Master Thesis Marketing Management

1

Master Thesis Marketing Management

“The Effect of Donation Magnitude and Face Valence on

Consumer Response in Cause-Related Marketing”

2 ABSTRACT

Companies often engage in cause-related marketing (CM) activities as a way to improve their image as well as increase sales. Such campaigns often include the donation amount the company is ready to give to the charity. Additionally, a common tactic is to incorporate the face of the victim in the advertisement. While there is extensive information on the application of facial expressions within the domain of charitable giving, research on this topic within the field of CM is lacking. This research addressed three main objectives. First, we investigated whether higher donation amount would result in a more positive consumer response, which includes attitude towards the campaign, attitude towards the company and purchase intentions. Second, we focused on the possible roles of perceiv ed moral hypocrisy and moral emotions. More precisely, to what extent the consumers would perceive the company as being hypocrite and would experience negative moral emotions, knowing how much the company would donate, and how this would be translated into the consumer response. Third, we investigated the role of face valence. Particularly, how implementing either happy or sad face will be reflected on consumers’ reaction about the campaign and the company. In order to complete our study, we investigated th e interplay between all the above mentioned constructs: whether and how the effect of donation amount on consumer response depends on perceived moral hypocrisy and moral emotions, and what role the face valence have in this interplay. Findings suggest that higher donation amounts result in a more positive consumer response. Moreover, higher donation amount lowered the perception of moral hypocrisy and experienced of negative moral emotions. Positive face valence was found to have a significant effect on consumer response. However, we found no support for the moderated mediation, as the face valence did not influence consumer response as it was predicted. Nevertheless, this study produced meaningful contribution to the research field of cause-related marketing.

3 TABLE OF CONTENTS Abstract... 2 1. INTRODUCTION... 5 2. Theoretical framework ... 9 2.1 Cause-related Marketing... 9

2.2. Direct effects of donation amounts in CM campaigns... 10

2.3. The mediating effect of perceived moral hypocrisy and moral emotions ... 12

2.4. The moderating role of face valence ... 15

2.5. Moderated mediation & Conceptual Model ... 17

2.6 Overview of the study ... 18

3. Methodology ... 19

3.1. Design and participants ... 19

3.2. Procedure ... 19

3.3. Measures ... 20

3.3.1. Consumer attitude toward the campaign... 20

3.3.2. Consumer attitude towards the company ... 20

3.3.3. Purchase intentions ... 21

3.3.4. Perceived moral hypocrisy... 21

3.3.5. Moral emotions... 22

3.4. Control variables ... 22

3.4.1. Cause importance... 22

3.4.2. Consumer skepticism to CM in general ... 23

3.5. Manipulation checks... 23 4. Results ... 24 4.1. Manipulation checks... 24 4.2. Control variables ... 25 4.2.1 Gender... 25 4.2.2 Age ... 26 4.2.3. Cause importance... 26

4.2.4. Consumer skepticism to CM in general ... 26

4.3 Testing the hypotheses... 27

4

4.3.1.1. Consumer attitude towards the campaign ... 27

4.3.1.2. Consumer attitude towards the company ... 28

4.3.1.3. Purchase intention... 29

4.3.1.4. Perceived moral hypocrisy... 30

4.3.1.5. Moral emotions... 30

4.3.2. Moderated mediation analysis ... 32

4.3.2.1. Campaign attitude ... 33

4.3.2.2. Company attitude ... 34

4.3.2.3. Purchase intention... 35

5. Discussion ... 37

5.1 Managerial implications ... 40

5. Limitations and future research ... 40

References ... 43

Appendix A... 49

Appendix B... 60

5

1. I

Nowadays, individuals perform diverse shopping activities on a daily basis. When doing shopping, consumers may face products which are labelled as related to charity. An example of such a product can be Pampers. Since 2006, Pampers and UNICEF have been successfully collaborating, in order to provide millions of children with vaccines. When consumers purchase one package of Pampers, the company would provide one vaccine together with UNICEF. This example of collaboration between a for -profit company and a non-profit organization (NPO) is called cause-related marketing (CM). Cause-related Marketing is increasingly becoming a key instrument used by for -profit companies to improve customer well-being and enhance company’s image (Smith, 1994, Koschate-Fischer et al., 2012). According to Varadarajan and Menon (1988), cause-related marketing (CM) can be defined as follows: ‘the process of formulating and implementing marketing activities that are characterized by an offer from the firm to contribute a specified amount to a designated cause when customers engage in revenue-providing exchanges that satisfy organizational and individual objectives”. In other words, it is considered a new form of corporate philanthropy, when a company makes a monetary or other kind of contribution to a designated social cause each time consumers purchase a CM-labelled product or service (Pracejus, Olsen and Brown, 2003). The greatest advantage of CM campaigns is their ability to provide benefits for three parties: the company, the non-profit organization and consumers. Despite the fact that CM campaigns are a common marketing practice these days, the effectiveness of CM campaigns on consumer response and corporate image remains rather ambiguous.

6 participating in CM is perceived as extrinsic (i.e. self-interested) rather that intrinsic (i.e. altruistic). In case firms’ actions are perceived as insincere, they risk to be punished by consumers (Sen and Bhattacharya, 2001; Simmons and Becker-Olsen, 2004).

Consumers may use different elements of CM campaigns in order to make attributions about the motivation of the company. Research suggests that, for consumers to respond positively to CM, the campaign should contain elements that justify the belief that the company is rejecting its basic self-interested nature (Ellen et al., 2000). The majority of previous work indicates that companies benefit more from CM campaign if they sell hedonic product rather than utilitarian or luxury product (Strahilevitz and Myers, 1998, Dean, 2004). This is due to the fact that consumers associate the altruistic utility offered by charity incentives more with hedonic products. In this way, consumers can attribute positive altruistic motives to the company. Additionally, different elements of these campaigns, used on the posters, can make a difference (Webb & Mohr, 1998). As examples of these elements, we are going to focus on donations and faces denoted in the advertisements.

7

TABLE 1.

COMPARISON OF DONATION AMOUNTS

Product Product Donation amount

“Hand of Friendship” soap Lush 100% of sales

Coffee Coffees 3 – 10 % of sales

RED T-shirts GAP 50% of the profits

Icons Dream Icons 100% of sales

Beanie Baby Ty Warner 100% of the profit

Charity Pot Lush 100% of the price

Jockey catalogue jockey.com 1% of sales

Colorful glass candle Glassybaby 10% of the revenues

Shoes Toms One pair donated for 1 purchase

Pack of Pampers Pampers (P&G) 1 pack = 1 vaccine provided by Unicef

All in all, there is considerable variation in donation amounts and types, which can subsequently influence consumer’s perceptions of self-interested motives of the company. Hence, an important question is how and in which way donations from CM campaigns reflect on consumer’s response and the company’s image. Indeed, we suggest that consumers can relate the donated amounts to their perception of the self-interested motivation of companies, i.e. the lower the donation amount, the higher the perceived self-interested motivation would be, as companies are keeping the larger share of profits to themselves.

8 provided by consumers (Small & Verrochi, 2009). However, within the domain of traditional marketing campaigns, it has been found that a smiling face is more e ffective in eliciting a positive attitude towards the company using the happiness appeal compared to sad faces (Söderlund & Rosengren, 2003). However, the research on which face valence is more effective within CM is lacking.

CM campaigns are a combination of a marketing campaign and a charitable giving. This raises a dilemma for companies: which face valence can be used best in order to avoid creating hypocrite image? On the one hand, consumers can accuse companies of using sad and crying children in their campaigns for abusing those children’s bad situation in order for the companies to make profit. On the other hand, when seeing a happy child, consumers might think that help is not needed and that company is being only self -interested and profit-oriented. Observations suggest that companies tend to incorporate smiling and happy faces in their CM campaigns. For example, Warby Parker used a happy face in their glasses CM campaign, with the slogan “Buy a Pair, Give a Pair”, which suggest that for e very purchased pair of glasses, company distributes another pair to someone in need (Figure 1). Interestingly, little research has investigated how the face valence within CM of for -profit companies can influence consumer response and what influence this e lement can have on company’s corporate image.

Figure 1. Warby Parker cause marketing campaign with positive face valence

9 know whether there can be any mediating factors involved in the evaluation process. Consumers tend to perceive companies based on the dimensions of morality and competence (Aaker et al., 2010). For-profit firms are generally perceived as more competent rather than moral. While being perceived as competent is translated into highe r purchase intention, positioning the company as being moral can make consumers steer away from the company (Newman & Cain, 2014). CM campaigns can produce these negative effects if the underlying motivation is perceived as exploitative (Varadarajan & Meno n 1988; Webb & Mohr 1998). As a foundation for judgement about company’s corporate image and decision making of customers, moral emotions play a role (Kim & Johnson, 2012). We argue, that, if, based on the donated amount and face valence of the victim the campaign entails, consumers perceive the company as being hypocrite, they will experience negative moral emotions. This can produce negative consumer response. Thus, it is relevant to understand how customers respond to the actions of the company involved in CM campaign, particularly investigating the perception of moral hypocrisy and moral emotions that the campaign and its elements can evoke and how it reflects on company’s image.

The objective of this research is to investigate how participating in the CM campaigns and different elements of those campaigns can reflect on the firms’ corporate image by investigating consumer responses. Particularly, this research will examine how consumers react to different donation amounts in CM campaigns and how it is r eflected on the attitude towards the campaign, purchase intention and company evaluation. We will propose that these effects can be explained by perceived moral hypocrisy and experienced moral emotions. Moreover, the face valence used in the CM campaign can moderate these effects. In other words, the effect of the donation amount on perceived moral hypocrisy and moral emotions, will depend on whether the company is adopting happy or sad face in their CM campaign.

2.

Theoretical framework

2.1 Cause-related Marketing

10 Currently, it is one of the fastest growing marketing communications practices. This is due to its win-win-win situation for three different parties: for the company itself, for the cause or non-profit organization the contribution is donated to, and for the consumers. With the help of CM campaigns, companies can increase sales not only of the CM-labelled product, but also sales of other products in their portfolio (Krishna and Rajan, 2009). Additionally, they enhance their corporate image (Adkins, 2000, Koschate-Fischer et al., 2012). Non-profit organizations benefit from this cooperation as they potentially expand their fun ding and increase public awareness of their causes (Docherty and Hibbert, 2003; Wood, 1998). With regards to consumers, they might be initially interested in purchasing the product. Among the benefits of charitable giving, consumers are able to help society. They also experience the feeling of satisfaction from doing something good (Polonsky and Wood, 2001). A theory of “warm glow” suggests that consumers experience positive warm feeling from giving to charity (Andreoni, 1990; Crumpler and Grossman, 2008).

Despite its popularity, a growing body of research has demonstrated the presence of criticism towards CM activities. Consumers develop negative attitude towards CM practice due to small donations delivered to NPOs (Endacott, 2004) and perceived exploitation of the NPOs for the personal benefit of the for-profit firm (Varadarajan and Menon, 1988). The motivation of the for-profit organization cooperating on the CM initiative can be perceived as self-interested and exploitative, which results in a negative hypocritical image of the company (Barone, Norman and Miyazaki 2007; Newman and Cain, 2014).

2.2. Direct effects of donation amounts in CM campaigns

11 effectiveness. However, according to these authors, percentage as a unit of donation is used up to 5 times more often than absolute amounts in CM campaigns.

Previous research has demonstrated a positive and concave relationship between the amount donated and willingness to pay for the product (WTP): the higher the donation level, the smaller effect the additional cent of donation had on WTP (Koschate-Fischer et al., 2012). Additionally, the perception of the average firm has also been improved when it donated 100% of profits (Dean, 2004). However, companies tend to donate lower amounts, for example 5% or 50%. Considering the findings of Dean (2004), these donation amounts would produce a negative effect on company evaluation, compared to 100%. However, 50% donation can be an important factor to consider.

As it was mentioned, one of the main advantages of CM campaigns is their potential benefit to the company, the cause and consumers. It can be assumed that 50% donation can indicate an equal allocation, where the company receives half and the cause receives the other half of the profit. Previous research on allocation behavior revealed , that when the situation is ambiguous and complicated, such as computational complexity is present, the allocations tend to become more equal (Debusschere and Van Avermaet, 1984).

Additionally, when the goal of the people is to improve morale, they would prefer equal allocation (Stake 1985; Leung and Park 1986). This concept of equal allocation is called equality heuristic (Messick and Schell, 1992). We will apply this concept to the CM practice, arguing, that 50% donation amount can be an optimal allocation of profits between the company and the NPO, which consumers will perceive as fair, and will not attribute significantly negative motives to firm’s initiative to participate in CM campaign. Consumers can perceive the company as less hypocrite, compared to when 5% is donated, as well as they understand that companies have to make profits. This can potentially result in positive company evaluation.

12 results in poor approximation (Olsen, Pracejus & Brown, 2003). Examples o f biased estimations can be incorrect strategies in estimating which of the packages has the lower price-per-unit (Capon & Kugn, 1982). If a brand states that “5% of the profit will be donated”, consumers tend to assume that this percentage is calculated o f the price of the product rather than of the actual profit the company makes after selling it (Olsen, Pracejus & Brown, 2003). This common phenomenon among consumers is referred to as profit-equals-price (PEP) effect. This effect results in upward estimation of the donated amount, which can make consumers think that company is donating more than it actually does. Thus, it is important to study consumer’s perception of company depending on the donation amounts. Moreover, we analyze how these perceptions influence consumer response, namely attitude towards the CM campaign and behavioral intentions.

Consumer attitudes towards products and brands are considered powerful predictors of purchasing behavior (Aurifeille, Clerfeuille and Quester, 2001). They are formed based on consumers’ feelings, thoughts and expectations about the brand and are derived from three classes of information: affective/emotional, cognitive and behavioral. Behavioral component is said to be the best predictor of buying behavior. Therefore, it is important to evaluate consumer’s response to the CM initiative, by means of attitude towards the campaign and the company as well as purchase intention. Additionally, we are going to investigate affective reaction towards the campaign, as emotions can play a role in judgement about the company and decision making process with regards to purchasing the product.

A positive attitude towards the CM campaign and company should go hand in hand with a higher purchase intention. Therefore, we suggest that higher donation amount would result in (a) more positive attitude towards CM initiative, (b) more positive attitude towards the company and (c) higher purchase intentions (hypothesis 1).

13 consumers’ intention to participate in CM campaigns can be explained by the inferences consumers make about the firm motives. As previously indicated, companies benefit from CM campaigns in several ways, such as gaining a better reputation and increasing sales and profits. Therefore, companies are being charitable by participating in CM campaigns, at the same time satisfying their personal profit-driven interest. Such presence of self-interested motivation in charitable activity was labeled tainted altruism (Newman and Cain, 2013). Due to the effect of tainted altruism, company’s actions can be perceived as moral hypocrisy. The perceived moral hypocrisy in CM campaigns can negatively influence consumers’ willingness to purchase the product (Newman and Cain, 2013). Using these findings, we can hypothesize that higher donation amount would reduce perception of moral hypocrisy

(hypothesis 2). Additionally, in the presence of perceived moral hypocrisy, consumers are

likely to have more (a) negative attitude towards CM initiative, (b) negative attitude towards the company and (c) lower purchase intentions (hypothesis 3).

The effect of tainted altruism was also found to play a role in moral evaluations of companies participating in CM campaigns (Newman and Cain, 2013). In general, customers tend to evaluate companies based on the dimensions of morality and competence (Aaker et al., 2010). For-profit firms are generally perceived as more competent rather than moral. While being perceived as competent is translated into higher purchase intention, positioning the company as being moral can make consumers steer away from the company. If the altruistic motivation of the company is perceived as hypocritical, we argue that consumers can experience negative emotions. Particularly, moral emotions can play a role in forming corporate image of the company (Grappi, Romani & Bogozzi, 2013). Within CM, moral emotions have been found to significantly influence purchase intention for a social-cause product (Kim and Jonson, 2013). Before stating the hypotheses, more extensive explanation of moral emotions is presented.

14 concerning whether behaviors are good or bad (Haidt 2001, 2007; Monin et al. 2007; Shweder & Haidt 1993). Within the context of company-consumer relationships, previous research has demonstrated that feelings of moral anger can motivate consumers to boycott the company (Braunsberger and Buckler 2011; Friedman 1999), generate negative attitudes toward the company (Grappi et al. 2013) and generate negative word-of-mouth (Grappi et al. 2013; Lindenmeier et al. 2012). Within the research field of moral emotions, several categorizations of this construct are proposed. It is relevant to investigate different categorizations as it will give insight about how this concept is being measured. According to Haidt (2003), moral emotions can be categorized in 4 types: other -condemning emotions, self-conscious emotions, other-suffering emotions, and other-praising emotions. Table 2 provides and overview of moral emotions categorization proposed by Haidt and their explanation.

TABLE 2.

MORAL EMOTIONS CATEGORIZATION BY HAIDT

Category Explanation Example

Other-Condemning ‘‘the negative feelings about the actions or anger, contempt,

emotions character of others’’ and disgust

Self-conscious emotions self-evaluation or self-reflection guilt, shame,

Other-suffering experience of distress by others embarrassment, pride

emotions

motivate individuals to be engaged in

empathy Other-praising admirable and respectful deeds in their

emotions relationships with others elevation, gratitude

Note that these findings refer to interpersonal relationships, while CM campaigns address company-consumer relationships. We will use insights from the interpersonal relationships to develop our hypotheses.

15 the act of defending norms. This emotion is particularly relevant as it plays a role of a mediator between the perception of company’s misbehavior and consumers’ decision to punish the company (Lindenmeier et al. 2012). Combined, this literature suggests that higher donation amount can result in less negative moral emotions (hypothesis 4). Additionally, if consumers perceive the actions of the company as self-interested, they will experience negative moral emotions (e.g. moral outrage), which will negatively reflect on (a) attitude towards CM initiative, (b) attitude towards the company and (c) purchase intentions (hypothesis 5).

2.4. T

16 An important element of such cause-oriented campaigns is emotional appeal. In general, emotional appeals are widely used to arouse persuasive communication (Dickinson & Holmes 2008). One most common tactic to evoke emotions is through displaying the victim’s face in CM campaigns. Previous research has demonstrated that a sad face compared to happy or a neutral face of victims facilitates empathy and giving in traditional charitable giving contexts (Small & Verrochi 2009, Burt & Strongman, 2005). Emotional contagion theory suggests that people are able to “catch” others’ emotions (Neumann & Strack, 2000). Demonstrating a sad face will make the observer feel sad, which results in greater empathy and increased wingless to help. Therefore, emotional contagion re sulting from the sad face valence can translate into willingness to donate in traditional charities. Additionally, sad emotional expression on the advertisement can evoke the sense of guilt (Chang, 2011). In general, guilt appeals are a popular social influence mechanism in diverse social marketing contexts (Basil et la., 2008). Demonstrating sad face can prompt guilt-ridden emotions, which can foster giving within the CM. An increased sense of guilt can be

mitigated with a contribution to a charitable cause (Strahilevitz and Myers, 1998).

Demonstrating sad face is effective in traditional charities. Companies can well adopt this technique, as CM campaigns are aimed at raising funds for charities. However, next to it, CM campaigns are profit-driven and are aimed at increasing sales. Thus, due to the dual nature of CM activity, it is relevant to investigate what effect different face valence can have within CM practice.

17 implementing happy face in CM campaigns can have positive effect on consumer responses, namely: (a) more positive attitude towards CM initiative, (b) more positive attitude towards the company and (c) higher purchase intentions (hypothesis 6).

Due to that CM campaigns are a combination of charitable activity and profit-oriented activity, and the possible risk of being perceived as self-interested, companies should be careful in adopting either happy or sad face. When using a sad face, consumers might perceive company as exploitative, as it is trying to abuse sad children and their difficult situation in order to make profits and at the same time manipulating consumers by evoking strong emotions. On the other hand, when using a happy face, it can be perceived as happy children do not need help, but consumers might catch positive emotions and, subsequently, develop positive attitude towards the campaign and the company. Based on this rationale, we can hypothesize that the face valence of the victim demonstrated on the ad can have an effect on the perceived moral hypocrisy (hypothesis 7). Additionally, there is an effect of face valence on moral emotions (hypothesis 8). Overall, we propose that positive face valence will have a more positive effect in both cases. We also propose that positive face valence can moderate the relationship between the donation amount and (a) attitude towards the campaign, (b) attitude towards the company and (c) purchase intentions, (d) perceived moral hypocrisy and (e) moral emotions (hypothesis 9).

2.5. Moderated mediation & Conceptual Model

As a final objective of our study, we are going to investigate the possible moderated -mediation, where the strength of the indirect mediation effect depends on the level of the moderator. Namely, we expect the following: donation amount affects consumer attitude toward the campaign through indirect effects, such that the effect of donation amount on consumer attitude towards the campaign will be moderated by face valence and mediated by perceived moral hypocrisy (hypothesis 10a) and experienced moral emotions (hypothesis

11a). Moreover, we expect that donation amount affects purchasing intentions through

18 moderated by face valence and mediated by perceived moral hypocrisy (hypothesis 10b) and experienced moral emotions (hypothesis 11b). Finally, we expect that donation amount affects purchase intentions through indirect effects, such that the effect of donation amount on purchase intentions will be moderated by the face valence and mediated by perceived moral hypocrisy (hypothesis 10c) and experienced moral emotions (hypothesis 11c).

2.6 O

This study focuses on three objectives, which will allow to complete literature gaps within the field of cause-related marketing. First, we are investigating whether higher donation amount would result in a more positive consumer response, which includes attitude towards the campaign, attitude towards the company and purchase intentions. Second, we are focusing on the possible roles of perceived moral hypocrisy and mor al emotions. More precisely, to what extent the consumers would perceive the company as being hypocrite and would experience negative moral emotions, knowing how much the company would donate? And how this would be translated into the consumer response? Third, we are investigating the role of face valence. Particularly, how implementing either happy or sad face will be reflected on consumers’ reaction about the campaign and the company? In order to complete our study, we are going to investigate the interplay between all the above mentioned constructs: how the interaction between the donated amount and face valence can affect consumer response, when taking into account perceived moral hypocrisy and moral emotions as mediators. Figure 3. demonstrates our conc eptual model.

Donation Consumer

Response

Perceived Moral

Face Valence Hypocrisy

Moral Emotions

19

3. Methodology

3.1. Design and participants

The study employed a 3 (level of donation: 5%, 50%, 100%) x 2 (face valence: happy vs sad) between subject design. A total of 438 subjects participated in the study. Forty -eight participants did not pass manipulation check (see Results for details) and wer e eliminated from the analysis. The final sample consisted of 390 subjects (208 males and 182 females, mean age M = 36.78, SD = 11.32), which were included in the analyses. The study was placed on Mechanical Turk (MTurk) using USA citizens only, and it too k 4 minutes to complete. Participants were paid $0.60 to complete the survey.

3.2. Procedure

20

3.3. Measures

Five different variables were investigated in order to test the influence of donation amount and face valence on (1) consumer attitude toward the campaign, (2) consumer attitude towards the company, (3) purchase intentions, (4) perceived moral hypocrisy a nd (5) moral emotions. Additionally, several control variables were included: age and gender, as they can possibly affect the perception of moral hypocrisy; cause importance and consumer skepticism to CM in general, as these variables can potentially confound with our dependent variables. All variables were tested on validity and reliability. To check the construct validity, factor analysis was performed. Factors were included based on the content interpretation, the KMO (>.6), the eigenvalue (>1.0) and the scree plot. Next to factor analysis, reliability was tested by the Cronbach’s alpha (α).

3.3.1. Consumer attitude toward the campaign

Consumer attitude toward the campaign was measured on a three-item seven-point scale with general bipolar adjectives proposed by Nan and Heo (2007): negative (1) – positive (7), unfavorable (1) – favorable (7), dislike (1) – like (7). In order to sum up these items into one new variable, which would measure the construct, two tests were performed. Factor analysis yielded a KMO value of .79, which was significant p<0.001. Additionally, the reliability analysis was performed. Cronbach’s alpha showed a value of .98. These results allowed to create new composite variable to measure consumer attitude towards the campaign.

3.3.2. Consumer attitude towards the company

21 company”. We did not ask one statement from the original 8-item scale as it did not reflect consumer attitude towards the company, but rather employee attitude. Factor analysis showed a KMO value of .92, which was significant p<0.001. Cronbach’s alpha showed a value of .95. Thus, we could create one composed variable to measure corporate reputation.

Because the two newly composed scales measure the same concept and are highly correlated (r = .83, p < .001), we were able to create a new composite variable to measure consumer attitude towards the company. Also, both scales measure the same construct, but in a different way, which suggests that they can complement each other and provide a more comprehensive measure of consumer attitude towards the company. Scores ranged from 1 to 7, with higher scores indicating a more positive attitude toward the company. Factor analysis of the new constructed measure yielded a KMO value of .96, which is significant ( p < .001), and Cronbach’s alpha displayed a value of .95. These results strongly indicate that we had a good measure of the consumer attitude towards the company.

3.3.3. Purchase intentions

Purchase intentions were measured with 3 statements, adopted from Wang, Minor and Wei (2011) and were rated on a seven-point likert scale, ranging from (1) very low to (7) very high, e.g. “After seeing this advertisement, I intend to purchase this product”.

Factor analysis yielded a KMO value of .78, which was significant p<0.001. Cronbach’s alpha showed a value of .98. These results allowed to create a new composite variable to measure purchase intentions.

3.3.4. Perceived moral hypocrisy

22 “HeyChoco does not have a genuine concern for refugee kids”. Factor analysis yielded a KMO value of .80, which was significant p<0.001. Cronbach’s alpha showed a value of .83. Because the two newly composed scales measured the same concept and are highly correlated (r = .84, p < .001), we were able to create a new composite variable to measure consumer attitude towards the company. Also, combining two scales allows to measure different aspects of the construct. Scores ranged from 1 to 7 with higher scores indicating higher perception of moral hypocrisy. Factor analysis of the new constructed measure yielded a KMO value of .95 which is significant (p < .001), and Cronbach’s alpha displayed a value of .96.

3.3.5. Moral emotions

The measure of moral emotions was adopted from several categorizations of moral emotions proposed by Haidt (2003) and Tetlock (2000). The list of items measuring moral emotions included: anger, empathy (rev.), guilt, elevation (rev.), outrage. Higher score on moral emotions implied more negative emotions. Respondents were asked to evaluate to which extent they were experiencing these emotions. First, factor analysis yielded a KMO value of .64, which was significant (p<0.001). However, when reliability analysis was performed, it was shown that removing item “empathy” would improve the value of Cronbach’s alpha from .52, which is considered to be low, to .62. According to the general rule of thumb, scales with Cronbach’s alpha lower than 0.7 are considered to have low reliability. Deleting item “empathy” resulted in increased reliability, which was still low. However, we decided to include the scale in our analyses. We will consider this result in our discussion.

3.4. Control variables

23 3.4.1. Cause importance

Cause importance was measured with three items, anchored by a seven-point scale with bipolar adjectives proposed by Lafferty and Edmondson (2009): unimportant (1) - very important (7), means nothing to me (1) - means a lot to me (7), very insignificant (1) - very significant (7). Factor analysis yielded a KMO value of .62, which was significant (p<0.001). Additionally, the reliability analysis was performed. Cronbach’s alpha showed a value of .83. Items were summed up to one composite variable to measure cause importance.

3.4.2. Consumer skepticism to CM in general

Consumer skepticism to CM in general, was measured using four questions that were adapted from Mohr et al. (1998), e.g. “Most statements made by a company in advertising or package labels about supporting a cause are true”. The items were measured using a seven -point Likert scale, ranging from “strongly disagree” to “strongly agree”. Factor analysis yielded a KMO value of .78, which was significant (p < 0.001). Additionally, the reliability analysis was performed. Cronbach’s alpha showed a value of .85. Items were summed up to one composite variable to measure cause importance.

3.5. M

In the end of the questionnaire, respondents were asked some questions to see whether they perceived the advertisement correctly. They were asked which level of donation was mentioned on the advertisement presented in the beginning of the questionnaire: 5%, 50% or 100%.

24

4. R

4.1. M

A chi-square test was performed in order to evaluate to what extent participants paid attention to the manipulation. Results indicated a strong association between the actual donation amount presented (5%, 50% of 100%) and what participants remembered, χ2 (4)

= 616,304, p < .001. However, some respondents did not remember the manipulation correctly. When the presented donation was 5%, a total of 9 participants responded incorrectly, with 50% donation being presented, 12 was wrong, and with 100% donation being presented, 27 participants responded incorrectly. This resulted in N = 48 participants who were excluded for further analyses, leaving 390 respondents. The description of the sample can be found in Table 3.

TABLE 3.

CHARACTERISTICS OF THE SAMPLE

Characteristic Frequencies Percentage %

25 A Two-Way ANOVA was performed in order to check whether face valence, and donation amount, influenced the perceived face valence. Only a significant main effect of face valence was found (F (1, 388) = 811.702, p < .001, η2 = .679). Participants demonstrated

significant difference in their perception of happy (M = 5.24, SD = 1.26) and sad face (M = 1.77, SD = 1.01). As expected, there were insignificant effects of donatio n (F (1, 388) = .602,

p = .436, η2 = .004) and of interaction between donation and face valence (F (1, 388) = .497, p = .504, η2 = .004).

4.2. C

To control for other variables that could potentially influence the effect of the donation amount on consumer response, the effect of four control variables (gender, age, cause importance and consumer skepticism to CM in general) on the main DVs were tested.

4.2.1 G

In order to investigate whether the average attitude towards the campaign of men is different from the average attitude towards the campaign of women, an independent-samples t-test with gender and attitude was conducted. This yielded a significant difference on attitude towards the campaign between men (M = 5.15, SD = 1.57) and women (M = 5.58,

SD = 1.68); t (388) = -2.59, p = .010. These results suggest that gender has a significant effect

on attitude towards the campaign. The same result held for other dependent and mediating variables. Men (M = 4.96, SD = 1.29) significantly differed from women (M = 5.44,

SD = 1.21) in their attitude towards the company, t (388) = -3.62, p = .000; men (M = 3.78, SD = 1.98) significantly differed from women (M = 4.35, SD = 1.95) in their purchase

26

4.2.2 A

In order to investigate whether the average attitude towards the campaign is influenced by the age of respondents, simple linear regression was performed. The results indicate no relationship, F (1, 388) = .53, p = .466, β= -.04, R2 = .001. The same results held for other dependent and mediating variables. Age had no influence on attitude toward the company, F (1, 388) = .04, p = .844, β = -.01, R2 = .000; age had no influence on purchase intention, F (1, 388) = .69, p = .408, β = -.04, R2 =.002. Age had no influence on perceived moral hypocrisy, F (1, 388) = .63, p = .429, β = .04, R2 = .002. There was a negative significant relationship between age of the respondents and their reported experienced moral emotions F (1, 388) = 5.86, p = .016, β= -.122, R2 = .015. However, as the effect of age was found to be insignificant in most of the cases, it will not be used as a covariate in subsequent analyses.

4.2.3. C

In order to investigate whether cause importance influences dependent and mediating variables, simple linear regression was performed. The results suggested that all dependent variables were influenced by cause importance. Consumer attitude towards the campaign is positively influenced by cause importance, F (1, 388) = 174.16, p = .000, β = .56, R2 = .310;

consumer attitude towards the company is positively influences by cause importance, F (1, 388) = 206.55, p = .000, β = .59, R2 = .347; purchase intention is positively influences by cause importance, F (1, 388) = 166.08, p = .000, β = .55, R2 = .300. Perceived moral hypocrisy is positively influenced by cause importance, F (1, 388) = 143.126, p = .000, β = -.52, R2 = .269; moral emotions are negatively influenced by cause importance, F (1, 388) = 15.77, p = .000, β =- .20, R2 = .039. As the effect of cause importance was found to be significant, it will be used as a covariate in subsequent analyses.

4.2.4. C

CM

27 importance. Consumer attitude towards the campaign is negatively influenced by skepticism, F (1, 388) = 131.18, p = .000, β =-.50, R2 = .253; consumer attitude towards the company is negatively influenced by skepticism, F (1, 388) = 151.45, p = .000, β =-.53, R2 = .281; purchase intention is positively influenced by skepticism, F (1, 388) = 83.57, p = .000,

β =-.42, R2 = .177. Perceived moral hypocrisy is positively influenced by skepticism, F (1, 388) = 174.922, p = .000, β = .56, R2 = .311; moral emotions are positively influences by skepticism, F (1, 388) = 26.52, p = .000, β = .25, R2 = .064. As the effect of consumer skepticism was found to be significant, it will be used as a covariate in subsequent analyses.

4.3 T

As a first step to test our hypotheses, we investigated the effect of the two IVs, namely donation level and face valence, on (1) consumer attitude towards the campaign, (2) consumer attitude towards the company, (3) purchase intentions, (4) perceived moral hypocrisy and (5) moral emotions. As covariates, we included gender, cause importance and consumer skepticism towards CM in general.

28 4.3.1. DIRECT EFFECT OF DONATIONS AND FACE VALENCE ON DVS

4.3.1.1. CONSUMER ATTITUDE TOWARDS THE CAMPAIGN

In order to investigate the direct effect of donations and face valence, including their interaction, on consumer attitude towards the campaign, a Two-Way ANCOVA was performed. Based on the previous analyses, gender, cause importance and consumer skepticism towards CM were included as covariates. A significant main effect of donations on attitude toward the campaign was found, F (1, 389) =18.850, p < .001, η2 _{= .090. A post}

hoc analysis using Tukey HSD test revealed that respondents who were presented with a 5% level of donation (EMM = 4.90, SE = .10) significantly differed from respondents in the 50% (EMM = 5.34, SE = .10, p < .05) and in the 100% condition (EMM = 5.86, SE = .11, p <.001). The mean score for attitude towards the campaign increased with larger donation amount. There is also a significant difference between the 50% and the 100% conditions ( p < .05). The main effect of face valence was found to be significant F (1, 389) =16.965, p < .001, η2 _{= .043. Scores on attitude towards the campaign in happy face condition (EMM =}

5.62 SE = .09) were significantly higher than scores in a sad face condition (EMM = 5.13 SE = .09) (p < .001). The interaction effect of donation amount and face valence was insignificant (F (1, 389) = 2.394, p = .093, η2 _{= .012). Table 4 provides overview on the estimated}

marginal means for the interaction effect.

Finally, the following covariates were significant: cause importance was significant (F (1, 389) =105.101, p < .001, η2 _{= .216); skepticism towards CM (F (1, 389) = 66.869, p <}

.001, η2 _{= .149). The effect of gender was found to be insignificant (F (1, 389) = 2.796, p}

=.095, η2 _{= .007).}

4.3.1.2. CONSUMER ATTITUDE TOWARDS THE COMPANY

In order to investigate the direct effect of donations and face valence, and their interaction effect, on consumer attitude towards the company, a Two -Way ANCOVA was performed. A significant main effect of donations on attitude toward the company was found, F (1, 389) = 24.071, p < .001, η2 = .112. A post hoc analysis using Tukey HSD test revealed that respondents who were presented with a 5% level of donation (EMM

29 significant difference between the 50% and the 100% conditions (p < 0.05). The main effect of face valence was found to be significant (F (1, 389) = 8.496, p < .05, η2 = .022). People in the happy face condition had more positive attitudes towards the company (EMM = 5.35, SE = .06) than people in the sad face condition (EMM = 5.09 SE = .06) (p < .05). The interaction effect of donation amount and face valence was insignificant (F (1, 389) = .954, p = .386, η2 = .005).

The effects of cause importance was significant (F (1, 389) =122.723, p < .001, η2 = .244) as well as the effect of skepticism towards CM (F (1, 389) = 79.022, p < .001, η2 = .172). The effect of gender was found to be insignificant (F (1, 389) = 2.796, p = .095, η2 = .007).

4.3.1.3. PURCHASE INTENTION

In order to investigate the direct effect of donations and face valence, and their interaction effect, on purchase intention, a Two-Way ANCOVA was performed. A significant main effect of donations on purchase intentions was found, F (1, 389) = 12.569, p < .001, η2

= .062. A post hoc analysis using Tukey HSD test revealed that respondents who were presented with a 5% level of donation (EMM = 3.61, SE = .13) did not significantly differ from respondents in the 50% (EMM = 4.02, SE= .12, p = .092), but significantly differ from the 100% condition (EMM = 4.60, SE= .15, p < .001). There is also a significant difference between the 50% and the 100% conditions (p < 0.05). The mean score for purchase intentions increased with larger donation amount. The main effect of face valence was found to be insignificant (F (1, 389) = 3.144, p = .077, η2 _{= .008). Scores on purchase}

intention in happy face condition (EMM = 4.22, SE = .11) did not significantly differ from scores in a sad face condition (EMM = 3.94 SE = .077) (p = .077). The interaction effect of donation amount and face valence was insignificant (F (1, 389) = .595, p = .552, η2 _{= .005).}

The effect of cause importance was significant (F (1, 389) = 99.849, p < .001, η2 ₌

.208) as well as the effect of skepticism towards CM (F (1, 389) = 30.008, p < .001, η2 ₌

.073). The effect of gender was found to be insignificant (F (1, 389) = .631, p = .428, η2 ₌

30 4.3.1.4. PERCEIVED MORAL HYPOCRISY

In order to investigate the direct effect of donations and face valence, and their interaction effect, on perceived moral hypocrisy, a Two-Way ANCOVA was performed. A significant main effect of donations on perceived moral hypocrisy was found, F (1, 389) = 33.885, p < .001, η2 _{= .151. A post hoc analysis using Tukey HSD test revealed that}

respondents who were presented with a 5% level of donation (EMM = 3.92, SE = .07) did not significantly differ from respondents in the 50% (EMM = 3.50, SE= .08, p = .092), but significantly differed from the 100% condition (EMM = 3.02, SE= .08, p < .001). There is also a significant difference between the 50% and the 100% conditions (p < 0.05). The mean score for perceived moral hypocrisy decreased with larger donation amount. The main effect of face valence was found to be insignificant (F (1, 389) = 3.287, p = .071, η2 _{= .009).}

Scores on perceived moral hypocrisy in happy face condition (EMM = 4.22, SE = .11) did not significantly differ from scores in a sad face condition (EMM = 3.94 SE = .077) (p = .077). The interaction effect of donation amount and face valence was significant (F (1, 389) = 3.461, p < .05, η2 _{= .018).}

The effect of cause importance was significant (F (1, 389) = 74.069, p < .001, η2 = .163) as well as the effect of skepticism towards CM (F (1, 389) = 110.984, p < .001, η2 = .226). The effect of gender was found to be insignificant (F (1, 389) = 1.781, p = .183, η2 = .005).

4.3.1.5. MORAL EMOTIONS

31 found to be significant (F (1, 389) = 43.513, p < .001, η2 = .103). Scores on experienced negative moral emotions in happy face condition (EMM = 2.27, SE = .06) did not significantly differ from scores in a sad face condition (EMM = 2.82 SE = .06) (p = .077). The interaction effect of donation amount and face valence was insignificant (F (1, 389) = .383, p

= .682, η2 = .002).

The effect of cause importance was significant (F (1, 389) = 4.423, p < .05, η2 = .011) as well as the effect of skepticism towards CM (F (1, 389) = 14.199, p < .001, η2 = .036). The effect of gender was found to be insignificant (F (1, 389) = .000, p = .983, η2 = .000).

TABLE 4.

RESULTS ANCOVA: INTERACTIONS OF DONATIONS AND FACE VALENCE

Dependent Donation amount Face valence EEM (SE)

32

50% Happy 4.39 (.24)

Sad 3.99 (.23)

100% Happy 4.89 (.25)

Sad 4.43 (.25)

Perceived moral 5% Happy 3.68 (.10)

hypocrisy Sad 4.16 (.10) 50% Happy 3.51 (.11) Sad 3.49 (.10) 100% Happy 3.01 (.11) Sad 3.03 (.12) 5% Happy 2.37 (.09) Moral emotions Sad 2.99 (.09) 50% Happy 2.26 (.10) Sad 2.72 (.09) 100% Happy 2.18 (.10) Sad 2.74 (.11)

4.3.2. M

33 We included only 2 experimental groups for our main independent variable, because this model cannot be tested with PROCESS with more than two levels of independent variable. We chose to include 5% and 50% conditions, because from the post hoc analyses we learned that the 100% condition most often did not significantly differ from the 50% condition. Also, from a theoretical perspective it is more interesting to look at the 50% condition, as 50% of donation can potentially display the win -win situation for the company and the NPO, due to the equal allocation of the profits. Face valence was set as a moderator. Both independent variables are categorical, which requires to code them as dummies: donation and face valence have two levels, thus, one dummy variable was created for each IV. Perceived moral hypocrisy was set as mediator. We excluded moral emotions from the analyses, because, according to the results of the ANOVAs, no significant effect of donations and interaction effect of donations and face valence was found1_{. Cause importance and}

consumer skepticism were set as covariates. Gender was excluded from the analyses as the effects of this variable were also insignificant.

4.3.2.1. CAMPAIGN ATTITUDE

The first model tested the direct effect of donations on consumer attitude towards the campaign, as well as conditional indirect effect of donations on campaign attitude, which is mediated by perceived moral hypocrisy and moderated by the face valence. Based on the results of PROCESS macro analysis, we found that the direct effect was negative and insignificant (B = -.035, p = .759). The effect of donations on perceived moral hypocrisy was negative and insignificant (B = -.150, p = .340). We found significant positive effect of face valence (B = .491, p < .05) and significant negative interaction effect of donations and face valence on perceive moral hypocrisy (B = - .528, p< .05).

Finally, the conditional indirect effects of donations on campaign attitude, med iated by the perceived moral hypocrisy and moderated by the face valence, are as follows: there is an insignificant effect of perceived moral hypocrisy moderated by a happy face

1

34

(B = .152, 95% CI = [-.14; .47]), and a significant effect moderated by a sad face (B =

.685, 95% CI = [.35; 1.05]). These results suggest that in the condition, where the company donates 50% of the profit, there is a positive effect on the attitude towards the campaign, which is mediated by the perceived moral hypocrisy and moderated by the face valence. However, only the effect of sad face was found to be significant. Also, due to the insignificant effect of donations on perceived moral hypocrisy, we cannot confirm moderated-mediation hypothesis. The final model of the effects on campaign attitude can be found in Diagram 1. The bold line represents significant effects, the dashed line represents insignificant effects.

Donation Campaign

Attitude

Face Valence Moral Hypocrisy

Diagram 1. Moderated-Mediation model. Effects on Campaign Attitude

4.3.2.2. COMPANY ATTITUDE

35

positive effect of face valence (B = .491, p < .05) and significant negative interaction effect of donations and face valence on perceive moral hypocrisy (B = -.528, p < .05).

Finally, the conditional indirect effect of donations on company attitude, mediated by the perceived moral hypocrisy and moderated by the face valence, are as follows: there is an insignificant effect of perceived moral hypocrisy moderated by a happy face (B =

.122, 95% CI = [-.10; .37]), and a significant effect moderated by a sad face (B = .551,

95% CI = [.23; .84]). These results suggest that in the condition, where the company donates 50% of the profit, there is a positive effect on the attitude towards the company, which is mediated by the perceived moral hypocrisy and moderated by the face valence. However, only the effect of sad face was found to be significant. Also, due to the insignificant effect of donations on perceived moral hypocrisy, we cannot confirm moderated-mediation hypothesis. The final model of the effects on company attitude can be found in Diagram 2.

Donation Company

Attitude

Face Valence Moral Hypocrisy

Diagram 2. Moderated-Mediation model. Effects on Company Attitude

4.3.2.3. PURCHASE INTENTION

36 perceived moral hypocrisy and moderated by the face valence. Based on the results of PROCESS macro analysis, we found that the direct effect of donations was negative and insignificant (B = -.150, p = .340). We found significant positive effect of face valence (B

= .491, p < .05) and significant negative interaction effect of donations and face valence on perceive moral hypocrisy (B = -.528, p < .05).

Finally, the conditional indirect effect of donations on campaign attitude, mediated by the perceived moral hypocrisy and moderated by the face valence, are as follows: there is an insignificant effect of perceived moral hypocrisy moderated by a happy face (B =

.140, 95% CI = [-.14 .42]), and a significant effect moderated by a sad face (B = .633,

95% CI = [.32; .96]). These results suggest that in the condition, where the company donates 50% of the profit, there is a positive effect on the purchase intentions, which is mediated by the perceived moral hypocrisy and moderated by the face v alence. However, only the effect of sad face was found to be significant. Also, due to the insignificant effect of donations on perceived moral hypocrisy, we cannot confirm moderated-mediation hypothesis. The final model of the effects on purchase intention can be found in Diagram 3.

Donation Purchase

Intention

Face Valence Moral Hypocrisy

37

5. D

Companies frequently engage in cause-related marketing activities for a variety of reasons. It is a common belief that these activities allow companies to boost their image and increase profits. Additionally, CM campaigns can benefit consumers as well, as by purchasing charity-labelled products consumers can see themselves as being good moral citizens. However, CM practices are not left without criticism. Previous research has demonstrated that companies risk to be perceived as hypocrite when collaborating on a charitable cause, which can have negative consequences for the company (Barone, Norman and Miyazaki 2007; Newman and Cain, 2014). This research attempted to investigate how companies can better develop their advertisement in order to prevent this from happenin g. Particularly, we examined how consumers react to different donation amounts in CM campaigns and how it is reflected on the attitude towards the campaign, attitude towards the company and purchase intentions. We also investigated how these effects can be explained by perceived moral hypocrisy and experienced moral emotions. Moreover, the face valence used in the CM campaign was proposed to moderate these effects.

In terms of the first objective, we found that higher donation improved the attitude towards the campaign, attitude towards the company and purchase intention. This is in line with previous research, which suggests that larger donations produce more positive outcomes for the company (Dean, 2004; Koschate-Fischer et al., 2012).

In terms of our second objective, we found that, when the company donated higher amount, the perception of hypocrisy was decreasing, but the experienced moral emotions were not affected. While the effect on perceived hypocrisy is in line with previous research (Newman and Cain, 2013) and our key assumptions, the insignificant effect on moral emotions raises certain concerns. Moral emotions are an important construct within the domain of CM marketing (Kim and Jonson, 2013). The absence of the effects calls for more discussion about this construct. These findings suggested that the effect of donation amount on consumer attitudes towards the campaign and company, as well as purchase intentions, was mediated only by perceived moral hypocrisy.

38 intentions and perceived moral hypocrisy were not significant, overall, we can conclude that positive face valence is more effective in improving consumer response. This is in line with the pilot study by Kang and Leliveld (2016), where the happy face facilitated more positive perception of the campaign and the company itself.

We also investigated whether incorporating face valence can affect the relationship between donations, perceived moral hypocrisy and consumer response. We found no significant evidence that the effect of donation amount on consumer response can be mediated by perceived moral hypocrisy and moderated by the face valence. The results suggested that attitude towards the campaign, attitude towards the company and purchase intentions are indirectly influenced by the level of donation through the perceived moral hypocrisy. However, the effect of face valence was only partly significant. Happy face did not have significant effect on the indirect relationship between donations and consumer responses. Also, the effect of the sad face was larger, than the one of the happy face. In other words, participants who were presented with a 50% donation amount and a sad face, reported more positive attitude towards the campaign, more positive attitude towards the company and higher purchase intention, compared to those who were in the 5% condition. This is in line with the main hypothesis, that higher donation amount would improve consumer response. However, the significant effect of sad face requires more discussion.

39 If we included 100% instead of 50%, we could possibly acquire different results2. Yet, we were particularly interested in the 50% condition, as it could potentially provide new insights about how the company is perceived, when the profits are equally allocated between the company and the charity.

Another explanation can be referred to the nature of the equal allocation. As participants noticed that the company is donating 50% to the charity, it could be interpreted it as a fair allocation due to the equality heuristic: both parties were receiving equal share of profits. In this case, respondents might not have paid deliberate attention to the facial expression of the victim. Moreover, the effect of facial expression and emotional contagion is an automatic process (Small and Verrochi, 2009). If consumers engaged in deliberative thoughts about the campaign and the donation amount, the effect of the facial expression could have been diminished.

Overall, our results support previous findings that higher donation amount would result in a more positive attitude and consumer response towards the company. However, this response is dependent on the consumers’ perception of the firm’s motive. If the motivation of the firm is perceived as exploitative, which is the case with lower donation amounts, consumers develop negative attitude. However, by incorporating a positive face valence, companies can improve the way consumers perceive them, and contribute from a positive consumer response. The final conclusion is that more positive attitude towards the campaign, more positive attitudes towards the company and increased purchased intentions develop from a genuine motivation of the company to act altruistically.

40

5.1 M

This research provided support to the notion that companies, engaging in cause -related marketing activities, risk being perceived as hypocrite and insincere in their actions, which can harm their image as well as decrease consumer response. In ord er to avoid it, managers should be cautious about what elements they include in their campaigns and what effect these variables may produce.

First, companies should try to donate as much as possible. Our research has clearly demonstrated that higher level of donation positively affects attitude towards the campaign, attitude towards the company and purchase intention. Moreover, higher donations decreased consumer’s perception of the company being a hypocrite. Solely following the common idea that participating in charitable giving will benefit the company should be avoided. According to our findings, 5% donation was low, which negatively reflected on the consumer response, compared to 50% and 100%.

Second, if companies are willing to adopt a face of a victim in their campaign, they should regard it with caution. This research demonstrated that happy face can potentially increase consumer response and improve perception of the firm. However, the final model demonstrated a larger and more significant effect o f the sad face. This finding requires more investigation, as this effect was found in the 50% donation condition only.

Overall, we suggest that companies should engage in CM practices if they are ready to donate a larger amount. This will create a more altruistic image of the company, rather than exploitative. With regards to using faces in campaigns, companies should consider using a happy face, as it will improve consumer response towards the company. However, using sad face is also a possibility, but it has to be regarded with caution, as it might not be applicable to all cases.

5. L

41 emotions, which turned out to be an insignificant construct in our analysis, despite its importance in CM marketing. In general, reporting emotions via self -reports can be a difficult task for participants. Future research could focus on using more advanc ed methods of emotional recognition, such as fMRI. This would provide more accurate information on what respondents are experiencing while seeing the advertisement.

Second, our sample consistent of mainly US participants. This makes the sample homogeneous, which benefits the analysis. However, conducting research in a different culture could potentially yield different results, especially, if we take moral emotions into account. Previous research has demonstrated significant differences between American an d Korean participants on their reported moral emotions (Kim and Johnson, 2012). More specifically, within the American sample, ego-focused emotions significantly influenced purchase intentions, while in the Korean sample, other-focused emotions played a more significant role. As an explanation, they proposed that experience of moral emotions can vary across cultures and can be linked to the self-construal. In our research, we focused more on other-focused emotions, while we had participants exclusively from the USA. This inconsistency could potentially lead to the insignificant effect of moral emotions, as the scale was not entirely suitable for the sample. This was also confirmed by the validity and reliability analysis of the scale measuring moral emotions.

Third, our research can be limited due to the possible low product-cause fit, which is consumer’s perceived link between the firm’s product and the affiliated cause (Das et al., 2014). According to these authors, a good example of such fit can be a book store donating to literacy fund. In our case, we have a chocolate producer donating to refugee children. Although incorporating hedonic product is in line with suggestions from previous research (Strahilevitz and Myers, 1998, Dean, 2004), this might not be the best fit for the cause of helping refugee children. Future research could potentially focus on investigating whether there is a good product-cause fit, before exploring the effects of the interplay between the donation amount and face valence, and consumer response.

42 Considering other variables, such as altruistic motives of consumers or warm-glow motives, can produce different results. By participating in charitable giving, individuals can be driven by their willingness to be perceived as good moral citizens, rather than their ultimate support of the cause. Within the context of charity auctions, bidders were distinguished as being either warm-glow (altruistic) bidders or selfish bidders (Haruvy and Popkowski Leszczyc, 2009). The different results of the auction were driven by the discrepancy in the motives of the bidders. It can be well assumed that a similar pattern can occur in CM context.

Fifth, in this research we used a fictitious brand and product. This was done to avoid any existing judgments and opinions about the company. However, factors such as brand familiarity or brand image can have a significant impact on attitude formation (Lafferty and Edmondson, 2009). Some consumers can be familiar with the brand and be loyal to it, while others could have developed negative opinion about the same brand or company. In the latter case, the established negative opinion can affect the perception of company’s motives in a negative way. Controlling for these predispositions can reveal different effects on consumer attitudes towards the company.

Sixth, future research can continue investigating the nature of equal allocation of profit between the company and the charity. We hypothesized that 50% donation will be perceived by the consumers as fair, based on the equality heuristic (Messick and Schell, 1992) and potential willingness of consumers to improve morale (Debusschere and van

Avermaett 1984). However, this assumption was not tested. We also found different pattern

of results in this condition, namely, the effect of face valence. These limitations could potentially affect our results and require more thorough investigation in the future research.

43

R

Aaker, J., Vohs, K. D., & Mogilner, C. (2010). Nonprofits are seen as warm and for -profits as competent: Firm stereotypes matter. Journal of Consumer Research, 37(2), 224-237.

Antonetti, P., & Maklan, S. (2014). Feelings that make a difference: How guilt and pride convince consumers of the effectiveness of sustainable consumption choices. Journal

of Business Ethics, 124(1), 117-134.

Aurifeille, J. M., Clerfeuille, F., & Quester, P. (2001). Consumers’ attitudinal profiles: an examination at the congruence between cognitive, affective and conative spaces.

NA-Advances in Consumer Research Volume 28.

Barone, M. J., Miyazaki, A. D., & Taylor, K. A. (2000). The influence of cause-related

marketing on consumer choice: does one good turn deserve another? Journal of the

academy of marketing Science, 28(2), 248-262.

Barone, M. J., Norman, A. T., & Miyazaki, A. D. (2007). Consumer response to retailer use of cause-related marketing: Is more fit better? Journal of Retailing, 83(4), 437-445.

Basil, D. Z., Ridgway, N. M., & Basil, M. D. (2008). Guilt and giving: A process model of empathy and efficacy. Psychology & Marketing, 25(1), 1-23.

Becker-Olsen, K. L., Cudmore, B. A., & Hill, R. P. (2006). The impact of perceived corporate social responsibility on consumer behavior. Journal of business research, 59(1), 46-53.

Braunsberger, K., & Buckler, B. (2011). What motivates consumers to participate in boycotts: Lessons from the ongoing Canadian seafood boycott. Journal of Business