Secret Santa:
Does low self-esteem lead to covering up altruistic behavior to avoid the
anticipated threat of do-gooder derogation?
Master Thesis, MSc Marketing Management, University of Groningen, Faculty of Economics and Business
June 20th, 2016 Méral N. Steffens S1869817
Steenhouwerskade 2a1, 9718 DA Groningen Tel.: +31 (0)645380097
E-mail: m.n.steffens@student.rug.nl
Total word count: 8654 Supervisor: dr J.W. Bolderdijk
PREFACE
This thesis was written as part of the MSc Marketing of the University of Groningen in the period between February 2016 and June 2016. The process of writing this thesis got me fascinated with the sometimes mysterious behavior of the human being, consumers specifically. Despite some difficulties, I absolutely enjoyed the process of writing this thesis. Therefore, I would like to take this opportunity to thank some people who have made it possible for me to write this thesis to the fullest of my capabilities.
To my friends and family, especially my parents and little brother, thank you for helping me in every way possible. Your support put me in the position to fully employ my abilities, both during the process of writing this thesis as well as during the rest of my career as a student in the years before. A special thanks goes out to my supervisor, dr J.W. Bolderdijk, thank you for your guidance during this process. Your unlimited enthusiasm for the subject genuinely motivated me to go the extra mile. Also, dr W. Jager, thank you for reading and assessing my thesis. Furthermore, my fellow students, in particular Tobias Meyer and Julia Storch, thank you for your patience with me and your feedback.
The University of Groningen, has provided me with the means to pursue a career in marketing. This master, and especially courses like consumer psychology, got me more excited than any subject matter before. Since this thesis is also the same time the final project of my student career, I would like to thank the University of Groningen which has enabled me to aim for a successful career in marketing and gave me the opportunity to start my professional career with full confidence. I have found something that truly gets me excited about the future and for that I am grateful.
ABSTRACT
While one would expect that do-gooders, consumers who purchase green products out of altruistic reasons, would be appreciated by other consumers, recent research reveals that this is not always the case. Moral do-gooders can pose a threat to other people’s self-concept which results in the derogation of the do-gooders to minimalize the negative consequences for other people’s self-concept. I argue do-gooders anticipate this threat of derogation and want to avoid it. Thus, my prediction is that they will choose to display less altruistic behavior in public situations than private situations. Furthermore, the level of self-esteem is expected to have an influence here, as low self-esteem individuals are more likely to experience social anxiety, and thus anticipate more do-gooder derogation. A lab study is set up to analyse these expectations. While the means are in the expected direction, the model appears not to be significant except for the direct influence of self-esteem on anticipated threat of do-gooder derogation - low self-esteem individuals anticipate more threat of do-gooder derogation. Also, while the influence of anticipated threat of do-gooder derogation on the level of altruistic behavior has not been proven, the level of self-esteem does, unexpectedly, influence this relationship: low self-esteem individuals seem more bothered by the threat of do-gooder derogation.
TABLE OF CONTENT PREFACE ... 2 ABSTRACT ... 3 INTRODUCTION ... 5 DO-GOODER DEROGATION ... 6 PUBLIC VS PRIVATE ... 7 SELF-ESTEEM ... 8 CONCEPTUAL MODEL ... 10 METHODOLOGY... 11 PROCEDURE ... 12
MEASUREMENT AND MANIPULATION OF THE VARIABLES ... 13
Measurement of dependent variable: altruistic behavior ... 13
Manipulation of independent variable: situation ... 13
Measurement of moderator: level of self-esteem ... 14
Measurement of mediator: anticipated threat of do-gooder derogation ... 15
Manipulation and attention check ... 15
RESULTS ... 17
PARTICIPANTS ... 17
MEAN COMPARISON ... 17
Anticipated threat of do-gooder derogation. ... 17
Level of altruistic behavior. ... 18
MODERATED MEDIATION ANALYSIS ... 19
ADDITIONAL MODERATED MEDIATION ANALYSIS ... 20
DISCUSSION ... 23
Threat though no behavior change. ... 23
Self-esteem as incentive to care about threat. ... 24
CONCLUSION... 24
MANAGERIAL IMPLICATIONS ... 25
LIMITATIONS AND SUGGESTIONS FOR FURTHER RESEARCH ... 26
RESOURCES ... 27
APPENDIXES ... 31
APPENDIX A ... 31
APPENDIX B ... 33
INTRODUCTION
‘What is the use of living, if it be not to strive for noble causes and to make this muddled world a better place for those who will live in it after we are gone?’
Winston Churchill, American novelist, 1871- 1947
As the negative environmental consequences of our consumption behavior became more clearly visible worldwide over the last decades, consuming sustainable becomes more and more a ‘hot topic’. However, there appears to be a gap between consumer attitudes and consumer behavior when it comes to sustainable purchasing, in other words, the concerns of consumers about the environment do not translate into increased sustainable consumption (Hughner, McDonagh, Prothero, Shultz, & Stanton, 2007). The opposite is even true; consumption levels are still growing (Witt, 2011).
While there has been a significant amount of research on how to reduce this gap between attitude and behavior, there seem to be many factors influencing this behavior. The factor that this analysis will focus on is the influence of do-gooder derogation. While at first instance one would expect that people who buy sustainable consumption goods, do-gooders, would be appreciated by others, recent research reveals another facet of doing the right thing; do-gooder derogation. Doing the right thing can make a person a moral outlier when doing the right thing is not what the mainstream consumer does and people are said to dislike outliers (Cramwinckel, Van Dijk, Scheepers, & Van den Bos, 2013). Do-gooders who buy exceptionally green products, products that go beyond the norm and are not socially accepted, are most likely to experience derogation since this makes the do-gooder an outlier. So, rather than being appreciated, those who do good are being sanctioned because they are different. The disliking results in derogation of these outliers because fellow consumers who do not buy green products feel their self-concept is threatened since these outliers basically make the ‘better’ choice by purchasing sustainable.
of self-esteem of Leary, Tambor, Terdal & Downs (1995), it is a person’s self-esteem that monitors the possibility of social exclusion that reduces the chance of belonging to a group. Group belonging is a basic human need, which increases self-esteem and is proven to be healthy (Jetten, et al., 2015). So, being liked is a motivation for behavior. However, altruistic behavior can threaten group belonging if this results in the do-gooder being disliked by other people. As a result, wanting to be liked can be a motivation to decrease altruistic behavior. As altruistic behavior decreases, the adoption of sustainable goods becomes more difficult.
DO-GOODER DEROGATION
As explained, the risk of social exclusion results in decreased altruistic behavior and the adoption of sustainable goods. Then, the question remains why people dislike others who do good. Moral behavior is a central element in the pursuit of a positive self-image. How moral people perceive themself to be is a crucial element in defining their self-image (Dunning, 2007). Furthermore, according to social comparison theory, people prefer to compare themself with inferior persons since this makes them feel good about themselves; the evaluation of the self is based on social comparison (Festinger, 1954). Moreover, a downward social comparison can increase a person’s self-image. Comparing moral behavior can decrease a person’s self-image when another person was superior in this aspect. As found by Tasimi, Dominquez, & Wynn (2015), even children showed reduced preference for generous others in case of a social comparison. This has also been researched by Cramwinckel, Van Dijk, Scheepers, & Van den Bos (2013) who came to the conclusion that it is the “self-threatening effect” (p. 1049) that people experience in the presence of superior moral beings; moral threats negatively influence their moral self-concept in case of social comparison. As a result, do-gooders are disliked because they propose a threat to the self-concept of others.
PUBLIC VS PRIVATE
One of the requirements of do-gooder derogation is the public display of do-gooder behavior. The fact that people behave differently in public situations compared to private situation has been subject to a significant amount of research in the past. The reason for this difference in behavior is obvious; people care about what others think of them and hence change their behavior based on expectations. While behavior in private situations will be consistent with personal attitudes, in public situations behavior will reflect societal expectations (Froming, Walker, & Lopyan, 1982). Since people do not want to be different because of the risk of social exclusion, their behavior will to some degree be adapted to social norms. Relating this to do-gooders, it seems to be their altruistic behavior that does not fit the social norm. Social exclusion may be the result of not fitting the social norm.
Furthermore, do-gooder derogation calls for a public situation to display altruistic behavior. When a consumption choice is visible for other people, the opinion of these other people and their resulting reaction can influence the behavior and decisions of a do-gooder. Out of fear for other people’s reactions, derogation, do-gooders may adapt their behavior to social expectations. Hence, in public situations, may strategically try to avoid it by displaying undeviating behavior. For instance, they may shy away from purchasing green, ethically conscious, products that they personally endorse but consider uncommon. However, when the behavior of a do-gooder is not visible for others, this person will not take into account other people’s opinions and its true altruistic attitude may surface. Therefore, the expectation here is that consumers will be less likely to anticipate the threat of do-gooder derogation in private situations compared to public situations.
Undercover altruism is a reaction to the anticipated resentment of moral rebels (Minson & Monin, 2012). In this analysis, it can be seen when consumers display a lower level of altruistic behavior in public situations compared to private situations. Those customers choose to ‘cover up’ their altruistic behavior to avoid the anticipated threat of do-gooder derogation in public situations. In private situations this threat does not exist and therefore the level of altruistic behavior will differ.
This leads to the following hypothesis:
H1: Public situations, compared to private situations, lead to a higher level of anticipated threat of do-gooder derogation, which diminishes altruistic behavior.
In previous research, the publicity of purchase behavior has been shown to increase the choice for green products because altruistic behavior is used to publicly display status (Griskevicius, Tybur, & Van den Bergh, 2010). While the results of the research by Griskevicius, Tybur & Van den Bergh (2010) seems contradicting to the concept of ‘undercover altruism’, two explanations are suggested. Firstly, these researchers made use of common green products in their study. Since those products are common and therefore socially accepted, consumers who purchase them are not seen as moral outliers. Therefore, there is no threat of do-gooder derogation and those products can be used to display status. Secondly, it could be the level of self-esteem of an individual that determines whether or not the altruistic behavior is publicly displayed as a status symbol or whether it is covered up to avoid anticipated derogation.
SELF-ESTEEM
self-esteem (Hewitt, et al., 2003). Individuals with high levels of self-self-esteem experience less social anxiety since their self-esteem acts as a defense mechanism. Consequently, individuals who have high levels of social anxiety are more aware of threats of others and dedicate more attention to it (Bar-Haim, Lamy, Pergamin, Bakermans-Kranenburg, & van IJzendoorn, 2007). Hence, it can be expected that individuals with low levels of self-esteem perceive more threats of others than individuals with high levels of self-esteem, including the threat of do-gooder derogation, when their consumption choice is publicly visible. As a result, they will purchase less altruistically.
This leads to the following hypothesis:
CONCEPTUAL MODEL
Taking the hypotheses together, the following conceptual model visually represents the research objective of this analysis:
H2
H1
METHODOLOGY
This study aimed at measuring the level of altruistic behavior in view of anticipated threat of do-gooder derogation and the influence of self-esteem on this relationship in a public or private situation. Therefore, the method employed for this research was an experiment and the participants were randomly allocated to either of the two situations; they chose products in a private (where their choices would not be visible to others) or public (where their choices would be visible to others) setting. Therefore, it is a between-subjects design to avoid carry-over effects (Greenwald, 1976).
The data collection was done in a research lab; the lab of the Economics and Business faculty of the University of Groningen. This study was administered simultaneously with 3 other studies and it was second in the sequence. Because it is the study’s sole objective to uncover the influence of self-esteem and anticipated threat of do-gooder derogation on the level of altruistic behavior, the study took place in a situation in which the participants had the opportunity to display altruistic behavior using exceptional products. Those products are used because exceptional products are expected to have the highest influence on derogation since do-gooders choosing these products are labeled as moral outliers who propose a threat to the self-concept of other people.
PROCEDURE
The study asked for a situation in which individuals got the chance to show altruistic behavior, therefore, this study made use of an imitation online shop. The participant got the chance to choose within a range of three products which differ to which extent they are “green / ethically-conscious products” and was asked to choose as he/she would have done in a real shop. Each individual made several choices in different product categories. These categories consisted of products that are not so familiar to the average customer or household. Uncommon green/ethically-conscious products make the customer a moral outlier because purchasing those products is altruistic as
well as norm deviating. Individuals who wanted to display altruistic behavior would choose the organic products because this gave them the opportunity to signal that they are willing to incur costs, which do not directly translate to personal benefits and hence benefit others. Moral outliers can trigger do-gooder derogation since a person can experience a threat to their self-concept since the outlier made the moral choice while this person did not. Common products are less likely to trigger this reaction because the people purchasing ‘common’ green products are not seen as moral outliers. The products that chosen were trash bags, a baby playsuit and a toddler toy, the other products were filler products and not used during the analysis (see figure 2). Also, the green products were labeled (€€€) while the regular products were labeled (€) to indicate the price difference between the
products. A print out of the survey can be found in appendix C.
MEASUREMENT AND MANIPULATION OF THE VARIABLES
Measurement of dependent variable: altruistic behavior
The dependent variable, the level of altruistic behavior was measured on a ratio scale. The participants rated which product they would choose out of two possible options, one organic and one non-organic. A 7-point Likert scale was used to rate to what extent they would choose each product (see figure 2). For the analysis, the scores on the three uncommon products were averaged; the toddler toy, baby bodysuit, and the trash bags (α = .577, M = 3.13, SD = 1.42).
Manipulation of independent variable: situation
The independent variable, situation, consisted of two levels, private and public, and is hence a categorical, nominal variable. A distinction between the two conditions of this experiment, private and public, was made upfront. This was manipulated by providing the participant with information whether the questionnaire results would be public or private. In case of the private situation, emphasis was placed on the fact that the results would be anonymous and not revealed to anyone (see figure 3).
In the public situation, the participants were told that their product choices would also be judged by other participants who also had the opportunity to make product choices and they were asked to fill out their full name (see figure 4). To make this credible, participants were made aware of this by informing them that after they made their own choices, they would have to form an opinion about the choice of another participant of the experiment. While participants’ opinions about the other person will not be used for further analysis, it was used to cover up the actual intention of the research as well as inform the participant of the publicity of his/her choices.
One of the last questions of the survey will include a manipulation check to verify whether the manipulation succeeded. For the analysis, a dummy variable was created which had two possible values; (1) private and (2) public.
Measurement of moderator: level of self-esteem
The moderator, level of self-esteem, was measured using the self-esteem scale of Rosenberg (1965). This scale was chosen because it is easy to administer because of its limited number of items and it has been validated several times in the past (Robins, Hendin & Trzesniewski, 2001; Martín-Albo, Núñez, Navarro & Grijalvo, 2007). This scale consisted of 10 items rated on a 7-point Likert scale. Items 3, 5, 8, 9, and 10 are reverse scored. All scores were averaged where a higher score indicated a higher level of self-esteem. The items can be seen in table 1 (α = .88, M = 5.17, SD = .95).
Statement
1. I feel that I am a person of worth, at least on an equal plan with others 2. I feel that I have a number of good qualities
3. All in all, I am inclined to feel that I am a failure Reversed 4. I am able to do things as well as most other people
5. I feel I do not have much to be proud of Reversed
6. I take a positive attitude towards myself
7. On the whole, I am satisfied with myself
8. I wish I could have more respect for myself Reversed
9. I certainly feel useless at times Reversed
10. At times, I think I am no good at all Reversed
Table 1. Self-esteem scale (Adapted from Rosenberg, 1965)
Measurement of mediator: anticipated threat of do-gooder derogation
The mediator, anticipated threat of do-gooder derogation, was measured using a 7-point Likert scale containing items that represented statements about how other participants would feel about them when choosing the most ethically-conscious option. The participants were asked to what extent they agreed with statements that represented negative feelings others could have when they would see them choosing the most ethically-conscious option. The participants were asked to rate 6 statements using a response on a 7-point scale ranging from strongly agree to strongly disagree. The items were averaged and can be seen in table 2 (α = .814, M = 2.44, SD = .94).
Statement
People would think of me negatively when they see me choosing the most ethically-conscious option.
People would think I am an outlier when they see me choosing the most ethically-conscious option.
People would think I am acting stupid when they see me choosing the most ethically-conscious option.
People would think less of me when they see me choosing the most ethically-conscious option.
People will not accept me if they see me choosing the most ethically-conscious option. People will feel that I am different than them when they see me choosing the most ethically-conscious option.
Table 2. Scale Anticipated threat of do-gooder derogation
Manipulation and attention check
RESULTS
PARTICIPANTS
During the two weeks that the study was conducted, 262 participants filled out the survey. Thirteen participants failed the attention check, 73 participants did not pass the manipulation check, and 20 participants took less than 5 minutes to complete the survey. A pilot test demonstrated that the survey could not be fully read and filled out in less than 5 minutes; hence the results of 106 participants were excluded. The remaining 156 participants were on average 21 years old, 38% was male and 62% was female. The data of these 156 participants was used for analysis, 73 participants were allocated to the private condition whereas 83 participants were allocated to the public condition.
MEAN COMPARISON
Anticipated threat of do-gooder derogation.
To compare the means, an ANOVA test with anticipated threat of do-gooder derogation as the dependent variable, and self-esteem (median split) and situation as independent variables. The ANOVA test shows that while participants anticipate more threat of do-gooder derogation in public situations (M = 2.52, SD = .99) than private situations (M = 2.36, SD = .87), this difference was not significant (F(1,152) = .26, p = .61). Furthermore, participants with a low level of self-esteem anticipate higher threat of do-gooder derogation (M = 2.67, SD = .71) than participants with a high level of self-esteem (M = 2.32, SD = .91) (F(1,152) = 7.80, p = .01). Moreover, the difference in anticipated threat of do-gooder derogation for private and public situations does not depend on the level of self-esteem (F(1,152) = .31, p = .58). Therefore, the conclusion is drawn that the level of self-esteem does not moderate the relationship between the situation and the anticipated threat of do-gooder derogation. The results of the test are visualized in figure 5.
Level of altruistic behavior.
Furthermore, the results of an ANOVA test with altruistic behavior as dependent variable shows that participants in the private condition choose to display more altruistic behaviour (M = 3.33, SD = 1.47) than participants in the public condition (M = 2.95, SD = 1.35), though in line with reasoning, this difference is not significant (F(1,152) = 2.33, p = .13). Moreover, participants with a low level of self-esteem displayed a lower level of altruistic behavior (M =
Figure 5. Mean comparison mediator
3.04, SD = 1.37) than participants with a high level of self-esteem (M = 3.21, SD = 1.47), however this difference is not significant (F(1,152) = .19, p = .66). Moreover, the difference in altruistic behavior for private and public situations does not depend on the level of self-esteem (F(1,152) = .13, p = .72). The results are visualized in figure 6.
MODERATED MEDIATION ANALYSIS
Does self-esteem determine whether participants anticipate threat?
The complete conceptual model is tested using the PROCESS macro by Hayes (2013), template 7. By using this model, it can be tested whether 1) the anticipated threat of do-gooder derogation mediates the influence of a public or private situation on altruistic behavior, and 2) whether this mediating effect is more pronounced for people low in self-esteem. β = .19 β = -.10 p = .26 β = .08 p = .40 p = .62 Direct effect β = -.36 p = .11
Figure 7. Moderated mediation analysis with self-esteem determining whether threat is anticipated 1
1
Note: the unstandardized coefficient of the moderator represents the coefficient of the interaction between the moderator and the independent variable (situation * self-esteem).
It appears that there is no significant relationship between situation (private vs public) and the level of altruistic behavior, as well as between situation and anticipated threat of do-gooder derogation. As a result, it seems unlikely that mediation effect takes place. To conform this, conditional process analysis is performed to confirm or reject the expectation that the influence of situation on the level of altruistic behavior is mediated by anticipated threat of do-gooder derogation. The results are visualized in figure 7. Employing model 7 (bias-corrected, 1000 bootstrap samples) of the PROCESS macro (Hayes, 2013), it is found that the indirect effect of situation through anticipated threat on altruistic behaviour was not significant for participants low in self-esteem (-1SD, 95% CI [-0.03, 0.14]), nor for those high in self-esteem (+1SD, 95% CI [-0.21, 0.02]). Hence, anticipated threat of do-gooder derogation does not take on a mediating role, regardless of whether participants have low self-esteem or not.
ADDITIONAL MODERATED MEDIATION ANALYSIS
Does self-esteem determine whether participants care about threat?
Although I initially expected that self-esteem affected whether people do or do not anticipate threat of do-gooder derogation, research also suggests that self-esteem may have another impact; it determines whether people care or do not care about threat. Whereas high self-esteem individuals take more risk (Baumeister, Heatherton & Tice, 1993), low self-self-esteem individuals tend to avoid situations that may put them in a bad light (Leary, 2001) and are therefore expected to care more for the anticipated threat of do-gooder derogation. Although not hypothesized, I tested it. For this, the PROCESS macro of Hayes (2013) is employed again, template 14 (bias-corrected, 1000 bootstrap samples). The results are visualized in figure 8 and the SPSS output can be found in appendix B.
β = -.31 p = .02 β = .16 β = -.06 p = .28 p = .60 Direct effect β = -.26 p = .26
Figure 8. Moderated mediation analysis with self-esteem determining whether threat is cared for 2
However, as the results show, though the main effect of the level of self-esteem on the level of altruistic behavior is not significant (β = .16, p = .21), the interaction effect (anticipated threat * self-esteem) is significantly influencing the level of altruistic behavior (β = .-31, p = .02). Therefore, the conclusion can be drawn that the level of self-esteem in conjunction with anticipated threat of do-gooder derogation influences the level of altruistic behavior. To gain more understanding of this interaction, the Johnson-Neyman technique was used to pinpoint at which values of self-esteem the moderating influence on the relationship between anticipated threat of do-gooder derogation and level of altruistic behavior was significant. As the results show, only exceptionally low self-esteem and relatively high self-esteem resulted in a significant influence. It appeared that self-esteem scores lower than 2.92 (M = 5.17) and scores higher than 5.85 in conjunction with anticipated threat of do-gooder derogation affected the level of altruistic behavior, which entails the lower 3.2% and the upper 20.5% of
2
Note: the unstandardized coefficient of the moderator represents the coefficient of the interaction between the moderator and the independent variable (anticipated threat * self-esteem).
DISCUSSION
All in all, the analysis shows that while altruistic behavior is somewhat less in public situations compared to private situations, there is no significant mediating effect of the anticipated threat of do-gooder derogation. Therefore, hypothesis 1 (Public situations, compared to private situations, lead to a higher level of anticipated threat of do-gooder derogation which diminishes altruistic behavior) can be rejected. Furthermore, while self-esteem does have direct effect on the anticipated threat of do-gooder derogation, it does not significantly moderate the relationship between the situation and the anticipated threat. Also, the anticipated threat of do-gooder derogation is only somewhat higher in public situations for individuals with high self-esteem though not for low self-esteem individuals which contradicts expectations. Hence, hypothesis 2 (Individuals with low levels of self-esteem anticipate a higher threat of do-gooder derogation and will display a lower level of altruistic behavior in public situations) can be rejected as well. However, as the additional analysis showed, exceptionally low and relatively high levels of self-esteem, in conjunction with the anticipated threat of do-gooder derogation, influence the level of altruistic behavior. This entails that only those individuals with an exceptionally low and relatively high level of self-esteem seem to care about anticipated threat of do-gooder derogation.
Threat though no behavior change.
behavior are at least comparable to the risk of do-gooder derogation. This conclusion is drawn from the fact that behavior is not changed. Looking at the mean comparison of this analysis, these expectations do appear to be confirmed since there is no significant difference between altruistic behavior in public situations compared to private situations. Therefore, it is expected that the benefits of altruistic behavior, for instance an increase in status or self-esteem, equal the anticipated risk of do-gooder derogation, however, further research is necessary to support this expectation.
Self-esteem as incentive to care about threat.
The supplementary analysis showed that self-esteem was, unexpectedly, also found to affect the relationship between anticipated threat of do-gooder derogation and the level of altruistic behavior. Low self-esteem individuals let the anticipated threat of do-gooder derogation prevent them from buying altruistically in a public setting. On the other hand, high self-esteem individuals purchase more altruistically despite of their anticipation of threat. While this is influence was not expected, it could be explained by the phenomena that Bellezza, Gino, & Keinan (2014) uncovered; the red sneakers effect. According to these authors, displaying nonconforming behavior “can lead to inferences of status and competence” (p. 36). So, the possibility to show altruistic behavior, which is seen as nonconforming behavior, could give people the opportunity to show status and competence. Furthermore, the need for uniqueness, “the willful pursuit of differentness relative to others” (p. 38), positively influences norm deviating behavior (Bellezza, Gino, & Keinan, 2014). Therefore, a possible explanation could be that while low self-esteem individuals want to behave according to the norm to increase the chances of group-belonging, high self-esteem do not have this need and will therefore have a higher need for uniqueness. Hence, especially high self-esteem individuals will be likely to increase their altruistic behavior despite the anticipation of threat of do-gooder derogation.
CONCLUSION
not appear to be a moderator though it does directly influence the anticipated threat of do-gooder derogation while the lowest and highest levels of self-esteem also, in conjunction with anticipated threat of do-gooder derogation influence altruistic behavior unexpectedly. Also, a possible explanation for the fact that behavior does not change despite anticipated threat could be that consumers make a trade-off between the costs and benefits of altruistic behavior. For low self-esteem individuals, the anticipated cost of the threat of do-gooder derogation appears to be at least comparable to the benefits, one of which may be the pursuit of self-esteem. High self-esteem individuals in general anticipate less threat of do-gooder derogation. Therefore, their benefits of altruistic behavior, which may be the display of status, outweigh the costs.
MANAGERIAL IMPLICATIONS
Because individuals with high self-esteem anticipate less threat and as a consequence purchase more altruistically, marketers of physical stores are advised to use tools that increase consumer’s self-esteem if they want to increase the sales of green products. This would attract low esteem individuals who pursue an increase in their esteem as well as high self-esteem individuals who feel confirmed in their self-concept. For this, marketers need to focus on what tools to use to improve the esteem of their targeted consumers. For instance, self-esteem increases whenever people feel good about themselves. This can be used by marketers who can use compliments in their advertisements. Putting a focus on the possible positive consequences of a product or service for society that gives the consumer a positive feeling about contributing to the welfare of society.
LIMITATIONS AND SUGGESTIONS FOR FURTHER RESEARCH
One of the main limitations of this research can be found in the manipulation of the independent variable; public vs private situation. While a relatively large amount of participants were excluded as a consequence of failing the manipulation test, the possibility exists that other the participants passed the manipulation test by chance, so the manipulation was not effective. This doubt exists also because another participant seeing your product choices might not be considered to be publically enough to cause behavior change. For example, for this manipulation to be effective more than one person has to see your choices or the possibility of interaction with another person should exist. While this is a limitation of this research, it reveals at the same an opportunity for further research.
RESOURCES
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215.
Bar-Haim, Y., Lamy, D., Pergamin, L., Bakermans-Kranenburg, M. J. & van IJzendoorn, M. H. (2007) Threat-related attentional bias in anxious and nonanxious individuals: A meta-analytic study. Psychological Bulletin, 133(1), 1-24.
Baumeister, R.F., Boden, J.M. & Smart, L. (1997). Relation of threatened egotism to violence and aggression: The dark side of high self-esteem. Psychological Review, 103(1), 5-33.
Baumeister, R.F., Heatherton, T.F. & Tice, D.M. (1993). When ego threats lead to self-regulation failure: Negative consequences of high self-esteem. Journal of Personality and Social Psychology, 64(1), 141-156.
Bellezza, S., Gino, F. & Keinan, A. (2014). The red sneakers effect: Inferring status and competence from signals of nonconformity. Journal of Consumer Research, 41(1), 25-54.
Bourgeois, K.S. & Leary, M.R. (2001). Coping with rejection: Derogating those who choose us last. Motivation and Emotion, 25(2), 101-111.
Cialdini, R.B. & Goldstein, N.J. (2004). Social influence: Compliance and conformity. Annual Review of Psychology, 55(1974), 591–621.
Cramwinckel, F.M., Van Dijk, E., Scheepers, D. & Van den Bos, K. (2013). The threat of moral refusers for one's self-concept and the protective function of physical cleansing. Journal of Experimental Social Psychology, 49(6), 1049–1058.
Dunning, D. (2007). Self-image motives and consumer behavior: How sacrosanct self-beliefs sway preferences in the marketplace. Journal of Consumer Psychology, 17, 237–249.
Fein, S. & Spencer, J.S. (1997). Prejudice as self-image maintenance: Affirming the self through derogating others. Journal of Personality and Social Psychology, 73(1), 31-44.
Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117-140. Froming, W.J., Walker, G.R. & Lopyan, K.J. (1982). Public and private self-awareness:
When personal attitudes conflict with societal expectations. Journal of Experimental Social Psychology, 18, 476-487.
Greenberg, J., Pyszczynski, T., Burling, J., Simon, L., Solomon, S., Rosenblatt, A., Lyon, D. & Pinel, E. (1992). Why do people need esteem? Converging evidence that self-esteem serves an anxiety-buffering function. Journal of Personality and Social Psychology, 63(6), 913-922.
Greenwald, A.G. (1976). Within-subjects design: To use or not to use. Psychological Bulletin, 82(2), 314-320.
Griskevicius, V., Tybur, J.M. & Van den Bergh, B. (2010). Going green to be seen: Status, reputation, and conspicuous conservation. Journal of Personality and Social
Psychology, 98(3), 392-404.
Hewitt, P. L., Flett, G. L., Sherry, S. B., Habke, M., Parkin, M., Lam, R. W., McMurtry, B., Ediger, E., Fairlie, P. & Stein, M. B. (2003). The interpersonal expression of
perfection: Perfectionistic self-presentation and psychological distress. Journal of Personality and Social Psychology, 84(6), 1303-1325.
Kocovski, N.L. & Endler, N.S. (2000) Social anxiety, self-regulation, and fear of negative evaluation. European Journal of Personality, 14(4), 347-358.
Leary, M.R., Tambor, E.S., Terdal, S.K. & Downs, D.L. (1995). Self-esteem as an
interpersonal monitor: The sociometer hypothesis. Journal of Personality and Social Psychology, 68(3), 518-530.
Leary, M.R. (2001). Interpersonal rejection. Oxford, England: Oxford University Press.
Martín-Albo, J., Núñez, J.L., Navarro, J.G. & Grijalvo, H. (2007). The Rosenberg self-esteem scale: Translation and validation in university students. The Spanish Journal of
Psychology, 10(2), 458-467.
Monin, B., Marquez, M.J. & Sawyer, P.J. (2008). The rejection of moral rebels: Resenting those who do the right thing. Journal of Personality and Social Psychology, 95(1), 76– 93.
Minson, J.A. & Monin, B. (2012). Do-gooder derogation: Disparaging morally motivated minorities to defuse anticipated reproach. Social Psychological and Personality Science, 3(2), 200-207.
Robins, R.W., Hendin, H.M. & Trzesniewski, K.H. (2001). Measuring global self-esteem: Construct validation of a single-item measure and the Rosenberg self-esteem scale. Personality and Social Psychology Bulletin, 27(2), 151-161.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, N.J.: Princeton University Press.
Tasimi, A., Dominguez, A. & Wynn, K. (2015) Do-gooder derogation in children: The social costs of generosity. Frontiers in Psychology, 6, 1-6.
Turner, J.C., Brown, R.J. & Tajfel, H. (1979). Social comparison and group interest in ingroup favouritism. European Journal of Social Psychology, 9(2), 187–204.
Verhallen, T. M. M., & van Raaij, W. F. (1986). How consumers trade off behavioural costs and benefits. European Journal of Marketing, 20(3/4), 19-34.
APPENDIXES
APPENDIX A
Descriptive Statistics
Dependent Variable: AntThreat SEMedSplit Situatio n Mean Std. Deviation N Low private 2,6667 ,71146 28 public 2,6597 1,02853 48 Total 2,6623 ,91933 76 High private 2,1630 ,90627 45 public 2,3238 ,91461 35 Total 2,2333 ,90769 80 Total private 2,3562 ,86750 73 public 2,5181 ,99047 83 Total 2,4423 ,93549 156
Tests of Between-Subjects Effects
Dependent Variable: AntThreat
Source Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared
Corrected Model 7,681a 3 2,560 3,041 ,031 ,057 Intercept 897,121 1 897,121 1065,614 ,000 ,875 SEMedSplit 6,567 1 6,567 7,801 ,006 ,049 Situation ,221 1 ,221 ,262 ,609 ,002 SEMedSplit * Situation ,262 1 ,262 ,312 ,578 ,002 Error 127,966 152 ,842 Total 1066,167 156 Corrected Total 135,647 155
Descriptive Statistics
Dependent Variable: AltrBehav SEMedSplit Situatio n Mean Std. Deviation N Low private 3,2143 1,43505 28 public 2,9444 1,33037 48 Total 3,0439 1,36663 76 High private 3,4000 1,50856 45 public 2,9619 1,39272 35 Total 3,2083 1,46634 80 Total private 3,3288 1,47352 73 public 2,9518 1,34862 83 Total 3,1282 1,41647 156
Tests of Between-Subjects Effects
Dependent Variable: AltrBehav
Source Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared
Corrected Model 6,121a 3 2,040 1,017 ,387 ,020 Intercept 1460,448 1 1460,448 728,138 ,000 ,827 SEMedSplit ,385 1 ,385 ,192 ,662 ,001 Situation 4,669 1 4,669 2,328 ,129 ,015 SEMedSplit * Situation ,264 1 ,264 ,131 ,717 ,001 Error 304,871 152 2,006 Total 1837,556 156 Corrected Total 310,991 155
APPENDIX B
SPSS output PROCESS macro Hayes (2013) Model 7.
Run MATRIX procedure:
************** PROCESS Procedure for SPSS Release 2.15 ******************* Written by Andrew F. Hayes, Ph.D. www.afhayes.com
Documentation available in Hayes (2013). www.guilford.com/p/hayes3 ************************************************************************** Model = 7 Y = AltrB X = Situatio M = AntThrea W = SelfEst Sample size 156 ************************************************************************** Outcome: AntThrea Model Summary R R-sq MSE F df1 df2 p ,2083 ,0434 ,8537 2,2974 3,0000 152,0000 ,0799 Model
coeff se t p LLCI ULCI constant 2,4629 ,0762 32,3417 ,0000 2,3124 2,6133 Situatio ,0764 ,1528 ,5003 ,6176 -,2254 ,3782 SelfEst -,1810 ,0810 -2,2339 ,0270 -,3410 -,0209 int_1 ,1862 ,1638 1,1372 ,2572 -,1373 ,5098 Product terms key:
int_1 Situatio X SelfEst
************************************************************************** Outcome: AltrB Model Summary R R-sq MSE F df1 df2 p ,1492 ,0223 1,9874 1,7420 2,0000 153,0000 ,1786 Model
coeff se t p LLCI ULCI constant 3,3777 ,3175 10,6392 ,0000 2,7505 4,0049 AntThrea -,1022 ,1215 -,8408 ,4018 -,3422 ,1379 Situatio -,3604 ,2271 -1,5873 ,1145 -,8090 ,0882 ******************** DIRECT AND INDIRECT EFFECTS ************************* Direct effect of X on Y
Conditional indirect effect(s) of X on Y at values of the moderator(s): Mediator
SelfEst Effect Boot SE BootLLCI BootULCI AntThrea -,9509 ,0103 ,0342 -,0257 ,1424 AntThrea ,0000 -,0078 ,0265 -,1196 ,0179 AntThrea ,9509 -,0259 ,0484 -,2083 ,0248
Values for quantitative moderators are the mean and plus/minus one SD from mean. Values for dichotomous moderators are the two values of the moderator.
******************** INDEX OF MODERATED MEDIATION ************************ Mediator
Index SE(Boot) BootLLCI BootULCI AntThrea -,0190 ,0342 -,1432 ,0196
******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals: 1000
Level of confidence for all confidence intervals in output: 95,00
NOTE: The following variables were mean centered prior to analysis: Situatio SelfEst
SPSS output PROCESS macro Hayes (2013) Model 14.
Run MATRIX procedure:
************** PROCESS Procedure for SPSS Release 2.15 ******************* Written by Andrew F. Hayes, Ph.D. www.afhayes.com
Documentation available in Hayes (2013). www.guilford.com/p/hayes3 ************************************************************************** Model = 14 Y = AltrBeha X = Situatio M = AntThrea V = SelfEstm Sample size 156 ************************************************************************** Outcome: AntThrea Model Summary R R-sq MSE F df1 df2 p ,0866 ,0075 ,8742 1,1646 1,0000 154,0000 ,2822 Model
coeff se t p LLCI ULCI constant -,2481 ,2417 -1,0261 ,3064 -,7256 ,2295 Situatio ,1619 ,1500 1,0792 ,2822 -,1345 ,4583 ************************************************************************** Outcome: AltrBeha Model Summary R R-sq MSE F df1 df2 p ,2449 ,0600 1,9360 2,4087 4,0000 151,0000 ,0518 Model
coeff se t p LLCI ULCI constant 3,4774 ,3738 9,3028 ,0000 2,7388 4,2159 AntThrea -,0644 ,1219 -,5278 ,5984 -,3052 ,1765 Situatio -,2605 ,2310 -1,1277 ,2612 -,7170 ,1959 SelfEstm ,1561 ,1241 1,2575 ,2105 -,0891 ,4013 int_1 -,3105 ,1358 -2,2858 ,0237 -,5788 -,0421 Product terms key:
int_1 AntThrea X SelfEstm
******************** DIRECT AND INDIRECT EFFECTS ************************* Direct effect of X on Y
Mediator
SelfEstm Effect Boot SE BootLLCI BootULCI AntThrea -,9509 ,0374 ,0463 -,0241 ,1763 AntThrea ,0000 -,0104 ,0294 -,1141 ,0240 AntThrea ,9509 -,0582 ,0662 -,2281 ,0324
Values for quantitative moderators are the mean and plus/minus one SD from mean.
Values for dichotomous moderators are the two values of the moderator. ******************** INDEX OF MODERATED MEDIATION ************************ Mediator
Index SE(Boot) BootLLCI BootULCI AntThrea -,0503 ,0515 -,1789 ,0324
******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence
intervals: 1000
Level of confidence for all confidence intervals in output: 95,00
NOTE: The following variables were mean centered prior to analysis: AntThrea SelfEstm
SPSS output PROCESS macro Hayes (2013) Johnson-Neyman Technique.
********************* JOHNSON-NEYMAN TECHNIQUE ************************** Moderator value(s) defining Johnson-Neyman significance region(s)
Value % below % above -2,2534 3,2051 96,7949 ,6776 79,4872 20,5128
Conditional effect of X on Y at values of the moderator (M)
SelfEstm Effect se t p LLCI ULCI -3,3724 1,0295 ,4807 2,1416 ,0338 ,0797 1,9793 -3,1124 ,9448 ,4468 2,1146 ,0361 ,0620 1,8275 -2,8524 ,8600 ,4131 2,0821 ,0390 ,0439 1,6761 -2,5924 ,7753 ,3796 2,0424 ,0428 ,0253 1,5253 -2,3324 ,6905 ,3465 1,9931 ,0480 ,0060 1,3751 -2,2534 ,6648 ,3365 1,9757 ,0500 ,0000 1,3296 -2,0724 ,6058 ,3138 1,9306 ,0554 -,0141 1,2257 -1,8124 ,5211 ,2817 1,8497 ,0663 -,0355 1,0776 -1,5524 ,4363 ,2504 1,7422 ,0835 -,0585 ,9311 -1,2924 ,3516 ,2204 1,5954 ,1127 -,0838 ,7869 -1,0324 ,2668 ,1920 1,3896 ,1667 -,1125 ,6462 -,7724 ,1821 ,1663 1,0949 ,2753 -,1465 ,5106 -,5124 ,0973 ,1446 ,6730 ,5020 -,1884 ,3830 -,2524 ,0126 ,1290 ,0975 ,9224 -,2423 ,2675 ,0076 -,0722 ,1219 -,5921 ,5547 -,3129 ,1686 ,2676 -,1569 ,1246 -1,2590 ,2099 -,4031 ,0893 ,5276 -,2416 ,1367 -1,7679 ,0791 -,5117 ,0284 ,6776 -,2906 ,1471 -1,9757 ,0500 -,5811 ,0000 ,7876 -,3264 ,1559 -2,0934 ,0380 -,6344 -,0184 1,0476 -,4111 ,1800 -2,2838 ,0238 -,7668 -,0555 1,3076 -,4959 ,2073 -2,3919 ,0180 -,9055 -,0863 1,5676 -,5806 ,2367 -2,4530 ,0153 -1,0483 -,1130 1,8276 -,6654 ,2675 -2,4876 ,0139 -1,1938 -,1369 ******************** ANALYSIS NOTES AND WARNINGS ************************* Level of confidence for all confidence intervals in output:
95,00
NOTE: The following variables were mean centered prior to analysis: AntThrea SelfEstm
APPENDIX C
APPENDIX D
