The effect of subliminally primed happiness on giving behavior to profit- and non-profit organizations

‘The effect of subliminally primed happiness on giving

behavior to profit- and non-profit organizations’

Master thesis

MSc in Business Administration- Marketing Track

Jesse Bouwer – 10535292 University of Amsterdam

Master thesis

MSc in Business Administration- Marketing Track Supervisor: Tina Dudenhöffer

2

Statement of originality

This document is written by Student Jesse Bouwer who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

3

Abstract

Even though the influence of happiness on consumer behavior has been the topic of many researches, studies often oversimplify the concept of happiness. This study makes happiness measureable, by defining happiness as a construct consisting of happy emotions and overall life satisfaction. Moreover, current research elaborates on the knowledge related to priming, by investigating if subliminal priming can be an effective tool to provoke giving behavior. Given these points, the present research tests the effect of subliminally primed happiness towards giving behavior, and whether this effect varies based on an individual’s level of emotional understanding and life satisfaction. These effects are measured for both profit- and non-profit organizations, to determine if the effect is dependent on the type of organization.

The conceptual model was tested with the use of an online experiment. This study’s sample consists of 201 participants, of which 50 were recruited via Amazon Mechanical Turk (mTurk). The first half participated to the first study, investigating the effects for profit organizations, and the second study participated to the second study, investigating the effects for non-profit organizations. Both studies had an experimental simple between-subjects design with 2 conditions (priming/non-priming).

Current research was not able to meet the expectations of primed happiness towards giving behavior, for both profit- and non-profit organizations. This means that subliminally primed happy emotions do not affect giving behavior, and this effect does not vary based on emotional understanding and life satisfaction. The overall findings of this research show that priming does not automatically lead to behavioral changes. Future research should determine under which circumstances priming is effective. Furthermore, this study provides evidence for the need to study happiness more into detail.

4

Table of contents

4.5.1 Level of life satisfaction... 34

4.5.2 Level of emotional understanding... 35

4.5.3 Level of giving behavior... 36

4.5.4 Control variables... 37

5. Results... 38

5.1. Descriptive statistics... 38

5 5.2.1. Main effect of subliminally primed happy emotions towards

giving behavior to profit organizations... 39

5.2.2. Different effects of subliminally primed happy emotions towards giving behavior to profit organizations, based on individual’s level of life satisfaction... 41

5.2.3. Different effects of life satisfaction towards giving behavior to profit organizations, based on the individual’s level of emotional understanding... 44

5.3. Study 2: Hypotheses testing... 45

5.3.1. Main effect of subliminally primed happy emotions towards giving behavior to non-profit organizations... 45

5.3.2. Different effects of subliminally primed happy emotions towards giving behavior to non-profit organizations, based on individual’s level of life satisfaction... 47

5.3.3. Different effects of life satisfaction towards giving behavior to non-profit organizations, based on the individual’s level of emotional understanding... 49

6. Discussion and conclusion... 51

6.1. Hypothesis H1a and H2a... 51

6.2. Hypothesis H1b and H2b... 52

6.3. Hypothesis H1c and H2c... 53

6.4. Practical and theoretical implications... 54

6.5. Limitations and future research...55

References... 60

Appendices... 66

Appendix 1. Pre-test images (4)... 66

Appendix 2. Stimuli materials... 68

6

List of tables

Table 1. Key differences profit- and non-profits organizations... 21

Table 2. Summarized hypotheses... 26

Table 3. Sample characteristics (N = 201)... 29

Table 4. Overview of experimental conditions... 30

Table 5. Mean, Standard deviation and reliabilty of life satisfaction... 35

Table 6. Mean, Standard deviation and reliability of emotional understanding... 36

Table 7. Means, Standard Deviations, Correlations... 39

Table 8. Hierarchical regression model testing the relationship between priming and giving behavior to profit organizations... 40

Table 9. Testing mediating role of life satisfaction in the effect of priming towards giving behavior to profit organizations... 43

Table 10. Testing moderating role of emotional understanding in the effect of life satisfaction towards giving behavior to profit organizations... 45

Table 11. Hierarchical regression model testing the relationship between priming and giving behavior to non-profit organizations... 46

Table 12. Testing mediating role of life satisfaction in the effect of priming towards giving behavior to non-profit organizations... 49

Table 13. Testing moderating role of emotional understanding in the effect of life satisfaction towards giving behavior to non-profit organizations... 50

List of figures

Figure 2. Conceptual model... 26

Figure 3. Final design of the advertisements with the aim to support giving behavior... 32

Figure 4. Main effect for profit organizations... 40

Figure 5. Mediation effect for profit organizations... 41

Figure 6. Moderation effect for profit organizations... 44

Figure 7. Main effect for non-profit organizations... 45

Figure 8. Mediation effect for non-profit organizations... 47

7

1. Introduction

Do you also recognize the many commercials asking for donations? Commercials illustrating neglected children in third world countries, which absolutely need your help. Your donation will change their lives and their future is in your hands (OXFAM, 2014). Your donation can make the difference!

Maybe you have recognized that you respond to these commercials differently, depending on the time of day, the medium, and perhaps your current mood can alter the reaction to these commercials. If an individual’s mood alters the reaction, it is important to determine if that particular mood contributes to the conviction of people to donate. Besides, if the current mood is counter-effective, it would be a waste of money for charities to invest in these commercials.

This study investigates if happiness can be primed and how happiness affects giving behavior. Below is explained how happiness consists of both happy emotions and life satisfaction. Additionally, the moderating effect of an individual’s level of emotional understanding will be determined.Below, all relevant variables will be elaborated, to clarify the theoretical relevance of this study.

First, the impact of emotions on someone’s giving behavior has been studied a lot lately. Emotional state seems to affect charitable giving behavior. Ibanez, Moureau and Roussel (2016) found that emotional states can positively impact donation levels among those who donate, but there is no direct effect for individuals who do not donate yet. However, other studies claim that emotions do influence prosocial behavior including donations, for all individuals (Doorn, Kleef & Pligt, 2015).

8 In this research the effect of happiness towards charitable giving behavior will be elaborated. Previous research shows that happiness and charitable giving behavior do correlate (Boenigk & Mayr, 2015). Different definitions of happiness exist and many studies oversimplify the concept of happiness. Some researchers define happiness as ‘life satisfaction’ (DeNeve & Cooper, 1998), while others define happiness as a subjective state of well being at any moment in life (Baumeister, et al., 2013). Both definitions are completely different and the concept of happiness should be studied more specifically. This study regards happiness as a construct, consisting of positive emotions (i.e. joy, pride, love) and life satisfaction.

Secondly, this study views positive emotion as affective primed emotions that form happiness, which was not the case in previous research. Affective priming implies that positive and negative reactions can be evoked with minimal stimulus input and virtually no cognitive processing (Zajonc, 1980). Positive emotions that evoke happiness provide a happy feeling at an exact moment, thus the idea of affective priming is relevant to link to the relationship between happiness and giving behavior. Cohn et al. (2009) studied the effect of daily emotions before, but they studied emotions over a month time. Sometimes temporary emotions at an exact moment are important, which can be caused by affective priming. This has not been studied before and that is why the results of this study will be of theoretical relevance.

As a third point, giving behavior to both profit- and non-profit organizations will be investigated. Boenigk and Mayr (2015) support the need to transfer their found effect of happiness towards charitable giving behavior to other types of donations. While several studies already studied the relationship between happiness and giving behavior to charities, this effect has not been studied in relation to profit organizations yet. The current study will investigate giving behavior to both types of organizations. This choice is supported by Austin and Stevenson (2006) who analyzed commercial and social entrepreneurship. They concluded several future research suggestions including the need to determine what the relative

9 competitive advantages, disadvantages and interactive dynamics are in markets where nonprofit and profit organization are both operating. It could be possible that happiness mostly affects giving behavior, when both types of organization are operating. For instance, by means of Corporate Social Responsibility (CSR) profits incorporate in the support of charities, which results in a fading distinction between both types of organizations. This is supported by the following example: Target claims that they donate 5 percent of its profit to local communities. In education alone Target has donated more than $875 million since 2010 (Moreno, 2015). As a final point, Lichtenstein, Drumwright and Braig (2004) add evidence that CSR does affect customer donations to corporate-supported non-profit organizations. This holds that consumers are more likely to donate to a corporate supported by a non-profit organization. This research will determine for which types of organization the effect found by Boenigk and Mayr (2015) is accountable.

Building on this, type of organization possibly influences the manner which people feel after donating money. Hasford, Farmer and Waites (2015) show that emotional understanding influences donation behavior. Their research shows that individuals with a low level of emotional understanding rely more on a charity’s scope, than individuals with a high level of emotional understanding. Therefore, the effect of emotional understanding is interesting to elaborate in the field of giving behavior. For this reason, this study will contribute by investigating the moderating role of emotional understanding on the effect of life satisfaction towards giving behavior to both profit- and non-profit organizations. Under such circumstances, individuals who donate money to a non-profit organization may feel better, because non-profit organizations do not have commercial goals, in contrast to profit organizations.

Finally, this study determines if happiness can be primed to evoke giving behavior, which holds that the dependent variable is giving behavior. The selection of behavior as the dependent variable is supported by the Behavior Modification Perspective on marketing (Nord

10 & Peter, 1980). According to this perspective, it is important to focus on environmental factors that influence behavior. While dependent variables such as attitudes, motives and needs form speculations, the behavior modification perspective takes the prediction and control of behavior. Behavior can facilitate the development of a comprehensive set of strategies and tactics (environmental and situational factors), which directly influences behavior.

Practical contributions

This study also provides practical contributions. As mentioned earlier, the gap between profit- and non-profit organizations can be a little vague. That is why it is important for organizations to determine how the public judges their organization, especially when organizations ask the public to give or donate money. If people indeed base their attitude towards an organization on the type of organization (profit/non-profit), the organization should consider to corporate with another organization. For example, it could motivate profit organizations to engage in CSR. If organizations learn more about the perceptions of their customers, they could raise more money as a result of donations.

Moreover, the idea of mood induction provides practical contributions. When organizations air a commercial, it is not possible to control the viewers’mood at the time when they watch the commercial. This is why it is important for organizations to adapt to the context in which the commercial airs. Grisekevicius, et al. (2009) show that contexts in which a specific mood is evoked does influence the effectiveness of advertisements. In their study, some people were exposed to a television program evoking fear, while others were exposed to a television program evoking romance. Results show that the effectiveness of peripheral heuristics has a different effect for each context.

Lastly, happiness will be primed with happy cartoon images. This choice is based on research by Kobayashi and Hara (1996), who claim that people show the highest emotional

11 recognition for cartoon-like faces. Results of the current study provide practical information to organizations. Namely, that organizations can now determine in which context it is more effective to evoke giving behavior as a result of happy cartoon images. This is relevant for decisions with regard to when commercials will be aired.

After the introduction, a literature review is presented in which the concepts of ‘priming’, ‘happiness’, emotional understanding’, ‘type of organization’ and ‘giving behavior’will be further explained. Then the conceptual framework of the study is presented, including the research question. Subsequently, the sample, design, stimuli material, procedure and measures are motivated in the methods section. After the methods are explained, the results of the experiment are illustrated. Lastly, the discussion and conclusion sections are presented, in which the limitations, implication and suggestions for future research are discussed.

12

2. Literature review

2.1. Priming

Before subliminal priming and affective priming will be explained, the basic concept of priming should be clarified. Janiszewski and Wyer (2014) claim that priming can be characterized as an experimental procedure in which the processing of an initially encountered stimulus is shown to influence a response to a subsequently encountered stimulus. The prime stimulus can have different shapes, such as words, sounds and images (Moore, 1982). During the process of priming the prime stimulus makes content, and the cognitive processes used to manipulate and comprehend this content more accessible (Förster & Liberman, 2007). This turns out in influenced judgments, decisions and overt behavior. That is why priming can be used as unconscious influence technology.

However, it is not self-evident that priming affects the receiver. Convincing evidence of the effectiveness of priming was scarce for a long time (Broyles, 2006). Only recently, research shows that the phenomenon of priming only works under very specific circumstances. A prime is only personally applicable when receivers of the prime are motivated to adapt their own purposeful behavior to the prime concept (Strahan, Spencer & Zanna, 2002). Further, it is important that the primed concept has the highest priority for the receiver. If it is the case that the receiver possesses various targets, or lacks time, targets activated by the primed concept cannot be pursued. A higher target priority also increases the chances that the primed concept becomes more accessible in the mind, which results in the influence on specific attitudes, intentions and behavior (Strahan, Spencer & Zanna, 2002). In sum, priming can be effective, but the motivation to aim a specific target and the relevance of the priming stimuli in order to accomplish these targets are required.

Even if the primed concept is both accessible and applicable, this does not guarantee that the prime influences the receiver (Dijksterhuis & Bargh, 2001). Three more factors that

13 contribute to the effectiveness of priming will be explained. First of all, the receiver of the prime should unconsciously process the prime. If the receiver recognizes the priming process, which holds that the receiver realizes that the prime serves as an influencing strategy, the priming effect is eliminated (Strahan, Spencer & Zanna, 2002). Secondly, Kunst-Wilson and Zajonc (1980) claim that when primes are graphically projected, they are better processed in comparison to the projection of words. This can be declared by the way how people process information. When processing graphic stimuli, less brain effort is required compared to processing words. At last, familiarity with the priming stimulus also affects the effectiveness of priming (Coates, Butler & Berry, 2006). Bargh (2006) claims that this occurs because prior (primed) knowledge has already been created in the brain, which means that processing information requires less effort (Bargh, 2006).

2.1.1. Subliminal priming

Karremans, Stroebe and Claus (2006) distinguish two types of priming, namely subliminal and supraliminal priming. While stimulus material of supraliminal priming is consciously noticed, stimulus material of subliminal priming is not. In this study, subliminal priming occurs during the online experiment. This is why the concept of subliminal priming will be further elaborated.

The term subliminal priming comes from the Latin words sub (below) and limen (threshold). The threshold relates to the threshold of conscious awareness. In 1957 subliminal advertising became notorious through the publicity towards a study by James Vicary. He claimed that he secretly and subliminally flashed the words ‘Drink Coca Cola’ and ‘Eat popcorn’ at the screen in a movie theater, which resulted in an increase in the sales of Popcorn (54%) and Coca Cola (18%). However, nobody has ever replicated these findings and the study by James Vicary has never been published and appears to have been en publicity hoax (Pratkanis, 1992). Nevertheless, the James Vicary study raised a lot of attention to subliminal

14 priming. In the upcoming years many researchers investigated whether subliminal priming influences behavior or not.

A large amount of people buys subliminal self-help tapes, with the aim to improve their self-esteem or lose weight (Natale, 1988). Despite the high sales in subliminal self-help tapes, Greenwald, Spangenberg, Pratkanis and Eskenazi (1991) found that neither of these tapes influenced the claimed effects. However, research also shows that subliminal priming can be a powerful technique to unconsciously influence people’s behavior. Bargh, Chen and Burrows (1996) determined if subliminal priming influences behavior. In their study, some participants were subliminally primed with an African-American face, and others were subliminally primed with a Caucasian face. They concluded that participants who were subliminally primed with an African-American face responded with more hostility to an unpleasant request made by the experimenter, compared to participants who were subliminally primed with a Caucasian face. Recent studies prove the effectiveness of subliminal priming as well. In a study by Bermeitinger, Goels, Johr, Neumann, Ecker and Doerr (2009), dextrose pills were primed and the consumption of the pills was measured in relation to the level of tiredness of the participant. They conclude that people can be subliminally primed, under the following circumstances: people are in a motivational state and the primes have to be need-related. Thus, research shows also that subliminal priming can influence people’s behavior.

2.1.2. Affective priming

Priming sometimes occurs unconscious. This happens when cues in the environment make cognitive processes more accessible, without people’s awareness (Bargh & Chartrand, 2000). One form of priming which can occur unconscious is affective priming. The affective primacy hypothesis claims that positive and negative affective reactions can be evoked with minimal stimulus input and virtually no cognitive processing (Zajonc, 1980).

15 Before affective priming will be further explained, the concept of ‘affect’ will be defined. Affect consists of two dimensions, which are valence and arousal (Kensinger & Schacter, 2006). Valence refers to a subjective specification of how positive or negative an event is, whereas arousal implies to the intensity of an event. Arousal ranges from very calming to highly exciting

Direct affective priming implies that an affective prime makes affective states more accessible. These affective states can be moods, emotions or feeling states that can influence judgments about unrelated stimuli (Janiszewski & Wyer, 2014). The effect of unconscious affective priming is proved in a study by Murphy and Zajonc (1993). Participants were asked to judge unknown Chinese ideographs as positive or negative. Before the ideographs were showed, an emotional face, neutral polygon or an empty screen was represented, either for 4 ms (suboptimal presentation) or for 1000 ms (optimal presentation). Only the suboptimal presentations of emotional faces produce shifts in participants’ liking and good/bad judgments. This can be considered as affective priming, because participants judged random ideographs as more positive after seeing a happy face, and as more negative after seeing an angry face.

But, affective priming does not always occur and boundary conditions do exist. First of all, the affective state must be relevant to the target judgment/evaluation (Gorn, Goldberg & Basu, 1993). Secondly, people who claim to trust their feelings are more likely to process affections as information (Avernet, Pham & Stephen, 2012). Moreover, Chang and Pham (2013) add that feelings provoked by affective priming are more diagnostic for temporally proximate than temporally distant judgments. The last boundary condition discussed is the way in which people process information. If people tend to visualize using a product, they are more likely to use affections as information than people who tend to verbally represent a product (Pham, 1998). Altogether, affective priming effects do not always occur and some boundary conditions

16 indicate the opportunity of affective priming. Only if the affective state is relevant to the judgment and people do not try to discount the affective information, priming occurs.

Additionally, a recent study by Phaf and Rotteveel (2012) explains the mechanism of affective priming. They formed the affective monitoring hypothesis, which focuses on affect raised by solving conflicts while processing information. Affective monitoring is about determining the advantages and disadvantages, from an evolutionary perspective. If something is evolutionary beneficial it is associated with positive affect, whereas something is evolutionary detrimental, it is associated with negative affect. The affective monitoring hypothesis claims that both negative and positive affect requires conflict, which can be solved quickly in the case of positive affect (i.e. matches, novelty, congruency) or not in the case of negative affect (i.e. incongruence, familiarity, mismatches). Thus, the presence of conflict creates an affective reaction, because people do not pay attention for other information any longer. The affective monitoring hypothesis is stronger for suboptimal affective priming effects, than for optimal affective priming effects.

2.2. Happiness

As discussed in the introduction, the specific mood of happiness will be studied. Before linking the concept of happiness to charitable giving behavior, happiness should be defined. Different definitions of happiness are formulated in previous studies. Diener (1984) defines happiness as holding positive feelings over negative feelings when evaluating one’s life. Some researchers extend the definition of happiness to ‘life satisfaction’ (DeNeve & Cooper, 1998), which can be explained as the quality of an individual’s experiences during his or her life. However, happiness and life satisfaction are not exactly the same because happiness can be defined as a subjective state of well being in any moment in life (Baumeister, et al., 2013), while life

17 satisfaction can be considered as an overall evaluation of the quality of life (Peterson, Ruch, Beermann, Park & Seligman, 2007).

Cohn et al. (2009) show the relationship between both components of happiness: life satisfaction and positive emotions (i.e. amusement, compassion, joy, love, pride) during any moment in life. These positive emotions related to happiness prepare the body for different kinds of responses, by increasing the activity in a brain center that inhibits negative feelings and boosts the energy available (Goleman, 1995). Goleman (1995) claims that this creates general rest, readiness and enthusiasm. Furthermore, positive emotions can be defined as a powerful source of change and growth, because they predict both individuals’ judgments about life and individuals’ skills for living well (Cohn et al., 2009). In other words, their study provides evidence for the role of positive emotions predicting life satisfaction. Similar results derive from a study to the shifting basis of life satisfaction judgments across cultures by Suh et al. (1998). The importance of the influencing role of emotions and social norms is measured in different cultures (individual and collective cultures) and results show that emotions form life satisfaction in both individual and collective cultures. To sum up, both components of happiness influence each other and they need to be studied in relation to each other.

This research focuses on both components of happiness, which results in giving behavior as a result of primed happy emotions mediated by life satisfaction. Previous research shows inconsistent directions of the relationship between happiness and giving behavior. First, Bekkers and Wiepking (2011) claim that charitable giving leads to positive psychological benefits for the donor. This is called ‘warm glow’, because donors immediately experience a pleasurable feeling as a direct result of the act of giving (Andreoni, 1989). This idea holds that an individual experiences a feeling of happiness when donating money to charities. Moreover, spending money for others creates greater happiness than spending money on oneself (Dunn, et al., 2008). On the other hand, the opposite relationship between happiness and giving

18 behavior is determined. This can be explained by the positive affect theory by Isen (1987), which maintains that happy individuals are more emotionally capable to give money for others than relatively unhappy people. These happy individuals have an overall more optimistic personality that positively affects giving behavior (Wang & Graddy, 2008). Rosenhan et al. (1974) adds that the emotional capability for giving behavior can be a consequence of the tendency to engage in altruistic behavior.

Boenigk and Mary (2016) studied both perspectives on the relationship between happiness and giving behavior and conclude that both are positively related. However, the perspective that considers giving behavior as a result of happiness is supported as the dominant perspective.

2.3. Emotional understanding

The concept of emotional understanding is a dimension of emotional intelligence that is relatively new, and recent studies show the importance of studying the role of emotional understanding. Kidwell et al. (2008) extended the framework of emotional understanding to the domain of consumer behavior to combine emotional understanding with consumer outcomes/decisions. Kidwell, et al. (2008) define emotional understanding as: “a person’s ability to skillfully use emotional information to achieve a desired consumer outcome”. They add that emotional understanding consists of a “set of first-order emotional abilities that allow individuals to recognize the meanings of emotional patterns that underlie decision making and to reason and solve problems on the basis of these abilities”. Consumers are now able to understand how they will feel as a result of a decision, which implies that they can regulate their emotions in the future. When emotional understanding is high, people are expected to integrate emotions into reasoned judgments, while the opposite effect would occur when emotional understanding is low.

19 Additionally, literature shows that some people have a higher possibility to score high on emotional understanding. First of all, emotional understanding gets bigger with age (Later, Mayer & Salovey, 1997). This may imply that older people are willing to donate more as a result of their knowledge related to their level of happiness, which is caused by the donation, than younger people. Secondly, research shows that emotional understanding has a correlation with someone’s quality of social interactions (Lopes, Salovey & Straus, 2003). In other words, people with a high level of emotional understanding are more likely to be better in socializing, than people with a low level of emotional understanding (Lopes, Salovey, Straus, 2003).

Previous research by Gannon and Ranzijn (2005) investigated the link between emotional understanding and life satisfaction. They claim that emotional understanding can explain a unique variance in life satisfaction. In this research the effect of life satisfaction towards giving behavior is studied, so it is relevant to determine if emotional understanding influences the relationship between these two variables.

Recently, more research is focused on non-profit organizations, because of its unique characteristics. The role of emotional understanding gains more attention as well. Hasford, Farmer and Waites (2015) expect the feeling that occurs after donating money is the result of a donor’s emotional understanding conveyed by a charity. They studied the effect of emotional understanding as a moderator on the relationship of a charity’s scope on donation behavior. They show that individuals with low emotional understanding do more often rely on a charity ’s scope when making donations.

2.4. Type of organization

In this study, the correlations between the concepts mentioned above will be studied for two different types of organizations, profit- and non-profit organizations. Both types of organization

20 share the same essence, namely the underlying needs to survive to operate (Keyton, 2010). However, both types of organization have a distinct ideology and their own characteristics.

Several key differences between profit- and non-profits organizations will be explained. First of all, profit organizations are motivated by revenue, continuity and growth (Lux, 2003), while non-profit organizations are motivated by social goals (Hull & Lio, 2006). Hull and Lio (2006) add that both types of organizations differ in their vision and strategic constraints. For example, profit organizations possess high budgets and focus on gaining profit, and non-profit organizations are focused on creating social value and have smaller budgets. Furthermore, non-profits organizations’ objectives are not managed by economic objectives and do not measure success in terms of profit, in contrast to profit organizations (Wright, Chew & Hines, 2012). Non-profit organizations are characterized by offering ideas, social behavior and services, while profit organizations offer psychological products. Another difference is that within the non-profit sector no direct payment-receipt linkage occurs, in contrast to the non-profit sector. In other words, consumers’ investments are not directly related to a concrete item, whereas this is the case for profits organizations. As a final difference, non-profit organizations serve multiple constituencies (Wright, Chew & Hines, 2012). Profit organizations serve the users of their offerings and suppliers which fulfill organizational missions, while non-profit organizations serve more stakeholders like volunteers, government agencies, contracting partners and the wider community indirectly affected by strategies of these non-profit organizations. The key differences are summarized in Table 1.

21 Table 1. Key differences profit- and non-profits organizations

Profit organizations Non-profit organizations

Motivated by revenue, continuity and growth Motivated by social goals

Possess high budgets Have smaller budgets

Objectives are managed by economic objectives and do measure success in terms

of profit

Objectives are not managed by economic objectives and do not measure success in

terms of profit

They offer psychological products. They offer ideas, social behavior and services

Direct payment-receipt linkage occurs No direct payment-receipt linkage occurs Just serve the users of their offerings and

suppliers which fulfill organizational missions

Serve multiple constituencies

Sometimes, the gap between profit and non-profits organizations is a little vague. Gassman et al. (2012) claim that non-profit organizations are particularly mission driven entities formed by observed demand concerning the welfare of people, animals and nature. However, this is not exclusively applicable to non-profit organizations anymore (Austin & Stevenson, 2006). Besides the main focus on money and increasing business by profit organizations, they concern the welfare of people by creating job places or they aid to economy in taxes. Furthermore, non-profit organizations such as charities are becoming more important for profit organizations, to fight for certain goals that are aligned with the business goals. For example, profit organizations engage in corporate social responsibility (CSR) activities. CSR can be defined as “a commitment to improve (societal) well-being through discretionary business practices and contributions of corporate resources” (Du et al., 2010, p. 8). CSR positively affects the reputation of an organization, which results in better relationships with stakeholders and an improved image (Du et al., 2010). Thus, it is becoming harder for the public

22 to distinguish both types of organizations, because profit organizations adopt non-profit-like strategies to engage in social and environmental issues (Trapp, 2012).

2.5. Giving behavior

For a start, the dependent variable within this study will be defined. The definition of philanthropy by Schuyt, Bekkers and Smit (2010) will be used to define giving behavior. Giving behavior entails: “Voluntary contributions (money, goods, time/expertise) to the public good, given by individuals and organizations and dominantly benefitting the public good”. Exclusively monetary voluntary contributions will be investigated, for the guarantee of measurable hypotheses.

But why do people make monetary donations to organizations? And which personal characteristics predict giving behavior? Previous research concludes several predictors of giving behavior. First of all, demographics, such as gender, age, education, income, religion and ethnicity do influence giving behavior (Slyke & Brooks, 2004). Additionally, previous experiences with donations, the ability to give and available information do impact giving behavior as well. Not only demographics, but also motivation drives giving behavior. Mount (1996) distinguishes fifteen different motivations for people to donate money. This includes believing in the purpose of the organization, overall aspiration to help other people and resulting psychological reward. Van Slyke and Brooks (2004) summarized these predictors of giving behavior in a model, based on several socio-demographic characteristics and the motivations found by Mount (1996). The model is presented in Figure 1. The model shows that these predictors influences behavior in the form of giving, volunteering or joining a charitable organization. Moreover, several triggers based on the motivations found by Mount (1996) are presented in the model. These external triggers have the ability to activate philanthropic behavior.

23 Figure 1. Predictors of giving behavior determined by Slyke and Brooks (2004)

As shown in Figure 1, not only monetary contributions to organizations do exist. Instead of cash offerings to the organization, people can donate time (volunteering for the charity), services and new or used goods (Ferris, 1984). Moreover, monetary donations can be divided into fixed-term donations and one-time donations. One-time gifts are beneficial (especially if it contains a large amount of money), but in general, fixed-term result in higher donations on an annual term (Anik, Norton & Ariely, 2014). Fixed-term donations are beneficial because they result in increased yearly revenue predictability due to the frequency of giving and they improve the odds that the organization will retain donors over a longer period of time.

In this study, the literature related to non-profit organizations is transferred to profit organizations as well, for the reason that there is a lack of literature that specifically investigates donations to profit organizations. In this paragraph, the topic of donations to profit organizations will be elaborated. Profit organizations can enhance giving behavior from consumers to the organization with the use of cause-related marketing. Cause-related marketing can be defined as a type of marketing, where the company communicates that a certain percentage of the product’s or service’s retail price that consumers pay will be donated to charity (Varadarajan & Menon, 1988). For illustration, one of Proctor & Gamble’s brands, Pampers, started a collaboration with Unicef. The aim was to encourage consumers to buy

Background factors Demographic characteristics Experiences Information Ability to give Triggers Asked to give Tax benefit Sense of duty Behaviors Giving to charitable organizations Volunteering for charitable organizations Participating in civic organizations

24 Pampers, because for every purchase a child vaccination was provided in a developing country (Vanhamme, Lindgreen, Reast & Popering, 2012). Proctor & Gamble provided a certain percentage of money to Unicef.

25

3. Conceptual framework and

hypotheses

In this chapter the effects and relationships between the variables previously introduced are elaborated. First, the main question will be revealed, and after this the different subjects will be explained in the conceptual model. The model guides the hypotheses, which are addressed afterwards.

3.1. Main question

Previous research shows that emotional state influences giving behavior. The emotional state of happiness seems to be the most effective one to affect people’s giving behavior (Boenigk & Mayr, 2016). However, research is only focused on the effects of happiness on non-profit organizations, while profit organizations seem to focus more on people and environment related subjects these days. It is important to determine in which extent profit organizations need to collaborate with non-profit organizations (Lichtenstein, Drumwright & Braig, 2004) to convince people to actually give something back to these profit organizations.

“How does the effect of primed happy emotions differ between giving behavior to profit organizations and giving behavior to non-profits organizations, mediated by life satisfaction? And how does emotional understanding moderate the effect of life satisfaction on giving behavior to profit organizations and giving behavior to non-profit organizations?”

3.2. Conceptual model

The main question is visualized in a conceptual model, in Figure 2 below. The aim of the model is to make the subject of the research more understandable.

26 Figure 2. Conceptual model

3.3. Hypotheses

In Table 2, a clear presentation of all hypotheses is shown. Table 2. Summarized hypotheses

H1a Individuals who are affectively primed with happy emotions have a higher tendency for giving behavior to profit organizations compared to individuals who are not affectively primed with happy emotions.

H1b The effect of primed happy emotions towards giving behavior to profit organizations is more pronounced for individuals who score high in life satisfaction compared to individuals who score low in life satisfaction.

H1c The effect of life satisfaction towards giving behavior to profit organizations is more pronounced for individuals who score low in emotional understanding when being exposed to happy emotion primes.

H2a Individuals who are affectively primed with happy emotions have a higher tendency for giving behavior to non-profit organizations compared to individuals who are not affectively primed with happy emotions.

H2b The effect of primed happy emotions towards giving behavior to non-profit organizations is more pronounced for individuals who score high in life satisfaction compared to individuals who score low in life satisfaction. IV: Primed happy emotions Mediator: Life satisfaction DV1: Giving behavior to profit DV2: giving behavior to non-profit Moderator: Emotional understanding Happiness H1c H2c H1b H2b H1a H2a

27 H2c The effect of life satisfaction towards giving behavior to non-profit organizations

is more pronounced for individuals who score high in emotional understanding when being exposed to happy emotion primes.

28

4. Methodology

This chapter precisely explains the online experiment that is executed. Two studies are conduced and this section addresses how and where the data is obtained. Moreover, the selection of specific measurements is explained. It is important to elaborate the methodology so threats to the internal validity of data can be avoided. The sample, design, stimulus material, procedure and measurements will be explained below.

4.1. Sample

In this research both men and females from 18 years and older participated in the online experiment. The initial sample size of the online experiment consisted of 279 participants. At first, people were asked to participate in the online experiment on the Facebook profile of the researcher. This approach can be considered as the snowball method. The post in which people were asked to participate in the study was shared many times, which resulted in a more varied group of participants. After a period of two weeks 229 participants started the online experiment. To gather a sufficient amount of data for a valid study, 50 more participants were recruited from Amazon’s Mechanical Turk (MT).

From the initial sample, 71 participants did not finish the online experiment and were hence excluded from the analysis. Furthermore, participants who filled in the preliminary test were excluded from the experiment to overcome potential issues with internal and external validity. Saunders et al. (2012) state that interaction effects can occur when people participate in both the preliminary test and the main experiment. The internal validity can be threatened by learning effects (Schreuder Peters, 2011), when participants understand the purpose of the research after finishing the preliminary test. Because of this, participants who understood the aim of the study were excluded from the dataset. For example, some participants answered the question ‘Do you think you know the purpose of the study’, with the following answer: ‘

29 subliminal messaging and ‘playing on emotion’, and ‘The influence of happy primes (pictures of disney characters in video) on willingness to pay more for your next purchase at Phillips’. Eventually, the final sample consisted of 201 participants (37% male, 63% female, age ranges from 19 to 78), with an average age of 34. More sample characteristics are presented in Table 3. After all variables were checked for missing data, none of these participants showed missing data.

Table 3. Sample characteristics (N = 201)

Age Gender n M SD Male 75 – (37%) 32.70 14.45 Female 126 – (63%) 34.73 14.15 Total 201 – (100%) 33.98 14.26

4.2. Design

This research consists of two online experiments to answer the research question. In the first experiment the effect of primed happiness towards giving behaviors to profit organizations is measured, and in the second experiment the effect of primed happiness towards giving behavior to non-profit organizations is measured. Both experiments consist of the exact same design, namely an experimental simple between-subjects design with 2 conditions.

The first online experiment consists of two conditions that are exposed to a profit advertisement. Both conditions differ in the absence or presence of primed happy emotions before the advertisement is shown. The conditions of the second online experiment differ in the absence or presence of primed happy emotions as well, but in contrast to the first online

30 experiment, participants in the second study are exposed to a non-profit advertisement. Table 4 provides a clear overview of the conditions.

To conclude, the approach to answer the research question of this study will be motivated, namely an online experiment. With the purpose to show causal relationships between primed happiness and giving behavior to profit- and non-profit organizations, an experiment seemed to be the most decent approach. Moreover, the experiment will be online, because it is beneficial from a monetary and time-related point of view. With the use of an online-experiment, it is possible to reach a large group of people in a short time period.

Table 4. Overview of experimental conditions

Study 1

Profit

Study 2

Non - profit Condition 1 This condition determines giving

behavior to profit organizations, with the presence of primed happy

emotions

This condition determines giving behavior to non-profit organization,

with the presence of primed happy emotions

Condition 2 This condition determines giving behavior to profit organizations, with

the absence of primed happy emotions

This condition determines giving behavior to non-profit organizations,

with the absence of primed happy emotions

4.3. Stimulus material

In both online experiments the stimulus material exists of a neutral video of 20 seconds. Two different videos do exist, namely a video in which primed happiness is absent and a video in which primed happiness is present. Both videos are almost identical to each other, because they are soundless and show the same film scene, which is a streaming waterfall. The video can be considered as neutral, because people see the exact same scene for 20 seconds and nothing

31 noticeable (i.e. sounds or changes in the environment of the waterfall) did occur. The neutral scene is illustrated in Appendix 2. However, both videos do differ in the presence of primed happiness. The video without primed happiness, shows the video for 20 seconds. The video in which primed happiness is present, 4 happy cartoon images are presented by subliminal priming, for 20ms. These cartoon images are illustrated in Appendix 1. Images are used to prime happy emotions, because literature claims that graphically projected primes are better processed, in comparison to the projection of words (Kunst-Wilson & Zajonc, 1980).

Additionally, some of the participants did participate in study 1 in which giving behavior to profit organizations is measured, while others participated in the second study in which giving behavior to non-profit organizations was measured. Philips was chosen as a profit organization and WWF as a non-profit organization. It is chosen to use already existing brands, because people are familiar with these organizations and consider Philips as a typical profit organization and WWF as a typical non-profit organization. If this would not be the case, the type of organization should be extremely emphasized, which could result in disclosing the aim of the experiment. This would treat the internal validity of the experiment.

For both conditions, a different advertisement with the aim to support giving behavior was designed. Apart from the type of organization, both advertisements were identical in terms of layout and message (help to save the world), to exclude disturbing variables as much as possible.

Finally, participants who were exposed to the Philips advertisement, were asked to pay extra for their next purchase at Philips to make a difference. Participants who were exposed to the WWF advertisement were asked to donate now to WWF to make a difference. The advertisements are illustrated in Figure 3.

32 Figure 3. Final design of the advertisements with the aim to support giving behavior

4.4. Procedure

Prior to the study, a preliminary test was conducted to 48 participants. The aim of the preliminary test is to determine which four of a variation of ten cartoon images are considered as happiest among participants, to prime happiness. As discussed in the literate review, the construct of happiness consists of positive emotions and overall life satisfaction. The purpose of the cartoon images is to manipulate the presence of positive emotions by priming. Cohn et al. (2009) made use of the Modified Differential Emotions Scale, in which people rate several emotions on a 5-point scale (0 = not at all, 5 = extremely). Out of the 10 positive emotions used in the Modified Differential Emotions Scale, the emotions of joyful and amusement were the only relevant ones to rate the cartoon images on (i.e. people cannot rate a mickey mouse image on the emotions of hope and compassion). The four images that are considered as the happiest, are illustrated in Appendix 1. The images of Mickey Mouse (M=3.24, SD=0.63), SpongeBob (M=3.23, SD=1.10), Winnie the Pooh (M=3.75, SD=0.76) and Pluto (M=3.27, SD=0.67) scored generally high on the Modified Differential Emotions Scale and are used to prime happiness in the main study.

33 4.4.2. Main study

Participants were recruited through Facebook and direct mails. They entered the experiment after reading a short introduction to the online experiment. The introduction claims that the experiment is a part of a master graduation at the university of Amsterdam and participation would take 5-10 minutes of their time. Further, a good proficiency with the English language is required in the introduction. Finally, to motivate people to participate to my study, the introduction mentions that a 20-euro bol.com gift card will be raffled at the end of the experiment. After participants read the introduction, they clicked on the link and the online experiment opened in Qualtrics. At first, people were informed about the aim of the study, namely to understand the drivers of giving behavior, followed by an agreement on informed consent.

Subsequently, the experiment consists of main 4 parts. The first part of the online experiment consists of several questionnaires, which were filled in by participants. These questionnaires measured participants’ level of the mediator life satisfaction and the moderator emotional understanding. The level of life satisfaction was determined with The Satisfaction With Life scale (Diener, Emmons, Larsen & Griffin, 1985) and the level of emotional understanding was determined with the Schutte Self-report Emotional Intelligence Test (Schutte, Malouff & Bhullar, 2009).

In the second part the priming manipulation occurs. Participants were exposed to a short video, which could neither to be paused, nor replayed. Before the video was shown, participants were asked to watch the video in an environment in which they could not be distracted and to watch the video carefully. Half of the participants were exposed to a video in which happy emotions were primed with cartoon images, while the other participants were exposed to a

34 neutral video without primed happy emotions. During the video, four cartoon images that prime happy emotions were presented with a rapid speed of 20ms.

Then, when participants were (or were not) primed with happy emotions, in part 3 an advertisement was shown. Participants were asked to imagine themselves seeing the advertisement with the aim to support giving behavior. In study 1, participants were exposed to an advertisement by profit organization Philips and in study 2 participants were exposed to a non-profit advertisement by WWF. After the advertisement was presented, participants provided an amount they would be willing to donate. The question consisted of an open-ended item, so participants could enter any euro amount.

Lastly, in part 4, participants were asked to answer some questions about their general background information (i.e. age, gender), and a validity check was conducted by asking participants if they had noticed anything unusual during the experiment and if they know the purpose of the study. As a final point, people could fill in their email address to participate to the gift card lottery.

4.5. Measures

4.5.1. Level of life satisfaction

Participants’ level of life satisfaction was determined using a well-validated scale, namely the Satisfaction with Life Scale (Diener, Emmons, Larsen & Griffin, 1985). The Satisfaction with Life Scale has been widely used and is considered to be reliable and valid (Pavot & Diener, 2008). The brief scale consists of 5 items, such as ‘the conditions of my life are excellent’. All items are rated on a 7-points scale (1 = strongly disagree, 7 = strongly agree). See Appendix 3 for all the scale item questions. The average of the items was taken to a construct scale. The higher people score on the scale, the higher their level of life satisfaction. Scores ranged from 1.40 to 7.00 (M = 4.873, SD = 1.213). The Satisfaction with Life Scale has high reliability with

35 Cronbach’s Alpha = .866 and good correlations (score > .30) among all items. Also, none of the items would substantially affect reliability if they were deleted. A test for normality resulted in substantial skewness across items (skewness between -.1.0 and 1.0, kurtosis -.5 and .5). Furthermore, a principal axis factoring analysis (PAF) was conducted on the scale. The Kaiser-Meyer-Olkin measure verified the sampling adequancy for the analysis, KMO = .855. Barlett’s test of sphericity χ2 (10) = 496.111, p < 0.001, indicated that correlations between items were

sufficiently large for PAF.

Table 5. Mean, Standard deviation and reliabilty of life satisfaction

Variable Mean SD Cronbach’s Alpha

Emotional understanding 4.873 1.213 (0,87)

4.5.2. Level of emotional understanding

Participants’level of emotional understanding was determined using a well-validated scale, namely the Schutte Self-report Emotional Intelligence Test (Schutte, Malouff & Bhullar, 2009). This extensive scale consists of 33 items that measure emotional understanding, such as ‘I am aware of my emotions as I experience them’. See Appendix 3 for all the scale item questions. All items are rated on a 5-points scale (1 = Strongly disagree, 7 = Strongly agree). The development of the scale by Schutte is based on Salovey and Mayer’s (1990) well-known model of emotional intelligence. Their model claims that emotional intelligence consists of utilization of emotions, appraisal of emotion in the self and others, regulation of emotion in the self and expression of emotion. The average of the items were taken to a construct scale. The higher people score on the scale, the higher their level of emotional understanding. Scores ranged from 2.67 to 4.76 (M = 3.818, SD = 0.374). The Schutte Self-report Emotional Intelligence scale has high reliability with Cronbach’s Alpha = .855 and good correlations (score >.30) among all

36 items. Also, none of the items would substantially affect reliability if they were deleted. A test for normality resulted in substantial skewness across items (skewness between -.1.0 and 1.0, kurtosis -.5 and .5). Furthermore, a principal axis factoring analysis (PAF) was conducted on the scale. The Kaiser-Meyer-Olkin measure verified the sampling adequancy for the analysis, KMO = .817. Barlett’s test of sphericity χ2 (528) = 2095.449, p < 0.001, indicated that

correlations between items were sufficiently large for PAF.

Table 6. Mean, Standard deviation and reliability of emotional understanding

Variable Mean SD Cronbach’s Alpha

Emotional understanding 3.818 0.374 (0.86)

4.5.3. Level of giving behavior

The dependent variable in this research is giving behavior. During the online experiment, an advertisement was shown, in which participants were told that they could contribute to help to save the world. After the advertisement was presented, participants were asked how many euros they were willing to give to the advertisement’s organization. They were asked to fill in any amount of euros they would like to give. This holds that the level of giving behavior was measured with an open-ended item, so participants could enter any euro amount. Previous research demonstrates that the use of an open-ended item to measure giving behavior is an effective measure (Hasford, Farmer & Waites, 2015). The use of an open-ended item encourages the discovery of unanticipated answers. If several amounts of donations are presented as categories to measure giving behavior, participants will be driven to think about the amount of the donation in terms the experiment presents it. However, it is important to measure how participants really think and feel about the amount of the donation. This is why giving behavior is measured with an open-ended item.

37 Within this study, two different types of giving behavior are measured. In study 1, giving behavior to profit organization Philips is measured, while in study 2, giving behavior to non-profit organization WWF is measured. After the Philips advertisement in study 1 was presented, the following question was asked:

“Imagine seeing this advertisement while purchasing something at profit organization Philips. Philips asks you to help to save the world, by paying extra for your next purchase at Philips. How much are you willing to pay on top of the regular price of the item you are buying to help to save the world?”

In study 2, giving behavior to non-profit organization WWF was measured with the following question:

“Imagine seeing this advertisement by non-profit organization WWF. WWF asks you to donate now, to help to save the world. How much are you willing to donate to help to save the world?”

4.5.4. Control variables

Because giving behavior, life satisfaction and emotional understanding may depend on age and gender, these characteristics were measured as control variables for in later analyses.

Age: year of birth, measured on an interval scale. Gender: male/female, measured on a nominal scale.

38

5. Results

5.1. Descriptive statistics

A correlation analysis was conducted to start with a study of the relationship between the variables. The means, standard deviations and correlation coefficients are reported in the correlation matrix in Table 7. The correlation matrix shows that the attendance of priming does not appear to relate significantly to giving behavior to both profit- and non-profit organizations, with a Pearson Correlation coefficient of r=0.07 for profit organizations, and r=0.12 for non-profit organizations. Furthermore, the correlation matrix demonstrates that giving behavior does not appear to relate significantly with life satisfaction and emotional understanding, which are the mediator and moderator within this study. For profit organizations, giving behavior does not significantly relate with life satisfaction 0.07) and with emotional understanding (r=-0.17). Non-profit organizations also show statistically insignificant correlations with life satisfaction (r=-0.4) and emotional understanding (r=-0.8)

On the other hand, one significant correlation results out of the correlation matrix. Emotional understanding seems to significantly relate with life satisfaction, with a Pearson Correlation coefficient of r=0.38.

39 Table 7: Means, Standard Deviations, Correlations

Variables M SD 1 2 3 4 5 6 1. Gender 0.37 0.48 2. Age 33.98 14.26 -.07 3. Life Satisfaction 4.87 1.21 -.04 .07 4. Emotional understanding 3.82 0.37 .12 .11 .38** 5. Giving behavior 24.29 113.13 .10 -.09 .00 .06 6. Attendance of priming for

profit organizations

0.51 0.50 -.12 -.03 -.07 -.17 .07 7. Attendance of priming for

non-profits organizations

0.47 0.50 .17 .11 -.04 -.08 .12 .00 ** Correlation is significant at the 0.01 level (2-tailed)

* Correlation is significant at the 0.05 level (2-tailed)

5.2. Study 1: Hypotheses testing

In this section, the hypothesis with regard to profit organizations (H1a, H1b, H1c) will be tested. PROCESS macro for SPSS, developed by Hayes (2012), was used to test the proposed simple moderation and the indirect mediation relationships. This implies that the models 4 and 1 were used. To explain the main effect, a hierarchical multiple regression was conducted.

5.2.1. Main effect of subliminally primed happy emotions towards giving behavior to profit organizations

The aim of the first hypothesis is to test whether exposure to subliminally primed happy emotions influences participant’s giving behavior to profit organizations. This main effect is illustrated in Figure 4.

40 Figure 4. Main effect for profit organizations

H1a. Individuals who are affectively primed with happy emotions have a higher tendency for giving behavior to profit organizations compared to individuals who are not affectively primed with happy emotions.

To test if participants who were subliminally primed with happy emotions do significantly differ in their giving behavior to profit organizations from those who were not subliminally primed with happy emotions, the first hypothesis was tested. A hierarchical multiple regression was performed to investigate the ability of subliminally primed happy emotions to predict giving behavior to profit organizations. After controlling for age and gender, the results are presented in Table 8 below.

Table 8. Hierarchical regression model testing the relationship between priming and giving behavior to profit organizations

R R 2 R 2 Change B SE _β t Step 1 .20 .04 Age -.06 .16 -.04 -.38 Gender 8.46 4.48 .19 1.89 Step 2 .22 .05 .01 Age -.03 -.35 Gender .20* 1,98 Attendance of priming .09 .90

Note. Statistical significance: *p <.05; **p <.01; ***p <.001

In the first step of the hierarchical multiple regression analysis the two control variables were entered, age and gender. This model was statistically insignificant for giving behavior to profit organizations (F (2, 95) = 1.90, p ns= .15). In step 2, the attendance of priming as an independent variable was entered. Similarly, the model was statistically insignificant for giving behavior to profit organizations (F (1, 94) = .81, p ns= .37). This implies that participants who

Attendance of priming

Giving behavior to profit organizations

41 were subliminally primed with happy emotions (M = 6.86, SD = 28.27) or those who were not subliminally primed with happy emotions (M = 3.96, SD = 10.42) do not differ significantly in terms of their giving behavior to profit organizations. Moreover, while gender alone presents no significant effect towards giving behavior to profit organizations in step 1, when the attendance of priming is added in step 2, a change occurs. The model shows an interaction between the variables of gender and giving behavior to profit organizations, with the Beta value (b=.20, p<.05). Men (M = 38.79, SD = 142.78) seem to have a higher tendency to give money to profit organizations than women (M = 15.58, SD = 90.39).

Given these results, hypothesis H1a was not supported.

5.2.2. Different effects of subliminally primed happy emotions towards giving behavior to profit organizations, based on individual’s level of life satisfaction

The second hypothesis tests if an individual’s level of life satisfaction mediated the effect of primed happy emotions towards giving behavior to profit organizations. The mediating effect is illustrated in Figure 5. Furthermore, the total, direct, and indirect relationships between primed happy emotions, life satisfaction and giving behavior to profit organizations, are presented in Table 9 below.

Figure 5. Mediation effect for profit organizations c1’

a1 b1

H1b. The effect of primed happy emotions towards giving behavior to profit organizations is more pronounced for individuals who score high in life satisfaction compared to individuals who score low in life satisfaction.

Attendance of

priming profit organizations Giving behavior to