SPECTACULAR NATURE, AWE, RISK
AND CONSTRUAL LEVEL -
HOW THE FEELING OF AWE, EVOKED BY
SPECTACULAR NATURE, REDUCES RISK TAKING
SPECTACULAR NATURE, AWE, RISK AND CONSTRUAL LEVEL -
HOW THE FEELING OF AWE, EVOKED BY SPECTACULAR NATURE,
REDUCES RISK TAKING
Malve Fiedler University of Groningen Faculty of Economics and Business
MSc Marketing Management Master Thesis June 22nd, 2015 Sollbrueggenstrasse 67 47800 Krefeld Germany +491773435165 m.fiedler@student.rug.nl S2550105 Supervisors University of Groningen
Management summary
Although nature and nature imagery and its influence on the human mind has been researched extensively, literature on the use of imagery of spectacular nature in advertising on consumer behaviour is scarce. This paper aims at studying the effects of awe as an affective reaction to spectacular nature on financial risk taking as consumer behaviour.
Nature and nature imagery, specifically spectacular nature, is assumed to evoke the feeling of awe, which is defined by two central features, a perceived vastness, exceeding current mental structures, and in turn a need for accommodation of the experience within the mind. Then, awe is expected to negatively influence risk taking behaviour. This relation is assumed to be moderated by one’s construal level, the degree of abstractness with which one creates mental representations of information in mind. The higher the level of construal the more abstract and general are the representations. An experience with awe is expected to increase the construal level which then in turn increases risk taking behaviour, moderating the negative influence of awe on risk taking behaviour. Furthermore, one’s level of consumer self-confidence is assumed to moderate the relation between awe and risk taking as well as between construal level and risk taking. Additionally, differences due to gender are assumed to moderate the influence of awe on risk taking.
In order to analyse the proposed individual relations between the concepts,
quantitative research is conducted, making use of an online questionnaire. The gathered data is then analysed with a variety of statistical tests in SPSS.
Results provided evidence for spectacular nature as an elicitor of awe. However, neither a statistical significant influence of awe on risk taking behaviour, nor a moderating influence of construal level is found. Furthermore, no support is given for the assumptions made concerning a moderating influence of consumer self-confidence and gender differences on the proposed relations.
Despite the lack of significant results, recommendations are given. Since prior research identified a variety of beneficial effects of nature, including stress reduction, attention restoration, an effect on overall well-being, as well as a positive effect on brand recall and brand recognition when nature imagery is used in advertising, marketers are
advised to use nature imagery in advertising to reap these advantages. Furthermore, for future research it is advised to investigate the feeling of rapture in connection with nature.
Preface
TABLE OF CONTENTS 1. INTRODUCTION ... 1 2. THEORETICAL FRAMEWORK ... 5 2.1 Awe ... 5 2.2 Risk Taking ... 7 2.3 Construal Level ... 8 2.4 Gender Differences ... 12 2. 5 Consumer Self-Confidence ... 13 3. CONCEPTUAL FRAMEWORK ... 15 4. METHODOLOGY ... 15
4.1 Design and Participants ... 15
4.2 Materials ... 16
4.3 Procedure ... 19
4.4 Plan of Analysis ... 19
5. RESULTS ... 19
5.1 Sample Characteristics and Data Description ... 20
5.2 Hypothesis Testing ... 20
6. DISCUSSION ... 28
7. CONCLUSION ... 31
7.1 Managerial Implications ... 32
7.2 Limitations and Recommendations for Future Research ... 32
REFERENCES ... 35
APPENDICES ... 41
Appendix A: Construal Level Measurement ... 41
Appendix C: Consumer Self-Confidence Measurement ... 43
Appendix D: Boxplot to Detect Outliers ... 44
Appendix E: Excerpt of Answer’s to the Writing Task as Condition Stimulus ... 45
Appendix F: SPPS Output ANOVA H1 ... 47
Appendix G: SPSS Output ANOVA H1 (2) ... 49
Appendix H: SPSS Output ANOVA H1 (3) ... 50
Appendix I: SPSS Output ANOVA H2 ... 52
Appendix J: SPSS Output Correlations H2 ... 53
Appendix K: SPSS Output ANOVA H3a ... 55
Appendix L: SPSS Output Correlations H3a ... 56
Appendix M: SPSS Output ANOVA H3b ... 58
Appendix N: SPSS Output t-test ... 59
Appendix O: SPSS Output ANOVA H4 ... 60
Appendix P: SPSS Output Moderation H5a ... 61
1
1. INTRODUCTION
The use of nature imagery to advertise nature-unrelated products is not an uncommon phenomenon in advertising. Just think of campaigns by well-known beer brands or car manufacturers, in which mundane nature, and often impressive and spectacular nature, is combined with rather unspectacular, nature unrelated products. This leads to the question, what is the marketers reasoning behind using images of spectacular nature to promote nature unrelated products? And what are the effects of nature imagery on the consumer and his or her behaviour?
The beneficial effects of nature and representations of nature in form of images or videos on the human mind and overall well-being has been extensively documented in the environmental psychology literature (e.g. Howell et al. 2011; Kaplan 1995; Ulrich 1979). Ulrich (1979) developed the stress recovery theory, which holds that nature experiences reduce stress, fear and anger and increase positive emotions, feelings of affection friendliness, playfulness, and elation. Another major finding in research of nature experience is
summarized by the attention restoration theory. This theory emphasizes the positive effect of nature scenery on mental and cognitive capacities of attention (Berman, Jonides and Kaplan 2008; Kaplan 1995; Kaplan and Kaplan 1989; Tennessen and Cimprich 1995). According to the attention restoration theory, brief interactions with nature, such as spending a few minutes in nature or just looking at nature images can restore individuals’ fatigued attention
(Tennessen and Cimprich 1995) and increase vitality (Ryan et al. 2010). Overall, interactions with nature have a positive effect on well-being (Howell et al. 2011), including the restored attentional capacity, positive emotions, and an ability to reflect on a life problem (Mayer et al. 2009). Additionally, experiences with spectacular nature lead to people feeling closer to others (Weinstein, Przybylski and Ryan 2009) and increase prosociality (Joye and Bolderdijk 2015; Zhang et al. 2014).
2 environments. Overall, nature’s power to evoke positive emotions leads to a favourable ad and brand attitude (Hartmann and Apaolaza-Ibáñez 2010). Furthermore, the use of pleasant nature imagery was demonstrated to enhance brand recall and recognition (Hartmann, Apaolaza and Alija 2013).
One affective reaction to nature, specifically spectacular nature, such as panoramic views of mountains, the vast ocean or thunderstorms (Keltner and Haidt 2003), has been mostly ignored in the context of research into the use of nature in advertising and nature influence on consumer behaviour, namely the feeling of awe. Awe, defined as “a feeling of reverential respect, mixed with fear or wonder” (Oxford English Dictionary 2010, p.112) finds its antecedent in an emotional response to a powerful leader (Keltner and Haidt 2003). As one might already be able to point from the definition, awe is an ambivalent emotion that might be experienced as negative or positive (Keltner and Haidt 2003; Lazarus 1991). This ambivalence of awe, with its positive and negative facets, makes it a very complex emotion. As there is yet limited research on the effects of awe, additional questions arise on how and to what extent awe influences consumer behaviour, due to the fact that the valence of emotions influences consumer behaviour (Hoyer, MacInnis and Pieters 2006). Therefore, among the goals of this work is to answer the question, if experiences and viewings of spectacular nature evoke the feeling of awe and if, how and to what extend this feeling influences consumer behaviour?
In this study, the influence of awe-evoking nature on consumer behaviour, specifically risk taking, will be investigated. Perceived risk and risk taking has been established as a major part of consumer behaviour (Solomon, Russell-Bennett and Previte 2002; Taylor 1974). Every decision a consumer has to take is connected to some kind of uncertainty that could lead to losses, and might thus be risky. With regard to consumer decisions, there are several
dimensions of risk, the major ones being psychological, physical or health-related, social and financial or economic risk (Hoyer, MacInnis and Pieters 2006; Taylor 1974; Weber, Blais and Betz 2002). The current research focusses on the financial dimension of risk taking which is usually represented through investment or gambling decisions (Weber, Blais and Betz 2002). Financial risk is connected to uncertainty about the financial outcome of an event or decision, and with that the potential of financial harm or loss (Hoyer, MacInnis and Pieters 2006; Taylor 1974). An individual’s tendency towards taking or avoiding risks, one’s risk
3 al.’s (2001) risk-as-feelings hypothesis emphasizes the influence of emotions felt right before and during decision making. Emotional reactions to risky decisions often deviate, sometimes strikingly, from cognitive evaluations of the same situation. Fear and anxiety have been emphasized in their effects on consumer behaviour in risky decision situations (Loewenstein et al. 2001). The feeling of awe, as mentioned in the previous, might be flavoured by positive or negative emotions, such as enlightenment or fear (Keltner and Haidt 2003; Lazarus 1991). Therefore, this study also aims at answering if, how and to what extent awe influences
consumer behaviour in risky decisions?
In this research the influence of awe-evoking nature on risk taking is assumed to be driven by construal level. One’s construal level refers to an individual’s categorization of information in mind and the degree of abstractness of one’s thinking, reflected in
psychological distance (Trope and Liberman 2003). On the one hand, awe seems to lead to higher construal. Reasons for that could be the vastness, evoking a needed change in mental structures to accommodate this new experience, being the central features of awe (Keltner and Haidt 2003), which could be translated into psychological distance between the experience and the confines of one’s mental structures. This distance is related to the concept of construal level as psychological distance (Trope and Liberman 2003). On the other hand, construal level relates to risk taking. As mentioned, awe affects mental structures (Keltner and Haidt 2003), which play a major role in perceiving risk and therefore risk taking. For example, according to Chandran and Menon (2004), framing risks in different timeframes, according to bigger versus smaller psychological distance, and accordingly higher versus lower construal level, has an impact on perceived degree of risk. Therefore, this study also aims at answering, if awe influences one’s construal level? Is there a change in risk propensity and risk taking with increasing construal level?
4 overwhelming feeling of awe, possibly leading to a decrease in consumer self-confidence. Therefore, among the questions to answer in this study, is if, how and to what extent consumer self-confidence plays a role in the relation between awe and risk taking?
However, self-confidence in one’s abilities and decisions is known to be sensitive to time. The further into the future the imminent task or decision, the less confident one seems to be in a successful handling of the situation (Gilovich, Kerr and Husted Medvec 1993). This temporal distance relates to the concept of construal level. Events in the distant future are construed in a far more abstract way than events set in the near future (Liberman, Sagristano and Trope 2002; Liberman and Trope 1998; Trope & Liberman 2003). Furthermore,
hypothetical distance might play a paramount role in the relation between construal level and risk taking (Trope, Liberman and Wakslak 2007; Wakslak et al. 2006). Hypothetical distance refers to the probability of an event as psychological distance in the frame of construal level theory. The less probable an event is to happen, the bigger is the psychological distance and in turn the level of construal (Trope, Liberman and Wakslak 2007; Wakslak et al. 2006). Then, if an event shows low probability, construal level should be high, leading to abstract
representations of the event and its possible outcomes, just as it happens with temporal distance, which leads to higher confidence in one’s decisions. Therefore, in sum, the
relationship between the level of consumer self-confidence in abilities and decision making and psychological distance according to the construal level theory, leads to the question, if, how and to what extent one’s level of consumer self-confidence plays a role in the relation between construal level and risk taking?
Prior research of both risk propensity and emotions placed an emphasis on differences between genders. The stereotypes say, in risk propensity men seem to show stronger risk proneness than women (Powell and Ansic 1997; Grable 2000), whereas in emotions, women seem to be more affected than men (Plant et al. 2000). Therefore, this research will include gender as a control variable, testing for differences in the effect of awe on risk taking
according to gender. The question to answer here is, if gender influences the effect of awe on risk-aversion?
5 “To what extent does the feeling of awe as an (emotional) response to spectacular nature influence risk-taking, depending on one’s construal level, influenced by gender and level of consumer self-confidence?”
The paper is structured as follows: First, existing literature on the topic at hand will be reviewed and elaborated on, to ultimately derive hypotheses for subsequent analysis. All relations between the variables will be summarized and visualized in a conceptual framework. Next, the methodology, including data collection procedure and analysis will be explained, followed by a results section. Here, the results will be analysed, to be discussed in the last chapter. This paper will finish with practical implications for marketers, as well as limitations for the study and directions for future research.
2. THEORETICAL FRAMEWORK
The goal of this paper is to show the possible impact of the emotion of awe on risk taking as a consumer behaviour, and to what extent this effect is driven by one’s construal level.
Additionally, gender differences and the influence of consumer self-confidence will be
observed. Therefore, in the following the concept of awe, literature on risk and risk taking and the construal level theory will be elaborated on. Next, an introduction to gender differences and research concerning consumer self-confidence will be given.
2.1 Awe
6 accommodate the new experience. The vastness of an awe evoking stimulus can trigger a variety of feelings, including negative experiences of fear and disorientation, or positive feelings of enlightenment and rebirth, once the experience has been placed in one’s mental structures, fulfilling the need for accommodation. Additionally, Keltner and Haidt (2003) identified several themes that influence the valence of the awe experience. These themes, namely threat, beauty, ability, virtue and the supernatural, leave room for a negative interpretation of awe.
Subsequent literature and research supported the given definition of awe as being related to vastness and something bigger than the self, overwhelming and extending the human mind in a need for accommodation (Griskevicius, Shiota and Neufeld 2010; Shiota, Keltner and Mossmann 2007). However, unlike in Keltner and Haidt’s (2003) definition, in subsequent literature, awe is clearly characterized as a positive emotion, such as by
distinguishing it from horror when for example exposed to a threatening storm (Shiota, Keltner and Mossmann 2007). Additionally, awe was identified to de-emphasize the individual self, placing importance on the environment and “the self as a part of a greater whole” (Shiota, Keltner and Mossmann 2007, p.955). Since awe-evoking stimuli can exceed the structures of the human mind and therefore evoke the need for accommodation, they increase one’s fundamental uncertainty or rather decrease one’s tolerance for uncertainty (Valdesolo and Graham 2014). This uncertainty, elicited by awe, can be compensated by an increased belief in the supernatural. Awe can also increase religiosity and spirituality (Van Cappellen and Saroglou 2012). Awe has an impact on perceived time, by shifting the focus to the present and seemingly expanding it, leading to changes in one’s decisions and life
satisfaction (Rudd, Vohs and Aaker 2012). Effects of awe include a positive impact on patience and willingness to volunteer time to help other people, and a negative impact on materialism (Rudd, Vohs and Aaker 2012).
Placing awe in the context of advertising, awe, as part of a set of positive emotions, has negative effects on message persuasiveness, due to increased systematic processing, which leads to an increase in the amount of scrutiny with which an advertising message is evaluated (Griskevicius, Shiota and Neufeld 2010).
7 such as waves, usually provoke the awe feeling, with the need for accommodation. Therefore, the following hypothesis is derived:
H1: Experiences with spectacular nature elicit the feeling of awe.
2.2 Risk Taking
To define the term risk poses a rather difficult task, since it is principally subjective to each individual (Brachinger and Weber 1997). Yates and Stone (1992) reviewed the different definitions of risk in literature to derive a core concept, consisting of three core elements that influence the subjective value and desirability of decision alternatives – losses, their
significance and uncertainty. For a long time, the construct of risk represented a utility function, with its shape representing one’s risk attitude (Brachinger and Weber 1997; Weber and Milliman 1997). However Kahneman and Tversky (1979) extended this expected utility theory to the prospect theory, in which importance is placed on the value of gains and losses instead of the final outcome. Weber and Milliman (1997) concluded risk preference as a stable personality trait, influenced by situationally varying risk perceptions. Similarly, Hsee and Weber (1999) see risk preference as a theoretically neutral construct describing choice behaviour in decision situations, when influenced by different circumstances. Weber, Blais and Betz (2002) concluded a domain-specific form or risk attitude that is not connected to personality traits but rather to perceptions of risks and benefits of the decision at hand. Risk tendencies seem to be different, according to the domain of the decision and situational influences. Loewenstein et al. (2001) chose a very different approach to risk, attributing feelings to an individual’s risk tendencies. The risk-as-feeling hypothesis stresses the
importance of affect and emotions felt right before or during the decision process. Cognitive evaluations depend on probabilities of outcomes and their intensity, whereas emotional reactions are influenced by a variety of factors: the vividness in imagining consequences of the decision and past personal experiences and exposure to outcomes and conditioning. The risk-as-feeling hypothesis is supported by evidence that emotions might arise independently from cognitive evaluations, that cognitive evaluations elicit emotions and in turn emotions induce cognitive evaluations and that emotions significantly shape behaviour (Loewenstein et al. 2001). Therefore, this risk-as-feeling hypothesis might be of importance in explaining the influence of awe on risk proneness.
8 focus from one’s self to the greater good and to the present environment through experiences of vastness that lead to feelings of insignificance, smallness and a connection with the world (Shiota, Keltner and Mossmann 2007). Experiences with spectacular nature, identified as one major driver of awe, were shown to increase prosociality and led to a closer connection and feeling of oneness with others (Joye and Bolderdijk 2015; Shiota, Keltner and Mossmann 2007; Van Cappellen and Saroglou 2012). This new focus might lead to more social awareness and therefore to decreased willingness to take risks. Additionally, awe was attributed to a decrease in materialistic tendencies (Rudd, Vohs and Aaker 2012; Shiota, Keltner and Mossmann 2007). Rudd, Vohs and Aaker (2012) provided evidence that these changes in materialistic tendencies are due to a heightened momentary life satisfaction triggered through the increased perceived time availability after an experience of awe. A higher life satisfaction and decreased materialistic tendencies could be translated into a weakened desire for financial wealth, leading to less risk proneness in a lottery decision situation. Furthermore, evidentially, awe leads to more systematic processing (Griskevicius, Shiota and Neufeld 2010). Systematic processing is associated with more thorough and detailed evaluation of arguments, generating the need for stronger arguments and leading to less impulsive decisions (Fennis and Stroebe 2010). Consequently, one could assume that with the increase of systematic processing, in this case triggered by awe, the tendency towards risk aversion increases. According to the previous elaborations, the following assumption is made:
H2: The feeling of awe has a significant negative effect on risk taking.
2.3 Construal Level
9 the supermarket. In a high construal, you think of it as “going to the supermarket”, whereas in a low construal, you might think of it as “buying milk”. Construal level theory therefore explains the increasing abstractness due to increased psychological distance and its effects on consumer behaviour (Trope, Liberman and Wakslak 2007).
Several subordinate dimensions of construal level as psychological distance have been identified. The paramount distinction depicts temporal distance (Liberman, Sagristano and Trope 2002; Trope and Liberman 1998; Trope and Liberman 2003). According to the
temporal construal, events in the distant future are more abstract, less detailed representations in broader categories, corresponding to the high construal level. Correspondingly, timely close events, happening in the near future, are construed in more detail, as in the low construal level. These differences in the mental construal of events changes decisions and judgements of individuals (Trope and Liberman 2003). Similar results are found for spatial or physical distance, where the abstractness increases with spatial distance (Fujita et al. 2006a).
Another important form of psychological distance is documented as social distance, which includes self-construal. Self-construal refers to two inner selves, the independent versus the interdependent self (Mandel 2003). Depending on which self is more salient, one acts more (or less) social. The final dimension is the hypothetical distance, the probability or likelihood of occurrence. The less likely the occurrence, the more abstract is the
representation in mind and vice versa (Wakslak et al. 2006).
Placing construal level theory in the frame of consumer behaviour, a low construal level has been affiliated with a stronger preference stability, preference-behaviour
10 reverse the depletion effect (Agrawal and Wan 2009). High-level construal leads the attention away from the depletion towards the overall goal (Agrawal and Wan 2009), increases control and increases the preference for delayed payoffs (Fujita et al. 2006b). Exerted self-control in turn leads to a low construal level (Bruyneel and Dewitte 2012). A low construal level leads to less procrastination (McCrea et al. 2008). Construal level in connection with politeness demonstrated an increase in politeness according to the increase in the degree of abstractness, temporal and spatial distance and vice versa (Stephan, Liberman and Trope 2010). The level of confidence was shown to significantly influence the level of abstractness, namely high confidence leads to a high construal level (Wan and Rucker 2012). A high-level construal, in turn, leads to a more abstract representation of a task, which increases one’s confidence about successfully performing the task (Gilovich, Kerr and Husted Medvec1993; Wan and Rucker 2012)
Several factors within the concept of construal level give reason to believe that an experience of awe and one’s construal level are in some way related. It is assumed that an experience of awe expands mental structures due to an overwhelming vastness (Keltner and Haidt 2003; Shiota, Keltner and Mossmann 2007). This vastness, evoked for example by a panoramic view of spectacular nature, or experiences with power (Keltner and Haidt 2003), exceeds mental structures, possibly leading to a feeling of psychological distance. The core of the construal level concept depicts increasing abstractness of mental representations due to psychological distance (Trope, Liberman and Wakslak 2007). Thus, one could assume, that the stronger that feeling of vastness, the more an experience exceeds mental structures, the bigger is the connected feeling of psychological distance and in turn the higher the level of construal. Another indication for a relation between awe and one’s construal level is the salience of physical height in both concepts (Aggarwal and Zhao 2015; Keltner and Haidt 2003). Anything bigger than the self was identified as an elicitor of awe (Keltner and Haidt 2003). This includes physical height, for example the height and vastness of a mountain chain. In regard to construal level, perceived physical height is set to evoke the corresponding level of construal (Aggarwal and Zhao 2015). Considering the information derived from literature review given above, the following relation is proposed:
H3a: An experience of awe leads to higher level construals.
However, besides the elaborated influence of awe on risk taking, the reviewed
11 according to the temporal dimension of psychological distance (Trope and Liberman 2003). With bigger distance, the core features and superordinate goals of an event and decision become more salient in mental representations (Liberman and Trope 1998; Trope and Liberman 2003). Accordingly, the construal level affects the desirability of outcomes and feasibility of attaining these outcomes (Liberman and Trope 1998; Sagristano, Trope and Liberman 2002; Trope and Liberman 2010). A high construal level in terms of temporal distance showed an increase in the importance of the desirability of outcomes, whereas in near future decisions, framed in a low-level construal, a higher feasibility of attaining these
outcomes gains importance. So, when a risky situation is represented in a gambling situation, asking the consumer to decide between two or more options with different payoffs and
varying probabilities, the payoffs, gains versus losses, serve as outcomes and the probabilities equal the feasibility of attaining the outcome. An option with a high payoff at a low
probability is seen as more risky than an option with reduced payoff at a higher probability. As explained before, according to construal level theory, with increasing level of construal, the amount of payoff gains importance over the probability of obtaining these payoffs, leading to the assumption that risk taking tendencies increase with construal level. This notion is supported by the hypothetical dimension of psychological distance in construal level theory. Hypothetical distance refers to the probability and uncertainty about an event (Trope, Liberman and Wakslak 2007; Wakslak et al. 2006). With increasing uncertainty and
decreasing probability, hypothetical distance increases (Trope, Liberman and Wakslak 2007; Wakslak et al. 2006). In other words, high payoffs with low probability are linked to a high construal. Therefore, one could assume that the higher a consumer’s level of construal, the more willing he or she is to take a risk in a gamble. Besides the connections explained so far, Chandran and Menon (2004) provided additional evidence for the relation between
psychological distance and risk perceptions. They demonstrated that seemingly closer risks, as in the nearer future with low temporal distance, pose a bigger threat as the same risk framed in a high construal of distant future.
In short, a high construal level is linked to more abstract representations and goal-related thinking (Trope and Liberman 2003), greater physical, temporal and hypothetical distance (Trope, Liberman and Wakslak 2007) and a preference for higher, but less probable payoffs (Sagristano, Trope and Liberman 2002), which is identified as more risky.
12 influence of awe on risk taking, leading to the assumption that risk aversion, supported
through a feeling of awe, decreases with an increase in construal level, or in other words:
H3b: A high construal level has a significant positive effect on risk taking.
2.4 Gender Differences
An ongoing discussion in literature concerns gender differences, both in connection to risk taking and in connection with emotions. According to stereotypes, women are more risk-averse than men, and women are more receptive to emotions (Grable 2000; Plant et al. 2000). In regard to risk taking behaviour differences due to gender the majority of findings support the stereotype that men exhibit not only a stronger tendency towards risk than women, but also show less discernment for obvious uncertain situations (Byrnes, Miller and Schafer 1999; Faff, Mulino and Chai 2008; Jianakoplos and Bernasek 1998; Powell and Ansic 1997; Watson and McNaughton 2007). Besides, several studies support the notion that women tend to be more conservative in financially risky decisions such as investment decisions or gambles (Bernasek and Shwiff 2001; Charness and Gneezy 2012; Dwyer, Gilkeson and List 2002; Faff, Mulino and Chai 2008; Grable 2000; Powell and Ansic 1997).
The evidence regarding gender differences in emotions is rather inconsistent. This is mainly due to the complexity of this research subject, in turn due to the variety of diverse variables – biological, social, cultural, cognitive and personality, influencing gender
13 Thus, the reviewed literature gives reason to believe that gender influences the impact of awe on risk behaviour. Therefore, the following assumption is made:
H4: The negative effect of awe on risk taking is significantly stronger for women than for men.
2. 5 Consumer Self-Confidence
At the basis of financial decision making is the evaluation of uncertainty and outcomes (Sitkin and Pablo 1992; Taylor 1974). The potential of financial harm or loss creates a risky decision situation, asking individuals to evaluate benefits and risks (Hoyer, MacInnis and Pieters 2006; Taylor 1974). Besides the previously hypothesized influence of gender differences on risk taking in decision situations, reviewed literature indicates a relation between consumer self-confidence and risk taking. “Consumer self-confidence is defined as the extent to which an individual feels capable and assured with respect to his or her
marketplace decisions and behaviours” (Bearden, Hardesty and Rose 2001, p.122). Confidence affects consumer behaviour through its role in attitude formation. Generally, attitudes are seen as reliable predictors of consumer behaviour. The more confident one is in his attitudes, the better do these attitudes predict behaviour and the more likely is one to act according to these attitudes (Ajzen 1991; Ajzen and Fishbein 1977; Hoyer, MacInnis and Pieters 2006). Because, with increasing confidence one has in one’s attitudes, one’s belief in the accuracy of these attitudes derived from judgements increases, leading to more certainty in right and successful decisions (See et al. 2011). The concept of self-efficacy is closely related to self-confidence. Self-efficacy is “the perceived ability to make good decision choices” (Krueger and Dickson 1994, p.386). Levels of self-efficacy lead to changes in perception of opportunities and threats: The higher the perceived self-efficacy, the stronger is the perception of opportunities and the weaker the perception of threats (Krueger and Dickson 1994). Thus, a high level of confidence and self-efficacy leads to increased salience of an opportunity, leading to higher risk-taking. Therefore, one could assume that the higher one’s consumer self-confidence, the less strong is the negative effect of awe on risk taking.
Hypothesis 5a: The negative effect of awe on risk taking is moderated by one’s level of consumer self-confidence.
14 self-confidence. The further the timely distance from an event or task, the bigger is one’s confidence in successfully managing that task or situation. This timely distance depicts the temporal dimension of construal level theory. High level temporal construal refers to events in the distant future, with general, abstract, goal-related mental representations (Liberman, Sagristano and Trope 2002; Liberman and Trope 1998; Trope and Liberman 2003). Thus, one could assume that a high construal leads to higher confidence in success. In case of a lottery this would lead to choosing the more risky option, offering a higher payoff at a lower
probability. Here, a connection to hypothetical distance can be drawn. Hypothetical distance represents an essential dimension of psychological distance in the frame of construal level theory (Trope, Liberman and Wakslak 2007). A high level construal in terms of hypothetical distance is elicited by a low probability of an event to happen, connected with a high
uncertainty concerning the outcomes of a decision (Wakslak et al. 2006; Trope, Liberman and Wakslak 2007). The higher the uncertainty surrounding an event, the greater is the connected psychological distance (Wakslak et al. 2006). This distance in turn influences consumers’ judgements and decisions. With increasing construal level, superordinate concerns, payoffs and outcomes become more important than subordinate concerns, feasibility and probability (Liberman and Trope 1998; Sagristano, Trope and Liberman 2002; Wan and Rucker 2013). Furthermore, Wan and Rucker (2013) believe confidence influences construal level as a psychological state through connecting confidence with more abstract, goal related thinking. The higher the level of confidence, the more abstract is the framing of information in mind and the stronger is one’s tendency towards processing of those abstractly framed, high level construed messages (Wan and Rucker 2013). Accordingly, with high confidence and in turn high construal of information, the attractiveness of high outcomes over probability, which constitutes great hypothetical distance and higher risk, increases (Wan and Rucker 2013). These findings are closely related to Nussbaum, Liberman and Trope’s (2006) findings which include evidence for a positive influence of abstractly framed outcomes on confidence.
In sum, abstractly framed messages in terms of psychological distance are believed to be those which offer high payoffs at a low probability. Due to the felt psychological distance with these decisions, confidence in success increases and in turn choosing the option seen as more risky becomes more attractive. Therefore, the following relation is proposed: The higher the construal level in terms of temporal or hypothetical distance, the higher is one’s
confidence in success. Ergo:
15
3. CONCEPTUAL FRAMEWORK
Based on the previous elaborations, the following figure gives an overview of the different factors (hypothetically) influencing risk taking.
Figure 1: Conceptual framework
4. METHODOLOGY
After the previous gave an overview of relevant literature, from which hypotheses were derived, the following will give an overview of the methodology used to test these. The empirical research will be explained, including research design, data collection and survey design. Next, measurements and the plan of analysis will be discussed.
4.1 Design and Participants
16 The chosen between participants design made sure that each participant serves in one
condition, randomly assigned (Aronson, Wilson and Brewer 1998). Additionally, two moderators were included, gender and ‘level of confidence’. The survey was published via Facebook to recruit participants. Using an online questionnaire offers several advantages, including flexibility and digital access (Malhotra 2006). Additionally, publishing the survey via social media has the advantage of facilitated spreading of the survey to reach the number of participants necessary for significant results in a short period of time. Furthermore, participation is not limited geographically. Participation was voluntary and anonymous. Average execution time for the survey for each participant was estimated at about 10 minutes.
4.2 Materials
Awe inducement
To induce awe (Keltner and Haidt 2003), participants were asked to retrieve their last memory of awe evoked by nature, adapted from Griskevicius, Shiota and Neufeld (2010, p.194). To replicate an experience with spectacular nature, participants read the following: “I would like you to recall an event in your life when you saw a particular panoramic view for the first time. Some examples might be seeing the Grand Canyon, seeing the view from high up on a
mountain, or seeing the vast ocean. Please recall a specific event when you saw this view for the first time, rather than a general period of time. Now, after you have thought about it thoroughly, please describe briefly the event and what you felt in that situation in three to six lines in the space below.”
Griskevicius, Shiota and Neufeld (2010) created these instructions without the use of the specific word ‘awe’ or any other ‘emotion label’ in order to avoid undesirable side effects due to specific ideas about the emotion label any participant may have, possibly differing severely from the researcher’s idea. Furthermore, emotions are often situation specific, elicited by certain triggers. Therefore, asking participants to remember a prototypical
situation, such as an experience with spectacular nature, should elicit a certain emotion, in this case ‘awe’.
17 For either condition, the answer required at least 250 characters in order to be sufficiently long to actually induce an emotion.
Measures
Mood measurements. The participant’s mood was measured by a self-reporting task. Participants were asked to indicate their current mood on a sliding scale ranging from 0, ‘the worst ever’, to 100, ‘the best ever’ (Sherman, Haidt and Coan 2009).
Emotion measurement. Adopted from Shiota, Keltner and Mossmann (2007, p.952), participants were asked to emotionally evaluate the situation they described in the writing task. In order to determine the specific emotions elicited through the writing task, and especially to ensure that the feeling of awe was evoked, participants were asked to rate specific emotions on a 7-point Likert scale, with 1 meaning ‘not appropriate at all’ up to 7, meaning ‘very appropriate’. The emotions listed for evaluation were awe, contentment, excitement, fear, joy, love, pride, rapture, sadness and surprise. Furthermore, participants were asked to rate thoughts and appraisals during the experience, specifically how intense, personally meaningful, novel, challenging and tiring the experience was, on a similar 7 point Likert scale, ranging from ‘not appropriate at all’ to ‘very appropriate’.
Additionally, participants rated on a 7 point scale from 1, ‘not true at all’, to 7, ‘very true’ to what extent the following statements were true to them: “I felt small or insignificant”; “I felt the presence of something greater than Myself”; “I was unaware of my day-to-day concerns”; “I felt closely connected to the world around me”; “I did not want the experience to end”; “I was aware of my personal values” and “I felt closely connected to my culture” (Shiota, Keltner and Mossmann,2007, p.952). The ratings and statements listed above all relate to the definition of awe, explained in the previous.
Reliability of the scales and with that internal consistency of the scales was tested and given (Emotions: α = .75; Thoughts: α = .76; Statements: α = .65), with each Cronbach’s alpha higher than .6. Therefore, the statements were combined into one variable, the “awe scale” for further analysis.
18 can choose one description they see as fittingly explaining the situation. High levels frame the action in terms of its ends, whereas low levels refer to the means of the action (Fujita et al. 2006). An example for such a task would be “drinking water” as the action, then identified as “appease thirst” (low level) versus “being healthy” (high level). The full list of items can be found in Appendix A. Low-level items were coded with 0, high-level items with a 1.
Summing these item scores up leads to an index for construal level ranging from 0 to 13. The higher the score, the higher is accordingly the construal level (see Fujita et al. 2006a).
Risk taking measurement. To measure participants risk taking, a lottery task was adopted from Trautmann and van de Kuilen (2011). To simplify the survey, participants only had to choose between two options. The first option is characterized by high uncertainty in terms of low probability in combination with high desirability in terms of high payoffs, the second option vice versa, low uncertainty and low desirability. For example, participants were asked to choose between Option 1: gain €3.600 at a probability of 7% and Option 2: gain €300 at a probability of 84%. All four lotteries are displayed in Appendix B. Each
participant’s choices for each lottery were summed up to identify his or her risk taking behaviour on a level from 0 to 4, with 0 being completely risk averse to 4 being completely risk prone (accordingly 1 = risk averse, 2 = neutral, 3 = risk prone).
Confidence measurement. To measure the construct of confidence, the scale developed by Bearden, Hardesty and Rose (2001) was used. They developed a scale consisting of six dimensions, all related to consumer confidence. The six dimensions of consumer self-confidence, information acquisition, consideration-set formation, personal outcomes decision making, social outcomes decision making, persuasion knowledge and marketplace interfaces, were measured by asking participants to rate statements on a 5-point Likert scale, from 1, ‘extremely uncharacteristic’, to 5, ‘extremely characteristic’. Examples of these statements include “I am confident in my ability to recognize a brand worth considering” in the
19 combined (α = .72), the dimensions were combined into one construct “confidence” for
further analysis.
4.3 Procedure
The survey opens with a few general questions, probing the participant’s gender and age, as well as a mood measurement. Then, participants were randomly assigned to either the neutral (n = 77) or the nature condition (n = 58). According to their assigned condition, they were asked to fulfil the writing task to stimulate awe or a neutral emotion, as explained in the previous. Next, for a manipulation check, participants were asked to rate their emotions felt during and evaluate the situation they described in the writing task. In the following step, participants were asked to choose between two diverging descriptions of a certain given situation or task, to measure the level of construal. Then, they were asked to rate on scales how characteristic certain statements are for them personally, measuring the level of confidence. In the final step of the survey, participants were asked to choose in a lottery. In total, participants had to choose in four lotteries. Each lottery consisted of two options, one with high payoff (gains) at a low probability, the risky option, and one with lower payoffs at a higher probability, the less risky option.
4.4 Plan of Analysis
To gain insights from the data collected, a plan of analysis was created. The plan of analysis consist of two parts. First, a descriptive analysis of the dataset and its respondents is
conducted, to describe the sample and obtain basic insights about the data. Information about the sample size, gender ratio, average age of the respondents and condition distribution are given. Then, the hypothesis of the conceptual model are tested. Analyses in SPSS include one-way and two-way analyses of variance (ANOVA) and a moderation analysis, making use of the PROCESS macro, developed by Hayes (2013).
5. RESULTS
20 individually, with a confidence level of 95% in adopting the hypotheses, displayed in a
significance level equal or lower than p = .05.
5.1 Sample Characteristics and Data Description
Demographics
Within a period of six days, 339 respondents started the survey. However, only 135
respondents completed the survey, leading to a drop-out rate of 60%. For the analysis, only completed surveys were used. A check for outliers was done by creating a box plot (see Appendix D). Although the boxplot revealed mild outliers for construal level and
self-confidence, they were not excluded from the survey due to non-extreme values. Thus, further analysis was continued with a data set of 135 respondents. Of those 135 who completed the survey, 98 were female respondents (72.6 %) and 37 male (27.4 %). Age of the respondents varied from 18 to 57, however the majority of respondents were young adults, aged between 23 and 28, making up about 78% percent of the respondents (Mage = 25.87; SD = 5.33).
Participants’ mood was overall better than average with an average rating of Mmood = 64.81;
SD = 17.14 and a range from 9 to 95 out of 1 to 100.The accessibility of the survey online enabled participation worldwide. The sample includes respondents from The Netherlands, Germany, the US and Australia.
Through randomization, 77 participants were assigned to the neutral condition, 58 to the nature condition, respectively 57% answered the writing task in the neutral condition and 43% in the nature condition. Of the 58 participants in the nature condition, 39 were women (67%), 19 men (33%). The 77 respondents in the neutral condition were 59 women (77%) and 18 men (23%). All written answers were checked for fit to the task and approved, so no survey had to be excluded from the data set. Appendix E gives an excerpt of answers given by participants in the nature condition.
5.2 Hypothesis Testing
Spectacular nature as an elicitor of awe (Hypothesis 1)
21 tiring). For the majority of emotions the analysis showed significant differences between conditions, with a significance level p ≤ .05. However, contentment and love were not significant, as can be seen in Table 1 (and Appendix F) which provides an overview of the results, listing the mean and standard deviation in each condition, the F-statistics and significance level. Results show, there was a statistically significant effect of the
environmental condition on the feeling of awe, F(1,134) = 66.76, p < .001. As expected, people in the nature condition experienced awe more strongly than participants in the neutral condition (nature condition: M = 5.22, SD = 1.92; neutral condition: M = 2.75, SD = 1.59).
NATURE/AWE NEUTRAL F p Sadness 2.02 (1.59) 1.57 (1.03) 3.91 .050 Fear 2.62 (1.94) 1.68 (1.21) 12.12 .001 Joy 6.19 (1.26) 5.52 (1.36) 8.51 .004 Contentment 5.60 (1.52) 5.08 (1.61) 3.69 .057 Awe* 5.22 (1.92) 2.75 (1.59) 66.76 .000 Excitement 5.72 (1.27) 5.18 (1.47) 5.07 .026 Love 4.64 (1.94) 4.08 (1.84) 2.92 .090 Pride 3.90 (1.94) 2.90 (1.69) 10.18 .002 Surprise 4.72 (1.73) 3.82 (1.83) 8.50 .004 Rapture 5.95 (1.34) 4.38 (1.80) 31.15 .000 novel 4.93 (1.74) 3.26 (1.69) 31.63 .000 challenging 3.98 (1.84) 2.84 (1.75) 13.42 .000 personally meaningful 5.67 (1.54) 3.75 (1.78) 43.21 .000 intense 5.97 (1.20) 3.61 (1.68) 81.83 .000 tiring 2.41 (1.90) 2.69 (1.76) .77 .383
Table 1: Overview of results - means (standard deviations) per condition, F-Statistics and significance level.
22
Figure 2: Mean ratings of emotions in each condition, with exact means in nature condition (numbers above the bars).
Regarding thoughts and appraisals, only tiring (p = .383) showed no statistical significance. As expected, participants in the nature condition rated their experience as significantly more novel, challenging, personally meaningful and intense than individuals in the neutral
condition. Figure 3 shows a summary of the means for each appraisal, differentiated by condition. Again, it becomes apparent, participants in the nature condition averagely rated the appraisals as more appropriate than the ones in the neutral condition. Solely tiring was
attributed to the neutral condition. Mean differences between conditions were especially strong for the rating of personally meaningful and intense as Figure 3 illustrates.
Figure 3: Mean ratings of appraisals in each condition.
Next, a one-way ANOVA was conducted for the statements related to the definition of awe, combined into the “awe scale”, as explained in the previous, which served as dependent variable, with “environmental condition” as independent variable. The analysis showed a statistically significant effect of the environmental condition on the awe scale, F(1,134) =
2.02 2.62 6.19 5.6 5.22 5.72 4.64 3.9 4.72 5.95 0 1 2 3 4 5 6 7 Nature Neutral 4.93 3.98 5.67 5.97 2.41 0 1 2 3 4 5 6 7
novel challenging personally meaningful
intense tiring
23 90.31; p = .000 (see also Appendix G). Again, individuals in the nature condition felt a
stronger feeling of awe than the ones in the neutral condition.
Additional insights were gained by analysing the emotional evaluation, specifically each statement of the “awe scale”, individually. Each statement served as dependent variable, while the condition again served as independent variable. The analysis showed significance for five of the seven statements, with no statistical significance for “I was aware of my personal values” (p = .076) and “I felt closely connected to my culture” (p = .418). All
statements showing statistical significance were also rated higher in the awe condition than in the neutral condition, providing additional support for the assumption that the nature
condition elicits a feeling of awe. Table 2 gives an overview of the results, including means per condition, F statistics and the significance level (see also Appendix H).
NATURE/AWE NEUTRAL F p
“I felt closely connected to the world around me.” 5.6 (1.54) 4.26 (1.45) 26.96 .000 “I was aware of my personal values.” 4.74 (1.48) 4.27 (1.53) 3.20 .076 “I felt small or insignificant.” 4.43 (2.04) 2.42 (1.33) 47.93 .000 “I did not want the experience to end.” 5.17 (1.81) 3.34 (1.74) 35.48 .000 “I was unaware of my day-to-day concerns.” 5.47 (1.66) 3.78 (1.64) 34.59 .000 “I felt closely connected to my culture.” 3.26 (1.91) 3.51 (1.64) .66 .418 “I felt the presence of something greater than Myslef.” 4.74 (2.07) 2.52 (1.60) 49.33 .000
Table 2: Overview of results - means (standard deviations) per condition, F-Statistics and significance level.
Overall, these results provide a manipulation check for the writing task and support the assumptions made about spectacular nature being an elicitor of the feeling of awe. Thus, Hypothesis 1 is supported.
The influence of awe on risk taking (Hypothesis 2)
The aim of this paper is to research the influence of the emotion of awe, elicited by spectacular nature, on consumer risk taking. As shown in the previous, a scenario of spectacular nature elicits the feeling of awe. To analyse the effect of awe on risk taking, a one-way ANOVA was conducted with “risk taking” as the dependent variable and
“environmental condition” as the between-subjects variable. The analysis shows no statistically significant difference between the two conditions, meaning environmental
24 willingness to choose risky options than those in the control condition (Mneutral = 1.05, SD =
1.12). Therefore, the feeling of awe does not reduce risk taking. Hypothesis 2 is not supported and has to be rejected.
To further analyse the relationship between the elicited emotions and risk taking, a Pearson’s correlation was run. Results show no significant correlation between the emotions and risk taking, except for Rapture, which shows a small negative statistically significant correlation with risk taking (r = -.19, p = .028), thus showing that the more rapture one
experiences, the less is one willing to take risks. Figure 5 gives an overview of all correlations between the individual emotions and risk taking (see also Appendix J).
2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 1. Risk Taking -.055 -.189* .019 .146 -.077 -.107 -.134 .007 -.163 -.047 2. Awe - .310** .115 .247** .185* .288** .152 .296** .343** .334** 3. Rapture - -.046 .101 .400** .409** .328** .415** .416** .559** 4. Sadness - .401** -.169* -.096 -.119 .053 .008 -.189* 5. Fear - -.079 -.018 -.185* .181* .134 .017 6. Contentment - .391** .371** .190* .050 .626** 7. Excitement - .294** .424** .490** .543** 8. Love - .335** .259** .371** 9. Pride - .512** .364** 10. Surprise - .241** 11. Joy - Note: * p ≤ .05; ** p ≤ .01.
Figure 5: Correlations between emotions and risk taking.
Relationship between awe and construal level (Hypothesis 3a)
Extensive literature review lead to the assumption that an experience with awe affects one’s construal level. This relationship was tested by conducting a one-way ANOVA, with “construal level” as the dependent variable and “environmental condition” (nature versus neutral) as the between-subjects variable. The measuring of the construal level showed a tendency towards preference for high level construals within the group of respondents, with an average of Mconstrual = 9.16, SD = 2.08 out of a scale from 0 to 13, with 13 being the highest
25 construal than those in the neutral condition (Mneutral = 9.32, SD = 1.98). Thus, no support for
Hypothesis 3a is given.
A Pearson’s correlation was run to gain further insights into the relation between the evaluated emotions and construal level. Figure 6 gives an overview of the results (see also Appendix L). However, results showed statistical significance solely for fear, with fear showing a small negative correlation with construal level (r = -.21, p = .013), showing that with increasing fear one experiences, one’s level of construal decreases.
Construal Level 1. Awe -.049 2. Rapture -.014 3. Sadness -.046 4. Fear -.214* 5. Contentment -.067 6. Excitement .125 7. Love -.028 8. Pride .032 9. Surprise .007 10. Joy -.036 Note: * p ≤ .05.
Figure 6: Correlations between emotions and construal level.
Relationship between construal level and risk taking (Hypothesis 3b)
Besides an effect of awe on construal level, an effect of construal level on risk taking was assumed. An experience of awe was expected to have a negative influence on risk taking. However, furthermore, an increase in construal level was assumed to increase tendencies towards risk proneness, thus having a positive effect on risk taking. A Pearson’s correlation was run to determine the relationship between an individual’s construal level and risk taking behaviour. No statistically significant correlation was found with r = -.01, p = .914 (see also Appendix M). The level of construal did not significantly affect risk taking behaviour. Thus, Hypothesis 3b has to be rejected.
Effects of gender differences on the relationship between awe and risk taking (Hypothesis 4)
26 than for men. First, an independent samples t-test with gender and risk taking was performed, in order to analyse whether or not the average risk taking of men is different from the average risk taking of women. The independent sample t-test was not significant, t(133) = -1.49, p = .140 (see also Appendix N). The average risk taking of men Mrisk/male = 1.32, SD = 1.27, does
not differ significantly from the average risk taking of women Mrisk/female = 1.00, SD = 1.08.
Furthermore, in order to analyse this influence of gender differences and awe on risk taking, a two-way ANOVA was conducted, with “risk taking” as the dependent variable, “environmental condition” (nature versus neutral) and gender (male versus female) as the between-subjects variables. Results of the analysis show no statistically significant effects, neither of the environmental condition and with it awe on risk taking (F(1,135) = .02, p = .879), nor gender on risk taking (F(1,135) = 2.04, p = .156) (see also Appendix O).
Participants in the nature condition (Mnature = 1.14, SD = 1.16) did not show a significantly
different risk taking behaviour than participants in the neutral condition (Mneutral = 1.05, SD =
1.12). The interaction effect was also shown to not be statistically significant (F(1,135) = .05, p = .817). Thus, there is no evidence for differences due to gender on the effect of awe on risk taking and Hypothesis 4 has to be rejected.
Effects of confidence on the relationship between awe and risk taking (Hypothesis 5a)
Based on the literature review, there is reason to believe that one’s level of consumer self-confidence has a moderating influence on the relationship between awe and risk taking. On average, participants showed medium to high level of confidence, with a mean of Mconfidence =
3.27, SD = .30, a minimum of 2.23 and a maximum of 4.29.
To test for a significant moderation the SPSS macro PROCESS, developed by Hayes (2013), was used. “Environmental condition” was entered as the independent variable, “risk taking” as the dependent, with consumer self-confidence as moderator. Table 3 gives an overview of the results (see also Appendix P). The analysis showed no statistically significant interaction between awe and confidence (p = .730). The model summary shows R2 = .007, which
27 b SEB t p Constant .50 [-2.40; 3.40] 1.47 .34 p = .735 Confidence .17 [-.71; 1.04] .44 .38 p = .705 Awe -.67 [-5.14; 3.80] 2.26 -.30 p = .768 Confidence x Awe .24 [-1.13; 1.61] .70 .35 p = .730 Note: R2 = .007.
Table 3: Overview of moderation results (Hypothesis 5a).
Effects of confidence on the relationship between construal level and risk taking (Hypothesis 5b)
Based on the literature review, there is reason to believe that one’s level of self-confidence as consumer has a moderating influence on the relationship between construal level and risk taking. Again, a moderation with the help of the PROCESS macro was run. Results, as summarized in table 4, show no statistically significant moderating influence of the
interaction between construal level and confidence (p = .957) (see also Appendix Q). One’s level of consumer self-confidence does not moderate the assumed positive influence of construal level on risk-taking. The bias-corrected 95% confidence interval included zero (-.30 to .31), giving no support or the assumption made, leading to rejection of Hypothesis 5b.
b SEB t p Constant .59 [-9.04; 10.21] 4.87 .12 p = .91 Confidence .18 [-2.79; 3.15] 1.50 .12 p = .91 Construal Level -.03 [-1.02; .96] .50 -.06 p = .95 Confidence x Construal Level .01 [-.30; .31] .15 .04 p = .97 Note: R2 = .004.
28 Below, table 5 gives an overview over the results of the hypotheses testing:
Assumption Adopted Rejected
H1 Experiences with spectacular nature elicit the feeling of awe.
H2 The feeling of awe has a significant negative effect on risk taking.
X
H3a An experience of awe leads to higher level construals.
X
H3b A high construal level has a significant positive effect on risk taking.
X
H4 The negative effect of awe on risk taking is significantlystronger for women than for men.
X
H5a The negative effect of awe on risk taking is moderated by one’slevel of confidence.
X
H5b The influence of construal level on risk taking is significantlyhigher with a higher level of confidence.
X
Table 5: Overview of the hypotheses and their outcome.
6. DISCUSSION
Although there has been extensive research on the effects and benefits of nature on the human mind in diverse contexts, literature on the topic of nature imagery in advertising and its effects on the consumer and his or her behaviour is scarce. Prior research identifies a variety of affective reactions to nature, as well as a variety of benefits of nature exposure. This study looked at the specific emotion of awe as a reaction to nature exposure and its effects on consumer behaviour, specifically risk taking in a consumer decision. The current research tested if, how and to what extent the emotion of awe, evoked by an exposure to spectacular nature, influenced individuals in their willingness to take risks in a consumer decision, represented by a lottery choice task.
29 used a lot, all somewhat related to the elaborated concept of awe. Appendix E holds an
excerpt of answers written by participants in the nature condition.
Furthermore, not only did participants in the nature condition feel the emotion of awe more strongly than participants in the neutral condition, but also felt their described
experiences to be more novel, challenging, intense and personally meaningful, all related to characteristics of experiences with awe (Shiota, Keltner and Mossmann 2007), further
supporting the definition of awe derived by Keltner and Haidt (2003). Finally, the evaluations showed, participants felt small or insignificant, the presence of something greater than
themselves and did not want the experience to end, supporting the notion that awe is a positively experienced emotional reaction to spectacular nature characterized by vastness (Keltner and Haidt 2003; Shiota Keltner Mossmann 2007).
Since a scenario and subsequent writing task was used to elicit awe, one could interpret that nature as an elicitor of awe is not limited to direct first-hand experiences with nature but that the feeling of awe can also be evoked through memories of nature experiences. This finding is supplementary to previous research proving that beneficial effects of nature can also be reaped through images of nature (Hartmann and Apaolaza-Ibáñez 2010).
Further analysis of the emotional evaluation of experiences with spectacular nature resulted in additional insights. Participants did not experience the described situations as sad or fearful, contradicting Keltner and Haidt’s (2003) idea of a possible negative connotation of awe, attributing to Shiota, Keltner and Mossmann’s (2007) notion that awe is a pure positive emotion. Supplementary to these findings, awe is shown to positively correlate with joy, contentment, excitement, pride, surprise and rapture, relating to prior research that found awe to be among a variety of positive emotions, to increase life satisfaction and to be novel and challenging (Keltner and Haidt 2003; Rudd, Vohs and Aaker 2012; Shiota, Keltner and Mossmann 2007). The mean ratings as part of the emotional evaluation showed exceptionally high ratings of joy and rapture in the nature/awe condition. Rapture is defined as “a feeling of intense pleasure or joy,” related to “expressions of intense pleasure or enthusiasm about something” (Oxford Dictionaries, 2015). From this definition one can see the connection between awe and rapture, both characterized by intensity, an exceedance of mental structures and positivity.
30 Bolderdijk 2015; Zhang et al. 2014), but also reduce stress and negative emotions such as fear and anger (Ulrich 1979). The positive evaluation of the described experience with nature, shown through high ratings of joy and rapture, could be interpreted as evidence for the positive effects of nature on mood and well-being.
After providing evidence for the assumption that spectacular nature is an elicitor of awe, the influence of this emotion on risk taking was investigated. The reviewed literature gave reason to believe that an experience with awe would lead individuals to behave less risky, choosing the ‘safer’ option in the provided lottery. This claim was not supported. An experience of awe does not lead one to show exceptional risk taking tendencies.
Despite the fact that no influence of awe on risk taking was found, it was further investigated how one’s construal level is affected by awe and in turn affects risk taking behaviour. It was assumed that an experience with awe would lead to higher level construals, due to the translation of overwhelming vastness of an experience with awe into psychological distance, which constitutes construal level theory. The claim that awe expands construal level was not supported. However, the average construal level of all participants was shown to be relatively high. Therefore, it might be possible that this prohibited the effect of awe on construal level to fully unfold. Likewise, it is possible that awe affects construal level in the opposite way as assumed. The idea that awe is evoked by an overwhelming vastness that needs to be accommodated by adapting mental structures and confines, could lead to an individual being more grounded and down-to-earth after successfully placing an experience with awe in one’s mental structures. The feeling of vastness and with it (psychological) distance might be replaced by a feeling of ‘grounding’, leading to a clearer focus, becoming more reasonable and realistic, more down-to-earth, which could translated into low level construals.
Although the results were unable to confirm an influence of awe on construal level, further analysis concerning the relationship between construal level and risk taking was conducted. Higher level construals frame information in more abstract mental representations, related to goal-related thinking and a stronger preference for higher, but less probable payoffs, considered more risky. This led to the assumption that the higher the construal level the higher the willingness to take risks. However, this assumption proved itself to not be supported. No evidence for an influence of construal level on risk taking was found.
Despite the absence of evidence for any influences of awe on risk taking, the proposed differences according to gender were analysed. Conforming with stereotypes and prior
31 than for men. Reason to assume this was given by the claim that women are more
conservative in risky situations, as well as feel negative emotions more strongly than men and therefore avoid negative outcomes as self-protection (Fessler, Pillsworth and Flamson 2004; Grable 2000; Plant et al. 2000). In line with prior results of this study, no evidence supporting these claims was found. Instead, a salient equivalence between men and women in all
emotions was found. Regarding risk tendencies, male participants showed a slightly stronger tendency towards risk taking. Although this difference is very small, it supports the notion of prior research that men are more willing to take risks and are more careless in risky decisions.
Next to the main effects as well as differences due to gender, it was tested if how and to what extent consumer self-confidence influences the relationship between awe and risk taking, as well as the relationship between construal level and risk taking. It was assumed that a higher level of self-confidence leads to a decreased effect of awe on risk taking, due to the increased believe in one’s own judgements with high self-confidence. Furthermore, the reviewed literature gave reason to believe that a high level of self-confidence in the accuracy of one’s judgements in risky decisions was related to a high construal level which is
characterized by abstractly framed mental representation, connected with seemingly lower risk (Gilovich, Kerr ad Husted Medvec 1993; Trope and Liberman 2003; Wakslak et al. 2006; Wan and Rucker 2013). However, the results were not able to confirm a relationship between consumer self-confidence, construal level and risk taking.
7. CONCLUSION
In the context of spectacular nature in advertising, the aim of this research was to investigate if, how and to what extent the feeling of awe, as an affective reaction to spectacular nature in advertising, influences consumer behaviour in financially risky decisions, depending on one’s construal level, self-confidence and gender. Overall, findings did not support the assumptions made in this research. The sole exception was evidence provided for nature as an elicitor of awe. Spectacular nature, such as panoramic views of impressive mountains or the vast ocean, was identified as an elicitor of the feeling of awe. Awe is characterised by an overwhelming experience, so novel that it creates a need for accommodation of itself in one’s current mental structures. Prior research gave reason to believe that awe would affect risk taking as a
32 between awe and risk taking was influenced by one’s construal level, the abstractness of mental information representation. Neither of these assumptions were supported by the
analysed data. Additionally, consumer self-confidence was included in the analysis, due to the belief that it would play an important role in the relation between awe and risk taking and construal level and risk taking. However, this influence could not be confirmed. Additionally, gender differences in all relationships were investigated, without any significant results.
7.1 Managerial Implications
Although the results of this study did not support the assumptions made after literature review, implications for marketers can be derived after all. The supported assumption that experiences with spectacular nature lead to a feeling of awe, as well as the finding that nature experiences increase joy and rapture, which is in line with prior research’s findings that nature experiences increase mood and well-being, can be used to marketers advantage. In order to distinguish one’s advertising from the advertising clutter, marketers could make use of experiences with spectacular nature, either in form of images or by triggering consumer’s memories of experiences with such nature. The mood enhancing power of nature could lead consumers to more positive evaluations of the advertising message and in turn the product.
7.2 Limitations and Recommendations for Future Research
Like any research, this study is subject to a number of limitations and shortcomings. The following will name a few and offer recommendations how future research can avoid these shortcomings.
In general, validity and representation of the data sample might be limited due to the fact that most respondents are in the age group of young adults (23 to 28, 78 % of the sample). Furthermore, a ratio of 27.4 % male versus 72.6 % female respondents might not give valid insights into gender differences. Therefore, future research should include a broader sample of age groups and equal numbers of male and female respondents.
