Using disparaging humor in webcare: a study on the effect of the use of disparaging humor in webcare on brand attitude

(1)

Ginger Reinhoudt – 10718869 Master’s Thesis Graduate School of Communication Master’s programme Communication Science prof. dr. S.J.H.M. (Bas) van den Putte 29 June 2018

Using disparaging humor in webcare:

A study on the effect of the use of disparaging humor in webcare on

brand attitude.

(2)

Abstract

The use of disparaging humor in webcare is not unspoken of: various brands apply this insult-based humor within their online tone-of-voice. Yet, knowledge on the effects of disparaging humor in brand communication is limited, especially with regard to webcare. This study therefore explores how the use of disparaging humor influences brand attitudes of social media users exposed to the webcare interaction. Following a 3 (disparaging humor, non-disparaging humor and non-humorous response) x 2 (familiar and unfamiliar brand) factorial between-subjects, experimental design (n = 204), the current study aims to test whether brand familiarity moderates the relationship between a disparaging humorous reply and brand attitude, and whether this relationship is mediated by perceived appropriateness. The current study additionally aims to test whether perceived funniness of a disparaging humorous reply will increase perceived funniness, provided the reply is perceived as benign; whether viewers’ power motivation moderates this effect; and whether perceived funniness increases interaction attitude, ultimately increasing brand attitude.

Results confirm moderated mediation for brand familiarity and appropriateness and confirm moderated serial mediation for power motivation and perceived moral violation, perceived funniness and interaction attitude. Against expectations, results indicate perceived moral violation of a benign disparaging reply does not explicitly increase perceived funniness. In general, results suggest brand attitude is not positively influenced by the use of disparaging humor. The study provides advice for practitioners using webcare in their brand communication. Finally, theoretical implications and suggestions for future research are given. Keywords: webcare, disparaging humor, perceived appropriateness, perceived moral violation, power motivation, brand familiarity, perceived funniness.

(3)

Abstract ... 1

Introduction ... 3

Theoretical framework ... 5

Webcare ...6

Disparaging humor and appropriateness ...7

Familiarity, brand attitude and appropriateness ...8

Perceived violation, perceived funniness and interaction attitude ...9

Power motivation and perceived funniness ... 10

Perceived funniness, attitude towards the interaction and attitude towards the brand ... 10

Method ... 11 Procedure ... 11 Sample ... 12 Measures ... 13 Independent variables ... 13 Dependent variables... 14

Manipulation check variables ... 15

Results ... 16

Appropriateness and familiarity ... 18

Perceived moral violation and perceived funniness ... 19

Perceived funniness and attitude towards the interaction ... 20

Conclusion and discussion ... 22

References ... 25

Appendices ... 30

Appendix A. – Stimulus materials ... 30

Appendix B. – Tables... 32

(4)

Introduction

Since the emergence of web 2.0, consumers no longer merely interact with brands offline. Where questions, concerns and complaints about a service or product may formerly have been voiced at in-store helpdesks or through written and phone contact (Stone, Hobbs, & Khaleeli, 2002), customer-initiated touchpoints with brands now increasingly take place online (Muntinga, Moorman, & Smit, 2011). Depending on what a company offers, consumers have various outlets at their disposal from where they can reach out to a brand, including e-mail, live chat, direct (private) messages on social media (Turel & Connelly, 2013) and public messages on social media (van Noort & Willemsen, 2012). The handling of customer-initiated touchpoints through these outlets poses opportunities for a company to establish, strengthen or repair a consumer’s brand attitude. Here, the specifically public nature of the latter outlet may cause the interaction to be exposed to more consumers than just the initiating customer. Hence, handling customers’ public messages poses the opportunity to influence brand attitude for whomever is exposed to the interaction: a potentially much larger audience (Ward & Ostrom, 2006). It is for this reason companies should provide webcare: interacting with initiating customers online by engaging in dialogue to address their consumer feedback (van Noort & Willemsen, 2012). Upon seeing a reaction and deeming it positive, those exposed to the interaction will likely value the brand in a more positive manner. In fact, when engaging in reactive webcare as opposed to not replying at all, consumers’ brand attitude is positively influenced (van Noort & Willemsen, 2012), thus maintaining the advice for practitioners to ensure they adequately participate in customer-initiated interactions.

Over the past years, it has become apparent that several brands have come to take on a strategic tone-of-voice in which disparaging – or putdown – humor is used in their reactive webcare. Rather than providing a response under the traditional notion of “the customer is king”, such webcare responses attempt to evoke laughter from the potential audience, based on

(5)

infirmities of the initiating customer. For example, on January 3rd_{2017, Twitter user Phono}

voiced a complaint at fast-food chain Wendy’s through a tweet: “you’re food is trash”. Rather than ignoring the message or replying considerately, Wendy’s webcare team engaged in the interaction with the following reply: “No, your opinion is though”, resulting in over 1500 retweets, over 4500 post-likes and numerous positive comments in favor of the Wendy’s brand. Examples like this suggest the use of disparaging humor in reactive webcare may have a positive effect on brand attitude for those exposed to the online interaction. If this is in fact the case, practitioners may benefit from applying the use of such humor. However, as of the moment of writing, this effect has yet to be established within the field of webcare. In order to shed light on this matter, this study therefore addresses the research question: ‘To what extent does disparaging humor in reactive webcare influence brand attitude?’

Results may facilitate sound advice for practitioners in terms of whether to apply disparaging humor in webcare or whether to refrain from doing so. Moreover, outcomes of this research will provide a scientific contribution to the existing literature on how brand attitude can be affected through webcare interactions and through the use of disparaging humor.

To further this, a mediating effect of perceived appropriateness will be evaluated, considering disparaging humor is at risk of being disapproved of (Ford & Ferguson, 2004) and perceived appropriateness appears to mediate the effect of informal brand responses on brand attitude. Moreover, a moderating effect of brand familiarity will be assessed, as brand familiarity is found to moderate the effect of an informal message on perceives appropriateness (Gretry, Horváth, Belei, & van Riel, 2017). Furthermore, this study aims to test for a second path, in which a possible mediating effect of perceived severity of moral violation on perceived funniness will be evaluated, applying the benign-violation theory as proposed by McGraw and Warren (2010). Moreover, work by Newton, Wong and Newton (2016) indicate a moderating effect for power motivation, in that those who demonstrate high power motivation perceive

(6)

disparaging jokes in advertising as more humorous, compared to those who demonstrate low power motivation. This study will therefore also test for a moderating effect by power motivation. Finally, with a positive, mediating effect of perceived funniness on attitude towards advertisements (Gelb & Pickett, 1983), a similar effect is expected with a positive effect of perceived funniness on attitude towards the webcare interaction. In turn, this is expected to lead to a more positive attitude towards the webcare interaction, ultimately eliciting a more positive attitude towards the brand (MacKenzie, Lutz, & Belch, 1986).

Theoretical framework

Over the past three decades, companies have progressively shifted their marketing communication from traditional advertising and promotion to a more overarching approach, by adopting an Integrated Marketing Communication (IMC) strategy (Belch & Belch, 2018). IMC is a business strategy in which all planning, developing, executing and evaluating elements of a brand’s communication are centralized. Within IMC, many companies take on a touchpoint perspective, through which they determine how and where to present their brand based on all potential circumstances the brand may interact with the target audience. All circumstances in which consumers may see or hear about a brand and/or have an experience with it are considered touchpoints and can be categorized in fourfold: (1) company created touchpoints - all marketing communication messages as deliberately planned by the company, like advertisements, press releases, websites and sale promotions; (2) intrinsic touchpoints – all interactions with a brand within the purchasing process or whilst using the product/service, like conversations with retail sales personnel or customer service representatives; (3) unexpected touchpoints – all unanticipated information about a brand that consumers are exposed to beyond the company’s control, like word-of-mouth (WOM) messages and product/service reviews platforms not hosted by the brand; and (4) customer-initiated

(7)

touchpoints – all interactions initiated by the consumer like complaints or questions posed directly to the company, often handled through customer service departments (Duncan, 2005).

All but the first of these touchpoints can be found when consumers express themselves about brands and their products and services online, through the newfound platform of social networking sites (Muntinga et al., 2011). Consumers’ online brand-related activities are predominantly customer-initiated touchpoints and can showcase both consumers’ positive feedback and negative concern (Hsieh, Hsieh, & Tang, 2012). Within this, the latter typically represents complaints (Gensler, Völckner, Liu-Thompkins, & Wiertz, 2013), addressing consumers’ dissatisfaction with a brand’s product or service (Singh & Widing, 1991). Online complaints voiced anywhere other than in private messages to the brand can create unexpected touchpoints for other consumers and are thus considered negative electronic word of mouth (NWOM) (van Noort & Willemsen, 2012). Subsequently, intrinsic touchpoints occur through a company’s handling of these complaints by providing customer service.

Webcare

Customer acquisition and retention are majorly impacted by how a company handles customer-initiated touchpoints (Belch & Belch, 2018), making adequate customer service crucial. In fact, it may be more important now than ever: where customers formerly could merely express their dissatisfaction to a relatively limited audience offline, their opinions can now easily be exposed to much larger audiences (Ward & Ostrom, 2006) through NWOM. With the possibility of the audience’s attitude towards the brand being negatively effected through NWOM, brands are at risk of their reputation being harmed to a much greater extent (Huibers & Verhoeven, 2014; Ward & Ostrom, 2006). Hence, companies’ sufficient understanding of if and how to respond to NWOM messages is essential (Fournier & Avery, 2011). In doing so, one must not overlook the fact that NWOM does not pose a threat in all circumstances. It may definitely present an

(8)

Osmonbekov, & Czaplewski, 2006) through webcare: addressing online consumer complaints and feedback by company representatives (van Noort & Willemsen, 2012). In fact, results by van Noort and Willemsen (2012) suggest brand attitude is positively affected when a brand responds to an NWOM message by an initiating customer, rather than if the brand were not to respond. In addition, the more a brand is perceived to respond with a conversational human voice (CHV), the stronger this effect is.

Using a CHV, an organization communicates with consumers in an engaging and natural manner (Kelleher, 2009), essentially enforcing the notion that by communicating with the organization, consumers are communicating with a fellow human. Next to familiar public relation qualities like promptly responding and welcoming conversational dialogue, CHV is also characterized by elements that may not be recognized as traditional corporate communication – one of which is the use of humor (Kelleher & Miller, 2006; van Noort, Willemsen, Kerkhof, & Verhoeven, 2014).

Disparaging humor and appropriateness

According to Shabbir and Thwaites (2007), three categories of humor theory are relevant in advertising: (1) incongruity – evoking laughter through unexpectedness; (2) arousal-safety – the notion of providing escape from a form of constraint (i.e. social norms) through humor; and (3) disparagement – humor based on laughing at the infirmities of others. Considering the public nature of webcare interactions to be similar to that of advertising, one can argue these categories are similarly relevant for webcare. Referring back to the Wendy’s webcare interaction as discussed in the introduction, it appears the initiating customer is deliberately joked about; hence disparagement is predominantly applied in this example.

The use of disparaging humor, however, does not go without risk. Ford and Ferguson (2004) argue that for a disparaging message to be considered humorous, the recipient must approve of it, intrinsically consenting to an implicit normative standard that there is no need to

(9)

consider the insult in a serious or critical manner. In other words, recipients must switch to a non-serious humor mindset to take the insult lightly and consider it humorous. If they do not do so – and with that do not approve of the disparagement – exposure to disparaging humor can have negative social consequences in which the intended joke is considered inappropriate (Ford & Ferguson, 2004). Considering this risk, the first hypothesis focuses on the effect of exposure to a brand reply with disparaging humor versus a brand replywithout disparaging humor:

H1: A brand reply with disparaging humor will lead to lower perceived appropriateness than a brand replywithout disparaging humor.

Familiarity, brand attitude and appropriateness

Taking a self-proclaimed new approach to CHV, Gretry et al. (2017) discuss communication style in terms of formality. Results indicated that when familiar with a brand, an informal communication style increases perceived appropriateness, which consequently increases brand trust. When unfamiliar with a brand, an informal style decreases perceived appropriateness, decreasing brand trust. Brand familiarity was therefore found to moderate the effect of communication style (formal vs. informal) on brand trust, which in turn was mediated by perceived appropriateness.

Taking into account humor is generally associated with informality (Bielenia-Grajewska, 2017; Rogerson-Revell, 2007; Yarwood, 1995) the findings by Gretry et al. (2017) are expected to be applicable to the relationship between perceived appropriateness and brand attitude through the following hypotheses:

H2a: When familiar with the brand, a brand replywith disparaging humor will lead to higher brand attitude than when unfamiliar with the brand.

(10)

Perceived violation, perceived funniness and interaction attitude

Benign-violation theory as proposed by McGraw and Warren (2010) explains how anything that violates a norm – threatening one’s sense of how the world ‘ought to be’ – will be considered humorous, provided that the violation simultaneously seems benign. In disparaging humor, the goal of evoking laughter due to the infirmities of another may challenge the social norm of respecting others. This brings forth the following hypothesis:

H3a: A brand reply with disparaging humor will lead to higher perceived moral violation than a brand replywithout disparaging humor

Similar to the need for a recipient to consent to an implicit normative standard for a disparaging joke to be considered humorous (Ford & Ferguson, 2004), a moral violation of a norm is found to be humorous if a different norm suggested the message was acceptable. Moral violations also appear amusing when the recipient is weakly committed or feels psychologically distant from it (McGraw & Warren, 2010). Thus, provided that a violation is still perceived as benign and is understood to be playful, perceived moral violation seems to positively effect perceived funniness.

Testing this, Kruschke and Vollmer (2014), found an inverted U-shape effect where low (n.b.: not violating) and high (n.b.: no longer benign) violations were not perceived as humorous, but violations scored between these extremes were perceived as humorous. If a requirement of benignity were to have been applied to this analysis – and the high violations were thus to be excluded – one can hypothesize a linear effect from low violation to moderate violation, increasing perceived funniness. Therefore, setting the requirement that the violation is still considered to be benign, the following hypothesis is proposed:

(11)

Power motivation and perceived funniness

In an empirical study looking into the effect of disparaging humor in advertisements on brand attitude, Newton et al. (2016) found a moderating effect for power motivation. Power motivation refers to one of three main motives for social behavior (McClelland, 1987): (1) the need for achievement – the desire to do things better; (2) the need for affiliation – the desire to establish, maintain or restore relationships with others; and (3) the need for power – the desire to have influence over others (Borges, Manuel, Elam, & Jones, 2010). In other words, those with high power motivation are motivated to perform certain behaviors in order to obtain power and superiority over others.

Newton et al. (2016) found that those with high power motivation perceive advertisements with disparaging humor to be more humorous than groups with low power motivation. Borrowing from this, power motivation is expected to have a similar moderating effect in the scope of disparaging humor in webcare:

H3c: Perceived moral violation of a disparaging webcare response will lead to higher perceived funniness for those who are more power motivated, than for those who are less power motivated.

Perceived funniness, attitude towards the interaction and attitude towards the brand

In terms of advertising, perceived funniness positively effects advertisement attitude, in that the more humorous an advertisement is considered to be, the more positive an advertisement is evaluated (Gelb & Pickett, 1983). Moreover, advertisement attitude has been found to increase brand attitude (Gresham & Shimp, 1985), mediating the relationship between perceived funniness and brand attitude (Zhang & Zinkhan, 2006). Similar effects are therefore expected for webcare interactions:

(12)

H4: Higher perceived funniness will lead to a more positive attitude towards the interaction.

H5: Attitude towards the interaction will lead to a more positive attitude towards the brand.

Furthermore, focusing on physically violent humor – which can be argued to be at least similar to disparaging humor (Brown, Bhadury, & Pope, 2010) –, an empirical study by (Swani et al., 2013) revealed a serial mediation path from perceived moral violation to brand attitude, through perceived funniness and interaction attitude. Here, it is noteworthy perceived moral violation was found to decrease perceived funniness. Upon further inspection, however, it appears perceived benignity was not included in the study (Swani et al., 2013). Thus, considering benign-violation theory (McGraw & Warren, 2010), this effect may have been positive if the violations were to have been perceived as benign.

Under the requirement the moral violations in the current study will be perceived as benign, a possible moderated serial mediation will be assessed predicting brand attitude due to a disparaging (vs. non-disparaging and non-humorous) brand reply, combining the path found by Swani et al. (2013) with the moderating effect of power motivation.

Method

A 3 between-subjects factorial design with disparaging humor (experimental: disparaging humor, non-disparaging humor and non-humor) x 2 brand familiarity (experimental: familiar and unfamiliar brand) design was used to test the hypotheses.

Procedure

Six webcare interactions were created as stimuli for the purpose of this study, with the same customer-initiated message and either a familiar, existing brand (Domino’s Pizza) or an

(13)

unfamiliar, fictional brand (Yahtzee’s Pizza), respectively representing a familiar and unfamiliar brand. Per brand, one interaction contained a reply from the brand with a message using disparaging humorous elements; one interaction contained a reply from the brand with a message using non-disparaging humorous elements; and one interaction contained a reply from the brand without any humorous elements.

All participants were presented with a consent form at the beginning of the questionnaire asking them to confirm their willful participation in the experiment. Following consent, participants were randomly assigned to a condition and were shown the corresponding Facebook interaction. Next, participants filled in scales regarding brand familiarity; perceived funniness; perceived appropriateness; perceived benignity; and perceived moral violation. Participants then filled in the scales for brand attitude and interaction attitude, after which items followed with regard to whether the reply was perceived to be disparaging; whether the brand replied in an informal manner; which brand replied; and with which response the brand replied. This lead to the page barring demographic questions targeting gender, age, nationality and education. Finally, participants filled in the power motivation scale. The very last page of the questionnaire presented an acknowledgement of appreciation for their participation in the study and a debriefing. At this point the study was completed.

Sample

Participants. Two hundred and four people agreed to participate in the study. Participants’ ages ranged from 19 to 59 (M = 27.43, SD = 7.82). Of the participants, 59.3% was female, 39.2% was male and 1.5% answered Other (3 participants). The vast majority (82.4%) completed the experiment in under ten minutes. Moreover, no straight-liners were identified within the dataset. Hence, analyses were conducted over the full sample of 204 participants.

(14)

Measures

Independent variables

The six original conditions were recoded into two variables: brand reply and brand familiarity. Brand reply. Regardless of brand name, participants were either exposed to a brand replywith disparaging humor, a brand replywith non-disparaging humor or a non-humorous brand reply. A new variable was computed accordingly, assigning participants to one out of three corresponding groups (0 = non-humorous, 1 = non-disparaging humor, 2 = disparaging humor).

Brand familiarity. A second variable recoded the six original conditions according to the brand that replied. Those who had seen an interaction with Yahtzee’s Pizza were assigned to the unfamiliar brand group and those who had seen an interaction with Domino’s Pizza were assigned to the familiar group (0 = unfamiliar, 1 = familiar).

Power motivation. As applied by Newton et al. (2016) the Dominance and Status aspiration sub-scales from Cassidy and Lynn (1989) were merged to create a scale for power motivation using 18 items measured on 5-point scales, for example: “I think I would enjoy having authority of other people”, “I find satisfaction in having influence over others because of my position in the community” and “I would like an important job where people looked up to me” (1 = strongly disagree, 5 = strongly agree). This scale was deemed reliable with Cronbach’s Alpha at .81 (M = 3.51, SD = .47). This variable was then dummy coded, based on the scale’s median of 3.50. In other words, power motivation values under 3.50 were recoded as low power motivation, values 3.50 and above were recoded as high power motivation (0 = low power motivation, 1 = high power motivation).

(15)

Dependent variables

Perceived funniness. Participants’ perceived funniness of the brand replywas measured through three items (Kruschke & Vollmer, 2014): “The response was funny / amusing / humorous” (1 = strongly disagree, 5 = strongly agree). This scale proved reliable with a Cronbach’s Alpha of .94 (M = 3.17, SD = 1.19).

Perceived appropriateness. An existing measure was used to measure perceived appropriateness of the brand’s communication style. Participants indicated whether they agreed that the way the brand responded “meets my expectations”, “corresponds with how I expect the brand to communicate with me” and “is appropriate” (1 = strongly disagree, 5 = strongly agree). This scale proved reliable with a Cronbach’s Alpha of .83 (M = 2.66, SD = 1.00).

Interaction attitude. Based on scales developed to determine attitude towards an advertisement (Dahlén, 2005), five items were included on 5-point semantic differential scales: “bad/good” “unpleasant/pleasant” and “unfavorable/favorable”. This scale proved reliable with a Cronbach’s Alpha of .89 (M = 3.08, SD = 1.01).

Brand attitude. To determine attitude towards the brand, participants were asked to indicate their overall feelings on the brand on 5-point semantic differential scales, following Gretry et al. (2017): “bad/good”, “unfavorable/favorable”, “negative, positive”, “undesirable/desirable”, and “unattractive/attractive”. This scale proved reliable with a Cronbach’s Alpha of .92 (M = 3.03, SD = .89).

Severity of moral violation. Six items from previous studies (Kruschke & Vollmer, 2014, McGraw et al., 2015) were combined to create a scale for severity of moral violation: “How aversive / upsetting / threatening / disturbing did you find the response?” and “The response was immoral (i.e.: wrong) / crossed a line” (1 = strongly disagree, 5 = strongly agree). Given these items had not yet been combined in previous work to represent a single latent variable, exploratory factor analysis was conducted to ensure all items could be used to create

(16)

the scale. Results indicated the scale was indeed unidimensional, explaining 75.19% of the variance in the six original items. This scale proved reliable with a Cronbach’s Alpha of .93 (M = 2.14, SD = .99).

Perceived appropriateness. An existing scale of 3 items was used to measure perceived appropriateness of the brand’s communication style (Gretry et al., 2017). Participants indicated on 5-point scales whether they agreed that the way the brand responded “meets my expectations”, “corresponds with how I expect the brand to communicate with me” and “is appropriate” (1 = strongly disagree, 5 = strongly agree). This scale proved reliable with a Cronbach’s Alpha of .83 (M = 2.66, SD = .99).

Manipulation check variables

Perceived disparagement. Following work by Newton et al. (2016), a 3-item scale was used to determine whether the reply was correctly perceived to contain disparaging humor. On a 5-point scale, participants indicated whether the initiating customer was made to feel “bad”, “humiliated” and “stupid” (1 = not at all, 5 = to a great degree). This scale proved reliable with a Cronbach’s Alpha of .89 (M = 2.84, SD = 1.15).

Perceived informality. Using a 3-item scale from Gretry et al. (2017), the perceived informality of the brand’s communication style was measured to ensure the brand replies were correctly perceived as informal. Participants indicated on 5-point scales whether they agreed that the brand “communicates in a casual way”, “communicates in an easygoing way” and “communicates in an informal way” (1 = strongly disagree, 5 = strongly agree). A reliability test for the three items revealed the scale to improve reliability (Cronbach’s Alpha from .64 to .77) if the latter item would be excluded. This may be caused by a more positively perceived connotation of the words “casual” and “easygoing” in the first two items. To prevent distorting

(17)

the concept of informality with these items, the brand “communicates in an informal way”-item was selected for further analysis (M = 4.05, SD = 1.11).

Perceived benignity. A 3-item scale from McGraw et al. (2015) was used to measure perceived benignity of the two humorous brand replies: “It seems playful”, “It seems serious” and “It expresses concern”, whereby the latter two were reverse scored. (1 = strongly disagree, 5 = strongly agree). This scale proved reliable with a Cronbach’s Alpha of .68 (M = 3.77, SD = 0.87).

Perceived familiarity. To ensure the brands were recognized as familiar, perceived familiarity of the brand was measured through a 3-item scale: “I am familiar with the brand”, “I am knowledgeable about the brand” and “I recognize the brand” (1 = strongly disagree, 5 = strongly agree). An exploratory factor analysis was performed, given the items had not yet been combined to create a scale in previous work. Results indicated the scale was unidimensional, explaining 94.94% of the variance in the three original items. This scale proved reliable with a Cronbach’s Alpha of .97 (M = 3.11, SD = 1.26).

Results

Randomization check. To check if random assignment to the conditions was successful, first, a One-Way ANOVA with the condition as independent variable and age as dependent variable was conducted. Results indicated no main effect for the conditions, F(5, 198) = 0.80, p = .553, η2_{= .02, on age. This indicates that no differences existed between the conditions on age; so,}

randomization on this variable was successful. Chi-square tests indicated no significant differences between the conditions on both gender, Χ2_{(10) = 11.07, p = .352, and education,}

Χ2_{(20) = 16.00, p = .716. Finally, a One-Way ANOVA for power motivation across the}

conditions also proved insignificant, F(5, 198) = 1.40, p = .226, η2_{= .03. Based on these}

(18)

Manipulation check. To check if participants had correctly acknowledged the brand in the stimulus material, a Chi-square test for the original conditions was conducted with the manipulation check variable for brand name. This manipulation check reached significance, Χ2

(10) = 200.11, p < .001, meaning the participants perceived a clear difference in the brand name used in the stimulus materials. A following Chi-square test with the manipulation check variable for brand reply reached significance, Χ2_{(25) = 398.54, p < .001. Ergo, also this}

manipulation was successful. A One-Way ANOVA manipulation check for perceived disparagement confirmed significant differences across the three brand reply types F(2, 201) = 33.26, p < .001, η2_{= .25, confirming the manipulation for disparagement was successful.}

Furthermore, all three reply types reached mean values above 3 for perceived informality (see Appendix B, Table 1 for M and SD), confirming the brand responses met the requirement of being perceived as informal. Both the non-disparaging (M = 4.17, SD = 0.88) and disparaging (M = 4.06, SD = 0.63) reached mean values above 3, confirming the brand responses met the requirement of being perceived as benign. Finally, a One-Way ANOVA manipulation check for brand familiarity confirmed significant differences across the two brands F(1, 202) = 1257.10, p < .001, η2_{= .86 (see Appendix B, Table 2 for M and SD), confirming Yahtzee’s}

Pizza was correctly perceived as unfamiliar and Domino’s Pizza was correctly perceived as familiar.

Direct effect. A One-Way ANOVA with brand reply as independent variable and brand attitude as dependent variable indicated significant differences across the three brand responses, F(2, 201) = 7.94, p <.001, η2_{= .07. The disparaging reply elicited significantly less}

brand attitude than the non-disparaging and non-humorous reply, yet no significant difference was found between the non-disparaging and non-humorous reply (See Appendix B, Table 3 for M and SD).

(19)

Appropriateness and familiarity

The first two hypotheses were tested using Andrew Hayes’ PROCESS v3.0 macro in SPSS. This method estimates the percentile bootstrap confidence intervals through 5,000 bootstrapped samples (Preacher & Hayes, 2008). Using model 7, brand reply was selected as multicategorical independent variable. Using indicator coding, this variable was recoded by PROCESS into X1 (0 = disparaging and humorous, 1 = disparaging) and X2 (0 = non-disparaging and non-humorous, 1 = non-disparaging). For this analysis, results for X2 were interpreted, meaning the independent variable essentially was a dichotomous variable comparing the disparaging reply to the non-disparaging and non-humorous replies combined. Brand attitude was selected as dependent variable, perceived appropriateness as mediator and brand familiarity was selected as moderator (see Appendix B, Table 4 for results). The proposed moderated mediation model is presented in Figure 1 .

Figure 1. Model of moderated mediation effect of brand familiarity and perceived appropriateness with unstandardized coefficients; n=204; *p<.05, **p<.01, ***p<.001

Results show a significant negative effect on perceived appropriateness, when disparaging humor is used in the reply (b = -1.75, p = <.001). This indicates a reply with disparaging humor leads to lower perceived appropriateness than replies without disparaging humor (non-disparaging and non-humorous). Thus, H1 is accepted. Furthermore, results show a significant positive interaction effect of disparagement and familiarity on appropriateness (b = .64, p = .025), indicating a familiar brand increases the effect of a disparaging reply on appropriateness

Brand reply 0 = non-disparaging and non-humorous 1 = disparaging Brand attitude R2_{= .20} Perceived appropriateness R2_{= .35} Brand familiarity 0 = unfamiliar brand 1 = familiar brand -1.75*** .64* .02 .39*

(20)

Overall, moderated mediation is significant (index of moderated mediation: .24, SE = .13, 95% CI [.03, .53]), meaning that a reply using disparaging humor leads to higher brand attitude through increased perceived appropriateness when a brand is familiar, compared to when a brand is unfamiliar. Considering this, H2 is also accepted.

Perceived moral violation and perceived funniness

To test the next path, a custom model was created through PROCESS v3.0 (see Appendix B, Table 5 for results). Also here, brand reply was selected as multicategorical variable, after which results were interpreted for X2. The proposed moderated serial mediation model is presented in Figure 2.

Figure 2. Model of power motivation moderated serial mediation effect by moral violation, perceived funniness and interaction attitude with unstandardized coefficients; n=204; *p<.05, **p<.01, ***p<.001

Results indicated a brand reply with disparaging humor (vs. disparaging and non-humorous) increased perceived moral violation (b = 1.44, p <.001), thus H3a is accepted.

A simple linear regression was calculated to predict perceived funniness based on perceived moral violation. A near-significant regression equation was found, (F(1, 202) = 3.71,

Brand reply 0 = non-disparaging and non-humorous 1 = disparaging Brand attitude R2_{= .37} Perceived moral violation R2_{= .40} Power motivation 0 = low 1 = high .41* .02 .56*** Perceived funniness R2_{= .05} Interaction attitude R2_{= .15} 1.44*** -.35** .33***

(21)

p = .055), with an R2 _{of .02. Were this to be considered significant, participants’ predicted}

perceived funniness would have been equal to 3.51 - .16*(perceived moral violation). In other words, perceived funniness would decrease by .16 for each additional reported Likert point indicating perceived moral violation. Considering the insignificance of this model, H3b is rejected. Yet even if the model were to be significant, H3b would have been rejected due to perceived funniness decreasing with every rise in perceived violation, rather than increasing as hypothesized.

Though a direct effect between perceived moral violation and perceived funniness could not be confirmed, results do show a significant positive interaction effect of perceived violation and high power motivation on perceived funniness (b = .41, p = .014). Thus, higher levels of moral violation lead to higher perceived funniness for those who highly power motivated, compared to those who are not or less power motivated. H3c can therefore be accepted.

Perceived funniness and attitude towards the interaction

Further results indicated a significant, positive effect for perceived funniness on interaction attitude (b = .33, p <.001), meaning the more a brand reply was perceived as funny, the more positive participants indicated their attitude towards the interaction. H4 can therefore be accepted.

A simple linear regression was calculated to predict brand attitude based on attitude towards the interaction. A significant regression equation was found, (F(1, 202) = 117.02, p < .001), with an R2 _{of .37. Participants’ brand attitude is equal to 1.39 + .53*(interaction attitude).}

In other words, brand attitude increased by .53 for each additional reported Likert point indicating interaction attitude. H5 is therefore accepted.

(22)

Overall, moderated mediation is significant (index of moderated mediation: .06, SE = .03, 95% CI [.01, .13]), meaning that a reply using disparaging humor (vs. non-disparaging and non-humorous) leads to higher brand attitude through increased perceived violation, perceived funniness and interaction attitude for those with high power motivation, compared to those with low power motivation.

Additional analysis

In an attempt to better understand why H3b was rejected, additional analysis was conducted. Rather than assuming the brand replies were considered benign due to the mean values being above 3, this analysis controlled for perceived benignity. A linear regression was calculated to predict perceived funniness based on perceived violation and perceived benignity. A significant regression equation was found, (F(2, 201) = 44.94, p <.001, with an R2 _{of .31. For this}

regression, participants’ perceived funniness is equal to .91 - .26*(perceived violation) + .75(*perceived benignity). Even so, an increase in perceived violation still appears to decrease perceived funniness.

Furthermore, a One-Way ANOVA for perceived funniness indicated significant differences across the three brand responses F(2, 201) = 45.30, p <.00, η2_{= .31. The}

non-disparaging response was perceived as the funniest, followed by the non-disparaging response and finally the non-humorous response (See Appendix B, Table 6 for M and SD). Moreover, a One-Way ANOVA for perceived moral violation also indicated significant differences across the three brand responses F(2, 201) = 67.31, p <.00, η2_{= .40. The disparaging response was}

perceived as the most morally violating, followed by the non-disparaging response and finally the non-humorous response (See Appendix B, Table 7 for M and SD), though the mean differences between the non-disparaging and non-humorous response were insignificant.

(23)

Two independent samples t-tests were conducted for only the disparaging response condition (n = 71). Results indicated no significant differences for brand attitude between the high (M = 2.85, SD = .95) and low (M = 2.56, SD = .96) power motivation groups, t(69) = -1.28, p = .206. Significant differences for brand attitude were found between the unfamiliar (M = 2.26, SD = .90) and familiar (M = 3.10, SD = .85) brand, t(69) = -4.02, p <.001. Two further independent samples t-tests were therefore done for only the non-disparaging response condition (n = 69) and only for the non-humorous response condition (n = 64)

Results for the non-disparaging response condition indicated no significant differences on brand attitude between the unfamiliar (M = 3.19, SD = .79) and familiar (M = 3.16, SD = .84) brand, t(67) = .18, p = .857. There were also no significant differences between the unfamiliar (M = 3.25, SD = .63) and familiar (M = 3.20, SD = .92) brand, for the non-humorous response, t(62) = .28, p = .785. These results indicate brand familiarity does specifically influence brand attitude when using a disparaging response, but familiarity does not influence brand attitude when using a non-disparaging or non-humorous response.

Conclusion and discussion

Academic research on the use of disparaging humor in external brand communication is limited, with only a few studies looking into the effects of disparaging humor in advertisements (Brown et al., 2010; Newton et al., 2016; Swani et al., 2013). At the same time, several studies address webcare and its effect on brand attitude (Huibers & Verhoeven, 2014; van Noort et al., 2014; van Noort & Willemsen, 2012). The combination of disparaging humor and webcare, however, appeared to be unchartered territory. Ergo, the main purpose of this study was to shed light on how disparaging humor in webcare influences brand attitude. This relationship was essentially examined in twofold. The first path examined the effect of a disparaging reply on

(24)

of perceived appropriateness to ultimately predict brand attitude. The second path examined the relationship between perceived moral violation of a disparaging reply and perceived funniness; the moderating effect of power motivation on this relationship; the influence of perceived funniness on interaction attitude; and the influence of interaction attitude on brand attitude.

For the first path, results from the conducted experiment confirmed that brand familiarity moderated the effect of a disparaging reply on perceived appropriateness, which in turn mediated the effect on brand attitude. These findings are in line with the expectations of the study and existing literature; H1 and H2 were accepted accordingly.

For the second path, results regarding the effect of perceived moral violation on perceived funniness contradicted expectations, resulting in the rejection of H3b. Though all brand replies were considered to be benign, an increase in perceived moral violation did not significantly increase perceived funniness: results even indicated an (insignificant) negative relationship. Looking further into this, an additional analysis was conducted where perceived benignity was controlled for, rather than assumed. These results were significant, yet still indicated that an increase in perceived moral violation decreased perceived funniness. This is also reflected in the general finding that the non-disparaging response was perceived as the funniest response, whilst less morally violating than the disparaging response. Nevertheless, these findings are not in line with the expectation that under benign-violation theory (McGraw & Warren, 2010) a linear effect will occur between violations and funniness as long as the violation is perceived as benign. Next, H3c was accepted: high power motivation was found to increase the effect of perceived moral violation of a disparaging response on perceived funniness. In fact, when highly power motivated, perceived moral violation did increase perceived funniness, as was initially expected through benign-violation theory. The finding that this was not the case for those with low power motivation suggests benign-violation theory

(25)

may be subject to different degrees of power motivation and/or other character traits, though further research is necessary to confirm and elaborate on this. Furthermore, perceived funniness was found to positively influence interaction attitude, which in turn positively influenced brand attitude. Both relationships were consistent with the study’s expectations and with existing literature originally focusing on advertising: H4 and H5 were accepted. Finally, results found moderated serial mediation was significant: the effect of a disparaging response on brand attitude is mediated by perceived moral violation, perceived funniness and interaction attitude, where power motivation moderates the relationship between perceived moral violation and perceived funniness.

In conclusion, results of this study indicate the use of disparaging humor in webcare does not typically have a positive effect on brand attitude. Though power motivation and brand familiarity can moderate effects that mediate the relationship between disparaging humor and brand attitude, only brand familiarity alone do not weighs-in sufficiently for a disparaging reply to increase brand attitude. Practitioners must therefore be wary when deciding whether to use disparaging humor in webcare. Based on the findings of this study, it is advisable for brands to refrain from using disparaging humor in webcare if the (target) audience is expected to not yet be familiar with the brand. Moreover, if the target audience is not specifically profiled as highly power motivated, brands must tread lightly when using disparaging humor in webcare.

In terms of theoretical implications, this study gives insight into various mechanisms through which the use of disparaging humor in webcare predicts brand attitude. Results are in line with the moderated mediation of informal brand communication on brand attitude (Gretry et al., 2017) and in that extend it to the academic literature on webcare (van Noort et al., 2014; van Noort & Willemsen, 2012) and on disparaging humor (Ford & Ferguson, 2004). Moreover, results regarding the relationship between perceived funniness, interaction attitude and brand attitude are also in line with existing literature which originally focused on advertising (Brown

(26)

et al., 2010; Gelb & Pickett, 1983; Swani et al., 2013; Zhang & Zinkhan, 2006). This implies that effects found with regard to advertising may be similarly applicable in the field of webcare, though further research must be conducted to confirm this. Furthermore, though results are not explicitly consistent with benign-violation theory (McGraw & Warren, 2010), the seemingly moderating effect of power motivation provides relevant argumentation for academics to conduct further research: benign-violation might be subject to varying degrees of character traits.

This study is limited in that only three responses were used within the stimuli. If a pre-test were to have been conducted in search for the response with the highest (and lowest) levels of perceived funniness, perceived moral violation and perceived appropriateness, results may have had more distinct differences. This would have also allowed for an inclusion of non-benign response, enabling a test for the non-benign-violation theory’s original inverted U-shaped effect. Moreover, the study tests for interaction and brand attitude. Given webcare takes place on social media, relevant outcome variables may have also been related to engagement on social media, such as liking, commenting or sharing the interaction and/or following the brand.

Despite these limitations, the results of this study provide sound advice for practitioners and contribute to academic literature on humor, specifically disparaging humor and benign-violation theory; and brand communication, specifically webcare.

References

Belch, G. E., & Belch, M. A. (2018). Advertising and promotion: an integrated marketing communications perspective.

Bielenia-Grajewska, M. (2017). Informal communication. In C. R. Scott, J. R. Barker, T. Kuhn, J. Keyton, P. K. Turner, & L. K. Lewis (Eds.), The International Encyclopedia

(27)

of Organizational Communication (pp. 1–15). Hoboken, NJ, USA: John Wiley & Sons, Inc. https://doi.org/10.1002/9781118955567.wbieoc108

Borges, N. J., Manuel, R. S., Elam, C. L., & Jones, B. J. (2010). Differences in motives between millennial and generation X medical students. Medical Education, 44(6), 570–576. https://doi.org/10.1111/j.1365-2923.2010.03633.x

Brown, M. R., Bhadury, R. K., & Pope, N. K. L. (2010). The impact of comedic violence on viral advertising effectiveness. Journal of Advertising, 39(1), 49–66.

https://doi.org/10.2753/JOA0091-3367390104

Duncan, T. (2005). Principles of advertising & IMC. New York: McGraw-Hill Irwin. Ford, T. E., & Ferguson, M. A. (2004). Social consequences of disparagement humor: a

prejudiced norm theory. Personality and Social Psychology Review, 8(1), 79–94. https://doi.org/10.1207/S15327957PSPR0801_4

Fournier, S., & Avery, J. (2011). The uninvited brand. Business Horizons, 54(3), 193–207. Gelb, P. B. D., & Pickett, C. M. (1983). Attitude-toward-the-ad: links to humor and to

Advertising Effectiveness. Journal of Advertising, 12(2), 34–42. https://doi.org/10.1080/00913367.1983.10672838

Gensler, S., Völckner, F., Liu-Thompkins, Y., & Wiertz, C. (2013). Managing brands in the social media environment. Journal of Interactive Marketing, 27(4), 242–256.

https://doi.org/10.1016/j.intmar.2013.09.004

Gresham, L. G., & Shimp, T. A. (1985). Attitude toward the advertisement and brand attitudes: a classical conditioning perspective. Journal of Advertising, 14(1), 10–49. https://doi.org/10.1080/00913367.1985.10672924

Gretry, A., Horváth, C., Belei, N., & van Riel, A. C. R. (2017). Don’t pretend to be my friend! when an informal brand communication style backfires on social media.

(28)

Journal of Business Research, 74, 77–89. https://doi.org/10.1016/j.jbusres.2017.01.012

Gruen, T. W., Osmonbekov, T., & Czaplewski, A. J. (2006). eWOM: the impact of customer-to-customer online know-how exchange on customer value and loyalty. Journal of Business Research, 59(4), 449–456. https://doi.org/10.1016/j.jbusres.2005.10.004 Hsieh, J.-K., Hsieh, Y.-C., & Tang, Y.-C. (2012). Exploring the disseminating behaviors of

eWOM marketing: persuasion in online video. Electronic Commerce Research, 12(2), 201–224. https://doi.org/10.1007/s10660-012-9091-y

Huibers, J., & Verhoeven, J. (2014). Webcare als online reputatiemanagement. Tijdschrift Voor Communicatiewetenschap, 42(2), 165.

Kelleher, T. (2009). Conversational voice, communicated commitment, and public relations outcomes in interactive online communication. Journal of Communication, 59(1), 172–188. https://doi.org/10.1111/j.1460-2466.2008.01410.x

Kelleher, T., & Miller, B. M. (2006). Organizational blogs and the human voice: relational strategies and relational outcomes. Journal of Computer-Mediated Communication, 11(2), 395–414. https://doi.org/10.1111/j.1083-6101.2006.00019.x

Kruschke, J. K., & Vollmer, A. (2014). Moral foundation sensitivity and perceived humor. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.2519218

MacKenzie, S. B., Lutz, R. J., & Belch, G. E. (1986). The role of attitude toward the ad as a mediator of advertising effectiveness: a test of competing explanations. Journal of Marketing Research, 23(2), 130. https://doi.org/10.2307/3151660

McClelland, D. C. (1987). Human motivation (paperback re-issue). Cambridge: Cambridge University Press.

McGraw, A. P., & Warren, C. (2010). Benign violations: making immoral behavior funny. Psychological Science, 21(8), 1141–1149. https://doi.org/10.1177/0956797610376073

(29)

Muntinga, D., Moorman, M., & Smit, E. (2011). Introducing COBRAs. International Journal of Advertising. 30. 13-46.

Newton, J. D., Wong, J., & Newton, F. J. (2016). Listerine – for the bridesmaid who’s never a bride: disparaging humour increases brand attitude and recall among the powerless. European Journal of Marketing, 50(7/8), 1137–1158. https://doi.org/10.1108/EJM-06-2015-0321

Preacher, K. J., & Hayes, A. F. (2008). Contemporary approaches to assessing mediation in communication research. In The Sage sourcebook of advanced data analysis methods for communication research (pp. 13–54). Thousand Oaks, CA, US: Sage Publications, Inc. https://doi.org/10.4135/9781452272054.n2

Rogerson-Revell, P. (2007). Humour in business: a double-edged sword. Journal of Pragmatics, 39(1), 4–28. https://doi.org/10.1016/j.pragma.2006.09.005

Shabbir, H., & Thwaites, D. (2007). The use of humor to mask deceptive advertising: it’s no laughing matter. Journal of Advertising, 36(2), 75–85.

https://doi.org/10.2753/JOA0091-3367360205

Singh, J., & Widing, R. E. (1991). What occurs once consumers complain? A theoretical model for understanding satisfaction/ dissatisfaction outcomes of complaint responses. European Journal of Marketing, 25(5), 30–46.

https://doi.org/10.1108/03090569110140489

Stone, M., Hobbs, M., & Khaleeli, M. (2002). Multichannel customer management: the benefits and challenges. Journal of Database Marketing & Customer Strategy Management, 10(1), 39–52.

Swani, K., Weinberger, M. G., & Gulas, C. S. (2013). The impact of violent humor on advertising success: a gender perspective. Journal of Advertising, 42(4), 308–319. https://doi.org/10.1080/00913367.2013.795121

(30)

Turel, O., & Connelly, C. E. (2013). Too busy to help: antecedents and outcomes of interactional justice in web-based service encounters. International Journal of Information Management, 33(4), 674–683.

https://doi.org/10.1016/j.ijinfomgt.2013.03.005

van Noort, G., & Willemsen, L. M. (2012). Online damage control: the effects of proactive versus reactive webcare interventions in consumer-generated and brand-generated platforms. Journal of Interactive Marketing, 26(3), 131–140.

https://doi.org/10.1016/j.intmar.2011.07.001

van Noort, G., Willemsen, L., Kerkhof, P., & Verhoeven, J. (2014). Webcare as an

integrative tool for customer care, reputation management, and online marketing: a literature review. In: Philip J. Kitchen & Ebru Uzunoglu (eds.), Integrated

Communications in the Post-Modern Era, p. 77-99. https://doi.org/10.1057/9781137388551.0008

Ward, J. C., & Ostrom, A. L. (2006). Complaining to the Masses: The Role of Protest Framing in Customer-Created Complaint Web Sites. Journal of Consumer Research, 33(2), 220–230. https://doi.org/10.1086/506303

Yarwood, D. L. (1995). Humor and administration: a serious inquiry into unofficial organizational communication. Public Administration Review, 55(1), 81–90. https://doi.org/10.2307/976830

Zhang, Y., & Zinkhan, G. M. (2006). Responses to humorous ads: does audience involvement matter? Journal of Advertising, 35(4), 113–127.

(31)

Appendices

Appendix A. – Stimulus materials

Figure 1. Facebook interaction with non-humorous brand reply from an unfamiliar brand.

Figure 2. Facebook interaction with non-disparaging brand reply from an unfamiliar brand.

(32)

Figure 4. Facebook interaction with non-humorous brand reply from a familiar brand.

Figure 5. Facebook interaction with non-disparaging brand reply from a familiar brand.

Figure 6. Facebook interaction with disparaging brand reply from a familiar brand.

(33)

Appendix B. – Tables

Table 1. Descriptive statistics for brand attitude per brand reply

Group M SD n

Non-humorous 3.23A _.79 ₆₄

Non-disparaging 3.17A _.81 ₆₉

Disparaging 2.70 .96 71

Note. Means sharing a letter in their superscript are not significantly different at the .05 level according to Tukey HSD test.

Table 2. Descriptive statistics for perceived informality per brand reply

Group M SD n

Non-humorous 3.53 1.08 64

Non-disparaging 4.25A _1.04 ₆₉

Disparaging 4.32A _1.05 ₇₁

Table 3. Descriptive statistics for perceived familiarity per brand name

Group M SD n

Unfamiliar 1.20 0.64 99

(34)

Table 4. Unstandardized OLS regression coefficients with confidence intervals (standard errors in parentheses) estimating perceived appropriateness and brand attitude

Predictor variables Dependent variables

Perceived appropriateness (M) Brand attitude (Y)

Coeff. 95% CI Coeff. 95% CI Brand reply (X) -1.75*** (.21) -2.15, -1.34 .02 (.17) -.32, .35 Perceived appropriateness (M) .39*** (.07) .25, .52 Brand familiarity (W) -.26 (.20) -.66, .15 X x W .64* (.28) .08, 1.19 Constant 3.52*** (.15) 3.23, 3.82 1.92*** (.25) 1.42, 2.42 R2_{= .35} _R2_{= .20} F(5, 198) = 21.33, p < .001 F(3, 200) = 16.54, p < .001 Note. *p < .05, ***p < .001

Table 5. Unstandardized OLS regression coefficients with confidence intervals (standard errors in parentheses) estimating perceived violation, perceived funniness, interaction attitude and brand attitude

Predictor variables Dependent variables

Perceived violation (M1) Perceived funniness (M2) Interaction attitude (M3) Brand attitude (Y)

Coeff. 95% CI Coeff. 95% CI Coeff 95% CI Coeff 95% CI

Brand reply (X) 1.44*** (.13) 1.18, 1.71 .02 (.13) -.25, .28

Perceived moral violation (M1) -.35**(.11) -.57, -.12

Perceived funniness (M2) .33*** (.06) .22, .44 Interaction attitude (M3) .55*** (.06) .44, .67 Power motivation (W) -.77+ _(.39) _{-1.54, .01} M1 x W .41* (.17) .08, .74 Constant 1.54*** (.10) 1.35, 1.73 3.88*** (.28) 3.32, 4.44 2.03*** (.19) 1.66, 2.40 1.36*** (.12) .95, 1.78 R2_{= .40} _R2_{= .05} _R2_{= .39} _R2_{= .61} F(2, 201) = 67.31, p < .001 F(3, 200) = 3.48, p = .017 F(1, 202) = 35.73, p < .001 F(3, 200) = 39.53, p < .001 Note. +_{p < .10, *p < .05, **p <.005 ***p < .001}

(35)

Table 6. Descriptive statistics for perceived funniness per brand reply

Group M SD n

Non-humorous 2.39 1.02 64

Non-disparaging 3.98 .81 69

Disparaging 3.15 1.12 71

Table 7. Descriptive statistics for perceived moral violation per brand reply

Group M SD n

Non-humorous 1.54A _.69 ₆₄

Non-disparaging 1.84A _.71 ₆₉

Disparaging 2.98 .88 71

(36)

Appendix C. – Experiment questionnaire

Dear participant, Welcome to this survey and thank you for participating. This survey is part of a research study for my Master thesis at the University of

Amsterdam. Firstly, you will be shown an interaction on social media and some questions regarding this interaction will be asked. Next, you will be asked some demographic questions. Finally, you will be asked to answer some statements about yourself. Your responses and the reports of the data will be anonymous. Please take note that there are no wrong or right answers. Answer the questions according to your feelings and opinion. The survey will take about 5 minutes. If you have any questions about the process or the study in general, please do not hesitate to contact me. Kind regards, Ginger Reinhoudt ginger.reinhoudt@student.uva.nl

Before continuing, please read the following consent:

I agree to voluntarily participate in this study. I am free to withdraw at any time, without giving a reason. Also, until 7 days after participation, I can withdraw my permission to use my answers for this research. If my results are used in scientific publications, or are published in any other way, my data will be completely anonymous. My personal data will not be read by third parties without my explicit permission. If I want more information, now or in the future, I can contact Ginger Reinhoudt (ginger.reinhoudt@student.uva.nl).

If I have any complaints about this research, I can contact the designated member of the Ethics Committee representing the ASCoR:

ASCoR secretariat, Ethics Committee, University of Amsterdam, Nieuwe Achtergracht 166, 1018 WV Amsterdam

020-52 53 680; ascor-secr-fmg@uva.nl

o

I understand the above and I agree to participate in the study

o

I do not agree and do not wish to participate

Please take a close look at the following Facebook interaction and imagine coming across this interaction whilst using Facebook. On the next pages, we will ask you some questions about the initiating customer and the brand's response.

(37)

I...

Not at all Very

am familiar

with the brand

o

am knowledgeable about the brand

o

recognize the brand.

o

The brand's response was...

Not at all Very

funny

_o

amusing

_o

humorous

_o

The brand's product/service is...

Not at all Very

important to

me

o

of concern to

me

o

(38)

The way the brand responded... Strongly

disagree Strongly agree

meets my expectations

o

is appropriate

_o

seems playful

_o

seems serious

_o

expresses concern

o

is aversive

_o

is upsetting

_o

is threatening

_o

is disturbing

_o

is immoral (i.e.: wrong)

o

is okay (i.e.: not wrong)

o

crossed a line

_o

corresponds with how I expect the brand to communicate with me

o

(39)

Please indicate whether you consider the brand to be... 1 2 3 4 5 Bad

_o

Good Unfavorable

_o

Favorable Negative

_o

Positive Undesirable

_o

Desirable Unattractive

_o

Attractive

Please indicate whether you considered the Facebook interaction to be...

1 2 3 4 5

Bad

_o

Good

Unpleasant

_o

Pleasant

Unfavorable

_o

Favorable

Unrealistic

_o

Realistic

In the Facebook interaction, the initiating customer was made to feel...

Not at all To a great _degree

bad

_o

humiliated

_o

(40)

In the Facebook interaction, the brand communicated in... Strongly

a casual way

_o

an easygoing

way

o

an informal

way

o

The brand in this interaction was

o

Domino's Pizza

o

Yahtzee's Pizza

o

I don't remember

What did the brand's response include?

o

Something about Trump

o

Something about sperm

o

Something about changing the customers mind

o

Something about discount

o

Something about cereal boxes

o

Neither of these options

(41)

What country are you from (i.e.: your nationality)? ▼ Afghanistan ... Zimbabwe

What is your age? (e.g.: 25)

________________________________________________________________

What is your gender?

o

Female

o

Male

o

Other: ________________________________________________

What is your highest or current level of education?

o

No education

o

Primary school

o

High school

o

Secondary Vocational Education (MBO)

o

University of Applied Sciences (HBO)

o

University

o

Other: ________________________________________________

This is the last part! You have almost completed the survey.

Here, we are interested in the following statements about you. Please indicate to what extent you agree with the statements.

(42)

Strongly

I think I would enjoy having authority of other people

o

If given the chance I would make a good leader of people

o

I think I am usually a leader in my group

o

I enjoy planning things and deciding what other people should do

o

I like to give orders and get

things going

o

People take notice of what I say

o

When a group I belong to plans an activity, I would rather direct it myself

than just help out and have someone else organize it

o

I would like an important job where people looked up to me

o

I like talking to people who are

(43)

I want to be an important person in the community

o

I really admire people who have fought their way to the top

o

If I had enough money I would not work

o

Even if I won a great deal of money on the pools I would prefer to continue to work

o

If unemployment benefit was really high I would still prefer to work

o

I like to be admired for my achievements

o

I dislike being the centre of attention

o

I like to have people come to me for advice

o

I find satisfaction in having influence over others because of my position in the community