Effects Dependent Variable: Log10Average Idea Quality Source Type III Sum of

Squares

df Mean Square F Sig. Partial Eta Squared Corrected Model ,078a ₆ _,013 _,501 _,807 _,020 Intercept 3,429 1 3,429 132,818 ,000 ,480 Gender ,032 1 ,032 1,227 ,270 ,008 Age ,017 1 ,017 ,640 ,425 ,004 CPS 3,151E-005 1 3,151E-005 ,001 ,972 ,000 Time of Day ,015 1 ,015 ,581 ,447 ,004 Nr Visits ,009 1 ,009 ,362 ,548 ,003 Type Group ,001 1 ,001 ,055 ,815 ,000 Error 3,718 144 ,026 Total 545,111 151 Corrected Total 3,795 150 a. R Squared = ,020 (Adjusted R Squared = -,020)

Humorous Stimulus on Humorous Mood

The influence of the humorous mood on someone’s cheerfulness, seriousness, and bad mood states was analyzed to examine if exposure to a humorous stimulus led to a more humorous mood. As mentioned before, the states are analyzed separately since combining them into one variable is not in line with the theoretical background. Firstly, the influence of the humorous stimulus on someone’s cheerful state was analyzed. Levene’s test showed no significance (.93) indicating that the variances between the groups (neutral and humour) are homogeneous. The results of the ANCOVA analysis in table 6 show that there is a significant difference in the participants’ cheerful state between the humour and neutral group F(1, 147)= 92.46, p= .00). Additionally, the type of group (neutral or humour) explains

approximately 38.6% of the variance. The estimated mean values show that the humorous stimulus leads to a significantly higher cheerful state (humour=19.62, neutral=15.50).

Secondly, the influence of the humorous stimulus on one’s bad mood state was analyzed. Levene’s test was non-significant (.08). The results of the ANCOVA analysis in table 7 show that there is a significant difference in the participants’ bad mood state between the humour and neutral group F(1, 147)=36.48, p=.00), which explains approximately 20% of the variance. The estimated mean values show that the humour group has a significantly lower bad mood state than the neutral group (humour=7.82, neutral=10.53).

Thirdly, the influence of the humorous stimulus on the participants’ serious state was measured. Levene’s test of equality of error

Table 5. Tests of Between-Subjects Effects Dependent Variable: Best Idea Quality Source Type III Sum of

Squares

(24)

24 variance shows significance (.00) indicating that the variances between the groups (neutral and humour) are not homogeneous. Although the assumption of homogeneity of variance is violated, ANCOVA analysis is allowed since the ratio of the largest group variance is not more than 3 times the variance of the smallest group (3.48²/2.13²=2.67).

The results of the ANCOVA analysis in table 8 show that there is a significant difference in the participants’ serious state between the humour and neutral group F(1, 147)=71.77, p=.00), which explains approximately 32.8% of the variance. The estimated mean values show that the humour group has a significantly lower

state seriousness than the neutral group (humour=13.01, neutral=17.08). Since the humorous stimulus had a positive effect on one’s state cheerfulness and a negative effect on both state bad mood and state seriousness, the stimulus had a positive effect on one’s humorous mood. Hence, H2 is fully accepted. Another interesting result is that gender had a significant effect (p=.00) on someone’s serious state and explains approximately 5.7% of the variance in one’s seriousness state. The male participants scored higher on state seriousness both in the humour groups (male=13.74, female=11.83) and neutral groups (male=17.40, female=16.56).

Table 6. Tests of Between-Subjects Effects Dependent Variable: State CH Source Type III Sum of

Squares df Mean Square F Sig. Partial Eta Squared Corrected Model 616,558a ₃ _205,519 _33,472 _,000 _,406 Intercept 10130,434 1 10130,434 1649,882 ,000 ,918 Gender ,383 1 ,383 ,062 ,803 ,000 Time of Day 1,987 1 1,987 ,324 ,570 ,002 Type Group 567,735 1 567,735 92,464 ,000 ,386 Error 902,594 147 6,140 Total 47431,000 151 Corrected Total 1519,152 150 a. R Squared = ,406 (Adjusted R Squared = ,394)

Table 7. Tests of Between-Subjects Effects Dependent Variable: State BM Source Type III Sum of

Squares df Mean Square F Sig. Partial Eta Squared Corrected Model 255,259a ₃ _85,086 _12,670 _,000 _,205 Intercept 2861,034 1 2861,034 426,035 ,000 ,743 Gender 2,397 1 2,397 ,357 ,551 ,002 Time of Day 7,365 1 7,365 1,097 ,297 ,007 Type Group 244,960 1 244,960 36,477 ,000 ,199 Error 987,178 147 6,715 Total 14167,000 151 Corrected Total 1242,437 150 a. R Squared = ,205 (Adjusted R Squared = ,189)

Table 8. Tests of Between-Subjects Effects Dependent Variable: State SE Source Type III Sum of

(25)

25 Susceptibility to Humorous Stimuli

The possible moderating effects of sense of humour were analyzed using the interaction effect in ANCOVA analysis. A significant interaction effect in the F-statistic would suggest a moderating effect of sense of humour on the relationship between exposure to the stimulus and humorous mood. The moderating effect of sense of humour was measured by the underlying cheerfulness, bad mood, and seriousness traits. Firstly, the moderating effect of sense of humour, in terms of the cheerfulness trait, on the relationship between the humorous stimulus and humorous mood was analyzed. The cheerfulness trait had moderating effects for both the cheerfulness and bad mood states, but not for the seriousness state (table 9). Secondly, the moderating effect of the bad mood trait was analyzed. The bad mood trait had significant moderating effects on both the cheerfulness and bad mood states, but did not significantly

moderate the effect of the stimulus on one’s seriousness state (table 9). Finally, the moderating effect of the seriousness trait on one’s humorous mood was examined. The seriousness trait showed significant moderating effects for all three underlying concepts of humorous mood and therefore fully moderates the relationship between the humorous stimulus and humorous mood (table 9).

An interesting result is that gender is significant each time when the relationship between the stimulus and humorous mood is moderated by the seriousness trait. However, this might be the fact that the male participants are more serious which resulted from the results of the previous ANCOVA analysis.

Since not all underlying concepts of sense of humour have a significant moderating effect on the relationship between the humorous stimulus and humorous mood, hypothesis 3 is partially supported.

Table 9. Moderating Effect of Sense of Humour on Humorous Mood

State CH State BM State SE

F Sig. F Sig. F Sig.

Type Group 3.62 .06 1.78 .18 6.51 .01

Time of Day .73 .40 1.65 .20 .62 .43

Gender .19 .66 .01 .93 8.13 .01

Type Group * Trait CH

17.61 .00 10.20 .00 1.46 .24

Type Group 1.00 .32 3.88 .05 2.44 .12

Time of Day .54 .46 1.51 .22 .69 .41

Gender .09 .77 .45 .50 8.51 .00

Type Group * Trait BM

5.48 .01 6.32 .00 2.34 .10

Type Group 3.12 .08 .01 .90 3.67 .06

Time of Day .46 .50 2.05 .16 .99 .32

Gender .00 .97 .09 .76 8.41 .00

Type Group * Trait SE

(26)

26 Humorous Mood on Creative Performance

The influence of being in a humorous mood on creative performance is analyzed using regression analysis.

Amount of Ideas

It was hypothesized that being in a humorous mood would lead to a flow of ideas and therefore increase the amount of ideas generated. The results of the regression analysis in table 12 show that having a high state cheerfulness has a significant positive effect on the amount of ideas generated, B=.01, p=.03. Additionally, it was hypothesized that being in a humorous mood, and thus having a low bad mood state, would increase the amount of ideas generated. Results in table 10 show that there is a significant negative effect between being in a bad mood state and the amount of ideas generated, B=-.01,

p=.03. Hence, having a low bad mood state has a positive effect on the amount of ideas generated. Finally, the influence of state seriousness on the amount of ideas was measured. The results in table 10 show that there is a significant negative effect between being in a serious state and the amount of ideas generated, B=-.01, p=.02. Hence, being in a humorous mood, and thus in a low serious state, has a positive effect on the amount of ideas generated. Since a higher cheerful state and lower bad mood and seriousness states have a positive effect on the amount of ideas generated, hypothesis 4 is fully accepted. Additionally, the moment of the experiment had a significant effect on the amount of ideas generated indicating that people generated more ideas in the afternoon than in the morning sessions.

Table 10: Regression Coefficients of Humorous Mood on the Amount of Ideas Dependent Variable: Log10 Amount of Ideas Independent

Variable:

B Sig. B Sig. B Sig.

State CH .01* .03 State BM -.01* .03 State SE -.01* .02

Control Variables: Control Variables: Control Variables:

CPS .01 .10 CPS .01 .12 CPS .01 .08

Time of Day .07* .05 Time of Day .06 .07 Time of Day .08* .03 Nr. of Visits .01 .41 Nr. of Visits .01 .45 Nr. of Visits .01 .29

Gender -.04 .32 Gender -.03 .35 Gender -.02 .57

Age .01 .90 Age -.00 .89 Age -.00 .79

R² .08 R² .08 R² .08

Adjusted R² .04 Adjusted R² .04 Adjusted R² .05

(27)

27 Average Idea Quality

It is hypothesized that being in a humorous mood would lead to ideas of higher quality. Since almost all participants generated multiple ideas, the scores were averaged. Regression analysis in table 11 shows there is no significant positive effect between having a high cheerfulness state and the average quality of the ideas, transformed using a log10-transformation (B=-.00, p=.63). Additionally, the results of the regression analysis show that being in a lower bad mood state does not have a significant positive effect on the average quality of the generated ideas (B=.00, p=.49). Finally, the influence of being in a less serious state on the average quality was examined. It was hypothesized that being in a less serious state would positively influence the average quality of the ideas generated. The results show no significant effect between state seriousness and average idea quality (B=-.00, p=.44). Since a higher cheerful state and lower bad mood and seriousness states did not have a

significant effect on the average quality of the ideas generated, hypothesis 5 is fully rejected.

Best Idea Quality

Besides measuring the quality of the ideas in general, the quality of the best idea was measured in particular since it is more relevant to companies. State cheerfulness had no significant positive effect on the quality of the best idea (B=.97, p=.52). Secondly, the influence of a bad mood state had no significant negative effect on the quality of the best idea (B=-.12, p=.94). Finally, state seriousness was regressed onto best idea quality. The regression results in table 12 show that there is no significant negative effect of having a high seriousness state on the quality of the best idea (B=-2.11, p=.13). Since being in a higher cheerful and lower bad mood and seriousness states did not have a significant positive effect on the quality of the best ideas, hypothesis 6 is fully rejected.

Table 11: Regression Coefficients of Humorous Mood on the Average Idea Quality Dependent Variable: Log10 Average Idea Quality Independent

Variable:

State CH -.00 .63 State BM .00 .49 State SE -.00 .44

CPS

3.62E-005

.99 CPS .00 .96 CPS .00 .96

Time of Day .02 .53 Time of Day .02 .51 Time of Day .03 .36 Nr. of Visits .01 .54 Nr. of Visits .01 .53 Nr. of Visits .01 .49

Gender .03 .29 Gender .03 .30 Gender .04 .21

Age .01 .40 Age .01 .39 Age .01 .44

R² .02 R² .02 R² .02

Adjusted R² -.02 Adjusted R² -.02 Adjusted R² -.02 **p<.01

(28)

28

Table 12: Regression Coefficients of Humorous Mood on the Best Idea Quality Dependent Variable: Best Idea Quality

Independent Variable:

State CH .97 .52 State BM -.12 .94 State SE -2.11 .13

CPS .97 .56 CPS 1.04 .53 CPS .98 .55

Time of Day 13.93 .16 Time of Day 12.84 .18 Time of Day 16.92 .09 Nr. of Visits 3.06 .35 Nr. of Visits 3.06 .35 Nr. of Visits 3.55 .27

Gender -4.06 .69 Gender -4.38 .67 Gender -.83 .94

Age -.365 .88 Age -.34 .89 Age -.62 .79

R² .03 R² .02 R² .04

Adjusted R² -.02 Adjusted R² -.02 Adjusted R² -.00 **p<.01

*p<.05

Mediating Analysis

Since there is no significant direct effect between the independent variable Humorous Stimulus and the dependent variables Best Idea Quality, Average Idea Quality, and Amount of Ideas mediating analysis cannot be performed. Although mediating analysis cannot be performed, the results do show that the stimulus has an indirect effect on creative performance through inducing a humorous mood.

Additionally, there was no significant moderating effect of Sense of Humour on the relationship between exposure to a humorous stimulus and creative performance except for the moderating effect of the bad mood trait on the amount of ideas (table 13). This is unexpected since we have showed that being in a humorous mood, and thus having a low bad mood state, increases the amount of ideas generated and a bad mood trait raise the threshold for humorous mood induction

Table 13: Moderating Effects Sense of Humour on Creative Performance

Log10Amount Ideas Log10Average Best Idea Quality

Type Group .41 .52 1.62 .21 1.69 .20 Time of Day 4.09 .05 .31 .58 2.01 .16 Gender .85 .36 .88 .35 .18 .68 Age .11 .74 .93 .34 .03 .87 CPS 2.54 .11 .02 .90 .38 .54 Nr of Visits .50 .48 .45 .50 .88 .35

Type Group * Trait CH

.52 .59 1.23 .30 1.02 .36

Type Group 4.11 .04 1.11 .29 .63 .43 Time of Day 3.28 .07 .47 .50 2.17 .14 Gender 1.66 .20 1.12 .29 .26 .61 Age .01 .91 .34 .56 .23 .63 CPS 2.48 .12 .00 .97 .41 .52 Nr of Visits .05 .82 .24 .63 .39 .53

Type Group * Trait BM

5.11 .01 1.16 .32 2.03 .14

Type Group .00 .97 .00 .99 .50 .48 Time of Day 3.87 .05 .60 .44 2.41 .12 Gender 1.09 .30 .99 .32 .23 .64 Age .09 .77 .76 .39 .03 .87 CPS 2.92 .09 .07 .80 .55 .46 Nr of Visits .45 .50 .97 .33 1.14 .29

Type Group * Trait SE

(29)

29

Table 14: Summary Results of Hypotheses

Hypotheses Supported/ Not Supported Clarification

1: Being exposed to a humorous

stimulus has a positive effect on someone’s creative performance.

Not Supported Being exposed to a humorous stimulus had no significant direct effect on the quality of the best ideas, the average quality, and the amount of ideas generated.

2: Being exposed to a humorous

stimulus has a positive effect on one’s humorous mood.

Fully Supported Being exposed to a humorous stimulus increased one’s cheerful state and decreased one’s bad mood and serious states.

3: The effect of being exposed to a

humorous stimulus on one’s humorous mood is moderated by one’s sense of humour.

Partially Supported Not all underlying concepts of sense of humour had significant moderating effects.

4: Being in a humorous mood has

a positive effect on the amount of ideas generated.

Fully Supported A higher cheerful state and lower bad mood and serious states did increase the amount of the ideas generated.

a positive effect on average idea quality.

Not Supported A higher cheerful state and lower bad mood and serious states did not increase the average quality of the ideas generated.

a positive effect on the quality of the best ideas.

Laughing Out Loud: The Effects of a Humorous Stimulus on Creative Performance in the Front-End of the Innovation Process through Humorous Mood Induction

Laughing Out Loud: The Effects of a Humorous Stimulus on

Creative Performance in the Front-End of the Innovation

Process through Humorous Mood Induction

Rémon Lemmens

S1772678

MSc Strategic Innovation Management

Dr. W.G. Biemans

Dr. K.R.E. Huizingh

Faculty of Economics & Business

University of Groningen

30-03-2015

Table of Contents

Laughing Out Loud: The Effects of a Humorous Stimulus on

Creative Performance in the Front-End of the Innovation

Process through Humorous Mood Induction

Rémon Lemmens

Introduction

1.

Literature review

.

2.

Hypotheses & Conceptual Model

3.

Research Method

4.

Results